MARINE MICROBIOLOGY

The 3rd Edition of this book captures all the recent amazing advances in our understanding of the marine microbiology world
but still manages to present the concepts in a an easy, informative and entertaining way that will engage the novice to the expert.
What a great book and a fun read.
- David Bourne, James Cook University and The Australian Institute of Marine Science

It is great to see another edition of the book given that marine microbiology is such a fast moving and scientifically diverse field.
Munn’s new edition will be a great resource for new students and advanced scientists alike.
- Greta Reintjes, Max Planck Institute for Marine Microbiology, Bremen, Germany

Reading this textbook has made me realise how much the field of marine microbiology has progressed in recent decades.
I recommend this book also to biogeochemists and ecologists in search of the Big Picture of ocean functioning. The many details
of interactions emerging from the microbial world are amazing and shed light on the factors driving evolution of these ancient
ecosystems.
- Victor Smetacek, Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany

It has been astonishing to see the evolution of this book over the years. With its many ‘RESEARCH FOCUS’ boxes and
‘SIDEBARS’, this is more than your usual textbook. It is written in an enthusiastic, thought-provoking manner, encompassing
the most up-to-date concepts in marine microbiology. From planktonic tunicates involved in carbon cycling, to viruses infecting
other viruses, this essential read has it all!
- Jozef I. Nissimov, Lecturer in Microbial Oceanography, SAMS, Scottish Marine Institute, UK

Colin B. Munn obtained a BSc (Hons) degree from University College London and a PhD from the University of Birmingham. He
is an Honorary Fellow of the Marine Institute of University of Plymouth, England and was Associate Professor of Microbiology
and Admissions Tutor for Marine Biology programs until 2017. He has a long experience of research in various aspects of
microbiology, with particular interests in the interactions between symbiotic and pathogenic microorganisms and their hosts.
He has held former positions as Visiting Professor at the University of Victoria, Central University of Venezuela and St Georges
University, Grenada and a visiting researcher (Leverhulme Fellow) at James Cook University and the Australian Institute of
Marine Science. He has served as external examiner for Bachelor’s, Master’s, and PhD degrees in many countries and as a special
assessor for molecular and organismal biology for the UK Quality Assurance Agency for Higher Education. He has extensive
research experience in a range of microbiological topics, with special interest in interactions between pathogenic and symbiotic
microbes and their hosts.
MARINE MICROBIOLOGY
ECOLOGY & APPLICATIONS
Third Edition

Colin B. Munn
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2020 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-0-367-18356-1 (Paperback)

International Standard Book Number-13: 978-0-367-18359-2 (Hardback)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been
made to publish reliable data and information, but the author and publisher cannot assume responsibility for the
validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the
copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to
publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let
us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted,
or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, includ-
ing photocopying, microfilming, and recording, or in any information storage or retrieval system, without written
permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com
(http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers,
MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety
of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment
has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
Contents

Chapter 1 Microbes in the Marine Microbes in sea ice form an important part of the food
web in polar regions 22
Environment 1 Microbial activity underpins productive food webs in
ORIGINS AND SCOPE OF MARINE coral reefs 25
MICROBIOLOGY 2 Living organisms are the habitats of many microbes 25
Conclusions 25
Marine microbiology has developed into one of the most References and further reading 26
important areas of modern science 2
Microbes include microscopic cellular organisms and
non-cellular viruses 2 Chapter 2 Methods in Marine
Marine microorganisms are found in all three domains of Microbiology 29
cellular life 3
Horizontal gene transfer confounds our understanding of SAMPLING METHODS 30
evolution 4 Sampling the marine environment requires special
Viruses are non-cellular entities with great importance in techniques 30
marine ecosystems 4
Microbial processes shape the living world 5 IMAGING TECHNIQUES 32
Marine microbes show great variation in size 5 Light and electron microscopy are used to study
OCEAN HABITATS 6 morphology and structure of microbes 32
Epifluorescence light microscopy enables enumeration
The world’s oceans and seas form an interconnected of marine microbes 33
water system 6 Confocal laser scanning microscopy enables recognition
The upper surface of the ocean is in constant motion of living microbes within their habitat 35
owing to winds 8 Flow cytometry measures the number and
Deep-water circulation systems transport water between size of particles 35
the ocean basins 8 Fluorescent in situ hybridization (FISH) allows
Light and temperature have important effects on visualization and quantification of specific microbes 36
microbial processes 9
Microbes occur in all the varied habitats found in CULTIVATION OF MICROORGANISMS 38
the oceans 10 Different microorganisms require specific culture media
Seawater is a complex mixture of inorganic and organic and conditions for growth 38
compounds, colloids, and gels 10 Enrichment culture selects for microbes with specific
The sea surface is covered by a gelatinous biofilm 14 growth requirements 40
SEDIMENTS AND SURFACES 14 Phenotypic testing is used for characterization of many
cultured bacteria 40
Microbes play a major role in marine sediments 14 Analysis of microbial cell components can be used for
Deep marine sediments contain a vast reservoir of bacterial classification and identification 42
ancient microbes 16
Microbes colonize surfaces through formation of METHODS BASED ON DNA AND RNA ANALYSIS 42
biofilms and mats 16 Nucleic acid-based methods have transformed
SOME EXAMPLES OF SPECIAL HABITATS—THE understanding of marine microbial diversity
DEEP SEA, POLAR OCEANS, CORAL REEFS, AND and ecology 42
LIVING ORGANISMS 19 Amplification and sequencing of ribosomal RNA
genes is widely used in microbial systematics and
Microbial activity at hydrothermal vents fuels an oasis diversity studies 42
of life in the deep sea 19 Isolation of genomic DNA or RNA is the first step in all
Cold seeps also support diverse life based on nucleic acid-based investigations 45
chemosynthesis 21 The polymerase chain reaction (PCR) forms the basis of
Microbes inhabit the interface of brine pools in the deep sea 21 many techniques 45
vi CONTENTS

Genomic fingerprinting can be used to assess diversity SOURCES OF ENERGY AND CARBON 73
of cultured isolates 47
Microbes obtain energy from light or oxidation of
Determination of DNA properties is used in bacterial
compounds 73
and archaeal taxonomy 48
Microbes differ in their source of carbon to make
DNA sequence data are used for identification and
cellular material 74
phylogenetic analysis 48
DGGE and TRFLP can be used to assess composition PHOTOTROPHY AND CHEMOTROPHY 74
of microbial communities 49 Phototrophy involves conversion of light energy to
Advances in DNA sequencing enable improved microbial chemical energy 74
community analysis 50 Oxygenic photosynthesis involves two distinct but
Elucidating the full genome sequence of microbes coupled photosystems 76
provides insights into their functional roles 52 Anaerobic anoxygenic photosynthesis uses only one
Metabarcoding and metagenomics have led to major type of reaction center 76
advances in microbial community analysis 53 Aerobic anoxygenic phototrophy is widespread in
Omics technologies provide information about the planktonic bacteria 77
functional gene composition of a microbial community 55 Some phototrophs use rhodopsins as light-harvesting
Genomes can now be obtained from single cells in pigments 77
environmental samples 56 Chemolithotrophs use inorganic electron donors as
IN SITU ACTIVITY OF MICROBIAL COMMUNITIES 57 a source of energy and reducing power 79
Many bacteria oxidize sulfur compounds 80
Microelectrodes and biosensors measure microbial
Many chemolithotrophs use hydrogen as an
processes at the microhabitat scale 57
electron donor 81
Radioisotopes can be used to detect metabolic activity
Bacterial and archaeal nitrification is a major process
in a community 57
in the marine nitrogen cycle 81
Stable-isotope probing (SIP) tracks fluxes of nutrients in
Ammonia can also support anaerobic
communities 58
chemolithoautotrophy 82
NanoSIMS allows metabolic transfers to be measured at
subcellular levels 58 CARBON AND NITROGEN FIXATION 83
Microarrays enable assessment of gene activity in the The Calvin–Benson–Bassham (CBB) cycle is the main
environment 59 method of carbon fixation in autotrophs 83
Metatranscriptomics, metaproteomics, and metabolomics Some Archaea and Bacteria use alternative pathways
reveal microbial activities in the environment 59 to fix CO2 83
Microfluidics enables study of microscale processes 60 Fixation of nitrogen makes this essential element
Mesocosm experiments attempt to simulate natural available for building cellular material in all life 84
conditions 60
Remote sensing permits global analysis of microbial HETEROTROPHIC METABOLISM 85
activities 61 Many marine microbes obtain energy by the
Conclusions 62 fermentation of organic compounds 85
References and further reading 62 Anaerobic respiration has major importance in marine
processes 86
Chapter 3 Metabolic Diversity and Nitrate reduction and denitrification release nitrogen and
other gases 86
Ecophysiology 65 Sulfate reduction is a major process in marine sediments 86
A BRIEF OVERVIEW OF CELL STRUCTURE AND MICROBIAL PRODUCTION AND OXIDATION OF
FUNCTION 66 METHANE 87
Bacteria and archaea show a variety of cell forms and Methanogenesis is unique to the Archaea 87
structural features 66 Methane is produced in the surface ocean by bacterial
The cytoplasmic membrane controls cell processes via cleavage of phosphonates 88
transport of ions and molecules 66 Anaerobic oxidation of methane (AOM) in sediments is
Cells may contain organelles, microcompartments, and coupled to sulfate reduction 88
inclusion bodies 67 Many marine microbes oxidize methane and other C1
The nature of the cell envelope has a major effect on compounds 90
physiology 68 NUTRIENT ACQUISITION AND MICROBIAL
Genome size and organization determines bacterial and GROWTH 90
archaeal lifestyles 69
Microbes use a variety of mechanisms to regulate Microbial metabolism depends on nutrient uptake 90
cellular activities 73
 CONTENTS vii

Acquisition of iron is a major challenge for marine The order Caulobacterales contains prosthecate bacteria 121
microbes 92 Several alphaproteobacterial genera show magnetotaxis 123
Marine bacterioplankton use two trophic strategies 92 Magnetotaxis is also found in other classes and phyla 124
Growth rate and turnover of organic material depend on The order Betaproteobacteriales includes many
nutrient concentrations 93 rare OTUs 124
Copiotrophic marine bacteria may show rapid growth The Gammaproteobacteria is a very large and
in culture 93 diverse class 124
Bacteria adapt to starvation by coordinated changes to The Gammaproteobacteria includes many uncultivated
cell metabolism 94 species of sulfide-oxidizing bacteria (SOB) 126
Some bacteria enter a “viable but nonculturable” state The family Vibrionaceae includes many important
in the environment 95 pathogens and symbionts 127
Many bacteria use motility to search for nutrients and Members of the order Oceanospirillales break down
optimal conditions 95 complex organic compounds 127
Flagella also have a mechanosensory function 97 The family Thiotrichaceae includes some
Microbes also respond to light, magnetic fields, and important SOB 128
other stimuli 100 The proposed phylum Desulfobacterota contains
Gliding and twitching motility occur on surfaces 100 anaerobic sulfate- or sulfur-reducing bacteria (SRB) 129
Microbes colonize surfaces via formation of biofilms 101 The proposed phylum Epsilonbactereota contains major
Pili are important for bacterial attachment to surfaces contributors to productivity at hydrothermal vents 129
and genetic exchange 102 Myxobacteria have a complex life cycle 132
Antagonistic interactions between microbes occur on The Bdellovibrionales contains predatory bacteria 132
particles or surfaces 102 Members of the Zetaproteobacteria are microaerophilic
Quorum sensing is an intercellular communication iron-oxidizers 133
system for regulation of gene expression 102 Members of the Cyanobacteria carry out oxygenic
PHYSICAL EFFECTS ON MICROBIAL GROWTH photosynthesis 133
A genome-based classification of the Cyanobacteria is
AND SURVIVAL 105
under development 134
Most marine microbes grow at low temperatures 105 Prochlorococcus is the most abundant photosynthetic
Microbes growing in hydrothermal systems are organism 134
adapted to very high temperatures 105 Synechococcus spp. dominate the upper photic zone 136
Microbes that inhabit the deep ocean must withstand Some free-living and symbiotic cyanobacteria
a very high hydrostatic pressure 106 fix nitrogen 136
Ultraviolet irradiation has lethal and mutagenic effects 107 Filamentous cyanobacteria are important in the
Bacterial bioluminescence may protect bacteria from formation of microbial mats 139
ROS and UV damage 108 Members of the Planctomycetes have atypical cell
Microbes use various mechanisms to prevent structure 139
osmotic damage 108 The phylum Bacteroidetes has a major role in nutrient
Conclusions 109 cycling via degradation of polymers 140
References and further reading 109 Members of the phylum Chloroflexi are widespread but
poorly characterized 141
Chapter 4 Diversity of Marine The phyla Aquificae and Thermotogae are deeply
Bacteria 113 branching primitive thermophiles 141
The Firmicutes are a major branch of Gram-
OVERVIEW OF BACTERIAL DIVERSITY 114 positive Bacteria 142
Members of the Actinobacteria are a rich source of
Understanding of diversity has been revolutionized by secondary metabolites, including antibiotics 143
phylogenetic and genomic techniques 114 Conclusions 144
Bacterial systematics is in transition due to application References and further reading 144
of genomic methods 115
OTUs and ASVs are used to represent diversity in
community analyses 118 Chapter 5 Marine Archaea 149
Marine microbial communities show high alpha Several aspects of cell structure and function
diversity 119 distinguish the Archaea and Bacteria 150
A TOUR OF THE BACTERIAL AQUARIUM 120 New phylogenomic methods have led to recognition of
multiple phyla of the Archaea 150
The Proteobacteria account for about half of all
bacterial ocean diversity 121 PHYLUM EURYARCHAEOTA 151
Members of the class Alphaproteobacteria are the most Many members of the Euryarchaeota produce methane 151
abundant marine bacteria 121
viii CONTENTS

Anaerobic oxidation of methane (AOM) in sediments is Photosynthetic prasinophytes are abundant members
carried out by syntrophic archaea 153 of the picoplankton 184
The class Thermococci contains hyperthermophiles Amoebozoa are important grazers of particle-
found at hydrothermal vents 154 associated bacteria 184
Archaeoglobus and Ferroglobus are hyperthermophilic Radiolarians and foraminifera have highly diverse
sulfate-reducers and iron-oxidizers 155 morphologies with mineral shells 185
The Euryarchaeota contains extreme halophiles 155 MARINE FUNGI 186
Uncultivated members of the Euryarchaeota are
abundant in the plankton 155 The Fungi form a distinct monophyletic group on a
branch within the Nucletmycea 186
PHYLUM CRENARCHAEOTA 156
Fungi are increasingly recognized to be major
Members of the Crenarchaeota are thermophiles components of the marine microbiome 187
occurring in hydrothermal vents 156 Conclusions 189
PHYLUM THAUMARCHAEOTA 157 References and further reading 192

A single clade of ammonia-oxidizing archaea

comprises 20% of the picoplankton 157 Chapter 7 Marine Viruses 195
PHYLUM NANOARCHAEOTA 161 Viruses are highly diverse non-cellular microbes 196
Phages are viruses that infect bacterial and
Nanoarchaeum is an obligate parasite of another archaeal cells 200
archaeon 161 The life cycle of phages shows a number of
Conclusions 162 distinct stages 201
References and further reading 162 Lysogeny occurs when the phage genome is integrated
into the host genome 201
Chapter 6 Marine Eukaryotic Loss of viral infectivity arises from damage to the
nucleic acid or capsid 205
Microbes 165 Measurement of virus production rates is important for
quantifying virus-induced mortality 205
MARINE PROTISTS 166
Viral mortality “lubricates” the biological pump 206
Protists are a highly diverse collection of unicellular Nucleocytoplasmic large DNA viruses (NCLDVs) are
eukaryotic microbes 166 important pathogens of microalgae and other protists 206
Protists show enormous diversity and classification Other giant viruses are abundant pathogens of
systems are regularly revised 167 heterotrophic protists 210
The -omics approaches have some limitations for RNA viruses also infect protists 211
understanding protist diversity 168 Viral mortality plays a major role in structuring
Picoeukaryotes play a major role in ocean food webs 169 diversity of microbial communities 212
Heterotrophic flagellated protists play a major role in Marine viruses show enormous genetic diversity 214
grazing of other microbes 169 Viromes are creators of genetic diversity and exchange 215
Heterotrophic flagellated protists have different feeding Conclusions 215
mechanisms 170 References and further reading 215
Many protists are mixotrophic 172
The choanoflagellates have a unique morphology
Chapter 8 Microbes in Ocean
and feeding mechanism 172
Dinoflagellates have several critical roles in Processes—Carbon Cycling 219
marine systems 173 Physical factors and biotic processes determine the
Dinoflagellates and other protists undertake diel fate of carbon in the oceans 221
vertical migration 175 Marine phytoplankton are responsible for about half
Some dinoflagellates exhibit bioluminescence 176 of the global primary production 222
The ciliates are voracious grazers of other protists There are wide geographical and seasonal variations in
and bacteria 177 primary production 224
The haptophytes (prymnesiophytes) are some of the Dark ocean carbon fixation makes a major contribution
most abundant components of ocean phytoplankton 178 to primary production 226
Diatoms are extremely diverse and abundant primary Classic food chain and microbial loop processes occur
producers in the oceans 181 in the epipelagic 226
Other stramenopiles may cause harmful blooms 183 The microbial loop results in retention of dissolved
Thraustochytrids and labyrinthulids are active nutrients 227
degraders of organic matter 183 The biological pump transports fixed carbon to the
deep ocean and sediments 228
 CONTENTS ix

Carbon export of primary production may change due THE SILICON CYCLE 253
to ocean warming and acidification 228
Silicification of diatoms is an economic process for
Ingestion of bacteria by protists plays a key role in the
construction of a cell wall 253
microbial loop 229
Diatom blooms depend on the availability of silica
The viral shunt catalyzes nutrient regeneration in the
in the environment 254
upper ocean 230
Eutrophication alters the silicon balance in
Microbial processes alter the composition of DOM 231
coastal zones 254
Eutrophication of coastal waters stimulates
Conclusions 255
microbial activity 232
References and further reading 255
Conclusions 233
References and further reading 233
Chapter 10 Microbial Symbioses of
Chapter 9 Microbes in Ocean Marine Animals 259
Processes—Nitrogen, Sulfur, Iron, Symbioses occur in many forms 260
Phosphorus, and Silicon Cycling 235 Chemosynthetic bacterial endosymbionts of animals
were discovered at hydrothermal vents 260
NUTRIENT LIMITATION 236 A wide range of other chemosynthetic endosymbioses
occurs in the deep sea 264
Key elements act as limiting nutrients for Chemosynthetic symbioses are also widespread in
phytoplankton 236 shallow-water sediments 266
Productivity of surface waters shows marked Animals colonizing whale falls depend on autotrophic
geographical variations 236 and heterotrophic symbionts 269
Ocean microbes require iron 237 Sea squirts harbor photosynthetic bacteria 269
Terrestrial runoff, dust, and volcanic ash are major Endosymbionts of bryozoans produce compounds that
sources of iron input 237 protect the host from predation 270
Hydrothermal vents and glacial melting also supply Sponges contain dense communities of specific microbes 273
iron to the oceans 238 Many marine invertebrates depend on photosynthetic
Whales and seabirds play a major role in supply of endosymbionts 275
iron to phytoplankton 241 Zooxanthellae (Symbiodiniaceae) show extensive
THE NITROGEN CYCLE 241 genetic diversity and host specificity 275
Many corals are dependent on zooxanthellae for nutrition 276
There have been major shifts in our understanding
Coral bleaching occurs when the host–symbiont
of the marine nitrogen cycle 241
interactions are uncoupled 278
Diazotrophs incorporate atmospheric nitrogen into
The coral holobiont contains multiple microbial partners 280
organic material 241
Zooxanthellae boost the growth of giant clams 281
Fixed nitrogen is returned to the inorganic pool by
Some fish and invertebrates employ symbiotic bacteria
ammonification and nitrification 243
to make light 284
Assimilation of ammonium and nitrate fuels growth
The bobtail squid uses bacterial bioluminescence for
of phytoplankton and other microbes 244
camouflage 286
Nitrate reduction, denitrification, and anammox reactions
Conclusions 287
return nitrogen to its elemental form and other gases 244
References and further reading 288
Diverse microbial metabolic processes occur in oxygen
minimum zones (OMZs) 245
Microbial processes in sediments are a major Chapter 11 Microbial Diseases of
contributor to nitrogen cycling 247 Marine Organisms 291
THE SULFUR CYCLE 247 Diseases of marine organisms have major ecological
The oceans and sediments contain large quantities and economic impact 292
of sulfur compounds 247 DISEASES OF CORALS, SPONGES,
Metabolism of organic sulfur compounds is especially AND ECHINODERMS 292
important in surface waters 248
DMSP production leads to release of the climate-active Infectious diseases threaten the survival of corals 292
gas dimethyl sulfide (DMS) 248 Vibrios are associated with many coral diseases 293
The fungus Aspergillus sydowii caused mass mortality
THE PHOSPHORUS CYCLE 252 of sea fans in the Caribbean Sea 295
Phosphorus is often a limiting or co-limiting nutrient 252 Black band disease of corals is a disease of corals
Marine microbes are adapted to low and variable worldwide 295
levels of phosphorus 252
x CONTENTS

White plague and white pox are major diseases DISEASES OF MAMMALS 320
affecting Caribbean reefs 298
Dinoflagellate and diatom toxins affect
Protistan parasites may cause tissue necrosis and
marine mammals 320
skeletal erosion 299
Virus disease cause mass mortalities in cetaceans and
Viruses have a pivotal role in coral health 300
pinnipeds 321
Sponge disease is a poorly investigated global
Viruses from nine different families have been linked
phenomenon 301
to diseases of marine mammals 321
Mass mortalities of echinoderms have caused major
Several species of bacteria and fungi infect marine
shifts in reef and coastal ecology 303
mammals 323
DISEASES OF MOLLUSCS 304
DISEASES OF SEA TURTLES 323
Bacteria are a major cause of disease in molluscs 304
Sea turtles are affected by a virus promoting growth
Several protistan diseases affect culture of oysters
of tumors 323
and mussels 305
Virus infections are a major problem in oyster culture 306 DISEASES OF SEAGRASSES AND SEAWEEDS 324
DISEASES OF CRUSTACEANS 306 Heterokont protists cause ecologically important
mortality of seagrasses 324
Bacteria cause epizootics with high mortalities in
Bacteria, fungi, and viruses infect marine macroalgae 324
crustaceans 306
Conclusions 326
Expansion of crustacean aquaculture is threatened by
References and further reading 326
viral diseases 307
Parasitic dinoflagellates also cause
disease in crustaceans 309 Chapter 12 Marine Microbes as
DISEASES OF FISH 309 Agents of Human Disease 331
Microbial diseases of fish cause major losses in
MICROBIAL INFECTIONS 332
aquaculture and natural populations 309
Microbial infections of fish cause a variety of Pathogenic vibrios are common in marine and
disease signs 310 estuarine environments 332
Fish-pathogenic bacteria possess a range of virulence Vibrio cholerae is an autochthonous aquatic bacterium 332
mechanisms 311 Complex regulatory networks control human
Vibrios are responsible for some of the main infections colonization and virulence of V. cholerae 333
of marine fish 311 Mobile genetic elements play a major role in the
Pasteurellosis affects warm-water marine fish 313 biology of Vibrio spp. 333
Aeromonas salmonicida has a broad geographic range Non-O1 and non-O139 serotypes of Vibrio cholerae are
affecting fish in fresh and marine waters 314 widely distributed in coastal and estuarine waters 335
Marine flexibacteriosis is caused by a weakly virulent Vibrio vulnificus is a deadly opportunistic pathogen 337
opportunist pathogen 315 Pathogenicity of V. vulnificus is due to the interaction
Piscirickettsia and Francisella are intracellular of multiple gene products 338
proteobacteria infecting salmon and cod 316 Environmental factors affect the pathogenicity of
Intracellular Gram-positive bacteria cause chronic V. vulnificus 338
infections of fish 316 Vibrio parahaemolyticus is the leading cause of
Some Gram-positive cocci affect the central nervous seafood-associated gastroenteritis 339
system of fish 317 Microbes associated with fish and marine mammals
Viruses cause numerous diseases of marine fish 318 can be transmitted to humans 340
Infectious salmon anemia (ISA) is one of the most DISEASES CAUSED BY MARINE MICROBIAL
serious diseases in salmon culture 318 TOXINS 341
Viral hemorrhagic septicemia (VHS) virus infects many
species of wild fish 318 Scombroid fish poisoning results from bacterial
Lymphocystis virus causes chronic skin infection of fish 319 enzyme activity 341
Birnaviruses appear to be widespread in marine fish Botulism is a rare lethal intoxication from seafood 341
and invertebrates 319 Fugu poisoning is caused by a neurotoxin of
Viral nervous necrosis (VNN) is an emerging disease bacterial origin 342
with major impact 319 TTX is widespread amongst marine animals 343
Protists cause disease in fish via infections, toxins, Some dinoflagellates and diatoms produce harmful
and direct physical effects 319 toxins 343
 CONTENTS xi

Paralytic shellfish poisoning is caused by saxitoxins Shellfish from sewage-polluted waters can cause
produced by dinoflagellates 344 human infection 369
Brevetoxin causes illness via ingestion or inhalation Microbiological standards are used for classification of
during red tides 346 shellfish production areas 370
Dinophysiotoxins and azaspiracid toxins from shellfish Direct testing for pathogens in shellfish is possible with
result in gastrointestinal symptoms 346 molecular methods 371
Amnesic shellfish poisoning is caused by toxic diatoms 347 OIL AND OTHER CHEMICAL POLLUTION 372
Ciguatera fish poisoning has a major impact on the
health of tropical islanders 347 Oil pollution of the marine environment is a
Bacteria influence the production of HAB toxins 349 major problem 372
Dinoflagellate and diatom toxins may be antipredator Microbes naturally degrade oil in the sea 372
defense mechanisms 349 Physical and biological processes affect the fate
Complex factors affect the incidence of HABs and of oil spills 373
toxin-associated diseases 350 Bioremediation of oil spills may be enhanced by
Coastal waters must be regularly monitored to assess emulsifiers and nutrients 373
the development of HABs 351 Microbes can detoxify heavy metals from contaminated
Conclusions 351 sediments 376
References and further reading 352 Microbial systems can be used for
ecotoxicological testing 377
Microbial adsorption and metabolism affect
Chapter 13 Microbial Aspects of accumulation of mercury 377
Marine Biofouling, Biodeterioration, Microbial cycling is important in the distribution of
persistent organic pollutants 377
and Pollution 355 Plastic pollution of the oceans is a major global problem 378
Conclusions 379
BIOFOULING AND BIODETERIORATION 356 References and further reading 382
Microbial biofilms initiate the process of biofouling 356
Microbes induce corrosion of metals, alloys, and
composite materials 357
Chapter 14 Marine Microbial
Microbes cause biodeterioration of timber and marine Biotechnology 387
wooden structures 358 Enzymes From Marine Microbes Have Many
Microbial growth and metabolism cause spoilage of Applications 388
seafood products 359 DNA polymerases from hydrothermal vent organisms
Processing, packaging, and inhibitors of spoilage are are widely used in molecular biology 390
used to extend shelf-life 360 Metagenomics and bioinformatics lead to new
Some seafood products are made by deliberate biotechnological developments 390
manipulation of microbial activities 361 Polymers from marine bacteria have many applications 391
MARINE POLLUTION BY SEWAGE AND Microalgae can produce biofuels and edible oils 391
WASTEWATER 361 Marine microbes are a rich source of biomedical
products393
Coastal pollution by wastewater is a source of
Many bioactive compounds from marine invertebrates are
human disease 361
produced by microbes 393
Human viral pathogens occur in sewage-
With so much potential from the sea, why are there so
polluted seawater 362
few new drugs? 395
Fecal indicator organisms (FIOs) are used to assess
Study of complex microbial communities may lead to
public health risks 363
new antibiotics 395
Coliforms and E. coli are unreliable FIOs for seawater
Marine microbes provide various health-
monitoring 363
promoting products 396
Enterococci are more reliable FIOs for seawater
Marine microbes have applications in biomimetics,
monitoring 364
nanotechnology, and bioelectronics 396
Molecular-based methods permit quicker analysis of
Microbial biotechnology has many applications in
indicator organisms and microbial source tracking 365
aquaculture 397
Various alternative indicator species have been
Antimicrobial agents are widely used in aquaculture 397
investigated 366
Antimicrobial resistance (AMR) is a major problem in
Countries have different quality standards for
aquaculture 399
bathing waters 367
Vaccination of finfish is widely used in aquaculture 400
xii CONTENTS

Recombinant DNA technology is used to produce Conclusions 405

vaccines for diseases caused by viruses and References and further reading 406
some bacteria 401
Live attenuated vaccines are effective but not
widely used 402 Chapter 15 Concluding Remarks 409
DNA vaccination depends on fish cells expressing a
protective antigen 402
Probiotics, prebiotics, and immunostimulants are
widely used in aquaculture 403 Index411
List of Research Focus Boxes

1.1 Deep subsurface microbes—are they dead, 9.1  an artificial fertilization of the oceans with iron
C
dormant, or just staying alive? help to mitigate global warming?
1.2 What will happen when the sea-ice melts? 9.2  ransformations of DMSP are major drivers of
T
diverse ocean processes
2.1 Cultivating the uncultured
10.1 G
 enomic insights into host–symbiont
3.1 Microbes use solar power to top up energy levels
interactions: infection, colonization, and dissemination
3.2 H
 ow can microscale bacterial swimming affect
10.2  hemosynthetic bacteria fuel ecosystem-
C
global processes?
engineering bivalves
4.1 H
 igh-throughput DNA sequencing provides new
10.3 Do corals have a core microbiome?
insights into marine microbial diversity
10.4  enomic insights into the evolution of
G
4.2 Cable bacteria unite to conduct electricity
bioluminescent symbionts of fish
5.1 E
 volution of an evolutionary pathway—changing
11.1 The pathogenicity of Vibrio coralliilyticus
views of Archaea in the tree of life
11.2  aribbean corals under threat from a new
C
5.2 D
 eep-sea FISHing—discovery of the role of
disease syndrome
archaea in carbon and nitrogen cycling
12.1 Are human Vibrio infections a microbial
6.1 S
 ailing towards a better understanding of
barometer for climate change?
heterotrophic ocean protists
13.1 I s it safe to swim? The health effects of
6.2 Exciting times for marine mycology
recreational exposure to seawater
7.1 Models for the co-evolution of bacteria and their phages
13.2 H
 ow microbes cleaned up deep-sea oil after the
7.2 “ Bloom and bust”—the life cycle of Emiliania Deepwater Horizon (DWH) disaster
huxleyi and Coccolithovirus (EhV) 13.3 Microorganisms and microplastics
7.3 New ideas about the nature and evolution of viruses
14.1 “ The enemy of my enemy is my friend”—phage
8.1 The role of microbes in the ocean carbon cycle therapy for marine diseases
Preface
In the Foreword to the Second Edition of this textbook published in 2011, Farooq Azam,
Distinguished Professor of Scripps Institution of Oceanography, wrote:

When we think of the ocean we might think of whales, waves, dolphins, fish, the smell of
the sea, its blue color, and its vastness; most of us would not look out at the sea and think
of marine microbes, nor did marine scientists for over a century. They sailed the seas and
strenuously dragged plankton nets through the ocean’s pelagic zone to capture what they
judged would represent the marine biota. But they were unaware that the great majority of
the biota, perhaps 99 percent, easily streamed through the holes of their nets; the holes were
simply too big to capture these microbes. Even when membrane filters and microscopy
were used, and they revealed great diversity of microplankton and nanoplankton, most
microbes evaded detection. The view of the pelagic web of life that emerged, and became
entrenched for a century, was then based on a tiny fraction of marine biota. As a result, fish-
eries scientists used models that did not include marine microbes, as did marine chemists
and geochemists who studied how biological forces influenced the grand cycles of elements
in the ocean. Much had to be revised as the major roles of the microbes were discovered,
following the development of new concepts, incisive imaging, and molecular methods to
observe and study marine plankton.

In this Third Edition, it will become clear that astonishing advances in marine microbiology
have continued to propel the discipline to be one of the most important areas of modern sci-
ence. I hope that readers will share a sense of my excitement of learning about new discover-
ies in this fascinating and fast-moving subject.

This book is intended primarily for upper-level undergraduates and graduate students, but
I anticipate that it will also prove useful to researchers who are interested in some of the
broader aspects and applications outside of their specific area of investigation. University
courses often include some element of microbiology as a specialist option for oceanography
or marine biology majors who have little previous knowledge of microbiology, but marine
microbiology is lightly covered in most textbooks. I hope that this book may play some part
in rectifying this deficiency. I also hope that the book will be useful to microbiology majors
studying courses in environmental microbiology, who may have little knowledge about ocean
processes or the applications of the study of marine microorganisms. I have attempted to
make the book sufficiently self-contained to satisfy all of the various potential audiences.
Above all, I wanted to create a book that is enjoyable to read, with the overall aim of bring-
ing together an understanding of microbial biology and ecology with consideration of the
applications for environmental management, human welfare, health, and economic activity.

As will become evident, many common themes and recurring concepts link the activities,
diversity, ecology, and applications of marine microbes. In each chapter, I have attempted to
summarize the current state of knowledge about each aspect, with extensive cross-linking to
other sections. To improve readability, I rarely cite specific references in the main text, but
each chapter contains special interest boxes, which contain references to recent research.
Short boxes marked with the symbols or highlight important questions or interest-
ing facts that supplement the main text. The choice of these topics is entirely my personal
whim—they represent subjects that I think are particularly intriguing, exciting, controver-
sial, or sometimes fun. Each chapter contains one to three Research Focus Boxes—these are
intended to be relatively self-contained “mini-essays,” which explore in more detail some
hot topics of investigation. Examples include the impacts of rising CO2 levels on microbial
community structure and ocean processes, interactions of microbes with plastic pollution,
symbiotic interactions, and emerging diseases of marine life. They are not intended to be
exhaustive reviews, but I hope that they serve as a stimulus for students to follow up some of
the original research papers suggested and use these as a starting point for further inquiry or
xvi PREFACE

discussion in seminars. The reference list for each chapter contain numerous suggestions for
further reading.

The great advances in marine microbiology have occurred because of the development and
application of innovative new techniques. Therefore, an understanding of the main principles
of these methods is essential if the student is to make sense of research findings. Chapter 2
is written in a style that concentrates on the principles and avoids too much technical detail,
so I implore you to read this chapter right through at an early stage to gain an overview of
methods, referring back whenever a particular technique is mentioned later.

What’s new in this edition? The general aims and structure of the book are similar, but all
material has been updated and most sections have been completely rewritten and expanded to
take account of the many new discoveries in this field since the second edition was published
in 2011, especially the astonishing advances due to extensive application of high-through-
put sequencing, single cell genomics, and analysis of large datasets. Significant advances in
understanding the diversity and evolution of bacteria, archaea, fungi, protists, and viruses
are discussed and their importance in marine processes is explored in detail. There are many
new color diagrams, illustrations, and boxes to aid students’ interest and understanding. I
have also tried to incorporate the numerous helpful comments received from students, course
leaders, and reviewers about previous editions and early drafts of this one. In addition, the
book has a companion website, which provides additional online resources for instructors
and students, including a summary of key concepts and terminology for each chapter, links
to further resources, artwork, videos, and more. My personal blog at www.marinemicro.org
also contains news and longer research articles.

Despite an exhaustive review process, astute readers will undoubtedly spot some errors and
omissions, or have suggestions for improvement. I welcome your comments—please e-mail
me at [email protected] or via the publishers.
Acknowledgments
This book would not have been possible without the continued stimulation of ideas through
teaching my university courses and mentoring research students. It has been a source of
immense pleasure and pride to see my efforts come full circle through the authoritative input
of several of my former students who have gone on to establish successful research careers
in marine microbiology. I am indebted to them and to the many other scientific colleagues
listed below, who gave up their time to comment on the original proposal, review drafts, and
provide valuable suggestions to improve the text. I also thank all those colleagues who pro-
vided images, especially Davis Laundon for his spectacular cover design. I am particularly
grateful to Alice Oven, Senior Editor at CRC Press, for her enthusiasm and encouragement
to produce a Third Edition and for developing the new improved design for the book. I also
thank Sadé Lee, Damanpreet Kaur, Andrew Corrigan, and other members of the editorial
and design team at CRC Press for their expertise and help. Finally, to Sheila—thank you for
your constant love, support, and patience during this venture.

Colin B. Munn,
Marine Institute,
University of Plymouth, UK.

Reviewers
Mike Allen, University of Exeter and Plymouth Marine Laboratory; Rudolf Amann, Max
Planck Institute for Marine Microbiology; Craig Baker-Austin, CEFAS Laboratory; David
Bourne, AIMS and James Cook University; Mya Breitbart, University of South Florida;
Craig Carlson, University of California, Santa Barbara; Ian Cooper, University of Brighton;
Michael Cunliffe, Marine Biological Association of the UK; Jesse Dillon, California
State University, Long Beach; Stuart Donachie, University of Hawaii, Manoa; Alexander
Gruhl, Max Planck Institute for Marine Microbiology; Marcelo Gutiérrez, Universidad de
Concepción, Chile; Oliver Jäckle, Max Planck Institute for Marine Microbiology; Christina
Kellogg, US Geological Survey, St. Petersburg Coastal and Marine Science Center; Anne
Leonard, University of Exeter; Davis Laundon; Marine Biological Association of the UK;
Sophie Leterme, Flinders University; Lauren Messer, University of Queensland; Jozef
Nissimov, Scottish Association for Marine Science; Mircea Podar, Oak Ridge National
Laboratory; Elizabeth Robertson, Gothenburg University; Emma Ransome, Imperial
College London; Greta Reintjes, Max Planck Institute for Marine Microbiology; Wolfgang
Sand, University of Duisberg-Essen; Val Smith, University of St Andrews; Victor Smetacek,
Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research; Roman Stocker,
Environmental Microfluidics Group, ETH Zürich; Ben Temperton, Exeter University;
Jack Thomson, University of Liverpool; Malin Tietjen, Max Planck Institute for Marine
Microbiology; Richard Thompson, University of Plymouth; Rebecca Vega-Thurber, Oregon
State University; Jack Wang, University of Queensland; Joanna Warwick-Dugdale, Plymouth
Marine Laboratory and University of Exeter; Robyn Wright, University of Warwick; Willie
Wilson, Marine Biological Association of the UK; Erik Zettler, NIOZ Royal Netherlands
Institute for Sea Research and Utrecht University; Xiao-Hua Zhang, Ocean University of
China.
Chapter 1
Microbes in the
Marine Environment

Viewed from space, it is clear why our planet would be better named “Ocean” than “Earth.”
More than 70% of the planet’s surface is covered by interconnected bodies of water. Life
originated in the oceans about 3.5 billion years ago and microbes were the only form of life
for two-thirds of the planet’s existence. The development and maintenance of all other forms
of life depend absolutely on the past and present activities of marine microbes. Yet the vast
majority of humans—including many marine scientists—live their lives completely unaware
of the diversity and importance of marine microbes. Such understanding is vital, as we now
live in a period of rapid global change. This chapter introduces the scope of marine microbi-
ology, the different types of marine microbe (viruses, bacteria, archaea, fungi, and protists),
and their place in the living world. The activities of microbes in the many diverse habitats
found in the marine environment are introduced to provide the background for more detailed
consideration in later chapters.

Key Concepts
• Modern methods have led to new ideas about the diversity and evolution of microbial
life.
• Marine microbes are highly diverse and exist in huge numbers, forming a major com-
ponent of biomass on Earth.
• The most abundant marine microbes are exceptionally small.
• The oceans provide diverse specialized habitats, in which physical and chemical con-
ditions determine microbial activities.
• Planktonic microbes are responsible for primary productivity and recycling of organic
compounds in a continuum of dissolved and particulate matter.
• Microbes are important in the formation and fate of sediments and there is abundant
life below the seafloor.
• Microbes colonize the surfaces of inanimate objects and other living organisms by the
formation of biofilms and microbial mats.
2 Chapter 1

ORIGINS AND SCOPE OF MARINE MICROBIOLOGY

Marine microbiology has developed into one of
the most important areas of modern science
Ever since a detailed study of the microbial world began in the late nineteenth century, scien-
tists have asked questions about the diversity of microbial life in the sea, its role in ocean pro-
cesses, its interactions with other marine life, and its importance to humans. However, despite
excellent work by pioneering scientists, progress in understanding accumulated gradually and
some aspects were poorly understood until recently. However, toward the end of the twentieth
century, several factors conspired to propel marine microbiology to the forefront of “main-
stream” science. The involvement of more investigators and the subsequent application of new
technology mean that it is now one of the most exciting and fast-moving areas of investigation.
Today, our subject is characterized by multidisciplinary investigations and widespread appli-
cation of powerful new tools in molecular biology, information technology, remote sensing,
and deep-sea exploration, leading to astonishing discoveries of the abundance, diversity, and
interactions of marine microbial life and its role in global ecology. These continuing new dis-
coveries necessitate radical rethinking of our understanding of ocean processes. We now real-
ize the vital role that marine microbes play in the maintenance of our planet, a fact that will
have great bearing on our ability to respond to problems such as the increase in human popula-
tion, overexploitation of fisheries, climate change, ocean acidification, and marine pollution.
Studies of the interactions of marine microbes with other organisms are providing intrigu-
ing insights into the phenomena of food webs, symbiosis, pathogenicity, and the important
role microbiomes play in metazoan biology. Since some marine microbes produce disease or
damage, we need to study these processes and develop ways to control them. Finally, marine
microbes have beneficial properties such as the manufacture of new drugs and materials, and
the development of new processes in the growing field of marine biotechnology. This chapter
sets the scene for the discussion of all these topics in this book.

Microbes include microscopic cellular

organisms and non-cellular viruses
Defining the terms “microbiology” and “microorganism” is surprisingly difficult!
Microbiology is the study of very small organisms that are too small to be seen clearly with
the naked eye (i.e. less than about 1 mm diameter), but most microbiologists are concerned
with the activities or molecular properties of microbial communities rather than viewing indi-
vidual cells with a microscope. The term “microorganism” simply refers to a form of life that
falls within the microscopic size range, but there is a huge spectrum of diversity concealed by
i TINY MICROBES…
HUGE NUMBERS this all-encompassing term. Indeed, some “microorganisms” are large enough to see without
using a microscope, so this is not entirely satisfactory either. Some scientists would argue that
Whitman et al. (1998) estimated the distinguishing features of microorganisms are small size, unicellular organization, and
the total number of bacterial osmotrophy (feeding by absorption of nutrients). The osmotrophic characteristic is important
and archaeal cells in the oceans because diffusion processes are a major limitation to cell size, as discussed in the next sec-
to be about 1029. This figure was
tion. However, this characteristic would exclude many microscopic unicellular eukaryotes,
confirmed by Bar-On et al. (2018)
in a recent recalculation based on
many of which feed by phagotrophy (engulfment of particles). For many years, these micro-
analysis of many new datasets; organisms were studied by specialists who had a traditional background in botany or zoology
they also estimated the biomass and classified into “plant-like” (algae) or “animal-like” (protozoa) groups. However, many of
of marine bacteria and archaea these organisms are mixotrophic and can switch from photosynthesis to phagotrophic feed-
at 1.3 and 0.3 gigatons (Gt) of ing, so the “plant” or “animal” similarity is meaningless. This loose grouping of organisms
carbon, respectively. Suttle (2005) is therefore called “protists,” a diverse category encompassing most of the diversity within
calculated the number of viruses the domain Bacteria. Depending on their size (see below), they may also be referred to as
to be about 1030 —again, this was microeukaryotes or picoeukaryotes. The study of marine protists and recognition that they
confirmed by Bar-On et al. (2018). are microbes with a major role in ocean processes has lagged behind the study of bacteria
This is an unimaginably huge num-
until recently. Where do viruses fit? Viruses are obviously microscopic, so I consider them to
ber—1 million, million, million, mil-
be microbes. However, they are not cellular, so cannot be described as microorganisms and
lion, million. If all the marine virus
particles were placed end to end, many would argue that they are not living (this question is explored in depth in Chapter 7).
they would span about 10 million In summary, in this book I use the term “microbe” as a generic descriptor for microscopic
light years (100 times the distance cellular organisms including bacteria, archaea, fungi, and protists, together with the non-
across our own galaxy). cellular viruses.
 Microbes in the Marine Environment 3

Marine microorganisms are found in all

three domains of cellular life
Biologists usually rely on the study of morphology and physiological properties to classify living
organisms, but these characteristics have always proved frustratingly unhelpful when dealing
with microbes. Modern methods of classification group organisms by attempting to determine
the evolutionary relationships. Such phylogenetic systems of classification depend on compari-
sons of the genetic information contained in their macromolecules, especially nucleic acids and
proteins. If two organisms are very closely related, we expect the sequence of the individual units
(nucleotides or amino acids) in a macromolecule to be more similar than they would be in two
unrelated organisms. In the 1970s, Carl Woese and colleagues pioneered the use of ribosomal
RNA (rRNA) sequencing in order to develop a better view of microbial diversity. Our view of the
living world has since been revolutionized by advances in this approach, made possible because
of the parallel advances in molecular biological techniques and computer processing of the large
amounts of information generated. Because the secondary structure of rRNA is so important in
the ribosome and the vital cell function of protein synthesis, base sequence changes in the rRNA
molecule occur quite slowly in evolution. In fact, some parts of rRNA are highly conserved and
sequence comparisons can be used to ascertain the similarity of organisms on a broad basis. The
methods and applications of this major technique are described in Chapter 2.

In 1990, Woese identified three distinct lineages of cellular life, referred to as the domains
Archaea, Bacteria, and Eukarya. A phylogenetic “tree of life” based on rRNA sequences
envisaged divergence of these three domains from an original “universal ancestor”
(Figure 1.1A). A phylogenetic approach to classification is now widely accepted, although
some biologists prefer other systems. Microbiologists like it because we can say that we study
two entire domains of life and a significant proportion of the third! Traditionally, members of
the domains Bacteria and Archaea have been grouped together as “the prokaryotes,” because
they share a simple internal cellular structure without a nucleus. However, the most impor-
tant consequence of the three-domain tree of life is that we now realize that the Bacteria and
Archaea are completely different phylogenetic groups. Archaea are not a peculiar, specialized

Figure 1.1 Representations of the

three domains of life. A. Simple
tree based on early interpretation
of ribosomal RNA sequencing. In
this model, the root of the tree is
envisaged as a hypothetical universal
ancestor from which all life evolved.
B. A three-domain tree based on
evidence of extensive lateral gene
transfer, revealed by studies of other
genes. (Drawn before discovery of
other archaeal branches; see Box 5.1).
C. An artistic representation of major
divisions of the tree of life by Hug
et al. (2016). The numerous known
groups of the Bacteria are shown on
the left, with the large group of cur-
rently uncultivable Bacteria termed
the Candidate Phyla Radiation at
upper right. The Archaea are shown
at the left of the lower branch, with
the Eukarya at the lower right. (See
Figure 4.1 for an updated detailed
version of the tree). Credits: B. Gary
J. Olsen, University of Illinois, based
on concept of W. Ford Doolittle. C.
Zosia Rostomian, Berkeley Lab.
4 Chapter 1

group of bacteria as originally thought (for many years they were called the archaebacteria)

?
TWO DOMAINS but are in fact a completely separate group that actually has closer phylogenetic relation-
OR THREE? ships to the Eukarya than to the Bacteria. This concept has proved to be very influential
There are many differences of in shaping our thinking about the evolution of organisms. As new methods and knowledge
opinion about the relationships about genomes has developed, the simple three-domain tree has changed, as illustrated in
between organisms, especially Figures 1.1B and 1.1C. These developments are discussed in detail in Chapters 4 and 5.
when we try to explain deep
evolutionary branches. Some The members of the Eukarya domain are the protists, fungi, plants, and animals. Their cells
authors believe that new evidence are distinguished by a membrane-bound nucleus and organelles with specific functions.
about the evolution of rRNA and
Mitochondria occur in all eukaryotic cells, with the exception of a few anaerobic protozoa, and
the use of new models to compute
phylogenetic trees calls the three-
carry out the processes of respiratory electron transport. In photosynthetic eukaryotes, chlo-
domain tree concept into ques- roplasts carry out reactions for the transfer of light energy for cellular metabolism. Various
tion. Embley and Williams (2016) lines of evidence (especially the molecular analysis of the nucleic acids and proteins) support
argue that the discovery of new the “endosymbiosis theory” (originally developed by Lynn Margulis in the 1960s) that the
archaeal groups (see Chapter 4) organelles of eukaryotic cells have evolved by a process of endosymbiosis, in which one cell
supports the idea of a two-domain lives inside another cell for mutual benefit. This theory proposes that the original source of
“eocyte” tree to explain the origin mitochondria in eukaryotic cells occurred when primitive cells acquired respiratory bacteria
of the eukaryotes from within the (most closely related to proteobacteria) and that the chloroplasts evolved from endosymbiosis
Archaea. The “archaeal ances- with cyanobacteria. Such interactions between different types of cell have continued through-
tor hypothesis” for the origin of
out evolution, and Chapter 10 contains many examples of endosymbiosis involving microbes.
eukaryotes is gaining acceptance,
although Forterre (2015) and Nasir
et al. (2016) provide alternative
explanations. Horizontal gene transfer confounds
our understanding of evolution
The use of rRNA sequences as a basis for phylogenetic classification has revolutionized
our understanding of microbial diversity and phylogeny. However, since advances in DNA
sequencing (see Chapter 2) made it possible to study the sequences of many other genes,
we have found increasing evidence of extensive horizontal gene transfer (HGT, also known
as lateral gene transfer, LGT) between microbes. Such transfers occur most commonly
between related organisms but transfers across bigger genetic distances also occur—even
between domains. Members of the Bacteria and Archaea contain some genes with very simi-
lar sequences, and members of the Eukarya contain genes from both of the other domains.
Some members of the domain Bacteria have even been shown to contain eukaryotic genes.
Previously, evolution was explained only by the processes of mutation and sexual recombina-
tion, but we now know that the pace of evolution is accelerated by the transfer and acquisition
of modules of genetic information. This phenomenon is widespread in modern members
of the Bacteria and Archaea and can occur via three processes. During the process known
as transformation, cells may take up and express naked DNA from the environment, while
conjugation relies on cell–cell contact mediated by pili. The most important source of HGT
is the process of transduction by phages (viruses infecting bacterial or archaeal cells and
introducing “foreign” DNA); this is explored in detail in Chapter 7. The enormous diversity
of marine viruses and the identification of a viral origin of genes in many marine organisms
indicate how important this process has been throughout evolution.

Viruses are non-cellular entities with great

importance in marine ecosystems
Virus particles (virions) consist of a protein capsid containing the viral genome composed of
either RNA or DNA. Because they only contain one type of nucleic acid, viruses must infect
living cells and take over the host’s cellular machinery in order to replicate. It is often thought
that viruses could have evolved (perhaps from bacteria) as obligate parasites that have progres-
sively lost genetic information until they consist of only a few genes, or that they represent
fragments of host-cell RNA or DNA that have gained independence from cellular control. New
ideas about the evolution of viruses are discussed in Chapter 7. The genome of viruses often
contains sequences that are equivalent to specific sequences in the host cell. Viruses exist for
every major group of cellular organisms (Bacteria, Archaea, Fungi, protists, plants, and ani-
mals), but at present we have knowledge of only a tiny proportion of the viruses infecting
marine life. As discussed in Chapter 7, recognition of the abundance and diversity of marine
 Microbes in the Marine Environment 5

viruses, and the role that they play in biogeochemical cycles and the control of diversity in
marine microbial communities, has been one of the most important discoveries of recent years.
? TIME TO CHOP DOWN
THE “TREE OF LIFE”?
The idea that relationships
Microbial processes shape the living world between all living organisms can
be represented as a tree of life
Probably the most important overriding features of microbes are their exceptional diversity
helped to shape Darwin’s theory
and ability to occupy every conceivable habitat for life. Indeed, what we consider “conceiv-
of evolution by natural selection
able” is challenged constantly by the discovery of new microbial communities in habitats and has been deeply embed-
previously thought of as inhospitable or carrying out processes that we had no idea were ded in the philosophy of biology
microbial in nature. Bacteria and archaea have shaped the subsequent development of life for more than 150 years. As the
on Earth ever since their first appearance—the metabolic processes that they carry out in importance of endosymbiosis and
the transformation of elements, degradation of organic matter, and recycling of nutrients HGT became better understood,
play a central role in innumerable activities that affect the support and maintenance of all some evolutionary scientists began
other forms of life. Microbial life and the Earth have evolved together, and the activities of to question the validity of the
microbes have affected the physical and geochemical properties of the planet. Indeed, they “tree of life” concept. A seminal
are actually the driving forces responsible for major planetary processes like changes in the paper by Doolittle (1999) argued
that “Molecular phylogeneticists
composition of the atmosphere, oceans, soil, and rocks. This is especially relevant to our
will have failed to find the ‘true
consideration of the marine environment, in view of the huge proportion of the biosphere tree’, not because their methods
that this constitutes. Despite the preponderance of microbes and the importance of their are inadequate or because they
activities, they are unseen in everyday human experience. Microbes live and grow almost have chosen the wrong genes, but
everywhere, using a huge range of resources, whereas plants and animals occupy only a small because the history of life cannot
subset of possible environments and have a comparatively narrow range of lifestyles. properly be represented as a tree.”
Relationships are now envisaged
as complex intertwined branches,
Marine microbes show great variation in size more like a web (Figure 1.1B) or
network of genomes (Dagan and
Table 1.1 shows the range of dimensions and volumes of some representative marine microbes. Martin, 2009). However, this
Even by the usual standards of microbiology, the most abundant microbes found in seawater remains a controversial topic, and
are exceptionally small—much smaller than implied by the common textbook or internet some have argued that analysis
statement that “bacteria are typically a few micrometers in length.” Their very small size is of genome sequences for “core
the main reason that appreciation of their abundance eluded us for so long. As described in genes” still supports the idea of a
Chapter 2, recognition of the abundance of marine microbes depended on the development common ancestor and branching
of fine-pore filters and direct counting methods using epifluorescence microscopy and flow tree (Ciccarelli et al., 2006)—an
cytometry. Small cell size has great significance in terms of the physical processes that affect approach that worked successfully
for Hug et al. (2016) to develop a
life. At the microscale, the rate of molecular diffusion becomes the most important mecha-
new tree of all major groups of life
nism for transport of substances into and out of the cell. Small cells feeding by absorption
(Figure 1.1C).
(osmotrophy) can take up nutrients more efficiently than larger cells. The surface area to
volume ratio (SA/V) is the critical factor because as cell size increases, volume (V) increases
more quickly than surface area (SA), as shown in Figure 1.2A.

The most abundant ocean bacteria and archaea have very small cell volumes and large SA/V
ratios. The majority are smaller than about 0.6 μm in their largest dimension, and many are
less than 0.3 μm, with cell volumes as low as 0.003 μm3. Indeed, the most abundant type
of organism in the oceans (the SAR11 clade, Figure 1.3A) has some of the smallest known
cells. If nutrients are severely limiting, as they are in most of the oceans, selection will favor
small cells. Since the first description of such small cells, termed “ultramicrobacteria,” their
size has provoked considerable controversy. Such extremely small cells could result from a
genetically fixed phenotype maintained throughout the cell cycle or because of physiologi-
cal changes associated with starvation. The latter explanation is supported by the fact that
some cultured bacteria become much smaller when starved. Most naturally occurring bacteria
(identified only by their genetic “signature”) have been impossible to grow in culture—this
is a central problem in marine microbiology, which we shall return to on several occasions
in future chapters. Because of this, it has been difficult to determine whether small size is a
genotypically determined condition for marine bacteria. However, studies with some recently
cultured marine bacteria from low-nutrient (oligotrophic) ocean environments show that addi-
tion of nutrients does not cause an increase in cell size. Cells use various strategies to increase
the SA/V ratio and thus improve efficiency of diffusion and transport. In fact, spherical cells
are the least efficient shape for nutrient uptake, and many marine bacteria and archaea are
thin rods or filaments or may have appendages such as stalks or buds. Figure 1.2B shows
examples of the diverse morphology of marine bacteria in a sample of ocean water. Many of
6 Chapter 1

Table 1.1 Size range of some representative marine microbes

Organism Characteristics Size Volume

(μm)a (μm3)b

Brevidensovirus Icosahedral DNA virus infecting shrimp 0.02 0.000004

Coccolithovirus Icosahedral DNA virus infecting Emiliania huxleyi 0.17 0.003

Thermodiscus Disc-shaped. Hyperthermophilic archaeon 0.2 × 0.08 0.003

“Ca. Pelagibacter ubique”c Crescent-shaped bacterium ubiquitous in ocean plankton (cultured 0.1 × 0.9 0.01
example of SAR11 clade)

Megavirus chilensis Giant virus infecting marine amoebae 0.44 0.045

Prochlorococcus Cocci. Dominant photosynthetic ocean bacterium 0.6 0.1

Ostreococcus Cocci. Prasinophyte alga. Smallest known eukaryote 0.8 0.3

Vibrio Curved rods. Bacteria common in coastal environments and associated 1×2 2
with animals and human diseases

Pelagomonas calceolata Photosynthetic flagellate adapted to low light 2 24

Pseudo-nitzschia Pennate diatom which produces toxic domoic acid 5 × 80 1600

Staphylothermus marinus Cocci. Hyperthermophilic archaeon 15 1800

Thioploca auraucae Filamentous. Sulfur bacterium 30 × 43 30000

Lingulodinium polyedra Bioluminescent bloom-forming dinoflagellate 50 65000

Beggiatoa Filamentous. Sulfur bacterium 50 × 160 314000

Epulopiscium fishelsoni Rods. Bacteria symbiotic in fish gut 80 × 600 3000000

Thiomargarita namibiensis Cocci. Sulfur bacterium 300d 14137100

aApproximate diameter × length; where one value is given, this is the diameter of spherical virus particles or cells. bApproximate values calculated

assuming spherical or cylindrical shapes. cCandidatus, provisional taxonomic name—see Chapter 4. dCells up to 750 μm have been recorded.

the larger organisms overcome the problems of diffusion by having extensive invaginations of
the cytoplasmic membrane or large intracellular vacuoles, increasing the SA. Small cell size
also has important implications for mechanisms of active motility and chemotaxis, because of
the microscale effects of Brownian movement (bombardment by water molecules) and shear
forces. Small marine bacteria have mechanisms of motility and chemotaxis quite unlike those
with which we are familiar from laboratory studies of organisms such as Escherichia coli.

As shown in Table 1.1, marine eukaryotic microbes also show a considerable variation in
size. Many protists have cellular dimensions that are more typical of the familiar bacteria.
The smallest known eukaryote is Ostreococcus tauri, which is only about 0.8 μm in diameter
(Figure 1.3B). Again, the realization that such small cells (now referred to as “picoeukary-
otes”) play a key role in ocean processes escaped attention until quite recently. Many small
protists seem capable of engulfing bacteria of almost the same size as themselves or can prey
on much larger organisms. Many groups of the flagellates, ciliates, diatoms, and dinoflagel-
lates are somewhat larger, reaching sizes up to 200 μm, and amoeboid types (radiolarians and
foraminifera) can be millimeters in diameter. Finally, a few types of bacteria can be bigger
than many protists. The largest of these is Thiomargarita namibiensis (Figure 1.3C). Further
discussion of the size range of microbes is given in the section on plankton below.

OCEAN HABITATS
The world’s oceans and seas form an
interconnected water system
The oceans cover 3.6 × 108 km2 (71% of the Earth’s surface) and contain 1.4 × 1021 liters of
water (97% of the total on Earth). The average depth of the oceans is 3.8 km, with a number
of deep-sea trenches, the deepest of which is the Marianas Trench in the Pacific (11 km). The
 Microbes in the Marine Environment 7

Figure 1.2 (a) Diagrammatic rep-

resentation of three spherical cells
showing a reduction in the ratio
of surface area (SA) to volume (V)
as size increases. V is a function of
r = 1.0 µm r = 2.0 µm r = 3.0 µm the cube of the radius (V = (4 / 3)πr 3 )
SA = 12.6 µm2 SA = 50.3 µm2 SA = 113.1 µm
2

V = 4.2 µm3 V = 33.5 µm3 V = 113.1 µm

3 whereas SA is a function of the
square of the radius (SA = 4πr2). Cells
with large SA/V ratios are more
efficient at obtaining scarce nutrients
by absorption across the membrane.
(b) Scanning electron micrograph
of picoplankton, showing various
cell morphologies of marine bac-
SA/V = 3.0 SA/V = 1.5 SA/V = 1.0 teria (cells are artificially colorized
(a) for effect). Bar represents ~ 1 μm.
Credit: Ed DeLong, Massachusetts
Institute of Technology.

(b)

ocean floor contains large mountain ranges and is the site of almost all the volcanic activity
on Earth. More than 80% of the area and 90% of the volume of the oceans occurs beyond the
continental shelf. Most of the deep-sea remains unexplored. It is usual to recognize five major
ocean basins, although they actually form one interconnected water system.

The Pacific is the deepest and largest ocean, almost as large as all the others combined. This
single body of water has an area of 1.6 × 108 km2. The ocean floor in the eastern Pacific is
dominated by the East Pacific Rise, while the western Pacific is dissected by deep trenches.
The Atlantic Ocean is the second largest with an area of 7.7 × 107 km2 lying between Africa,
Europe, the Southern Ocean, and the Americas. The mid-Atlantic Ridge is an underwater
mountain range stretching down the entire Atlantic basin and the deepest point is the Puerto
Rico Trench (8.1 km). The Indian Ocean has an area of 6.9 × 107 km2 and lies between Africa,
the Southern Ocean, Asia, and Australia. A series of ocean ridges cross the basin, and the
deepest point is the Java Trench (7.3 km). The Southern Ocean is the body of water between
60°S and Antarctica. It covers 2.0 × 107 km2 and has a fairly uniform depth of 4–5 km, with
a continual eastward water movement called the Atlantic Circumpolar Current. The Arctic
Ocean, lying north of the Arctic Circle, is the smallest ocean, with an area of 1.4 × 107 km2. As
well as the major oceans, there are marginal seas, including the Mediterranean, Caribbean,
Baltic, Bering, South China Seas, and many others.

At the margins of major landmasses, the ocean is shallow and lies over an extension of the
land called the continental shelf. This extends offshore for a distance ranging from a few
kilometers to several hundred kilometers and slopes gently to a depth of about 100–200 m,
8 Chapter 1

Figure 1.3 Extremes of size in

marine microbes (note different scale
bars). A. Cryo-electron tomography
of “Candidatus Pelagibacter ubique”
cells, one of the smallest bacteria
known (a cultivated representative
of the abundant SAR11 clade). Left:
a tomographic slice of a typical log-
phase cell. Right: the 3D isosurface-
rendered model of the same cell
reveals internal cellular organization.
Model coloring: outer membrane
(blue), inner membrane (cyan),
peptidoglycan (white), cytoplasm
(orange), nucleoid (red), ribosome-
like particles are represented by
yellow spheres. B. Transmission
electron micrograph (TEM) of sec-
tion of Ostreococcus tauri, the small-
est known eukaryote. N = nucleus,
m = mitochondrion, c = chloroplast.
C. Light micrograph of Thiomargarita
namibiensis, the largest known
bacterium, showing sulfur gran-
ules. Credits: A. Xiaowei Zhao and
Daniela Nicastro, University of
Texas SW Medical Center, Stephen before there is a steeper drop-off to become the continental slope. The abyssal plain covers
Giovannoni, Oregon State University, much of the ocean floor; this is a mostly flat surface with few features, but is broken in vari-
and J. Richard McIntosh, University ous places by ocean ridges, deep-sea trenches, undersea mountains, and volcanic sites.
of Colorado. B. Reprinted from
Henderson et al. (2007), CC-BY-2.0.
C. Heide Schulz-Vogt, Max Planck The upper surface of the ocean is in
Institute for Marine Microbiology, constant motion owing to winds
Bremen.
Rotation of the Earth deflects moving air and water in a phenomenon known as the Coriolis
Effect. Wind belts created by differential heating of air masses move the surface water, and in
combination with the Coriolis Effect they generate major surface current systems. This leads
to large circular gyres that move clockwise in the northern hemisphere and anticlockwise
in the southern hemisphere. Such gyres and currents affect the distribution of nutrients and
marine organisms. On the basis of surface ocean temperatures, the marine ecosystem can be
divided into four major biogeographical zones, namely polar, cold temperate, warm temper-
ate (subtropical), and tropical (equatorial). The boundaries between these zones are roughly
defined by latitude but are strongly affected by surface currents moving heat away from the
equator, as well as varying with the season.

Deep-water circulation systems transport

water between the ocean basins
Below a depth of about 200 m, ocean water is less affected by mixing and wind-generated
currents. However, a system of vast undersea rivers transports water around the globe and
has a major influence on the distribution of nutrients (Figures 1.4 and 8.1). This ther-
mohaline circulation system—often referred to as the “global ocean conveyor belt”—is
formed by the effects of temperature and salinity causing differences in the density of
water. Surface water in the North Atlantic flows toward the pole as the Gulf Stream. Water
is removed to the atmosphere by evaporation and during the formation of sea ice in high
latitudes, resulting in higher salinity. The cold, salty water becomes denser and sinks to
form a deep pool, which then flows south toward Antarctica, where more cold, dense water
is added. The current then splits, with one branch going toward the Indian Ocean and the
other to the Pacific Ocean. As the current nears the equator it warms and becomes less
dense, so upwelling occurs. The warmer waters loop back to the Atlantic Ocean, where
they start the cycle again.
 Microbes in the Marine Environment 9

Figure 1.4 A. The thermohaline

­circulation system (global ocean
“conveyor belt”). (Credit: NASA/JPL.)
B. Approximate locations of the
major warm (red) and cold (blue)
ocean currents and the gyres in the
South Pacific (SPG), North Pacific
(NPG), North Atlantic (NAG), South
Atlantic (SAG), and Indian Ocean
(IOG).

Light and temperature have important

effects on microbial processes
Light is of fundamental importance in the ecology of microbes that use light energy for photo-
synthesis and other functions, thus affecting primary productivity. The extent to which light of
different wavelengths penetrates seawater depends on a number of factors, notably cloud cover,
the polar ice caps, dust in the atmosphere, and variation of the incident angle of solar radiation
according to season and location on the Earth’s surface. Light is absorbed or scattered by organ-
isms and suspended particles. Even in the clearest ocean water, photosynthesis is restricted by
the availability of light of sufficient intensity to the upper 150–200 m (Figure 1.5). This is termed
the photic or euphotic zone (from the Greek for “well lit”). Blue light has the deepest penetration,
and photosynthetic microbes at the lower part of the photic zone have light-harvesting systems
that are tuned to collect blue light most efficiently (p.134). In turbid coastal waters, during sea-
sonal plankton blooms, the euphotic zone may be only a few meters deep, and green and yellow
light have the deepest penetration. Solar radiation also heats surface waters and leads to thermal
stratification of seawater. In tropical seas, the continual input of energy from sunlight leads to
warming of the surface waters to 25–30°C, causing a considerable difference in density from
that of deeper waters. Thus, throughout the year, there is a marked thermocline at about 100–150
m, below which there is a sudden reduction in temperature to 10°C or less. Little mixing occurs
between these layers. In polar seas, the water is permanently cold except for a brief period in
the summer, when a slight thermocline develops. Apart from this period, turbulence created by
surface winds generates mixing of the water to considerable depths. Temperate seas show the
greatest seasonal variation in the thermocline, with strong winds and low temperatures leading
to extensive mixing in the winter. The thermocline develops in the spring, leading to a marked
shallow surface layer of warmer water in summer. As the sea cools and wind increases, the ther-
mocline breaks down again in the autumn. Combined with seasonal variations in light intensity,
these effects of temperature stratification and vertical mixing have a great impact on rates of
photosynthesis and other microbial activities that affect the entire trophic system.
10 Chapter 1

Figure 1.5 Penetration of light of

different wavelengths in seawater.
Credit: Kyle Carothers, NOAA-OE.

Microbes occur in all the varied

habitats found in the oceans
Various ecological zones can be recognized in the marine environment, as shown in
Figure 1.6. Microbes are found everywhere—as members of the plankton, associated with
suspended particles and colloidal materials, attached to surfaces like rocks and submerged
structures, in association with plants and animals, or in sediments.

Plankton is a general term in marine biology referring to organisms suspended in the water
column that do not have sufficient power of locomotion to resist large-scale water currents (in
contrast to the nekton, which are strong-swimming animals). Traditionally, biologists refer
to phytoplankton (plants) and zooplankton (animals). Using this approach, we can add the
terms bacterioplankton for bacteria, virioplankton for viruses, and mycoplankton for fungi.
Traditional concepts of “plant” and “animal” are now unsatisfactory, and the term phytoplank-

? WHY IS THE
SEA SALTY?
ton therefore refers to all photosynthetic microbes, including cyanobacteria, as well as algae
and other eukaryotic protists. Of course, phytoplankton is only active in the photic zone, but
The constant percolation of rain- heterotrophic protists are found at all depths, where active bacterial and archaeal production
water through soil and rocks leads provides their food source. Another approach to classifying the plankton is in terms of size
to weathering, in which some of classes, for which a logarithmic scale ranging from megaplankton (>20 mm) to femtoplankton
the minerals are dissolved. Ground (<0.2 μm) has been devised. Table 1.2 shows the size classes that encompass marine microbes.
water has very low levels of salts Thus, the viruses constitute the femtoplankton, and bacteria and archaea mainly occur in the
and we cannot taste it in the water picoplankton. The recently discovered giant viruses (see Chapter 7) may also be considered
we drink. The addition of salts to to be part of the picoplankton size range. While protists (eukaryotic microbes) have a wide
the oceans from rivers is thus a size range and occur in the picoplankton, nanoplankton, and microplankton, we now know
very slow process, but evaporation
that in most marine samples, the majority of protistan cells are in the picoplankton range. This
of water from the oceans to form
clouds means that the salt concen- system of tenfold progression is not rigorously adhered to, and many investigators define pico-
tration has increased to its present plankton as organisms ≤3 µm. (This cut-off value provides a more coherent pattern in surveys
level over hundreds of millions of to estimate seasonal or geographic changes in abundance of small eukaryotes measured by
years. Seawater also percolates into filtration). Because individual taxa of photosynthetic and heterotrophic protists can span these
the ocean crust where it becomes three size ranges, this system is of little use in identification and classification. However, it is
heated and dissolves minerals, useful for specifying the size ranges that are likely to be collected using filters or meshes with
emerging at hydrothermal vents different cut-off values. It is also useful when considering the feeding of heterotrophic protists,
(Figure 1.13). Submarine volcanoes which generally graze by phagocytosis of organisms in the next size class down. Although
also result in reactions between there are important exceptions, the predator–prey size ratio is typically about 10:1.
seawater and hot rock, resulting in
the release of salts. The salt con-
centration in the oceans appears to Seawater is a complex mixture of inorganic and
be stable, with deposition of salts
in sediments balancing the inputs organic compounds, colloids, and gels
from weathering, hydrothermal
Seawater is a slightly alkaline (pH 7.5–8.4) aqueous solution—a complex mixture of more
vents, and volcanic activity.
than 80 solid elements, gases, and dissolved organic substances. The concentration of these
 Microbes in the Marine Environment 11

Estuarine and intertidal Figure 1.6 The major ecological

Epibionts: inanimate objects, detritus zones of the oceans and marine
Estuarine and intertidal Photic zone Plankton and marine snow microbial habitats (not to scale).
Epi- and endobionts: Epi- and endobionts: Epi- and endosymbionts,
coastal plants and animals algae and animals pelagic animals
Depth (m)
Neuston
0
Littoral Cold seeps Epipelagic
(coastal) 100–200
Shelf Mesopelagic
10°C 700–1000
Slope Bathypelagic
<4°C 2000–4000
Hydrothermal vents Abyssopelagic
Sediments, reefs, vents
and seeps. Epi- and Abyssal
endobionts: benthic animals 6000
plain
Hadalpelagic

Neritic (shelf) Deep-sea

10 000
trench

varies considerably according to geographic and physical factors, and it is customary to refer
to the salinity of seawater in parts per thousand (‰) to indicate the concentration of dissolved
substances. The open ocean has a salinity in the range 34–37‰, with differences in different
regions due to dilution by rainfall and evaporation. Oceans in subtropical latitudes have the
highest salinity as a result of higher temperatures, while temperate oceans have lower salinity
as a result of less evaporation. In coastal regions, seawater is diluted considerably by fresh-
water from rivers and terrestrial runoff and is in the range 10–32‰. Conversely, in enclosed
areas such as the Red Sea and Arabian Gulf, the salinity may be as high as 44‰. In polar
areas, the removal of freshwater by the formation of ice also leads to increased salinity. The
major ionic components of seawater are sodium (55% w/v), chloride (31%), sulfate (8%), mag-
nesium (4%), calcium (1%), and potassium (1%). Together, these constitute more than 99%
of the weight of salts. There are four minor ions—namely bicarbonate, bromide, borate, and
strontium—which together make up just less than 1% of seawater. Many other elements are
present in trace amounts (<0.01%), including key nutrients such as nitrate, phosphate, silicate,
and iron. The concentration of these is crucial in determining the growth of marine microbes
and the net productivity of marine systems, as discussed in Chapter 9.

Table 1.2 Classification of plankton by size

Size category Size Examples of microbial groups

range
(μm)

Femtoplankton 0.01–0.2 Virusesa

Picoplankton 0.2–2 (3)b Bacteriac, archaea,

prasinophytes, haptophytes,
some flagellates

Nanoplankton 2–20 Coccolithophores, diatoms,

dinoflagellates, flagellates

Microplankton 20–200 Ciliates, diatoms,

dinoflagellates, foraminifera,
yeasts

aSome giant viruses are >1 µm long. bA value of ≤3 µm is most commonly used, see text. cSome
filamentous cyanobacteria and sulfur-oxidizing bacteria occur in larger size classes (see Table 1.1).
12 Chapter 1

The concentration of salts has a marked effect on the physical properties of seawater. The
freezing point of seawater at 35‰ is −1.9°C, and seawater increases in density up to this point.
As noted above, this results in the formation of masses of cold, dense water in polar regions,
which sink to the bottom of the ocean basins and are dispersed by deep-water circulation cur-
rents. Differences in the density of seawater create a discontinuity called the pycnocline, which
separates the top few hundred meters of the water column from deeper water. This has great
significance, because the gases oxygen and carbon dioxide are more soluble in cold water.

Oxygen is at its highest concentrations in the top 10–20 m of water, owing to exchange
with the atmosphere and production of oxygen by photosynthetic plankton. Concentration
decreases with distance from the surface until it reaches a minimum between 200 and 1000
m, and bacterial decomposition of organic matter may create conditions that are almost
anoxic. Below this, the oxygen content increases again as a result of the presence of dense
water (with increased solubility at lower temperature) that has sunk from polar regions and
been transported on the thermohaline circulation system. This oxygen gradient varies greatly
in different regions, and there are several regions where large bodies of hypoxic water occur
at relatively shallow depths—these are the oxygen minimum zones (Figure 9.7).

The solubility of carbon dioxide is an important factor in controlling the exchange of carbon
between the atmosphere and the oceans and therefore is of huge significance in understand-
ing climatic processes, as discussed in Chapter 8. Only a very small proportion of dissolved
inorganic carbon (DIC) is present in the form of dissolved CO2 gas. Carbon dioxide reacts
with water to form carbonic acid, which rapidly dissociates to form bicarbonate, hydrogen
ions, and carbonate in the reactions:

CO2 (gas) + H 2O  H 2 CO3  H + + HCO3 −  2H + + CO32 −

These reactions tend to stay in equilibrium, buffering the pH of seawater within a narrow
range and constraining the amount of CO2 taken up from the atmosphere. However, the
IS “DISSOLVED
? ORGANIC MATTER”
REALLY DISSOLVED?
increasing atmospheric levels of CO2 since the industrial revolution are shifting the equilib-
rium and causing the pH to fall because of increased levels of H+ ions, leading to the phe-
nomenon known as ocean acidification, which may have major consequences for ocean life.
Oceanographers tradition-
ally describe organic matter as It is tempting to think of seawater as a homogeneous fluid, with planktonic microbes and nutri-
“dissolved” and “particulate’ ents evenly distributed within it. However, a growing body of evidence indicates that there
(DOM and POM, respectively).
is microscale heterogeneity in the distribution of nutrients around organisms and particles of
Measurements of concentrations
organic matter. Large-scale processes like productivity, nutrient recycling, and geochemical
and fluxes of DOM and its con-
stituent elements carbon (DOC), cycles are the result of microbial activity. In turn, physical processes like turbulence, pho-
nitrogen (DON), and phospho- ton flux, and gas exchange are translated down to the microscale level, affecting microbial
rus (DOP) are among the most behavior and metabolism. Physical factors such as diffusion, shear forces, and viscosity must
important factors in the study of be considered in this context. A pool of small, soluble organic molecules provides the starting
ocean processes. It is important material for bacterial productivity and recycling or carbon compounds that drives ocean food
to remember that the difference webs (see Chapter 8). Only molecules of less than about 600 Da can be assimilated across the
between “dissolved” and “particu- membrane—any larger molecules must be broken down by extracellular enzymes. Seawater
late” is a purely empirical distinc- contains vast amounts of polymers in the form of proteins, carbohydrates, and nucleic acids,
tion, reflecting the size of filters resulting from excretion by organisms or the release of their cellular material by lysis. Free
used in sample preparation. There
dissolved organic matter (DOM) in the form of polymeric macromolecules can spontane-
is no absolute definition, but most
filters used in studies of DOM and
ously assemble to form gels in surface waters, which coalesce to form larger aggregates
POM have pore sizes from about that diffuse into the bulk seawater. It is estimated that at least 10% of organic carbon in the
0.45 to 1.0 μm. Many bacteria oceans exists in this form. We can envisage seawater as a complex three-dimensional gel-like
and almost all viruses would pass network with a continuum between dissolved, colloidal, and particulate material (Figure 1.7).
through such filters and there- As a result, at a micrometer scale—the realm of the microbes—there is great patchiness in
fore appear in the DOM fraction! the distribution of nutrients and the physical properties of microenvironments.
Colloidal material and polymers
aggregate to form particles, and it These microgels can further coalesce into larger structures termed transparent exopolymeric
is only low-molecular-weight com- particles (TEP). These are especially important in the carbon cycle, because they are criti-
pounds like sugars and amino acids
cal in promoting the formation of particles that are sufficiently dense to sink through the
that are truly dissolved. Thus, DOM
water column, depositing carbon in the depths of the ocean and its sediments. This con-
and POM form a continuum, with
microbes spanning both fractions. tinuous shower of clumps and strings of material which falls through the water column is
termed “marine snow” because of its resemblance to falling snowflakes when illuminated
 Microbes in the Marine Environment 13

Figure 1.7 Representation of the

size continuum of marine particles,
indicating the size range of plank-
tonic microbes and methods used
to study the different fractions.
Reprinted from Verdugo et al. (2004)
with permission from Elsevier.

underwater. Marine snow consists of aggregates of plankton cells, detritus from dead or
dying plankton, zooplankton fecal material, and inorganic particles, glued together by the
matrix of TEP released from plankton (Figure 1.8). Most particles are 0.5 to a few microme-
ters in diameter, but they can grow to several centimeters in calm waters. Dissolved polymers
aggregate to form nanogels, stabilized by calcium binding. Larger aggregates form microgels
as a result of collision and coagulation of primary particles, and they increase in size as they
acquire more material through these physical processes. The nucleus for snow formation is
often the mucus-based feeding structures used by salps and larvaceans in the zooplankton.
Dying diatoms, at the end of a bloom, often precipitate large-scale snow formation owing to
the production of large amounts of mucopolysaccharides in their cell walls. The generation
of water currents during feeding by flagellates and ciliates colonizing the aggregate also col-
lects particles from the surrounding water and leads to growth of the snow particle.

Marine snow is mainly produced in the upper 100–200 m of the water column, and large par-
ticles can sink up to 100 m per day, allowing them to travel from the surface to the ocean deep
within a matter of days. This is the main mechanism by which a proportion of the photosyn-
thetically fixed carbon is transported from the surface layers of the ocean to deeper waters
and the seafloor. However, aggregates also contain active complex assemblages of bacteria and
protists that graze on them. Levels of microorganisms in marine snow are typically 108–109
mL−1, which are about 100–10000-fold higher than in the bulk water column. As particles sink,
organic material is degraded by extracellular enzymes produced by the resident microbial pop-
ulation. Microbial respiration creates anoxic conditions, so that diverse aerobic and anaerobic
microbes colonize different niches within the snow particle. The rate of solubilization exceeds
the rate of remineralization, so dissolved material leaks from snow particles, leaving a plume of
nutrients in its wake as it spreads by diffusion and advection. This may send chemical signals
that attract small zooplankton to consume the particle as food. The trailing plume also provides
a concentrated nutrient source for suspended planktonic bacteria, which may show chemotac-
tic behavior in order to remain within favorable concentrations. Thus, much of the carbon is
recycled during its descent, but some material reaches the ocean bottom, where it is consumed
by benthic organisms or leads to the formation of sediments. Photosynthesis by algae and bac-
teria leads to the formation of organic material through CO2 fixation, but viruses, heterotrophic
bacteria, and protists all play a part in the fate of this fixed carbon. The balance of their activi-
ties throughout the water column determines the proportions of fixed carbon that are reminer-
alized to CO2, transferred to higher trophic levels, or reach the seafloor. The discovery of this
mechanism, termed the microbial loop, was one of the most important conceptual advances in
biological oceanography, and its significance is considered further in Chapter 8. The sinking
rate of marine snow particles depends greatly in their composition—one area of considerable
current interest and importance is the extent to which microplastics and microbes can affect the
settlement of particles. This is explored in Box 13.1.
14 Chapter 1

Figure 1.8 Schematic diagram Surface Phytoplankton

showing the microbial processes Free-living microbes colonize
occurring in the formation and fate plume of DOM leaching
of a marine snow particle as it falls from sinking particle Metazoan
through the water column. The zooplankton
action of extracellular enzymes and
viral lysis leads to the release of dis- Grazing
solved organic material (DOM).
Bacteria, protists

Chemotaxis

Cellular detritus,
CO2
fecal pellets

Bacterial and
Mucus
protist
assemblages
within particle

Grazing

HOW DID
? CHERNOBYL FALLOUT
HELP THE STUDY OF
Deep ocean Metazoan
zooplankton
MARINE SNOW?
In 1986, a major environmental The sea surface is covered by a gelatinous biofilm
catastrophe occurred when the
nuclear reactor in Chernobyl, The interface between the sea surface and the atmosphere is the site of the exchange of gases,
Ukraine exploded, releasing hun- aerosols and trace elements, in both directions Figure 1.9). We know now that the sea surface
dreds of tons of radioactive par- microlayer (SML, typically up to 1 mm thick) has very different physicochemical and micro-
ticles. These were carried by wind bial composition from the underlying seawater. It contains high concentrations of lipids,
to many distant regions and settled proteins, and polysaccharides, much of it aggregated into TEPs, formed mainly from phy-
with rain on land and sea. Fowler toplankton as described above. The organisms associated with this surface layer are known
et al. (1987) were able to extract
as the neuston. Analysis of the microbial community of the SML shows that the bacterial
some benefit from this tragedy.
community composition contains distinct taxa, but these are interconnected to those in bulk
They had set up sediment traps
to measure vertical transport in seawater below. Surface wind, exposure to UV irradiation, deposition of air-borne dust, and
the Mediterranean Sea. Following other factors are important in determining the composition of the bacterioneuston, and these
the Chernobyl fallout, they found vary greatly according to location. The composition of the phytoneuston—microalgae found
that the pulse of radioactivity— in the SML—is also very different from the phytoplankton, especially in the diversity of
especially the rare nuclides 141Ce diatoms. Recently, fungi have been shown to be a significant component of the SML (p.187).
and 144Ce—was transported to The physical properties of the sea surface are altered by the presence of the SML through the
depths of over 200 m within a few production of surfactants, which modify phenomena such as turbulence and the formation of
days at a rate of 29 m per day. bubbles and micro-waves. This is often visible in the form of surface slicks of calmer water.
Physical processes alone could not This affects the transfer of microbes and organic compounds into the atmosphere via aerosols
account for such rapid settlement.
produced by wind action, which is linked to the formation of clouds and ice, affecting local
Fowler and colleagues concluded
that zooplankton were ingesting
and global climatic conditions (Figure 1.9). This topic is explored further in Chapter 9.
radioactive particles adsorbed to
their food source and repackaging
them as larger, denser particles in SEDIMENTS AND SURFACES
fecal pellets that aggregated with
marine snow to sink at a high rate. Microbes play a major role in marine sediments
Subsequent studies have shown
that the fecal pellets of different More than 70% of the Earth’s surface is covered by marine sediments. On the continental
zooplankton species vary greatly in shelf, sediments are formed due to the accumulation of eroded materials transported into the
their density and sinking rate. ocean as particles of mud, sand, or gravel, together with copious particulate organic matter,
 Microbes in the Marine Environment 15

Figure 1.9 Conceptual view of the

processes determining transport
across the sea surface microlayer
(SML). Reprinted from Engel et al.
(2017), CC-BY 3.0.

carbonate- and silica-rich compounds produced by biological activity. The mineral compo-
sition reflects the nature of the rocks and the type of weathering. Large rivers such as the
Amazon, Orinoco, or Ganges transfer millions of tons of fine sediments to the ocean each
year. Most of this mud settles along the continental margins or is funneled as dilute suspen-
sions by submarine canyons. Here, the sediments may be more than 10 km thick. By con-
trast, in much of the Atlantic and Pacific Oceans underneath the ocean gyres, the sediments
may only be 100–1000 m thick. Here, far from the continental land masses, abyssal clays are
formed by the deposition of fine sediments from the continents mixed with wind-blown dust,
volcanic ash, and cosmogenic dust from meteor impact. These accumulate very slowly—less
than 1 mm per 1000 years—while biogenous oozes accumulate at up to 4 cm per 1000 years.
Biogenous oozes contain over 30% of material of biological origin, mainly shells of protistan HOW MANY
plankton, mixed with clay. Oozes are usually insignificant in the shallow waters near conti-
nents. Calcareous oozes or muds cover nearly 50% of the ocean floor, especially in the Indian
? MICROBES ARE ON
A GRAIN OF SAND?
and Atlantic Oceans. They are formed by the deposition of the calcium carbonate shells (tests)
of two main types of protist: the coccolithophorids and the foraminifera (see Chapter 6). To answer this question, Probandt
et al. (2018) developed a method
Siliceous oozes are formed from the shells (frustules) of diatoms and radiolarians, which are
to examine individual grains of
composed of opaline silica (SiO2.nH2O). The rate of accumulation of biogenous oozes depends sand from a coastal sediment using
on the rate of production of organisms in the plankton, the rate of destruction during descent to fluorescent in situ hybridization
the seafloor, and the extent to which they are diluted by mixing with other sediments. (FISH) and confocal microscopy,
together with polymerase chain
In the case of coccolithophorids and foraminifera, depth has an important effect on reaction (PCR) amplification of
dissolution of the calcified scales or shells. At relatively high temperatures near the rRNA genes (see Chapter 2 for
surface, seawater is saturated with CaCO3. As calcareous shells sink, CaCO3 becomes methods). They found that indi-
more soluble as a result of the increased content of CO2 in water at lower temperatures vidual grains (just 202–635 µm
and higher pressures. The carbonate compensation depth is the depth at which carbon- diameter) were colonized by a
highly diverse community of 104 –
ate input from the surface waters is balanced by dissolution in deep waters; this varies
105 microbes (mostly bacteria, with
between 3000 m in polar waters and 5000 m in tropical waters. For this reason, calcare-
smaller numbers of archaea and a
ous oozes tend not to form in waters more than 5000 m deep. Similarly, not all of the few eukaryotes), densely packed in
silica in the frustules of diatoms reaches the ocean floor because bacterial action has a thin film in the indentations on
been shown to play a large part in the dissolution of diatom shells during their descent the sand grain. The wide range of
(p.182). The rate of deposition of protist remains to the seabed is much more rapid than taxa and functional groups found
would be assumed from their small size. This is because they are aggregated into larger shows that the variable physical
particles through egestion as fecal pellets after grazing by zooplankton and through the conditions and nutrient availability
formation of marine snow as described above. In shallower waters near the continen- in these grains of surface sediment
tal shelf, the high input of terrigenous sediments mixes with and dilutes sediments of clearly provide the right microen-
biogenous origin. vironments for many of the major
metabolic transformations carried
out by different microbial groups—
There is increasing recognition of the importance of microbial activities in the sediment– a microbial zoo on a tiny grain
water interface (SWI) and deep-sea benthic boundary layer (BBL), which is a layer of homo- of sand!
geneous water 10 m or more thick, adjacent to the sediment surface. The SWI includes high
16 Chapter 1

concentrations of particulate organic debris and dissolved organic compounds that become
adsorbed onto mineral particles.

In nutrient-rich areas with high rates of microbial activity, oxygen may be present in only the
top few millimeters or few centimeters of the sediment, but the structure and composition
of the microbial habitat is modified by benthic “storms” and the action of animals such as
worms and burrowing shrimps, which move and resuspend sediments, transporting oxygen
into deeper layers (bioturbation).

As well as the constant “snowfall” of plankton-derived material, concentrated nutrient inputs

reach the seabed in the form of large animal carcasses. For example, time-lapse photography
has shown how quickly fallen whale carcasses attract colonies of animals, and microbiologi-
cal studies accompanying these investigations have yielded novel bacteria, some with bio-
technological applications (p.390). The microbial communities and symbioses that develop
are similar to those found at hydrothermal vents and cold seeps. Other types of sediment that
provide special habitats for microbes include those in salt marshes, mangroves, and coral
reefs.

Studies of the extent to which carbon fixed in the photic zone finds its way to the seabed, and
its fate in sediments, are important in understanding the role of the oceans in the planetary
carbon cycle. Microbial processes such as production and oxidation of methane and oxidation
and reduction of sulfur compounds are of special interest. Studies of the diversity and activ-
ity of microbial life in the various types of sediment are yielding many new insights, mainly
because of the application of molecular techniques, and these are described in subsequent
chapters.

Deep marine sediments contain a vast

reservoir of ancient microbes
The microbiology of deep marine sediments and subsurface rocks is an area of current active
investigation using deep-core drilling and novel coring devices, and microbes have been
detected to a depth of 1.6 km in porous layers that were laid down as sediments tens or hun-
dreds of million years ago (MYA). Current estimates of the global distribution of bacteria
and archaea in deep sediments, obtained using a variety of techniques, give a consensus
value of about 3 × 1029 bacterial and archaeal cells. The deep biosphere remains one of the
most inaccessible ecosystems on Earth and investigation is expensive and technologically
challenging, but recent research is providing new insights leading to the conclusion that most
organisms are “barely alive” descendants of cells buried over millions of years, as discussed
in Box 1.1)

Microbes colonize surfaces through

formation of biofilms and mats
In the last few decades, the special phenomena that govern the colonization of surfaces by
microbes have come under intense scrutiny, with the growing recognition that such bio-
film formation involves complex physicochemical processes and community interactions.
Biofilms consist of a collection of microbes bound to a solid surface by their extracellular
products, which trap organic and inorganic components. In the marine environment, all kinds
of surfaces including other microbes, plants, animals, sediment particles, rocks, and fabri-
cated structures may become colonized by biofilms. Biofilm formation is considered in more
detail in Chapter 3, and its economic importance in biofouling is discussed in Chapter 13.

As a result of metabolic processes, ecological succession can result in the development of

multi-layered structures known as microbial mats. Mats can be several millimeters to a few
centimeters thick. Depending on the nutritional and environmental conditions, mats may
contain multiple types of bacteria, archaea, protists, and fungi (and their viruses) in combina-
tion with microbial polymers and sedimentary materials. These are particularly important in
shallow and intertidal waters, but they are also found in deeper nutrient-rich water. The com-
position of microbial mats is affected by physical factors such as light, temperature, water
 Microbes in the Marine Environment 17

Figure 1.10 Global distribution of

subseafloor sedimentary cell abun-
dance. A. Sedimentation rate. B.
Distance from shore. C. Integrated
number of cells. Reprinted from
Kallmeyer et al. (2012) with per-
mission from National Academy of
Sciences.

content, and flow rate; and by chemical factors such as pH, redox potential, the concentration
of molecular oxygen, sulfide, nitrate, iron, and dissolved organic compounds. Phototrophic
bacteria and diatoms are major components of stratified microbial mats in the photic zone,
and the species composition and zonation are determined by the intensity and wavelength
of light penetration into the mat. Light normally only penetrates about 1 mm into the mat
and anoxic conditions develop below this. The formation and diurnal variations of physico-
chemical gradients (especially of oxygen and sulfide) have a major effect on the distribution
of organisms in the mat. Microbial mats formed by chemosynthetic bacteria are common at
hydrothermal vents (Figure 1.12A).

Stromatolites are formed by the trapping of sedimentary particles, cemented together by

microbial exudates to form reefs or pillar-like structures. Fossils of stromatolites are common
in ancient rocks (Figure 1.12B). Cyanobacteria are active on the surface of the structures and
the underlying material is formed by a slow build-up of lithified remains of the former growth
(<1 mm per year). They occur today in a few shallow marine lagoons, such as Shark Bay in
Western Australia (Figure 1.12C).
18 Chapter 1

BOX 1.1 RESEARCH FOCUS

Deep subsurface microbes—are they dead, dormant, or just staying alive?

Studying deep subsurface sediments. Ensuring that samples are to explain the long-term maintenance of microbes in the sediments
recovered without contamination during drilling is a major chal- over millions of years. They concluded that the organisms in the deep
lenge, depending on multi-national consortia of microbiologists, oligotrophic sediments survive by using the low available energy to
geologists, and chemists, such as the Integrated Ocean Discovery just “staying alive,” rather than growing and reproducing. The rate
Programme (IODP) (Hoehler and Jørgensen, 2013) operating spe- of respiration in these deep sediments was found to be about 104
cialized drilling ships (Figure 1.11). Based on analysis of various times lower than at the seafloor and the density of organisms is only
studies using epifluorescence counts, Whitman et al. (1998) esti- about 100–1000 cells cm−3. Jørgensen (2011) estimates that cells with
mated of the number of bacteria and archaeal cells in the deep such a low metabolic rate might reproduce only once every several
marine subsurface sediments at 3.5 × 1030 bacterial and archaeal thousand years. One possibility is that cells in this nutrient-deprived,
cells. More recent calculations put the figure about ten times lower, semi-solid environment are not subject to predation by protists or
with an estimated biomass of 10 Gt (Pg) carbon (Bar-On et al., attack by viruses and this enables a stable population to exist, with
2018). What are all these organisms doing—are they dead, dor- cells devoting all their metabolic efforts to maintenance (turnover
mant, or alive? Different investigators have produced conflicting and repair of essential cell components like DNA, proteins, and
opinions on this issue (reviewed by Orcutt et al., 2013). membrane lipids) rather than growth and reproduction.

In ocean regions with high productivity, oxygen becomes quickly Could such cells be stimulated into a more active lifestyle? In
depleted below the seafloor and the anaerobic reduction of sulfate experiments conducted by Morono et al. (2011), deep sediments
coupled to the oxidation of organic matter is the dominant metabolic from beneath the Japan sea were incubated with isotope-labeled
process in sediments. However, under the extremely oligotrophic nutrients such as glucose or amino acids. They found that most of
subtropical ocean gyres, the rate of deposition of organic material the cells incorporated these nutrients, increasing their metabolic
reaching the seabed is so low (a few millimeters every 1000 years) rate 1000 times. This could be explained if most of the cells are
that oxygen can penetrate far below the seafloor (Røy et al., 2012). in a truly dormant state, such as endospores that become activated
The amount of energy available to sustain life under these conditions into vegetative growth by the addition of nutrients. This is known
is very limited. Bradley et al. (2019) developed a mathematical model to occur in laboratory studies of endospores. Although endospores
are found in deep sediments and could survive long periods of
environmental stress such as starvation, Jørgensen (2011) argues
that the amount of energy needed to germinate and return to an
active growth state far exceeds that found in these deeply buried,
ancient sediments. Trembath-Reichert et al. (2017) analyzed mate-
rial recovered from a coal seam 2.5 km below the seabed off the
coast of Japan. In this deep sample, the tiny microbial cells were
very scarce (10–100 cm−3) and DNA sequence analysis showed that
the bacterial community composition was more closely related to
that found in forest soil than in marine sediments. This indicates
that they originate from rich organic material from primeval forests
that was buried by subsidence into the ocean over 20 million years
ago and subsequently overlaid by 2 km thick shales. Studies of the
diversity of deep subsurface communities are still rather limited,
although metagenomic analyses are beginning to yield valuable
information. Based on studies of distinctive membrane lipids, sup-
ported by nucleic acid-based techniques, in sediments more than 1
m deep, it appears that archaea constitute a much greater proportion
of the biomass than bacteria (Lipp et al. 2008). Archaea may have a
selective advantage due to the nature of their cell membranes.

Do populations of these buried microbes adapt to the new

low-nutrient conditions by evolving? To answer this question,
Starnawski et al. (2017) investigated the assembly and evolution
of microbial communities in 8700-year-old 10 m thick sediments
in Arhus Bay, Denmark. They used DNA sequencing techniques
to show that a small number of specific bacterial types were pres-
Figure 1.11 A. The ocean drilling vessel JOIDES Resolution. ent throughout all sample depths. This subset represented a small
B. Sampling a sediment core for microbial analysis. Credit: proportion of the diversity near the surface but made up 40–50%
William Crawford, Integrated Ocean Drilling Program, US of the total microbial community in the deeper layers. The authors
Implementing Organization.
 Microbes in the Marine Environment 19

BOX 1.1 RESEARCH FOCUS

concluded that rare members of surface sediment microbial commu- concluded that as populations were separated over time (thousands)
nities become predominant as they become buried over thousands of years and space (tens of meters) during burial, there was hardly
of years. By comparing the whole genome sequences obtained from any accumulation of mutations. Given the extremely low generation
single cells from different depths, they showed that their genetic times of cells due to low-nutrient availability, they suggested that
diversification is minimal. Since the stable structure of the sedi- their findings can be generalized to the deepest sediments, which
ments prevents exchange of cells or genes, Starnawski et al. (2017) are millions of years old.

SOME EXAMPLES OF SPECIAL HABITATS—THE MICROBIAL MATS

DEEP SEA, POLAR OCEANS, CORAL i AND EVOLUTION
REEFS, AND LIVING ORGANISMS Some of Earth’s oldest sedimentary
rocks in Western Australia contain
Microbial activity at hydrothermal vents evidence of complex microbial
communities that existed 3.48
fuels an oasis of life in the deep sea billion years ago (BYA). These
Hydrothermal vents form a specialized and highly significant habitat for microbes. They fossilized microbially induced
occur mainly at the mid-ocean ridges at the boundary of the Earth’s tectonic plates, where sedimentary structures (MISS) and
seafloor spreading and formation of new ocean crust is occurring (Figure 1.13A). Over 300 stromatolites are believed to have
formed from mats colonizing an
such sites have been studied in the Pacific and Atlantic Oceans and many others are pre-
ancient shoreline or lagoon and
dicted on the basis of geological surveys. Seawater permeates through cracks and fissures
consist of layers of sediment and
in the crust and interacts with the heated underlying rocks, thereby changing the chemical organic material (Noffke et al.,
and physical characteristics of both the seawater and the rock. The permeability structure 2013). These ancient MISS prob-
of the ocean crust and the location of the heat source determine the circulation patterns of ably contained anaerobic photo-
hydrothermal fluids. As cold seawater penetrates into the ocean crust, it is gradually heated trophic bacteria. Cyanobacteria
along its flow path, leading to the removal of magnesium from the fluid into the rock, with evolved subsequently (~3 BYA),
production of acid during the process. This leads to the leaching of other major elements and and this led to the development
transition metals from the rock into the hydrothermal fluid, and sulfate in the seawater is of an oxygen-containing atmo-
removed by precipitation and reduction to hydrogen sulfide. As the percolating fluids reach sphere. This heralded the evolu-
tion of eukaryotic organisms (~2
the proximity of the magma heat source, extensive chemical reactions occur within the rock
BYA) and eventually, multicellular
and the pressurized fluids are heated to over 350°C, becoming buoyant and rising toward the
life (0.6 BYA). Until the start of
ocean floor. As they rise, decompression causes the fluids to cool slightly, and precipitation the Cambrian period (~0.54 BYA),
of metal sulfides and other compounds occurs en route. The hydrothermal fluid is injected microbial mats are believed to have
dominated the surface of the ocean
floor, but this is thought to have
ended with the diversification of
animals that began burrowing into
the seabed.

Figure 1.12 A. Chemosynthetic

microbial mats covering red algae
and coral in an area where hydro-
thermal vent and coral reef com-
munities overlap at 190 m depth. B.
Cross section of stromatolite fossil
(Eocene period, 56–34 MYA) from
Fort Laclede Bed, Wyoming showing
layered structure due to microbial
growth and sediment accumula-
tion. C. Stromatolites at the hyper-
saline Hamelin Pool, Shark Bay, W.
Australia. Credits: A. Submarine
Ring of Fire 2004 Exploration,
NOAA Vents Program; B. James St.
John, CC-BY-SA-2.0 via Wikimedia
Commons; C. Bryn Pnzauger. CC BY
2.0 via Wikimedia Commons.
20 Chapter 1

Figure 1.13 A. Schematic diagram

of processes occurring at hydro- Seawater Plume of superheated vent
thermal vent systems. Gradients of enters
and manganese oxides and
temperature and chemical elements ocean crust silicates precipitate in seawater
create a variety of habitats for diverse
microbial and animal communities. 350°C
B. A venting black smoker emit-
“Smoker” chimney
ting jets of particles of iron sulfide. formed by massive
Image shows a dense colony of Riftia Sedimentation
preciptation of
pachyptila giant tubeworms. Credit:
NOAA Pacific Marine Environmental Plume at warm vent
Laboratory. 6–23°C Seawater 2–4°C

Shallow rocks
10–200 m (20–100°C )

Permeat ion Deep rocks

1–3 km (> 350°C) enriched with Cu,
Mn, Fe, Zn, S, Si

Magma heat source

A.

B.

into the ocean as plumes of mineral-rich superheated water. The hottest plumes (up to 350°C)
are generally black, because of the high content of metal sulfide and sulfate particles, and
precipitation occurs as the hot plume mixes with the cold seawater. Some of these precipitates
form chimney structures called “black smokers” (Figure 1.13B) while others are dispersed
through the water and form sediments in the vicinity. In other parts of the vent field, the cir-
culation of hydrothermal fluid may be shallower, leading to diffuse plumes of water heated
to 6–23°C. The gradients of temperature and nutrients that exist at hydrothermal systems
provide a great diversity of habitats for microbes suspended in the surrounding heated waters,
in sediments, and attached to surfaces of the chimneys. Many of these are hyperthermo-
philic bacteria and archaea, which can grow at temperatures up to 121°C, while others grow
at lower temperatures further from the fluid emissions. Molecular studies are revealing an
 Microbes in the Marine Environment 21

astonishing diversity of such organisms, many of which have biotechnological applications.

HOW VULNERABLE
The microbiology of the deep subsurface rocks beneath vents is also now under investigation,
and many novel microbes and metabolic processes are being discovered. Microbial activity ? ARE VENT
COMMUNITIES?
in the deep subsurface contributes to the chemical changes in composition during circulation
of the hydrothermal fluids. Hydrothermal vent communities
are among the most productive
Hydrothermal vent systems were first described in 1977, when scientists aboard the submers- ecosystems on Earth, but they
ible Alvin, from Woods Hole Oceanographic Institution, were exploring the seabed about cover a relatively small area and
2500 m deep near the Galapagos Islands. The discovery of life around the vents was totally are very ephemeral, lasting only
months or a few years. The flow of
unexpected. The Alvin scientists observed dense communities of previously unknown ani-
vent fluids may gradually decline,
mals, including giant tubeworms, clams, anemones, crabs, and many others. Subsequent
or volcanic eruptions may destroy
research showed that the warm waters near hydrothermal vents contain large populations the site completely. The key ani-
of chemosynthetic bacteria and archaea, which fix CO2 using energy from the oxidation of mals inhabiting these sites depend
sulfides in the vent fluids. This metabolism supports a food chain with many trophic levels on production of large quantities
that is independent of photosynthesis. We now know that many of the animals at vent sites of larvae with sufficiently long
contain chemosynthetic bacteria as symbionts within their tissues or on their surfaces—these lifespans for them to be dispersed
relationships are discussed in Chapter 10. In addition, bacterial populations directly support via currents until a new vent site
the growth of filter-feeding animals, such as clams and mussels, or shrimp that graze on is located. Goffredi et al. (2017)
microbial mats. Previously, animal life was thought always to rely ultimately on the fixation showed that differences in the
of CO2 by photosynthesis, but the vent communities function without the input of material overlying sediment and the chem-
istry of vent fluids can result in
derived from the use of light energy, although the sulfide oxidation depends on dissolved
colonization of neighboring vents
oxygen in the water, and the origin of this is photosynthetic. by very different animal species.
Vent ecosystems clearly show resil-
ience to natural disturbance but
Cold seeps also support diverse life may be threatened by undersea
based on chemosynthesis mining for rare minerals deposited
around vents unless effective con-
Cold seeps are abundant along the continental shelf and slope, where the upwards percola- servation measures are put in place
tion of fluids through fissures in the sediments is caused by plate tectonic activity and other (Van Dover, 2014).
geological processes, allowing high concentrations of hydrocarbons to seep into the water
column. The temperature of emissions is the same as ambient seawater, or just a little higher,
but they were described as “cold” seeps to distinguish them from the warm and superheated
hydrothermal vents. Cold seeps were first discovered in 1983 in the Gulf of Mexico at a depth
of 3.2 km and have subsequently been found in many other regions. They are most common
along the continental margins, and although they usually occur in deep water (the deepest is
at 7.3 km in the Japan Trench), there are some shallow seeps off the coasts of Chile, northern
California, Oregon, and Denmark.

Methane is produced in anoxic sediments by a range of methanogenic archaea (p.87), and in

very deep sediments it is formed by through thermogenic chemical transformation of organic
matter at high pressure and temperature. Methane has a low density and rises to stability
zones with a specific combination of temperature and pressure under the seabed, where it
combines with water molecules combine to form a stable crystalline structure (clathrate).
Seeps are formed when fissures occur, destabilizing the clathrate due to changes in tempera-
ture or pressure. Some sites are associated with seeps of hypersaline brines or leakage of oil
or gas from hydrocarbons reservoirs. The methane supports prolific chemosynthetic com-
munities consisting of free-living methanotrophic bacteria and archaea, as well as those liv-
ing symbiotically with invertebrates, including bivalve mollusks from five different families
and siboglinid tubeworms (Chapter 10). Reef-like structures are created by the deposition of
calcium carbonate as a product of the anaerobic oxidation of methane (p.89) by consortia of
methanotrophic archaea and sulfate-reducing and bacteria. A wide range of other animals
including polychaete worms, sea stars, urchins, echinoderms, gastropods, crabs and lobsters,
and shrimp are sustained by this chemosynthetic community (Figure 1.14).

Microbes inhabit the interface of

brine pools in the deep sea
Distinct pools of hypersaline water occur in the Gulf of Mexico, Red Sea, and Eastern
Mediterranean Sea, created by movement of the Earth’s crust. For example, those in the
Mediterranean Sea are thought to have been formed when a large area of the sea became
22 Chapter 1

isolated from the Atlantic Ocean about 250 MYA. Regions of the sea evaporated, leaving a

i
LISTENING IN TO layer of rock salt, which later became covered by sediments after the Mediterranean basin
METHANE SEEPS reflooded. Movement of the tectonic plates has exposed the salt deposits in a few areas. Here,
To date, most methane seeps the salts have dissolved to form dense pockets of highly concentrated solutions of different
have been discovered via under- salts, separated from the overlying seawater by extremely sharp density gradients (pycno-
water exploration using manned clines) and environmental interfaces (chemoclines). These undersea brine pools have extreme
submersibles or remotely oper- physical and chemical conditions, notably the complete lack of oxygen and near-saturated
ated underwater vehicles (ROVs, solutions of salts (up to ten times the salinity of seawater). They are therefore termed Deep
p.32). Scientists recently detected
Sea, Hypersaline, Anoxic Basins (DHABs).
nearly 1000 seeps about 16 km
off the coast of Oregon by using a
hydrophone to record the sound
The interface is just a few meters thick, and accurate sampling within this chemocline at
of methane bubbles escaping from such a great depth requires great ingenuity. Conventional remotely operated vehicles (ROVs)
the seabed. Dziak et al. (2018) and manned submersibles cannot be used because of the damaging properties of the brines.
found that the frequency of the Because of the extreme conditions, organisms growing in such habitats require special adap-
sound emitted is related to the size tations, and it might be expected that the diversity of microbial types would be low. However,
of the gas bubble. The next step through the use of DNA-based identification methods, many previously unknown taxa of bac-
will be to fine-tune the detec- teria have been identified, occupying narrow niches in the highly stratified interface. Archaea
tion process, so that hydrophone appear to be less prevalent than bacteria. The low availability of organic compounds at such
methods can be used directly to depths suggests that chemoautotrophic metabolism sustains these communities and there is
estimate the size of the methane
evidence for production of methane and sulfur cycling. Eukaryotic microbes (heterotrophic
reservoirs.
protists) are also abundant at the interface, and there is extensive grazing of the bacteria. It
appears that some protists may be protected from the extreme anoxic and sulfidic chemical
conditions by symbiotic interactions or partnerships with bacteria. Each basin seems to be
inhabited by a distinctive set of microbial species, which have adapted through evolution to
the unique chemical conditions of each DHAB.

At the edge of some brine pools, there are communities of mussels containing chemosyn-
thetic bacteria and associated invertebrates, similar to those found at methane seeps.

Microbes in sea ice form an important part

of the food web in polar regions
At the poles, the temperature is so low during the winter that large areas of seawater freeze to
form sea ice, some of which forms adjacent to the coastal shoreline and some of which forms
floating masses of pack ice. Sea ice forms when the temperature is less than −1.9°C, the freez-
ing point of water at 35‰ salinity. The first stage in sea-ice formation is the accumulation of
minute crystals of frazil ice on the surface, which are driven by wind and wave action into
aggregated clumps called grease ice. These turn into pancake-shaped ice floes that freeze
together and form a solid ice cover. At the winter maxima, the combined coverage by sea ice
at the north and south polar regions is almost 10% of the Earth’s surface (1.8 × 107 km2 in
the Antarctic and 1.5 × 107 km2 in the Arctic). During the formation of frazil ice, planktonic
microbes become trapped between the ice crystals and wave motion transports more organ-
isms into the grease ice during its formation. Near the ice-air interface, temperatures may be

Figure 1.14 Mussels and shrimp at

a chemosynthetic cold-seep com-
munity, Gulf of Mexico. Credit:
NOAA-OE, Expedition to the Deep
Slope 2007.
 Microbes in the Marine Environment 23

as low as −20°C during the depths of the polar winter, while the temperature at the ice-water
interface remains fairly constant at about −2°C. When seawater freezes, it forms a crystal-
line lattice of pure water, excluding salts from the crystal structure. The salinity of the liquid
phase increases, and its freezing point drops still further. This very cold, high-density, high-
salinity (up to 150‰) water forms brine pockets or channels within the ice, which can remain
liquid to −35°C. The ice becomes less dense than seawater and rises above sea level, with the
channels draining brine through the ice to the underlying seawater. Thus, sea ice is very dif-
ferent from freshwater glacial ice. Loss of sea ice cover is a topic of great concern (Box 1.2).

The structure of sea ice provides a labyrinth of different microhabitats for microbes, with varia-
tions in temperature, salinity, nutrient concentration, and light penetration. This enables coloni-
zation and active metabolism by distinctive mixed communities of cold-adapted (psychrophilic)
photosynthetic and heterotrophic protists and bacteria, as well as viruses. Microbial activities
also alter the physicochemical conditions, mainly owing to the production of large amounts
of cryoprotectant compounds and extracellular polymers, leading to the creation of additional
microenvironments. The dominant photosynthetic organisms near the ice–sea interface are pen-
nate diatoms and small dinoflagellates (Figure 1.15A). The density of diatoms in sea ice may be
up to 1000 times that in surface waters. Through photosynthesis, the microalgae make a small,
but significant, contribution to primary productivity in the polar regions. For example, the con-
tribution of sea ice to primary productivity in the Southern Ocean is only about 5% of the total,
but it extends the short summer period of primary production and provides a concentrated food
source that sustains the food web during the winter. During the Antarctic winter, microalgae on
the undersurface of sea ice are the main source of food for grazing krill, shrimp-like crustaceans
that are the main diet of fish, birds, and mammals in the Southern Ocean. A wide range of pro-
tists and heterotrophic bacteria have been found in sea ice, including new species with biotechno-
logical potential. Some microbes remain active—albeit at a much-reduced metabolic rate—even
in the coldest parts of the ice and in “frost flowers” formed on the surface of the ice, where they
are trapped in pockets of very low temperatures and high-salinity (Figure 1.15B).

Figure 1.15 (a) Light microscopy

image of diatoms living in annual
sea ice, McMurdo Sound, Antarctica.
(b) Frost flowers over young sea ice
in the central Arctic Ocean. Credits:
(a) Gordon T. Taylor, NOAA Corps
Collection. (b) Matthias Wietz, MPI
Marine Microbiology, Bremen.
24 Chapter 1

BOX 1.2 RESEARCH FOCUS

What will happen when the sea-ice melts?

Global warming is affecting polar regions more rapidly than trophic levels. Thus, an increase in small Synechococcus in place
any other parts of the planet. Although there is a large natural of larger phytoplankton in a warmer Arctic Ocean could have major
annual variation in the thickness and extent of summer sea ice in the effects on transfer of productivity to other organisms.
Arctic Ocean, these have shown a dramatic decline since the 1970s.
In 2018, some of the thickest and oldest areas of ice to the north of Benthic microbes are also affected. Antje Boetius and colleagues
Greenland started to break up (Figure 1.16). Notz and Stroeve (2016) at the Alfred Wegener Institute and Max Planck Institute of Marine
provide compelling evidence that this is directly linked to CO2 emis- Microbiology are undertaking an intensive research program to inves-
sions as a result of human activity, and their models predict that the tigate the microbiology of current and archived samples of Arctic
Arctic Ocean will be free of summer ice by the end of this century. deep-sea sediments, revealing important information about changes
The loss of ice cover results in changes to wind patterns and warmer in microbial diversity and how the flux of particulate organic mat-
and less saline waters in the upper ocean. Also, we now know that ter is affected by the loss of permanent ice cover. Rapp et al. (2018)
sea ice absorbs large amounts of CO2 from the atmosphere (Søgaard report the analysis of microbial communities in areas affected by the
et al., 2013). Most significantly, loss of summer ice also reduces the summer ice melt in 2012, which was the largest decline to date. Vast
albedo effect resulting in a positive feedback, which increases the quantities of ice algae associated with the underside of the sea ice
absorption of solar energy by seawater rather than reflecting sun- became detached and sank to the seafloor at over 4000 m in depth.
light. This further accelerates the rise in temperature. Many scien- Using high-throughput DNA sequencing, Rapp and colleagues stud-
tific programmes are investigating the impact of these changes and ied the diversity of different groups of microbes in the sea ice, in
microbiological research is especially important. We urgently need melt water and in the surface sediment beneath the melting ice. The
to know how changes in microbial communities and their activities sea-ice algae aggregates were shown to be mostly composed of dia-
are altering the cycling of carbon and other elements. toms, especially Melosira arctica. The large filaments of this diatom
sank rapidly to the seabed. The sinking aggregates were shown to
Changes in Arctic picophytoplankton. One example of change transport bacteria such as Flavobacteriia and Gammaproteobacteria,
has been observed in apparent shifts of picophytoplankton diversity including Glaciecola and Paraglaciecola. These are known to be
in the Arctic Ocean. Paulsen et al. (2016) found high numbers of the actively involved in the breakdown of algal and other organic mate-
cyanobacterium Synechococcus at high latitudes. This was surpris- rial via the production of extracellular enzymes. Labyrinthulids were
ing, as Arctic water masses are usually thought to be dominated also detected—these protists are known to be active in degradation
by larger picoeukaryotes adapted to low temperatures, such as of algae (p.183). Rapp et al. found that the algal deposits changed
Micromonas spp. This could be attributed in part to a change in the the community composition on the seafloor by introducing bacteria
flow of water from the Atlantic into the Arctic Ocean due to global from the sea ice, although these appeared to be overgrown by spe-
warming, but analysis of the distribution of different genetic clades cific groups of indigenous deep-sea bacteria within a few months.
of the cyanobacteria suggests that some of them have adapted to If strong summer ice melting becomes a regular occurrence as pre-
Arctic conditions. Because populations of Synechococcus are con- dicted, the abrupt export of large amounts of sea-ice algae and asso-
trolled by grazing by small heterotrophic protists, Paulsen and col- ciated microbiota will become more frequent. The authors suggest
leagues concluded that much of their biomass would be recycled that this could lead to permanent changes in bacterial community
within the microbial loop (p.220), rather than transferring to higher composition, which may lead to significant changes in the cycling of
nutrients in the surface sediments.

Changes in Antarctic food webs. Warming is also occurring in

the Antarctic, where increased melting of the ice-sheets increases
the input of fresh water, altering the salinity and structure of the
surrounding water column. Consequent changes in ocean food web
structures will also affect benthic communities, as seasonal shifts
in different types of phytoplankton in the water column or underside
of sea ice will affect the flow of fixed carbon to the seafloor. Based
on a large-scale study of microbial diversity in Antarctic sediments,
Learman et al. (2016) found that community composition varied
greatly with geographic location and was strongly influenced by
the nature of organic matter reaching the seafloor. Warming could
lead to an increase in meltwater, stimulating increased phytoplank-
ton blooms transporting organic matter to the seabed. Learman and
Figure 1.16 Time series showing the Arctic ice remain- colleagues suggest that this might cause a shift from communi-
ing at the end each summer melt season since 1985. The ties dominated by lithotrophic organisms (such as some types of
area of the oldest ice has declined 16-fold. Credit: M. archaea) to a greater proportion of chemoheterotrophs. This could
Tschudi, S. Stewart, University of Colorado, and W. Meier, cause significant changes to biogeochemical cycling.
J. Stroeve, NSIDC.
 Microbes in the Marine Environment 25

Microbial activity underpins productive

food webs in coral reefs
The most familiar coral reefs are those found in the shallow photic zone, especially in tropi-
cal and subtropical areas. These present specialized habitats, in which symbiotic interactions
between corals and microbes are responsible for primary production. Symbiotic dinoflagel-
lates (zooxanthellae) are responsible for the fixation of CO2, which promotes growth of the
coral animal tissue and skeleton, but also leads to large excesses of fixed carbon that fuels
microbial processes in the seawater and sediments surrounding the reef. Corals produce large
amounts of mucus, much of which forms TEPs and gels which provide a major source of
nutrients sustaining the growth of planktonic bacteria or aggregate to form sinking particles
which fuel benthic activities. Thus, although coral reefs only occupy about 0.1% of the sea-
floor area and occur in ocean areas of generally low productivity (due to lack of nutrient
input), they are as productive as tropical rain forests and are estimated to contain a third of
all marine species. Recent studies have shown that many other microbes associated with cor-
als and other reef organisms are involved in the primary production and recycling of carbon,
nitrogen, and other key elements. We now recognize that major phase shifts in microbial
activity are responsible for the degradation of coral reefs, with profound implications for
marine life and the destabilization of safety and livelihood of millions of people. Climate
change is a major threat to the world’s reefs, with severe mass bleaching affecting many
tropical reefs (most notably, Australia’s Great Barrier Reef) in years when average sea sur-
face temperatures have been just 1°C or so above normal. As sea levels rise, reefs may not be
able to grow quickly enough to maintain optimum light levels, while ocean acidification will
reduce their ability to calcify. These issues are discussed in detail in Chapter 10.

Living organisms are the habitats of many microbes

Microbial biofilms also form on the surfaces of all kinds of animals, seaweeds, and coastal plants;
these provide a highly nutritive environment through secretion or leaching of organic compounds.
Many organisms seem selectively to enhance surface colonization by certain microbes and dis-
courage colonization by others. This may occur by the production of specific compounds that
inhibit attachment or growth of certain microbes. Once established, particular microbes may
themselves influence colonization by other types. These processes offer obvious applications in
the control of biofouling (p.356). As well as surface (epibiotic) associations, microorganisms can
form endosymbiotic associations within the body cavities, tissue, or cells of living organisms.

Many microalgae (such as diatoms, dinoflagellates, and prymnesiophytes) and other pro-
tists (such as ciliates) harbor bacteria on their surfaces, or as endosymbionts within their
cells (Figure 4.11). Intimate associations between bacterial and archaeal cells are also being
revealed by new imaging techniques (Figure 5.5). Seaweeds and seagrasses have dense popu-
lations of bacteria (up to 106 per cm2) on their surfaces, although this varies considerably with
species, geographic location, and climatic conditions.

The external surfaces and intestinal content of animals provide a variety of habitats to a wide
diversity of microbes. Such associations commonly lead to some mutual benefit for host and
microbe—examples of symbiotic interactions between animals and microbes are considered
in Chapter 10. Pathogenic interactions may also result—microbial diseases of marine organ-
isms are discussed in Chapter 11.

Conclusions
This chapter has introduced the various types of marine microbes and some of the wide range
of habitats that they occupy. The adaptations of marine microbes to the different physical,
chemical, and biological conditions encountered have led to the evolution of highly diverse
microbes. The discovery that these tiny microbes are present in such large numbers and
biomass—and that they are responsible for the biogeochemical processes that shape our
planet—can be viewed as one of the most important advances of modern science. Subsequent
chapters build on this introduction by exploring the mechanisms underlying this diversity of
form and function.
26 Chapter 1

References and further reading Ocean habitats

Origins and scope of marine microbiology Azam, F. and Long, R. A. (2001). Sea snow microcosms. Nature 414: 495–498.
Azam, F. and Malfatti, F. (2007). Microbial structuring of marine ecosys-
Bar-On, Y. M., Phillips, R., and Milo, R. (2018). The biomass distribution
tems. Nat. Rev. Microbiol. 5: 782–791.
on Earth. Proc. Nat. Acad. Sci. 115: 6506–6511.
Costello, M. J. and Breyer, S. (2017). Ocean depths: The mesopelagic and
Brown, J. W. (2015). Principles of Microbial Diversity. ASM Press.
implications for global warming. Curr. Biol. 27: R19–R41.
Caron, D. A., Worden, A. Z., Countway, P. D., et al. (2008). Protists are
Cunliffe, M., Engel, A., Frka, S., et al. (2013). Sea surface microlayers: A
microbes too: A perspective. ISME J. 3: 4–12.
unified physicochemical and biological perspective of the air–ocean inter-
Ciccarelli, F. D., Doerks, T., von Mering, C., et al. (2006). Toward auto- face. Prog. Oceanog. 109: 104–116.
matic reconstruction of a highly resolved tree of life. Science 312: 697.
Cunliffe, M. and Murrell, J. C. (2009). The sea-surface microlayer is a
Dagan, T. and Martin, W. (2009). Getting a better picture of microbial gelatinous biofilm. ISME J. 3: 1001–1003.
evolution en route to a network of genomes. Phil. Trans. Roy. Soc. B Biol.
Engel, A., Bange, H. W., Cunliffe, M., et al. (2017). The ocean’s vital skin:
Sci. 364: 2187–2196.
Toward an integrated understanding of the sea surface microlayer. Front.
Doolittle, W. F. (1999). Phylogenetic classification and the universal tree. Mar. Sci. 4: 165.
Science 286: 1433.
Fowler, S. W., Buat-Menard, P., Yokoyama, et al. (1987). Rapid removal of
Embley, M. and Williams T. (2016). Only two domains, not three: Chernobyl fallout from Mediterranean surface waters by biological activ-
Changing views on the tree of life. Microbiol. Today 2016: 70–73. ity. Nature 329: 56–58.
Forterre, P. (2015). The universal tree of life: An update. Front. Microbiol. Grossart, H. P., Kiorboe, T., Tang, K. W., et al. (2006). Interactions
6: 717. between marine snow and heterotrophic bacteria: Aggregate formation
Gasol, J. M. and Kirchman, D. (2018). Introduction: The evolution of and microbial dynamics. Aquat. Microb. Ecol. 42: 19–26.
microbial ecology of the ocean. In Microbial Ecology of the Oceans, ed. Kjørboe, T. (2001). Formation and fate of marine snow: Small- scale pro-
Gasol, J. M. and Kirchman, D. L., pp. 1–46, Wiley Blackwell. cesses with large-scale implications. Sci. Mar. 65: 57–71.
Grossart, H.-P. and Rojas-Jiminez, K. (2016). Aquatic fungi: Targeting the Mühlenbruch, M., Grossart, H. P., Eigemann, F., and Voss, M. (2018).
forgotten in microbial ecology. Curr. Opin. Microbiol. 31: 140–145. Phytoplankton‐derived polysaccharides in the marine environment and their
Henderson, G. P., Gan, L., and Jensen, G. J. (2007). 3-D ultrastructure interactions with heterotrophic bacteria. Environ. Microbiol. 20: 2671–2685.
of O. tauri: Electron cryotomography of an entire eukaryotic cell. PLoS Nissimov, J. I. and Bidle, K. D. (2017). Stress, death, and the biological
ONE 2: e749. glue of sinking matter. J. Phycol. 53: 241–244.
Hug, L. A., Baker, B. J., Anantharaman, K., et al. (2016). A new view of Seymour, J. R. and Stocker, R. (2018). The ocean’s microscale: A microbe’s
the tree of life. Nat. Microbiol. 1: 16048. view of the sea. In Microbial Ecology of the Oceans, ed. Gasol, J. M. and
Karl, D. M. and Proctor, L. M. (2008). Foundations of microbial oceanog- Kirchman, D. L., pp. 289–466, Wiley Blackwell.
raphy. Oceanography 20: 16–27. Verde, C., Giordano, D., Bellas, C. M., et al. (2016). Polar marine microor-
Keeling, P. J. and de Campo, J. (2017). Marine protists are not just big ganisms and climate change. Adv. Microb. Physiol. 69: 187–215.
bacteria. Curr. Biol. 27: R451–R459. Verdugo, P., Alldredge, A. L., Azam, F., et al. (2004). The oceanic gel
Kolter, R. and Maloy, S. (eds.) (2012). Microbes and Evolution: The World phase: A bridge in the DOM-POM continuum. Mar. Chem. 92: 67–85.
That Darwin Never Saw. ASM Press.
Lawton, G. (2009). Why Darwin was wrong about the tree of life. New Sediments and surfaces
Sci. 2692: 34–39. Betts, H. C., Puttick, M. N., Clark, J. W., et al. (2018). Integrated genomic
Martin, W. and Embley, T. M. (2004). Evolutionary biology: Early evolu- and fossil evidence illuminates life’s early evolution and eukaryote origin.
tion comes full circle. Nature 431: 134–137. Nat. Ecol. Evol. 2: 1556–1562.
Nasir, A., Kim, K. M., Da Cunha, V., and Caetano-Anollés, G. (2016). Bolhuis, H., Cretoiu, M. S. and Stal, L. J. (2014). Molecular ecology of
Arguments reinforcing the three-domain view of diversified cellular life. microbial mats. FEMS Microbiol. Ecol. 90: 335–350.
Archaea 2016: 1851865. Bradley, J. A., Amend, J. P., and LaRowe, D. E. (2019). Survival of the few-
Schulz, H. N. and Jørgensen, B. B. (2001). Big bacteria. Annu. Rev. est: Microbial dormancy and maintenance in marine sediments through
Microbiol. 55: 105–137. deep time. Geobiol. 17: 43–59.
Sherr, B. F., Sherr, E. B., Caron, D. A., et al. (2007). Oceanic protists. Corinaldesi, C. (2015). New perspectives in benthic deep-sea microbial
Oceanog. 20: 130–134. ecology. Front. Mar. Sci. 2: 5–17.
Suttle, C. (2005). The viriosphere: The greatest biological diversity on Danovaro, R., Corinaldesi, C., Dell’Anno, A., and Rastelli, E. (2017).
Earth and driver of global processes. Environ. Microbiol. 7: 481–482. Potential impact of global climate change on benthic deep-sea microbes.
FEMS Microbiol. Lett. 364: 23, fnx214.
Whitman, W. B., Coleman, D. C., Wiebe, W. J. (1998). Prokaryotes: The
unseen majority. Proc. Nat. Acad. Sci. USA 95: 6578–6583. Hoehler, T. M. and Jørgensen, B. B. (2013). Microbial life under extreme
energy limitation. Nat. Rev. Microbiol. 11: 83–94.
Yayanos, A. A. (2003). Marine microbiology at Scripps. Oceanography
16: 63–75. Jørgensen, B. B. (2011). Deep subseafloor microbial cells on physiological
standby. Proc. Natl. Acad. Sci. USA 108: 18193–18194.
Zhao, X., Schwartz, C. L., Pierson, J., et al. (2017). Three-dimensional
structure of the ultraoligotrophic marine bacterium “Candidatus Jørgensen, B. B. and Boetius, A. (2007). Feast and famine—Microbial life
Pelagibacter ubique”. Appl. Environ. Microbiol. 83: e02807–16. in the deep-sea bed. Nat. Rev. Microbiol. 5: 770–781.
 Microbes in the Marine Environment 27

Kallmeyer, J., Pockalny, R., Adhikari, R. R., et al. (2012). Global distribu- Ewert, M. and Deming, J. W. (2013). Sea ice microorganisms:
tion of microbial abundance and biomass in subseafloor sediment. Proc. Environmental constraints and extracellular responses. Biology 2:
Natl. Acad. Sci. USA 109: 16213–16216. 603–628.
Learman, D. R., Henson, M. W., Thrash, J. C., et al. (2016). Biogeochemical Garren, M. and Azam, F. (2012). New directions in coral reef microbial
and microbial variation across 5500 km of Antarctic surface sediment ecology. Environ. Microbiol. 14: 833–844.
implicates organic matter as a driver of benthic community structure. Goffredi, S. K., Johnson, S., Tunnicliffe, V., et al. (2017). Hydrothermal
Front. Microbiol. 7: 284. vent fields discovered in the southern Gulf of California clarify role of
Lipp, J. S., Morono, Y., Inagaki, F., and Hinrichs, K. U. (2008). Significant habitat in augmenting regional diversity. Proc. Roy Soc. B: Biol. Sci. 284:
contribution of Archaea to extant biomass in marine subsurface sedi- 20170817.
ments. Nature 454: 991–994. Hoegh-Guldberg, O., Hughes, T., Anthony, K., et al. (2009). Coral reefs
Morono, Y., Terada, T., Nishizawa, M., et al. (2011). Carbon and nitrogen and rapid climate change: Impacts, risks and implications for tropical soci-
assimilation in deep subseafloor microbial cells. Proc. Natl. Acad. Sci. eties. IOP Conf. Ser.:Earth Environ. Sci. 6: 302004.
USA 108: 18295–18300. Kallmeyer, J. ed. (2017). Life at Vents and Seeps. Walter de Gruyter,
Noffke, N., Christian, D., Wacey, D., and Hazen, R. M. (2013). Microbially GmbH, Berlin/Boston.
induced sedimentary structures recording an ancient ecosystem in the Kallmeyer, J., Pockalny, R., Adhikari, R. R., et al. (2012). Global distribu-
ca. 3.48 billion-year-old Dresser Formation, Pilbara, Western Australia. tion of microbial abundance and biomass in subseafloor sediment. Proc.
Astrobiology 13: 1103–1124. Natl. Acad. Sci. USA 109: 16213–16216.
Orsi, W. D. (2018). Ecology and evolution of seafloor and subseafloor Kirchman, D. L., Moran, X. A. G. and Ducklow, H. (2009). Microbial
microbial communities. Nat. Rev. Microbiol. 16: 671–683. growth in the polar oceans—Role of temperature and potential impact of
Prieto-Barajas, C., Valencia-Cantero, E. and Santoyo, G. (2018). Microbial climate change. Nat. Rev. Microbiol. 7: 451–459.
mat ecosystems: Structure types, functional diversity, and biotechnologi- Levin, L. A., Baco, A. R., Bowden, D. A., et al. (2016). Hydrothermal
cal application. Electron. J. Biotechnol. 31: 48–56. vents and methane Seeps: Rethinking the sphere of influence. Front. Mar.
Probandt, D., Eickhorst, T., Ellrott, A., et al. (2018). Microbial life on a Sci. 3: 72.
sand grain: From bulk sediment to single grains. ISME J. 12: 623–633. Martin, W., Baross, J., Kelley, D. and Russell, M. J. (2008). Hydrothermal
Røy, H., Kallmeyer, J., Adhikari, R. R., et al. (2012). Aerobic microbial res- vents and the origin of life. Nat. Rev. Microbiol. 6: 805–814.
piration in 86-million-year-old deep-sea red clay. Science 336: 922–925. Mock, T. and Thomas, D. N. (2005). Recent advances in sea-ice microbiol-
Schrenk, M. O., Huber, J. A., and Edwards, K. J. (2010). Microbial prov- ogy. Environ. Microbiol. 7: 605–619.
inces in the subseafloor. Ann. Rev. Mar. Sci. 2: 279–304. Notz, D. and Stroeve, J. (2016). Observed Arctic sea-ice loss directly fol-
Starnawski, P., Bataillon, T., Ettema, T. J. G., et al. (2017). Microbial com- lows anthropogenic CO2 emission. Science 354: 747–750.
munity assembly and evolution in subseafloor sediment. Proc. Natl. Acad. Orcutt, B. N., LaRowe, D. E., Biddle, J. F., et al. (2013). Microbial activity
Sci. USA 114: 2940–2945. in the marine deep biosphere: Progress and prospects. Front. Microbiol.
Trembath-Reichert, E., Morono, Y., Ijiri, A., et al. (2017). Methyl- 4: 189.
compound use and slow growth characterize microbial life in 2-km- Paulsen, M. L., Doré, H., Garczarek, L., et al. (2016). Synechococcus in
deep subseafloor coal and shale beds. Proc. Natl. Acad. Sci. USA 114: the Atlantic Gateway to the Arctic Ocean. Front. Mar. Sci. 3: 191.
E9206–E9215.
Rapp, J. Z., Fernández-Méndez, M., Bienhold, C., and Boetius, A. (2018).
Some special habitats—the deep sea, polar Effects of ice-algal aggregate export on the connectivity of bacterial com-
munities in the central Arctic Ocean. Front. Microbiol. 9: 1035.
oceans, coral reefs, and living organisms
Reysenbach, A.-L. and Cady, S. L. (2001). Microbiology of ancient and
Arrigo, K. R. and Thomas, D. N. (2004). Large scale importance of sea ice modern hydrothermal systems. Trends Microbiol. 9: 79–86.
biology in the Southern Ocean. Antarctic Sci. 16: 471–486.
Rowher, F. and Youle, M. (2010). Coral Reefs in the Microbial Seas. Plaid
Caron, D. A., Gast, R. J. and Garneau, M. (2017). Sea ice as a habitat for Press, San Diego, CA.
micrograzers. In Sea Ice, ed. Thomas, D. N. pp. 370–393. John Wiley &
Sons Ltd. Silveira, C. B., Cavalcanti, G. S., Walter, J. M., et al. (2017). Microbial
processes driving coral reef organic carbon flow. FEMS Microbiol. Rev.
Cavicchioli, R. (2015). Microbial ecology of Antarctic aquatic systems. 41: 575–595.
Nat. Rev. Microbiol. 13: 691–706.
Søgaard, D. H., Thomas, D. N., Rysgaard, S., et al. (2013). The relative
Deming, J. W. and Collins, R. E. (2017). Sea ice as a habitat for Bacteria, contributions of biological and abiotic processes to carbon dynamics in
Archaea and viruses. In Sea Ice, ed. Thomas, D. N., pp. 370–393. John subarctic sea ice. Polar Biol. 36, 1761–1777.
Wiley & Sons Ltd.
Torda, G., Donelson, J. M., Aranda, M., et al. (2017). Rapid adaptive
Dick, G. J., Anantharaman, K., Baker, B. J., et al. (2013). The microbiol- responses to climate change in corals. Nat. Clim. Change 7: 627–636.
ogy of deep-sea hydrothermal vent plumes: Ecological and biogeographic
linkages to seafloor and water column habitats. Front. Microbiol.4: 124. Van Dover, C. L. (2014). Impacts of anthropogenic disturbances at deep-
sea hydrothermal vent ecosystems: A review. Mar. Environ. Res. 102:
Dziak, R. P., Matsumoto, H., Embley, R. W., et al. (2018). Passive acous- 59–72.
tic records of seafloor methane bubble streams on the Oregon continental
margin. Deep Sea Res. II: Top. Stud. Oceanogr. 150: 210–217. Verde, C., Giordano, D., Bellas, C. M., et al. (2016). Polar marine microor-
ganisms and climate change. Adv. Microb. Physiol. 69: 187–215.
Chapter 2
Methods in Marine
Microbiology

The continuing development and application of new techniques has resulted in a huge recent
expansion of our knowledge of the diversity of marine microbes, their structure and func-
tions, and their roles in ecology and ocean processes. It is essential for the student to appre-
ciate the scope of available methods and their underlying principles, in order to understand
how they are applied in research. A key ambition for this book is to encourage you to read
the primary literature and explore the latest developments. This can be daunting, because
it is easy to be put off by a plethora of technical details, which can sometimes appear quite
impenetrable. This chapter doesn’t aim to describe in detail every possible technique you
might encounter, but instead aims to provide you with an introductory overview that will
help you to understand research described in later chapters and in your own discovery of
scientific papers. The chapter is divided into sections covering: (1) sampling methods; (2)
imaging techniques; (3) cultivation methods; (4) RNA- and DNA-based techniques; and (5)
approaches to studying the collective in situ activities and interactions of microbial commu-
nities in the environment. Today, we have methods that enable us to study microbial ecology
at multiple levels of scale—from the function of single cells, to community activities, to
global processes.

Key Concepts
• Sampling and experimental observation of marine microbes requires special
techniques.
• Epifluorescence microscopy and flow cytometry led to recognition of the abundance
and diversity of marine microbes in the oceans.
• Most marine microorganisms cannot currently be cultivated.
• The most significant recent advances in marine microbiology have occurred as a result
of the development of methods for the study of nucleic acids collected directly from
the environment.
• New sequencing technologies and bioinformatics tools are resulting in massive expan-
sion of knowledge of microbial ecology through metagenomics, metatranscriptomics,
metaproteomics, and metabolomics.
• Progress towards full understanding of marine microbial diversity, physiology, and
ecology requires the integration of field observations and measurements, imaging,
culture-based methods, and molecular biological approaches.
30 Chapter 2

SAMPLING METHODS
Sampling the marine environment
requires special techniques
The overall goal of microbial ecology is to understand how cells, populations, and com-
munities of microbes interact with each other and with their environment. A wide range of
methods in microbial ecology are designed to answer three fundamental questions about
microbes in the environment: “who’s there?” (biodiversity), “what are they doing?” and
“how are they doing it?” (metabolic functions and activities). The importance of scale can-
not be overemphasized: microorganisms carry out their activities in microenvironments.
Experimental approaches and measurement techniques must be designed to cause minimum
disturbance of the microenvironment that they are probing, because microbial communi-
ties may exist in a delicate three-dimensional structured organization, which can be easily
disrupted. At the micrometer scale, physicochemical gradients are very steep. For example,
anaerobic, microaerophilic, and aerobic conditions may occur in a sediment particle or a
microbial mat just a few hundred micrometers thick. Other conditions such as boundary
layer effects, diffusion, turbulence, and flow patterns will affect the microenvironment in
the vicinity of a microbe—a microbe swimming for a few seconds through a patch of sea-
water might encounter huge differences in physical conditions and concentrations of dif-
ferent nutrients. Most important of all, the microbes carry out biochemical reactions that
alter the physical and chemical conditions in their immediate vicinity. Bulk measurements
of substrates, metabolites, pH, or oxygen in marine samples may therefore not reflect the
heterogeneity of the natural environment. Increasingly, we recognize the importance of
gradients in the marine environment.

The traditional method of collecting plankton at sea is to tow a funnel-shaped net behind a
boat. The smallest nylon mesh used is about 60 µm, which is clearly too large to capture most
microbes. However, microbes will be attached to larger phytoplankton and zooplankton, and
some aggregates of free-living microbes, such as clumps of filamentous cyanobacteria or
diatoms, are also large enough to be collected in this way.

Obtaining samples of seawater and sediments is straightforward in shallow coastal waters,

with simple collection from beaches, piers, harbors, or small boats. Samples of sediments and
benthic organisms are obtained using grabs and corers from a boat, or by scuba divers who
can find specific locations for collection. For most microbiological work, samples can be col-
lected in sterile plastic bags or bottles, taking great care to prevent contamination. Sampling
in the open ocean requires the use of research vessels equipped with suitable sampling gear
and onboard laboratory facilities. Research investigations are facilitated through routine
monitoring of oceanographic and atmospheric data and research cruises coordinated through
international collaboration. Well-known examples of established long-term sampling pro-
grams include the Pacific Ocean time series site known as station ALOHA (A Long-term
Oligotrophic Habitat Assessment) off Hawaii, the Bermuda Atlantic Time Series (BATS)
site, and the Western Channel Observatory in the English Channel off Plymouth, UK.

It is important to collect water samples in a form suitable for subsequent analysis together
with appropriate environmental data about the sampling site. Because ocean water is highly
heterogeneous, proper attention must be paid to the replication, frequency, and location of
sampling. For sampling from specific depths within the water column, specialized hydro-
logical bottles are used. Niskin bottles are plastic cylinders that are open at both ends; they
are lowered into the sea on a wire until they reach the required depth; they are then closed
by a weighted trigger that is sent down the cable from the surface or by an electrical signal
or automatic pressure sensor. One of the most widely used systems is a circular rosette sys-
tem of up to 24 Niskin bottles holding 10 or 20 L, attached to a steel frame equipped with
sensors for conductivity, temperature, and depth (CTD), and sometimes other parameters
such as turbidity, dissolved oxygen, fluorescence, and photosynthetically active radiation,
so that the water properties can be monitored in real time at the sampling locations. It is
lowered by a winch from the research vessel and closed selectively at specific depths of up
to 6000 m (Figure 2.1A). Careful attention is needed in the choice of construction materi-
als for frames and sampling containers because many microbial processes are affected
 Methods in Marine Microbiology 31

Figure 2.1 Procedures for sampling

water and sediments. (a) CTD frame
and rosette full of water sampling
containers being recovered from
Antarctic waters. (b) Water filtra-
tion system on board RRS James
Cook. Samples of up to 200 L are
sequentially passed through fil-
ters of different pore sizes to trap
different fractions of the microbial
plankton. (c) Biomass collected by
filtration of Arctic seawater. (d) Box
corer (50 × 50 × 50 cm) for sam-
pling of sediments, being deployed
in the Southern Ocean from RV
Polarstern. Credits: (a)-(c) Jozef I.
Nissimov, Scottish Association of
Marine Science, Oban. (d) Haanes
Grobe, Alfred Wegener Institute, via
Wikimedia Commons, CC-BY-2.5.

by the presence of trace metals or organic materials in hoses and stoppers; specialized
systems are needed for some applications. For some investigations in biologically produc-
tive waters, it may be necessary to collect only a few liters of water (Figures 2.1B, 2.1C),
whereas in low productivity waters, hundreds of liters may be required. Samples are sieved
through a mesh to remove large aggregates and micro- or mesoplankton. Microbes and sus-
pended particles are usually obtained by filtration though a series of filters of varying pore
sizes depending on the size fractions of interest, from glass fiber prefilters of ~30 µm down
to 0.1 µm polycarbonate filters or 0.02 anodisc filters for viruses. Such selective filtration
is frequently used to denote divisions between particle-associated and free-living microbes
or to discriminate plankton size classes (pico-, nano-, microplankton, see Table 1.2). For
concentration of small microorganisms and viruses, tangential flow filtration is commonly
used, which allows for the filtration of thousands of liters without clogging of membranes.
Viruses can also be concentrated by flocculation with iron salts followed by large-pore-size
filtration.

Several methods can be used for sampling the sea surface microlayer (SML, p.14). A common
approach is to dip a glass plate into the water or to float a hydrophobic membrane filter or fine
nylon mesh screen on the surface of the water.

For collecting sediment samples, various types of grabs and corers are used. Small samples
of coastal mud can be collected simply with a cutoff syringe, but special designs are needed
for operation in deep waters and for collecting sands or loose sediments. Tubular corers
consist of weighted sections of steel tube that bore into the sediment and are sealed with
caps to keep the sample intact. Specialized drilling rigs allow the recovery of deep sediment
cores from hundreds of meters below the seabed (Figure 1.10). Box corers are designed to
collect undisturbed sediment samples, typically ~0.1–0.2 m3. A weighted box is lowered into
the sediment and spring-loaded flaps close the box to prevent disturbance during recovery
(Figure 2.1D).
32 Chapter 2

Investigations in deep water are increasingly carried out using remotely operated vehi-
CLOCKWORK
i SAMPLER STANDS
THE TEST OF TIME
cles (ROVs), which are usually equipped with cameras, lights, and robotic sampling arms
and linked by an umbilical cable to a surface vessel for control by a pilot (Figure 2.2A).
Improvements in battery technology and electronic systems are leading to the development
A device called the Continuous of fully autonomous underwater vehicles (AUVs, gliders, or rovers, Figure 2.2B) that can
Plankton Recorder (CPR) invented travel for several months along preplanned routes either in the water column or on the seabed.
by Alister Hardy in 1931 is designed These can be equipped with a range of measuring instruments to record environmental data
to be towed by ships at a depth and devices to collect samples of water, sediment, or benthic animals.
of 5–10 m. Its robust casing and
mechanical structure mean that it One example especially suited to plankton microbiology is the Environmental Sample
is highly reliable and can be towed
Processor, which uses an electromechanical fluidic system to collect water samples that
from merchant ships on regular
are filtered, preserved, and stored. The Environmental Sample Processor is carried aboard
routes. The CPR filters plankton
from the water over long distances the long-range AUV developed by Monterey Bay Aquarium Research Institute (MBARI),
on continuously moving bands of designed to take multiple samples over long periods from eddies of the ocean gyres that are
fine silk, which is wound through otherwise hard to investigate. Some processes such as automated microarrays or polymerase
the CPR on rollers turned by chain reaction (PCR; see below) can be carried out on board. The AUV Clio developed by
gears, powered by a propeller. On Woods Hole Oceanographic Institute (WHOI) can dive to a depth of 6000 m, collecting mul-
return to the laboratory, the silk is tiple water samples which are immediately filtered and sampled before the vessel returns to
removed from the mechanism and its mother ship. Free-falling lander systems may be deposited on the seabed; some of these
divided into samples represent- have mini-laboratories that collect samples for in situ molecular analysis and analysis of
ing ten nautical miles of towing microbial activity. Purpose-specific equipment has also been used in the vicinity of hydro-
and the plankton trapped on
thermal vents.
the silk is identified and enumer-
ated by microscopy. Since 1931,
Robotics and artificial intelligence systems are increasingly used in oceanographic research,
recorders have been towed over
5 million miles, producing more but there is no replacement for a human presence to explore the deep sea. The most famous
than a quarter-million analyses, of the human-operated submersible vehicles is DSV Alvin, owned by the US Navy and oper-
with the CPR survey producing one ated by WHOI (Figure 2.2C, D). Alvin has been rebuilt numerous times and has made >4800
of the longest archived biologi- deep-sea dives since its first launch in 1964. A typical dive can take two scientists and a pilot
cal datasets. Methods to extract down to 4500 m, taking about two hours for Alvin to reach the seafloor and another two to
and amplify DNA from preserved return to the surface, with four hours of carefully planned recording and sampling work on the
samples mean that retrospective seafloor. Viewing ports and video cameras allow direct observation so that specific samples
analyses of bacteria (Box 12.1) and can be taken. Several other agencies operate research programs with human-operated sub-
viruses associated with host cells mersibles capable of diving to greater depths; these include China National Deep Sea Center
or attached to particles have been
(Jialolong, 7000 m), Japan Agency for Marine-Earth Science & Technology (Shinkai, 6500
possible.
m), Shirshov Institute of Oceanology (Russia, MIR-1 and MIR-2, 6000 m), and IFREMER
(France, Nautile, 6000 m). Special technology is needed to collect samples from abyssal
depths at great pressure, and from high-temperature environments near hydrothermal vents.

IMAGING TECHNIQUES
Light and electron microscopy are used to study
morphology and structure of microbes
The study of microbes began in the mid-17th century with Antonie van Leeuwenhoek’s
observations of “animalcules” using a simple handheld microscope and development of the
light microscope by Robert Hooke. Microscopy remains an important method for the initial
examination of plankton samples and cultured microbes. Eukaryotic marine microbes like
diatoms, dinoflagellates, ciliates, and fungi are large enough for microscopy to be useful in
distinguishing morphological and structural features—in these groups, microscopic appear-
ance is usually the main criterion for classification (for examples, see figures in Chapter 6).
Direct light microscopy is also used for enumeration of eukaryotic plankton in seawater
samples. A device called the FlowCam is used for rapid, high-throughput analysis of plank-
ton samples; it uses a microsyringe to draw a sample through a cuvette, and a high-resolution
camera photographs each cell as it flows, recording size, shape, and concentration of plankton
samples. However, with bacteria and archaea, light microscopy reveals little more than the
general shape and morphology, although the use of special stains and illumination techniques
can improve the amount of information revealed. The wavelength of visible light limits the
effective magnification of the light microscope to ~1000–1500 times and it is not possible
to resolve objects or structures smaller than ~200 nm. Nevertheless, high-resolution phase-
contrast light microscopy linked to digital recording equipment provides the best method
 Methods in Marine Microbiology 33

Figure 2.2 Recovery of the AUV

AUTOSUB-3 following a seabed map-
ping mission in 2000 m deep water.
B. Recovery of the ROV ISIS following
a seabed sampling mission on 1500
m deep seamounts for collection of
targeted benthic invertebrates for
genetic and microbiological analysis.
C. DSV Alvin being deployed from
the RVV Atlantis. D. Alvin begin-
ning its descent. Credit: A, B:
Alex Nimmo-Smith, University of
Plymouth, NERC DEEPLINKS project.
C. Mountains in the Sea Research
Team; the IFE Crew; and NOAA/OAR/
OER. D. Gavin Eppard, NOAA Photo
Library, via Flickr CC-BY-2.0.

for observing microbial behavior, such as bacterial movement (Figure 3.14) and predation of
bacteria by protists.
“WE KNOW MORE
The development of the electron microscope enabled the study of the ultrastructure of cells i ABOUT THE SURFACE
OF THE MOON…”
and viruses. In the transmission electron microscope (TEM), a beam of electrons is focused
onto an ultrathin section of the specimen in a vacuum, usually after staining with lead or ura- While 12 humans have walked
nium salts. Electrons are scattered as they pass through the specimen according to different on the Moon, only four have
densities of material in the cell. Because the wavelength of an electron beam is much smaller been to the deepest part of the
than that of light, the TEM has an effective magnification up to 1 million times and objects ocean. In 1960, Jaques Piccard
as small as 0.5 nm can be resolved (e.g. Figure 4.4). Various techniques such as shadowing, and Don Walsh descended to the
Challenger Deep at the bottom
freeze-etching, and negative staining are used to visualize the membranes, internal struc-
of the Marianas Trench almost 11
tures, and surface appendages of cells. However, it must always be borne in mind that TEM
km deep in the Pacific Ocean in a
images are only obtained after staining and fixing the cells and observing them in a vacuum. bathyscape (Trieste) designed to
Therefore, the appearance of structures in sections may not reflect their actual organization in withstand the enormous pressures.
the living cell. Refinement of techniques for preparing ultrathin sections allows visualization The mission took 8 hours, and the
of the 3D ultrastructure of microbial cells (e.g. Figure 6.6). The recent development of cryo- men spent only 20 minutes on
electron tomography involves tilting the specimen in an electron beam to create a series of 2D the seabed. Trieste was expensive
projection images, which are then used to reconstruct a 3D image of the object. This reveals to maintain and had almost no
a life-like, frozen hydrated state with sufficient resolution to visualize the macromolecular scientific value, so further mis-
organization of intact cells (e.g. Figure 1.3A). In scanning electron microscopy (SEM), a very sions never took place and the
fine electron beam scans the surface of the object, generating a 3D image. Therefore, SEM is excitement of space exploration
attracted greater public interest
particularly useful for studying the structure of cell surfaces (e.g. Figure 6.7).
and geopolitical impetus than
exploring the ocean floor. In 2012
Other exciting modern developments of electron microscopy include atomic force micros- and 2019 respectively, film direc-
copy (AFM), in which a tiny probe is held in place very close to the surface of an object using tor James Cameron and investor/
weak atomic repulsion forces. Effective magnifications up to 100 million are possible and the explorer Victor Vescovo made
atomic structure of molecules such as DNA or proteins can be visualized. A major advantage privately funded solo trips to the
of this technique is that specimens do not need to be fixed or stained and can be examined in Deep. While the statement that
the living state. This technique has been used to reveal complex surface architectures at the we know more about the surface
nanometer to micrometer scale, including structures which connect planktonic microbes in of the Moon than that of our own
microscale networks (Figure 2.3). oceans has become a cliché and
is hard to justify objectively, it
does highlight the difference in
Epifluorescence light microscopy enables public expenditure. For example,
in 2013, the USA spent $23.7
enumeration of marine microbes million on NOAA marine explora-
Fluorescence occurs when material absorbs light at one wavelength (the excitation or tion research, but $3.8 billion on
absorption spectrum) and then re-emits it at a different wavelength (the emission spec- NASA space exploration research
(Conathan, 2013).
trum). The specimen is illuminated with a tungsten-halogen or mercury vapor lamp. In
34 Chapter 2

Figure 2.3 Atomic force micros-

copy. (a) Interactions between
pelagic bacteria; a heterotrophic bac-
terium (red) is closely associated with
a larger Synechococcus cell (yellow).
(b) Bacteria surrounded by an appar-
ent network of gel and nanometer-
sized particles. Scale bars indicates
cell height. Credit: Francesca Malfatti
and Farooq Azam, Scripps Institution
of Oceanography (see Malfatti &
Azam, 2009).

FLUORESCENCE
i MICROSCOPY AND
THE “GREAT PLATE
COUNT ANOMALY”
The development in 1977 of epi-
fluorescence microscopy for exami-
nation of seawater by John Hobbie
and colleagues at the Woods Hole direct fluorescence microscopy, a filter is placed between the light source and the speci-
Marine Biological Laboratory led men, allowing only light of the desired excitation wavelength to be transmitted, while a
to a major upheaval in microbial barrier filter placed between the specimen and the eyepiece transmits the emitted fluores-
oceanography. Routine measure- cence and absorbs longer wavelengths. Epifluorescence microscopy depends on the use of
ments became possible because dichroic mirrors as interference filters that transmit one set of wavelengths and reflect the
of improvements in the type of others. Water samples are usually fixed with paraformaldehyde or glutaraldehyde and fil-
filter used, revealing that bacterial
tered through membrane filters of a pore size appropriate to the microbial group under
numbers in plankton were a thou-
study—0.22 µm filters are most commonly used for bacteria. For enumeration of viruses,
sand or more times higher than
previously realized. This prompted larger particles are removed by pre-filtration before trapping the viruses on 0.02 µm alumi-
a major reappraisal of their role in num oxide filters. This method can therefore be used for observation and enumeration of
marine processes. The term “great all groups of marine microbes. The original stain (fluorochrome) used in plankton studies
plate count anomaly”—first used was acridine orange (AO, 3,6-bis[dimethylamino]acridinium chloride), which binds to DNA
by Staley and Konopka (1985)— and RNA. Problems arise with AO owing to background fluorescence and difficulties in
became embedded in microbial distinguishing microbes from inanimate particles, so the most widely used stain today is
ecology as a concept to explain the DAPI (4′,6′-diamidino-2-phenylindole), which binds to DNA and largely overcomes these
large difference between the total problems. DAPI is excited by ultraviolet light and emits bright blue light. The use of inci-
number of bacteria revealed in dent light also permits observation of microbes attached to suspended particles. Although
direct cell counts with those recov-
viruses are below the limits of resolution of the light microscope, they may be visualized
ered by culture techniques such as
viable counts on agar plates.
sufficiently for enumeration if stained with SYBR Green, which emits a very bright fluores-
cence (Figure 7.1). This method only works for DNA viruses.
 Methods in Marine Microbiology 35

Some staining methods can indicate the physiological state of individual cells within
THE “JELLYFISH GENE”
a population, especially to distinguish between living and dead cells based on differ-
ences in membrane permeability. CTC (5-cyano-2,3-dilotyl tetrazolium chloride) is a i IS A MAJOR TOOL IN
MOLECULAR BIOLOGY
fluorogenic redox dye that detects an active electron transport chain. Although there
have been some criticisms of the technique, the CTC assay has been used to estimate the Green fluorescent protein (GFP) is
distribution of cell-specific metabolic activity in natural assemblages of marine bacteria. produced in the jellyfish Aequorea
The Live–Dead kit is a widely used test for bacterial viability; a mixture of SYTO 9 and victoria and has very wide appli-
propidium iodide stains bacteria fluorescent green if they have intact cell membranes, cations in all areas of biology. Its
whereas bacteria with damaged membranes appear fluorescent red. This is widely used discovery and application by Tsien,
Chalfie, and Shimomura led to the
to assess the viability of bacterial populations following environmental changes or chem-
award of the Nobel Prize in 2008.
ical treatments. The gfp gene can be easily inserted
into the genome of many organ-
isms and used as a marker for the
Confocal laser scanning microscopy enables expression of specific genes, with
recognition of living microbes within their habitat many applications in cell biology
and genetic modification. Some
Confocal microscopy uses a laser light source coupled to an optical microscope and
examples of its use in marine
computer-aided digital imaging system. The beam is directed through a scanner and then microbiology include following
through an aperture that adjusts the focal plane of the beam, meaning that it is possible to the fate of gfp-marked bacteria
zoom in on particular vertical layers in the specimen. The advantage of confocal micros- after their ingestion by grazing
copy lies in its ability to be used on living specimens and its generation of a 3D image protists or the process of infec-
via the assembly of digital outputs from each layer. Because the magnification is much tion of animal hosts by symbiotic
less than that of electron microscopy, it is not as useful for revealing ultrastructural detail. or pathogenic bacteria. If specific
However, it is proving enormously useful in ecological studies of microbial communities, genes are tagged, it is possible to
especially when combined with fluorescent in situ hybridization (FISH) techniques (p.36, visualize when they are “switched
below), which enables specific microbes to be located on surfaces or within other organisms. on” during the development of a
microbial process.
In particular, understanding of the structure of biofilms and microbial mats has advanced
considerably using this technique.

Flow cytometry measures the number and size of particles

Flow cytometry was originally developed for biomedical uses and was first applied to marine
studies in the late 1970s. Since then, use of this technique has produced spectacular advances in
the enumeration and characterization of ocean microbes. A flow cytometer is an instrument that
can be used to identify cells according to a specific fluorescence signature, leading to a wide
range of physiological and ecological information. Flow cytometry simultaneously measures
and analyzes multiple physical characteristics of single particles, ranging in size from 0.2 to 150
µm, as they flow in a fluid stream through a beam of light. The coupling of the optical detection
system to an electronic analyzer records how the particle scatters incident laser light and emits
fluorescence. A flow cytometer is made up of three main components (Figure 2.4A).

The fluidics system injects particles into the instrument in a stream of fluid so that they
pass in single file through a laser beam. The optics system consists of lasers—there may be
four or more different wavelengths—to illuminate the particles in the sample stream and
optical filters to direct the resulting light signals to the appropriate detectors. Each particle
scatters light at different angles and may also emit fluorescence. The scattered and emitted
light signals are converted to electronic pulses that can be processed by the electronics sys-
tem. In a fluorescent-activated cell sorter (FACS), the electronics system can also transfer
a charge and deflect particles with certain characteristics, so that individual cells or differ-
ent cell populations can be separated and collected. Flow cytometry is used for a range of
specific measurements in marine microbiology, including quantifying different microbial
groups and their diversity, estimation of cell dimensions and volume, determination of DNA
content, and assessment of viability and growth rates. In marine studies, flow cytometry
was first used for the analysis of phytoplankton cells, as they are naturally labeled with
photosynthetic pigments such as chlorophyll and phycobilins, which are autofluorescent.
However, most microbes can be detected in flow cytometry by using the same principles
described for epifluorescence microscopy—labeling cells with fluorochromes that emit
light of various wavelengths when illuminated by the appropriate laser. Flow cytometry
has thus been particularly valuable in the analysis of bacteria and small eukaryotes in the
picoplankton, especially following the development of portable instruments with stable
36 Chapter 2

optical and electronic systems that can be deployed on ships for direct examination of water
samples. Improvements in the sensitivity of flow cytometry instruments and the introduc-
tion of new dyes mean that flow cytometry can now be used for detection and quantification
of marine viruses (see p.200 for discussion of precautions needed in the interpretation of
such counts). Table 2.1 lists examples of fluorochromes used in flow cytometry and epifluo-
rescence microscopy. Using an appropriate mixture of different fluorochromes, lasers, and
light of different wavelengths, different populations of microbes in aquatic ecosystems can
be analyzed based on size, abundance, and specific properties. Exciting new advances in
flow cytometry have been possible owing to a fusion with FISH technology (Figure 2.5),
in which fluorescently labeled oligonucleotides can be used to discriminate specific taxa in
heterogeneous natural marine microbial communities.
Figure 2.4 Flow cytometry. (a)
Schematic diagram of the compo-
nents of a flow cytometer. Particles Fluorescent in situ hybridization (FISH) allows
pass single file through a narrow visualization and quantification of specific microbes
channel and into the laser beam,
The acronym FISH is no accident. This technique permits an investigator to “fish” for
producing a single wavelength
of light at a specific frequency to
a specific nucleic acid sequence in a “pool” of unrelated sequences. It is widely used
activate the fluorescence signal. in biomedical imaging and its use in marine microbiology was pioneered by Rudolf
Detectors measure forward and Amann and Bernard Fuchs of the Max Plank Institute (MPI) for Marine Microbiology,
side scattering of light from fluo- Bremen. It has proved to be one of the most useful culture-independent techniques for
rochrome-stained cells. Different identification of particular organisms or groups of organisms in the marine environment.
light signals are split into different Another major advantage is the ability to quantify different members of a microbial com-
wavelengths by a series of filters munity and their dynamics because oligonucleotide probes can be designed to recognize
and mirrors for detection by specific
photomultiplying tubes. Signals are
converted by the electronic system
for visualization and interpretation.
(b), (c). Examples of typical flow
cytometry plots obtained during
a coastal mesocosm experiment
conducted at the Espegrend Marine
Biological Station, Bergen, Norway
in 2017. (b) The 585 nm (phyco-
erythrin) fluorescence vs. 692 nm
(chlorophyll) fluorescence of par-
ticles excited with a 488 nm laser.
Distinct phytoplankton groups can
be distinguished based on the types
and amounts of pigments present
within cells. (c) The nanophytoplank-
ton group is further distinguished by
side scatter (SSC) vs. forward scatter
(FSC). Elevated SSC is indicative of
cell surface roughness, granularity,
or light-scattering properties, which
can be caused by the presence of
calcifying cell wall features (e.g.
coccoliths on calcifying coccolitho-
phores). FSC is a proxy for cell size.
Using cultures of known calcified
coccolithophores a threshold gate
can be set to delineate calcifying and
non-calcifying nanophytoplankton.
Here, calcified nanophytoplankton
were enumerated by setting a gate
using calcified Emiliania huxleyi strain
DHB607 (Johns et al., 2019). Credits:
(a) Boster Biological Technology,
Pleasanton, Ca.; (b), (c) Brittany
Schieler and Kay Bidle, Rutgers
University, NJ.
 Methods in Marine Microbiology 37

Table 2.1 Some representative fluorochromes used in epifluorescence

DISCOVERY OF
microscopy and flow cytometry

Fluorochrome Excitation/emission (nm) Target

i THE PLANET’S
MOST ABUNDANT
PHOTOSYNTHETIC
Acridine orange 500/526; 460/650 DNA, RNA ORGANISM
DAPI 358/461 DNA In 1985, Sallie Chisholm and col-
leagues from the Massachusetts
Ethidium bromide 518/605 Double-stranded DNA, RNA
Institute of Technology (MIT)
FITC 495/520 General fluorescent and Woods Hole Oceanographic
Institution (WHOI) were among
Hoechst 33342 350/461 AT-rich DNA the first scientists to use a portable
flow cytometer on a research
Mithramycin 425/550 GC-rich DNA
vessel. They were studying the
Propidium iodide 535/617 Double-stranded DNA, RNA cyanobacterium Synechococcus
when they noticed unusual red
Rhodamine 123 480/540 Membrane potential fluorescence signals from very tiny
SYBR green 494/521 DNA, can be used for viruses cells present in huge numbers.
Chisholm et al. (1988) recognized
SYTOX Green 504/523 Cell viability that these cells represented a novel
phytoplankter, which possesses
AT, adenine and thymine; DAPI, 4′,6′-diamidino-2-phenylindole; FITC, fluorescein isothiocyanate; an unusual divinyl chlorophyll
GC, guanine and cytosine. responsible for the unexpected
fluorescence signal. It would
and hybridize to complementary sequences that are unique to specific microbial groups later be cultured and named as
(Figure 2.8A). The oligonucleotide probe is covalently linked to a fluorescent dye such Prochlorococcus, now considered
as fluorescein, Texas Red, or indocarbocyanines. Hybridization of the probe with the to be the commonest photosyn-
target sequence within a morphologically intact microbial cell can then be visualized thetic organism on Earth and
by fluorescence or confocal microscopy. The signal that is produced after hybridization responsible for production of at
least 20% of atmospheric oxygen
and washing away the excess unbound probe is proportional to the amount of target
(Partensky and Garczarek, 2010;
nucleic acid present in the sample. This allows an investigator to visualize and quantify
see p.135). It is remarkable that
the microbes in marine samples and to determine cell morphology and spatial distribu- such an abundant and important
tions of specific taxa in situ. FISH has been particularly useful for examining microbes organism escaped detection for so
attached to particles and spatial localization of microbes in biofilms and sediments, in long. Pennisi (2017) describes the
symbiotic associations, and in infected tissues. Probes for different taxa can be linked to story of Chisholm’s fascination with
alternative fluorochromes and used together, enabling two or more dissimilar organisms Prochlorococcus, which has led to
to be located within a sample. Examples of micrographs using this technique can be seen many major research discoveries.
in Figures 3.13 and 9.6; these show the importance of revealing associations between
different microbes in significant biogeochemical processes like anoxic methane oxida-
tion and nitrogen fixation. It can also be used to assess diversity of planktonic microbes
collected on filters. Figure 5.3 shows an important example, in which the relative abun-
dance of archaea and bacteria in the water column was assessed by FISH microscopy.
Small subunit (SSU) rRNA genes are most commonly used for diversity studies. Cells
usually contain thousands of copies of rRNA molecules, which are relatively stable.
Figure 2.5 Fluorescent in situ
However, many marine microbes have very small cells and may be slow growing or
hybridization (FISH). A. Principles
of the basic method. The sample is
fixed to stabilize cells and make the
membrane permeable to the oligo-
nucleotide probe, which attaches to
its intracellular targets. After wash-
ing, single cells can be identified by
epifluorescence microscopy or flow
cytometry. B. Principles of catalyzed
reporter deposition in CARD-FISH.
Hybridization involves a single oligo-
nucleotide that is covalently linked
to horseradish peroxidase (HRP). The
signal is amplified because multiple
tyramide molecules are radicalized
by a single HRP enzyme. Reprinted
from Amann and Fuchs (2008) with
permission of Springer Nature.
38 Chapter 2

dormant and this can result in variable results because they may have low numbers of
ribosomes or have membrane modifications that lower the accessibility of the probe. This
places some limits on the uses of FISH in investigation of oligotrophic ocean environ-
ments, but modifications have improved the efficiency of hybridization; multiple probes,
different sequences, and different fluorochromes can be applied to single samples to
improve sensitivity and to assess genetic diversity. Enzyme-linked methods such as cata-
lyzed reporter deposition (CARD) FISH (Figure 2.8B) can also provide greater signal
intensity. Here, hybridization involves an oligonucleotide that has been covalently cross-
linked to the enzyme horseradish peroxidase (HRP). Because the HRP molecule is much
larger than the fluorochromes, permeabilization requires harsher enzymatic or chemical
treatments, conducted in agarose gels to avoid loss of cell shape. Fluorescently labeled
tyramide (a derivative of the amino acid tyrosine) is then added, and multiple molecules
are activated by a single HRP enzyme which binds permanently to the cells, resulting in
amplification to give a strong fluorescent signal. Besides detection of rRNA, the FISH
technique can be extended to search for a wide range of specific gene sequences charac-
teristic of particular structures or metabolic functions, assuming that sufficient copies of
the targeted nucleic acid are present in the sample.

CULTIVATION OF MICROORGANISMS
Different microorganisms require specific
culture media and conditions for growth
As noted above, there is a large discrepancy between the number of microbes that can be
observed in direct counts of marine samples and those that can be cultured in the labora-
tory. Even those organisms that can be successfully cultured differ enormously in their
physiological properties (see Chapter 3), so the composition of culture media and environ-
mental conditions (temperature, pH, atmospheric conditions, and pressure) must be chosen
with great care.

Many autotrophic bacteria and archaea can be grown in a simple defined synthetic medium
containing bicarbonate, nitrate or ammonium salts, sulfates, phosphates, and trace metals.
Appropriate light conditions for photolithotrophs, and the requisite substrates as energy
sources and electron donors for chemolithotrophs must be provided. Chemoorganotrophic
(heterotrophic) bacteria, archaea, protists, and fungi can be grown on a similarly defined
medium with the addition of appropriate organic substrates such as sugars, organic acids,
or amino acids—organisms differ greatly in their preference for particular nutrients. Initial
isolation and routine culture are usually carried out using complex media, made from semi-
defined ingredients such as peptone (a proteolytic digest of meat or vegetable protein, which
provides amino acids) and yeast extract (which provides amino acids, purines, pyrimidines,
and vitamins). Many heterotrophs are more fastidious and require the addition of specific
growth substances. An example of a widely used medium for routine culture of many marine
bacteria is Zobell’s 2216E medium, which contains low concentrations of peptone and yeast
extract plus a mixture of various mineral salts at concentrations similar to those found in
seawater. It must be stressed that there is no single medium suitable for all microbial types,
and microbiologists employ a very wide range of recipes for different purposes. Despite this,
only a tiny proportion of the bacteria and archaea identified in gene surveys have been cul-
tured, but there is a renewed interest in finding appropriate media and incubation conditions
to rectify this (Box 2.1).

Microorganisms may be grown in liquid media (broth) or on media solidified with agar.
Cultivation on agar plates (Petri dishes) permits samples to be streaked out to obtain single
colonies, which can then be re-streaked to form pure cultures. In the shake tube and roll tube
techniques, samples are mixed with molten agar, which allows culture of some anaerobic
bacteria. Agar is a polysaccharide obtained from red algae such as Gracilaria and it has
several advantages as a gelling agent for microbiological work. It is colorless, transparent,
and remains solid at normal incubation temperatures, but can be held in a molten state above
42°C. Its main advantage in microbiology is that it does not normally affect the nutritional
status of the medium because most bacteria and fungi do not possess extracellular enzymes
for its degradation. However, not surprisingly, many marine microorganisms can degrade
 Methods in Marine Microbiology 39

this algal compound by production of agarase enzymes. When marine samples are plated
onto agar, craters are frequently observed surrounding the colonies as a result of enzymatic
digestion of the agar.

The concentration of cultivable microorganisms in a sample can be determined by plating

appropriate tenfold dilutions on agar plates, either by mixing with molten agar (pour plate),
spreading evenly over the surface of the agar (spread plate), or by pipetting small drops onto
the surface (drop count or Miles–Misra technique). Colonies are counted after incubation
at an appropriate temperature and the viable count is expressed as colony-forming units per
milliliter (CFU mL−1). When plating samples from natural sources, it is important to check
plates at regular intervals, as slow-growing colonies may take days or weeks to develop and
may be obscured by rapidly growing types. Even if the right culture conditions can be found,
some marine bacteria and archaea grow very slowly. To produce a visible colony on agar after
12–24 h incubation, bacteria must divide about once per hour, whereas bacteria with a dou-
bling time of 48 h would take several weeks to produce a visible colony, assuming they are
capable of growth under the specific conditions. Organisms in low abundance in the natural
environment are often out-competed by those which grow more rapidly or produce inhibi-
tory substances. Selective media containing antibiotics or other chemicals that inhibit the
growth of certain microorganisms can be used, thus allowing only selected groups of interest
to multiply. They may also include dyes or chromogenic substances that give a differential
reaction when metabolized by bacteria belonging to particular groups. These media are used
as an enrichment method in liquid culture or to select colonies of the desired type on agar
plates. In marine microbiology, selective media are used mainly for the isolation and growth
of animal and human pathogens and allochthonous indicators of pollution. Selective media
frequently contain bile or synthetic detergents, which select for enteric bacteria and related
groups by their effects on membrane function. For example, the medium thiosulfate-citrate-
bile-sucrose (TCBS) agar is widely used for isolation of vibrios, while lauryl sulfate broth
is used in the enumeration of coliform bacteria in assessing fecal pollution of seawater or i “BUT THEY ALL
LOOK THE SAME!”
shellfish (Figures 13.6 and 13.9). Selective antibiotics are often added to suppress the growth Many culture-based investigations
of bacteria on fungal isolation plates. in marine microbiology begin with
the collection of seawater samples,
As well as plating, successive dilution in liquid media can be used to separate individual sediments, material scraped from
bacteria to obtain pure cultures. This method has been used successfully to grow previ- surfaces, or extracts of animal or
ously nonculturable bacteria in seawater without the addition of nutrients (see Box 2.1). The seaweed tissues. These are plated
dilution-to-extinction method is also used in the most probable number (MPN) technique for on agar media and, after incuba-
tion, colonies are examined, and
quantification of microbial populations. One widely used application of this is in the deter-
different types can be distin-
mination of coliforms as indicators of fecal pollution in seawater or shellfish (Figure 13.9).
guished by noting the colony size,
color, form, elevation, texture, and
Pure culture methods for the cultivation of protists and fungi follow the same basic principles margin. Different types are then
described above. Many photosynthetic microalgae can be maintained in culture, requiring picked off, streaked onto fresh
mineral salts solution and appropriate conditions of temperature and light/dark cycles, and a plates and re-incubated to obtain
method of keeping cells in suspension through shaking or aerating. Large-scale cultures may pure colonies for further character-
be maintained with bioreactors (Figure 14.1). However, algal cultures can be very difficult ization. While some isolates have
to maintain and there are numerous recipes for special media for growing and maintaining bright colors or other distinctive
different species. Culture media can be made of defined chemicals or from natural seawater features (Figure 2.6), many marine
enriched with extracts from various sources. It is relatively easy to isolate single microalgae by bacteria have very similar-looking
colonies. Students starting an isola-
streaking or making serial dilutions to obtain a clonal population, or individual cells of larger
tion project are often frustrated
species can be isolated using a fine pipette. A recent advance in micromanipulation allows
that so many colonies appear as
the isolation of individual cells, which can be trapped using a highly focused laser beam off-white, circular, with entire mar-
which provides a force, depending on the relative refractive index between the cell and the gins. Patience, careful observation
surrounding medium. The cell can be separated off for subsequent pure culture. Such “optical of slight differences, and methodi-
tweezers” provide a very powerful technique when used in conjunction with gene probes or cal recording after different periods
antibody staining. Phagotrophic flagellates, dinoflagellates, and other heterotrophic protists of time are all essential. The chance
are much more difficult to culture, and established cultures of only a few types are avail- of obtaining a good representa-
able. Because the nutritional requirements can vary so widely, considerable experimentation is tive collection of isolates is further
needed to optimize conditions for isolation and maintenance of different types. For many pro- enhanced by using different culture
media and conditions, but the
tists, especially diatoms and dinoflagellates, it is very difficult to obtain pure axenic cultures
results will always be only a partial
because bacteria and fungi often occur in close association with their cells. For loose associa-
reflection of the true diversity in a
tions, axenic cultures can sometimes be obtained by extensive rinsing of the isolated protists, natural microbial community.
or treatment with antibiotics, but some bacteria or fungi are tightly associated (for example,
40 Chapter 2

within the frustules of diatoms) or occur as intracellular symbionts or parasites. Such extracel-
lular and intracellular associations are often symbiotic and necessary for successful growth of
the algae, but they add major complications to the analysis of the genomes of protists.

Enrichment culture selects for microbes

with specific growth requirements
Enrichment culture depends on the provision of particular nutrients and incubation conditions
to select for a certain group. For example, the addition of cellulose or chitin plus ammonium
salts to a marine sample under aerobic conditions will enrich for Cytophaga-like bacteria,
while sulfates with acetic or propionic acid under strictly anaerobic conditions will favor
growth of Desulfovibrio and related types. The Winogradsky column is especially useful in
investigating communities in marine sediments. A tall glass tube is filled with mud and sea-
water and various substrates added prior to incubation under appropriate conditions. A gradi-
ent of anaerobic to aerobic conditions will develop across the sediment and a succession of
microbial types will be enriched. In the light, microalgae and cyanobacteria will grow in the
upper parts of the column and will generate aerobic conditions by the production of oxygen.
Depending on the substrate added, the activity of various groups will lead to the production
of organic acids, alcohols, and hydrogen, which favor the growth of sulfate-reducing bacteria.
In turn, the sulfide produced leads to the growth of anaerobic phototrophs that use hydrogen
sulfide as an electron acceptor. A wide range of enrichment conditions may be adapted for
Figure 2.6 Iridescent colonies of specific groups, and the resulting communities may then be subject to further enrichment or
Cellulophaga lytica isolated from the selective plating to obtain pure cultures. The results of enrichment cultures must be inter-
anemone Actinia equine on marine preted carefully because they often lead to overrepresentation of groups that may not be the
agar at 25°C. A. The initial isolation dominant members of the natural community.
plate shows colored C. lytica colo-
nies together with agarolytic and
white bacterial colonies. B. A pure Phenotypic testing is used for characterization
culture of C. lytica observed under of many cultured bacteria
direct epi-illumination shows the
After isolation, the initial characterization of bacterial cultures usually begins with prepara-
intense structural “glitter-like” or
tion of a Gram stain and observation of cell morphology and grouping under the microscope,
“pointillist” iridescence. Subsequent
studies revealed that the iridescence
but this does not provide sufficient information for identification and taxonomy. Cultivated
is caused by self-organization of bacteria can be identified and characterized using a combination of growth characteristics
the bacterial cells within the colony and tests for the production of particular enzyme activities, particularly those involved in
biofilm, creating hexagonal photonic carbohydrate and amino acid metabolism. Table 2.2 shows a summary of the methods used
crystals (Kientz et al., 2013, 2016). for bacterial identification and classification. Once it is certain that a pure culture has been
Credit: courtesy of Eric Rosenfeld, obtained, it is usually easier to use 16S rRNA gene sequencing (p.43) to place a new isolate
Université de La Rochelle. into a particular phylogenetic group, before deciding which phenotypic tests should be per-
formed to provide more detailed information. Of course, phenotypic characterization based
on biochemical test methods is suitable only for the small proportion of heterotrophic marine
bacteria that can be easily cultured. Diagnostic tables and keys aid in the identification of
unknown bacteria, although the inherent strain-to-strain variation in individual charac-
teristics often causes problems for accurate identification and taxonomy. Commercial kits
employing a battery of tests (e.g. API®, Micro-Bact®) offer advantages for standardization of
methods and processing of results with the aid of databases developed using numerical taxo-
nomic principles. However, most biochemical methods were developed for the identification
of bacteria of medical importance and require adaptation for examination of marine envi-
ronmental samples. These methods are most useful in areas such as identification of patho-
gens or fish spoilage organisms. For example, there are extensive databases and established
diagnostic keys based on biochemical tests for the identification of Vibrio and related patho-
gens in aquaculture. Some tests are very characteristic of particular groups. For example, the
detection of the enzymes β-galactosidase and β-glucuronidase using fluorogenic substrates
is important in the identification of coliforms and E. coli in polluted waters (Figure 13.4).
The BIOLOG® system differentiates microbial taxa based on their nutritional requirements
and incorporates up to 95 different substrates as a sole source of carbon or nitrogen, together
with a tetrazolium salt. Different microbial groups use a range of substrates at characteristic
levels and rates. Positive reactions are tested via a color reaction based on reduction of the
tetrazolium salt. BIOLOG systems have been developed specifically for Gram-positive bac-
teria, Gram-negative bacteria, and yeasts. Again, this method is not suitable for detection of
 Methods in Marine Microbiology 41

BOX 2.1 RESEARCH FOCUS

Cultivating the uncultured

Many major divisions of the Bacteria and Archaea contain no physiological properties of this ubiquitous organism to be investi-
known cultured species. As discussed in Chapters 4 and 5, our gated for the first time and enabled determination of the genome
knowledge of bacterial and archaeal diversity in the oceans has sequence of the cultured isolate, which was named “Candidatus
been revolutionized by the application of molecular biology tech- Pelagibacter ubique” (Giovannoni et al., 2005). Even with the
niques, especially sequencing of 16S rRNA genes, which allow genome sequences, identifying the nutrients required for growth
recognition and relatedness of groups of organisms based solely proved difficult and cultivation of more strains of SAR11 and
on their genetic sequence. In the 1990s, as the number of previ- other Alphaproteobacteria required further modification of the
ously unknown taxa recognizable by direct sequencing of genes method by using ultraclean techniques and Teflon culture vessels
in environmental samples increased, many microbial ecologists (Stingl et al., 2007). Zengler et al. (2002) used encapsulation of
assumed that these organisms are inherently unculturable and that single bacteria in microdroplets of gel to allow large-scale culture
it was futile to attempt to grow them. In the excitement generated of individual cells. Concentrated seawater was emulsified in mol-
by the rapidly accumulating results produced by new techniques, ten agarose to form very small droplets, and flow cytometry was
many molecular microbiologists became vociferous in their dis- used to sort droplets containing single bacteria. Various media
missal of the traditional methods applied to laboratory cultivation, and growth conditions were evaluated and microbial diversity in
considering them old-fashioned, distinctly “uncool”, and of little the microcolonies within the droplets was examined using 16S
further use. The phrase “more than 99% of bacteria are uncultur- rRNA gene typing. When unsupplemented filtered seawater was
able” became dogma. As pointed out by Donachie et al. (2007), this used as a growth substrate, a broad range of bacteria, including
attitude ignores important lessons from the history of microbiol- previously uncultured types, were isolated and grown.
ogy; Beijerinck, Winogradsky and other pioneers of general micro-
biology used numerous approaches to overcome the difficulties of Why is cultivation so important in the age of metagenomics?
microbial cultivation more than 100 years ago. Provided sufficient Zengler (2009) emphasized the importance of developing new
effort and ingenuity are applied, there is every reason to be optimis- cultivation strategies to enable detailed studies of cell physiol-
tic that we will be able to culture many more marine microbes. As ogy and genetics, essential to understand community-level pro-
Carini (2019) reminds us: “unculturable” is a frame of mind, not a cesses. Clearly, cultivation-independent and cultivation-dependent
state of microbiology—we should regard these organisms as “not approaches to microbial community assessment are complemen-
yet cultured” instead. tary. But as Carini (2019) points out, it is not always possible to
predict the activity of microbes and their ecosystem function from
Overcoming the problem of extreme oligotrophy. The most genome sequences alone. He gives the analogy “… one cannot
likely reason that so many marine bacteria are so difficult to understand the experience of driving a Ferrari from the list of its
grow is that they are extreme oligotrophs, adapted to growing components; a parts list does not convey the handling, the sound, or
very slowly at low cell densities with very low-nutrient concen- the driver’s connection with the machine.” Carini emphasizes the
trations (p.92). It is very difficult to reproduce these conditions great value obtained from current cultivation-independent genomic
in the laboratory, especially if we try to make organisms grow methods but reminds us that “… genomes are a parts list: we seek
quickly at thousands of times the density at which they exist in an understanding of the emergent principles of the cells themselves
nature. Common laboratory media used for the culture of marine and the communities that they constitute.” Furthermore, metage-
heterotrophic bacteria typically contain nutrients at levels 10 6 –109 nomic analysis of microbial communities relies heavily on data
times higher than those found in seawater. Button et al. (1993) obtained from the sequencing of genomes of cultivated species.
developed techniques for culturing slow-growing marine bacte- To interpret the likely activities of new genes identified in metage-
ria by making successive dilutions of natural seawater in filtered, nomes, we need genes from cultivated organisms to compare them
sterilized seawater, so that only one or two bacteria per tube are with. We can use molecular data to identify which uncultured taxa
obtained. This leads to a successful enrichment of the dominant we should target for cultivation. How abundant are they? Do they
cell types, which can grow to densities of about 105 –10 6 mL −1, diverge substantially from cultured taxa? Might they have biotech-
similar to those occurring in seawater. Schut et al. (1997) used nological potential? Do they have a key role in ecology or biogeo-
this technique to isolate an oligotrophic Sphingomonas sp. This chemical processes? Only by studying the growth, physiology, and
approach was developed in order to culture representatives of metabolism of live bacteria and archaea in culture can we deter-
the SAR11 clade, which was discovered using 16S rRNA gene mine the properties of the whole organism and relate them to its
sequencing in the Sargasso Sea in 1990 (Giovannoni et al., 1990) ecosystem function. One of the most compelling examples of the
and later recognized as the most abundant ocean bacteria (Morris benefits of ingenuity, patience, and persistence is the successful
et al., 2002; Giovannoni, 2017). Rappé et al. (2002) developed cultivation —after painstaking attempts spanning 12 years—of an
a high-throughput method in microtiter trays containing sterile extremely slow-growing Asgard archaeon found in sediments; an
seawater supplemented with phosphate, ammonia, or low levels organism that that could hold the key to the origin of eukaryotic
of organic compounds. Bringing SAR11 into culture allowed the cells, as described in Box 5.1.
42 Chapter 2

nonculturable microorganisms, but it provides a less stressful environment than the surface
of solid growth media and some investigators have used BIOLOG systems for community
characterization of metabolic activities in sediments or surface layers by direct inoculation
without an intervening culture step.

Analysis of microbial cell components can be used

for bacterial classification and identification
Analysis of the composition of bacterial membranes via gas chromatography of fatty acid
methyl esters (FAME) is a powerful technique that can detect very small differences between
species and strains. Individual taxa have a distinct fatty acid fingerprint. However, careful
standardization of growth conditions is needed because membrane composition is strongly
affected by environmental conditions and the nature of the culture medium.

Analysis of the protein profile of microbes can also be used for comparison of microbial
species. Extracted proteins are dissociated with the detergent sodium dodecyl sulfate and
separated by polyacrylamide gel electrophoresis (SDS-PAGE). However, gene expression and
the resulting protein pattern produced are extremely dependent on environmental conditions
and the technique needs to be used with caution in identification and taxonomy. Indeed, its
main use is in assessing the effect of various factors on the synthesis of particular proteins,
such as bacterial colonization or virulence factors during infection of a host, or responses to
temperature or nutritional shifts.
BREAKTHROUGH—
i DIRECT ISOLATION
OF NUCLEIC ACIDS
Pyrolysis mass spectrometry (PyMS) generates a chemical fingerprint of the whole microbe
by thermal degradation of complex organic material in a vacuum to generate a mixture of
FROM THE SEA low-molecular-weight organic compounds that are separated by mass spectrometry. PyMS is
Following Woese’s use of rRNA for easily automated and a high throughput of samples can be processed, but the equipment is
studies of relationships between expensive and is found only in a few specialized laboratories.
organisms, parallel advances in
nucleic acid sequencing tech-
nologies and computing led to the METHODS BASED ON DNA AND RNA ANALYSIS
rapid generation of large databases
of information linking sequences Nucleic acid-based methods have transformed
from known organisms to their understanding of marine microbial diversity and ecology
taxonomic position. Because
it was clear that many marine Marine microbiologists use a wide range of techniques based on the isolation and analysis
microbes could not be cultured, of nucleic acids. Many of these can be applied to the characterization of microbes in cul-
microbial ecologists asked whether ture, for example, for accurate identification and taxonomy or for diagnosis of disease, as
it would be possible to obtain shown in Table 2.2. The complete genome sequence of many cultured microbes has also been
sequence information directly determined, as shown in Table 3.1. However, the most dramatic advances in marine micro-
from environmental samples. The bial ecology have occurred as a result of the development of “environmental genomics”—
first application of direct sequenc- methods for the detection and identification of microbes based on direct extraction of genetic
ing of nucleic acids from the
material from the environment, without the need for culture. First introduced in the mid-
marine environment was in 1986
1980s, these culture-independent molecular techniques are very sensitive and can allow the
by Norman Pace and colleagues
at the University of Colorado. recognition of very small numbers of specific organisms or viruses among many thousands
Development of the PCR technique of others. Direct analysis of the genes that are present in an environmental sample or com-
in 1988 enabled rapid application parison of community profiles using “genetic fingerprinting” techniques allows us to make
in many different diversity studies inferences about the diversity, abundance, and activity of microbes in the oceans. Since 2000,
cataloging ocean microbes. A there have been rapid developments in techniques for isolation, amplification, and sequenc-
number of research teams, most ing of DNA, together with advances in bioinformatics—the mathematical and computing
notably including those led by approaches needed to analyze, interpret, and manipulate the genomic data generated. It has
Stephen Giovannoni (Oregon State become possible to assemble and analyze the genetic information obtained from microbial
University), Ed DeLong (then at communities in the environment in a way comparable to the analysis of single genomes, lead-
Monterey Bay Aquarium Research
ing to what we now term “metagenomics.”
Institute), and Jed Fuhrman
(University of Southern California),
pioneered research in this field in Amplification and sequencing of ribosomal RNA genes is
the 1990s, providing the break-
through that paved the way for widely used in microbial systematics and diversity studies
the current importance of marine Chapter 1 introduced the pioneering work of Woese and colleagues based on sequencing of
microbiology. ribosomal RNA, which led to revision of ideas about evolution of the major domains of life.
 Methods in Marine Microbiology 43

Table 2.2 Summary of techniques used for the identification and classification
of cultured bacteriaa

Technique Usefulness at different taxonomic levels

Genome properties

DNA sequencing All levels, for phylogenetic analysis

DNA–DNA hybridization All levels above species, 70% hybridization

level is used for species definition

GC ratios Comparison of species known to be closely

related

PCR-based fingerprinting (e.g. RAPD, Good discrimination at strain level

AFLP, rep-PCR)

RFLP, ribotyping Strain differentiation, but prone to variability

MLST Reliable and robust; good discrimination of

species and strains

16S rRNA gene sequencing Higher taxonomic levels, for phylogenetic

analysis

ITS region sequencing Strain differentiation

Phenotypic characteristics

Bacteriophage typing Strain differentiation, highly specific

Biochemical tests (e.g. API® system, Species and strain differentiation, routine
BIOLOG Phenotypic Microarrays™) identification, community analysis

Morphology Limited information except in some groups

Plasmid and protein profiles Strain differentiation but highly variable

Serology Strain (serotype) differentiation

Chemotaxonomic markers

FAME Rapid typing to genus level; further

discrimination may be possible with careful
standardization

PyMS Rapid typing to genus level; further

discrimination may be possible with careful
standardization

Quinones and other biomarkers Good differentiation above species level

aMany of these methods can also be used for archaea, fungi, and protists. AFLP, amplified fragment

length polymorphism; FAME, fatty acid methyl ester analysis; GC ratio, guanine–cytosine base pair
ratio; ITS, internal transcribed spacer; MLST, multilocus sequence typing; PCR, polymerase chain
reaction; PyMS, pyrolysis mass spectroscopy; RAPD, random amplification of polymorphic DNA;
RFLP, restriction fragment length polymorphism.

The ribosome is composed of a large number of proteins and rRNA molecules, as shown
in Figure 2.7. The rRNA molecules (16S for members of the Bacteria and Archaea; 18S for
Eurkarya) in the small ribosomal subunit (SSU) quickly became the first choice for microbial
community analysis and diversity studies for several reasons. First, the rRNAs are universally
present in all organisms, because of the role of the ribosome in the essential function of pro-
tein synthesis. Second, the rRNA molecule has a complex secondary structure and mutations
in parts of the gene that affect critical aspects of structure and function in the ribosome are
often lethal; therefore, changes in rRNA occur slowly over evolutionary time. Some parts of
the rRNA molecules—and consequently the genes that encode them—are highly conserved,
while others have a high degree of variability. These genetic differences act as a proxy for dif-
ferences between genomes and form the basis of modern phylogenetic classification systems.
44 Chapter 2

Figure 2.7 Structure of the ribo- Bacteria and Archaea Eukarya

somes in the three domains of life.
The subunits, proteins, and rRNA
molecules are classified by the
Svedberg unit (S), a measure of 70S ribosome 80S ribosome
6 6
the rate of sedimentation during 2.8 × 10 Da 4.2 × 10 Da
centrifugation. Sedimentation rate
depends on the mass and the shape
of the particles; hence when the SSU LSU SSU LSU
large (LSU) and small (SSU) subunits
are measured separately, they do not 30S subunit 50S subunit 40S subunit 60S subunit
add up to the S value of the intact 6
0.9 × 10 Da 1.8 × 106 Da 6
1.4 × 10 Da 2.8 × 106 Da
ribosome. The 16S (Bacteria and
Archaea) and 18S rRNA (Eukarya)
molecules in the SSU are most com-
monly analyzed. Molecular masses of 16S rRNA 23S rRNA 18S rRNA 28S rRNA
the ribosomes and their subunits are ~1540 nucleotides ~2900 nucleotides ~1870 nucleotides ~4700 nucleotides
shown in daltons (Da). + + + +

21 polypeptides 5S rRNA 33 polypeptides 5.8S rRNA

~120 nucleotides ~160 nucleotides
+ +
31 polypeptides 49 polypeptides

Another advantage is that growing microbes often contain multiple copies of the rRNA genes,
so use of the PCR (see below) allows them to be amplified from very small amounts of DNA.

PCR amplification and Sanger sequencing (described below) still remains a reliable and
inexpensive method in identification and classification of cultivated bacteria and archaea.
Because PCR is carried out with primers directed against specific sequences in the SSU
rRNA genes, it can also be used to selectively amplify gene fragments from different organ-
isms in an environmental sample. Most environmental studies use broad-spectrum (“uni-
versal”) primers directed against highly conserved regions in the SSU rRNA genes while
amplifying the more variable sequences between the primer annealing sites. Specific prim-
ers can be designed to target particular taxonomic groups that share a known sequence. The
PCR-amplification products can be cloned in E. coli to create a gene library. Proprietary
cloning vector kits are widely used; these ensure high efficiency of cloning and allow detec-
tion of transformant colonies containing DNA inserts, which are purified by re-amplifi-
cation before sequencing. After determination of DNA sequences, data are compared to
generate phylogenetic trees. A brief description of each of the main stages in this process is
given in the following sections. In large-scale environmental surveys, these methods have
been replaced by high-throughput sequencing (HTS), but they still remain an important part
of many research projects.

Although tremendous advances have been made using SSU rRNA gene sequencing—this
method dominates much of the discussion about diversity in this book—it should be remem-
bered that it is not the only approach and it has some inherent limitations. The highly con-
served nature and the small size of rRNA genes (1500–1800 bases) means that their useful
information content is relatively low. SSU rRNA gene sequencing is best suited to compari-
son of higher-level microbial taxa, and there can be drawbacks when it is used to delineate
genera and species. In theory, genes encoding proteins contain more information because
the genetic code, based on four nucleotides, generates a protein sequence based on 20 amino
acids. Increasingly, the sequences of genes encoding key structural proteins (e.g. protein
synthesis elongation factors) or enzymes (e.g. DNA polymerase or ATPase) are determined
for detailed phylogenetic comparisons. In addition, the use of probes directed against spe-
cific metabolic genes can provide information about the importance of microbial activities
in the environment. For very fine discrimination between closely related lower taxonomic
levels (such as between species, or between strains within a species) it is often preferable to
use noncoding regions between genes, such as the internal transcribed spacer (ITS) regions
of nonfunctional RNA that occurs between the structural rRNA genes. These regions of
 Methods in Marine Microbiology 45

DNA are not under the same constraints as functional genes and are often hypervariable.
Multilocus sequence typing (MLST) is a technique for the typing of multiple loci, using
7–10 “housekeeping” genes for major metabolic processes. Internal 450–500-bp fragments
are sequenced on both strands and separated by electrophoresis, providing allelic profiles
that allow high discriminatory power. MLST remains useful, especially for epidemiological
studies of pathogenic bacteria, enabling characterization of isolates from different hosts and
geographic regions, but it is less commonly used since the advent of HTS. Examples of recent
research findings using HTS in studies of microbial diversity are discussed in Box 4.1.

Isolation of genomic DNA or RNA is the first

step in all nucleic acid-based investigations
Protocols for DNA or RNA extraction from cells in culture are usually straightforward and
reproducible. By contrast, isolations from environmental samples are often problematic and
require considerable optimization. Samples must usually be processed rapidly or preserved
in such a way that the nucleic acids are protected from degradation (e.g. immediate freezing
in liquid nitrogen or transfer to a −80°C freezer). RNA is particularly prone to degradation
by nucleases, and special solutions for sample preservation (e.g. RNAlater) and avoidance of
contamination in the laboratory are needed. For extraction from planktonic microbes, it may
be necessary to concentrate biomass by centrifugation or tangential flow filtration of large
volumes of water. Cells are lysed using an ionic detergent (SDS) and chelating agent (EDTA)
plus proteinase K to break the membranes and dissociate proteins. Gram-positive bacteria
require treatment with lysozyme to disrupt the cell wall. RNA is removed with ribonuclease.

i
With a few techniques such as rep-PCR, preparation may involve simple procedures such SECRETS FOR SUCCESS
as boiling a sample of culture, but for most methods, good-quality purified RNA or DNA WITH THE PCR
is needed. The traditional method is based on phenol-chloroform extraction, which results
An exhibition called Chain Reaction
in separation of DNA from proteins, based on their relative polarity. However, most studies
held in Canterbury UK in 2013 cel-
employ commercially available kits in which DNA is collected by ethanol precipitation onto ebrated 30 years since the inven-
silica or anion exchange resin inserts in spin column microcentrifuge tubes and washed with tion of the PCR. The co-curator of
buffers containing chaotropic agents and high salt concentration, before eluting in pure water the exhibition, Charlotte Sleigh
or low salt buffer. These provide advantages in speed and reproducibility. All nucleic-acid- (Reader in the History of Science
based methods for analysis of diversity in the environment depend on the assumption that at the University of Kent) has com-
DNA is extracted in a more or less pure state equally from the different members of the com- mented on how the exhibition was
munity, but there are several reasons why this may not be the case. Microbes differ greatly used to show how the PCR has led
in their sensitivity to the treatments used to break open the cells. In particular, the nature to huge advances in genetics and
of the cell envelope varies greatly, and cells may be protected by association with organic biological understanding, despite
its simple process and relatively
material or because they are inside the cells of other organisms as symbionts or pathogens,
humble equipment—“it’s smaller
meaning that DNA may be extracted preferentially from some types according to the treat-
than a microwave, and even less
ment used. Techniques such as freeze-thawing, bead beating (rapid shaking of the mixture exciting to look at” (Sleigh, 2013).
with tiny glass beads), ultrasonic disintegration, or use of powerful lytic chemicals enhance The article also highlights the ritu-
the recovery of DNA but can result in extensive degradation. Problems often arise when try- als and superstitions of laboratory
ing to amplify DNA from samples such as sediments, microbial mats, or animal tissue due work with the PCR. Sometimes it
to the presence of PCR inhibitors. Again, specialized proprietary kits may overcome many doesn’t work, and a tedious and
of these problems. frustrating cycle of troubleshoot-
ing will ensue. Sleigh reports how
When investigating microbial symbionts or parasites, extra steps such as selective lysis or many scientists devise comic rou-
differential centrifugation may be needed to separate microbial cells from host material. tines and superstitions to “ensure”
success. She reports of a scientist
Special treatments may be needed to remove inhibitors of DNA extraction, amplification,
who had to see a “lucky clon-
and sequencing. Filtration or fluorescence-activated cell sorting (FACS) can be used if only ing rabbit” from the lab window
part of the community is under investigation; for example, viruses or a particular size class before starting the thermocycler.
of plankton. In my lab, there were students who
had to wear a “lucky lab coat” or
The polymerase chain reaction (PCR) turned the PCR hood into a mock
altar, where they made offerings of
forms the basis of many techniques reagents to the “PCR god” and car-
ried out ritual cleaning with bleach
PCR is a method of amplifying specific regions of DNA, depending on the hybridization
wipes. And, woe betide anyone
of specific DNA primers to complementary sequences in the target organism’s DNA. It
who touched a lab colleague’s
is a routine process in many laboratories, but a methodical approach and avoidance of micropipettes!
contamination are essential. The target DNA is mixed with a DNA polymerase, a pair
46 Chapter 2

Figure 2.8 The polymerase chain

reaction (PCR). The initial stages are
shown. (1) The reaction is initiated
by heating (usually 94–98°C for 1
min), so that the double-stranded
DNA dissociates into single strands.
(2) The temperature is then lowered
(usually 50–60°C for 0.5–1 min), so
that the primers can attach to the
complementary sequences on the
DNA (the annealing temperature is
calculated to be close to the Tm to
ensure efficient but specific binding).
(3) The temperature is raised to 72°C of oligonucleotide primers and a mixture of the four deoxyribonucleoside triphosphates
(the optimum for Taq polymerase), (dNTPs). The PCR depends on the use of a thermostable DNA polymerase which is able
and primer extension begins. After to function during the repeated cycles of heating and cooling in the reaction, as illustrated
the first few cycles, the number of in Figure 2.8. The original and most widely used enzyme, Taq polymerase, is a thermo-
short duplexes containing the ampli- stable enzyme isolated from the hot-spring thermophilic bacterium Thermus aquaticus.
fied sequence will increase expo-
Alternative thermostable polymerases, including some from deep-sea thermophilic bacte-
nentially; one target molecule will
ria, are now also used (see Table 14.3). The primers are designed to anneal to opposite
become 2n molecules after n cycles
strands of the target DNA so that the polymerase extends the sequences toward each other
(~1 billion after 30 cycles). Credit:
by addition of nucleotides from the 3′ ends, by base pairing using the target DNA sequence
modified from Enzoklop (CC-BY-SA
3.0) via Wikimedia Commons. as a template for synthesis. The heating and cooling cycles of the PCR are carried out in
specially designed electronically controlled thermal cyclers, in which the reaction tubes
are placed in blocks designed to ensure rapid and precise heating and cooling. Usually, ~30
cycles are used, resulting in amplification of a single target sequence to over 250 million
PCR products (usually slightly less than the theoretical maximum). The outcome of the
reaction can be checked by agarose gel electrophoresis, which separates the DNA molecules
according to size; bands are then visualized under ultraviolet light after staining the gel with
ethidium bromide, Safe-Red, or SYBR Green dye. Because of the high degree of conser-
vation in the SSU gene (or other conserved genes used as targets), the PCR products will
all have approximately the same size and should migrate as a single band. The molecular
weight of the amplicon can be assessed by comparison with a “ladder” of DNA fragments
that is separated in the same gel. Before submitting it for sequencing, further purification
steps are usually carried out, before checking the quantity and purity of the DNA by a drop-
based spectrophotometer that measures microliter volumes.

Choosing a primer for use in the PCR obviously depends on knowing the sequence of the tar-
get DNA, but several factors need to be considered in order to ensure a successful PCR. For
frequently studied genes, such as those for rRNA, there are standard primers and PCR condi-
tions that are widely used. For studying novel genes, it is necessary to compare sequences in
the databases to find a consensus sequence that recognizes conserved regions of a gene. An
oligonucleotide primer, usually about 17–28 bases in length, is made using a DNA synthe-
sizer—this is often carried out by commercial services. The ratio of G + C bases should usu-
ally be about 50–60%, depending on the gene, and primers should end in G or C to increase
efficiency. Care needs to be taken to avoid complementary sequences within the primer,
and between the pair of primers, as this can lead to secondary structures such as “hair-
pins” or preferential synthesis of primer dimers. Alternative nucleotide bases can be placed
at specific positions, giving a set of degenerate primers that will allow for strain variation in
DNA sequence for a particular gene. Computer programs are used to design the appropriate
sequence of bases to give good results and to calculate the temperatures to be used in the
thermal cycler program.

Suitable positive and negative controls are always included in the PCR run and each applica-
tion of the PCR must be carefully optimized. This is especially important when trying to
amplify sequences from a large mixed population of DNA, such as direct examination of
environmental samples. Conditions such as the annealing temperature, amount of template
DNA, and concentration of Mg2+ are critical. For example, Taq polymerase is inactive in
the absence of Mg2+, but at excess concentration the fidelity of the polymerase is impaired
and nonspecific amplification can result. Designing a PCR protocol to amplify a new gene
requires extensive optimization.
 Methods in Marine Microbiology 47

There are many variations of the basic PCR technique. In nested PCR, the level of specific-

i
ity and efficiency of amplification is improved by carrying out a PCR for 15–30 cycles with THE PCR CAN HAVE
one primer set, and then an additional 15–30 cycles with a second set of primers that anneal INHERENT BIAS
within the region of DNA amplified by the first primer set. Multiplex PCR involves the use of
AND LIMITATIONS
multiple sets of primers and results in the production of multiple products. This is often used A considerable amount of prior
as a quick screening method for the presence of certain organisms within water or infected investigation is needed to ensure
tissue. Reverse transcriptase PCR (RT-PCR) detects gene expression by quantifying messen- efficient extraction and ampli-
ger RNA (mRNA). A DNA copy of mRNA (cDNA) is made with reverse transcriptase, before fication of DNA from samples.
running the PCR. This technique requires great care in sample preparation and handling, So-called “universal” primers
targeted against specific groups
since mRNA is very short-lived, and the technique is prone to contamination from genomic
can hybridize more favorably with
DNA and nucleases. In quantitative real-time PCR (Q-PCR), the formation of a fluorescent some templates; for example,
reporter is measured during the reaction process. By recording the amount of fluorescence DNA templates from thermo-
emission at each cycle and incorporation of appropriate standards, it is possible to monitor philic organisms with a high
the PCR reaction during the exponential phase in which the first significant increase in the content of G + C base pairs may
amount of PCR product correlates to the initial amount of target template. Q-PCR offers denature less rapidly than those
significant advantages for the diagnostic detection of microbes in the environment and in from mesophiles. An additional
infected tissue samples, and handheld instruments for use in the field are now available. problem arises from the produc-
tion of chimeric PCR products.
These are artifacts caused by the
Genomic fingerprinting can be used to joining together of the amplifica-
tion products of two separate
assess diversity of cultured isolates sequences and, if not detected,
Investigation of diversity among closely related individual species, or strains within a spe- such sequences could suggest the
cies, of bacterial or archaeal cultures can be achieved by a variety of genomic fingerprinting presence of a novel nonexistent
techniques. The most widely used methods make use of differences in genetic sequences organism. This is a particular pitfall
when multi-template PCR is car-
revealed by the action of restriction enzymes, which cut DNA at specific sites determined
ried out using DNA derived from
by short sequences of nucleotide bases. All of these fingerprinting techniques produce band environmental samples. Various
patterns that lend themselves to computer-assisted pattern analysis, leading to the generation techniques are available to limit the
of databases and phylogenetic trees useful for studying diversity among cultured isolates. formation of chimeras, and soft-
ware to detect chimeric sequences
Ribotyping is a technique used for bacterial identification in which the PCR products from should be used. Analysis of rRNA
rRNA gene amplification are cut with restriction endonucleases. These enzymes recognize databases such as GENBANK has
short sequences in the DNA and cut the molecule wherever those sequences occur. The DNA revealed that numerous chimeric
fragments are separated by agarose electrophoresis and the resulting band pattern gives a sequences have been deposited in
rapid indication of similarities and differences between strains. While this method obviously the past, leading to confusion for
phylogenetic analysis.
lacks the resolution of sequence analysis, it is quick and inexpensive.

Restriction fragment length polymorphism (RFLP) analysis involves cutting DNA with a
combination of restriction endonucleases and separating the fragments according to size
by gel electrophoresis. Fragments containing a specific base sequence can be identified by
Southern blotting, which involves transfer to a sheet of nitrocellulose and addition of a radio-
actively labeled complementary DNA probe. Hybridization of the probe is detected using
autoradiography or with an enzyme-linked color reaction. This enables identification of the
restriction fragment with a sequence complementary to that of the DNA probe. Fragments
of different length caused by mutations in duplicate copies of the gene can be detected and
used for strain differentiation. The RFLP method can be combined with PCR amplification
of specific genes, such as SSU rRNA genes, followed by restriction endonuclease digestion
and electrophoresis. This avoids the need for blotting and radiography but is not always suc-
cessful in providing useful markers for strain differentiation.

The RAPD (random amplified polymorphic DNA) technique is a more commonly used PCR
marker technique for strain differentiation. It uses a single short primer in a PCR reaction,
leading to a fingerprint of multiple bands generated by single nucleotide differences between
individuals that prevent or allow primer binding. The method is quick and easy to perform
but does not always give reliable results for detecting strain differences. AFLP (amplified
fragment length polymorphism) gives better resolution than either RFLP or RAPD analysis.
Genomic DNA is digested using a pair of restriction endonucleases, one of which cuts at
common sites and the other at rare sites. Adapter sequences are ligated to the resulting frag-
ments and a PCR is performed with primers homologous to the adapters plus selected addi-
tional bases. A subset of the restriction fragments is amplified, resulting in a large but distinct
set of bands on a polyacrylamide gel. Band patterns are compared using imaging systems,
48 Chapter 2

and similarities and differences between strains can be computed to generate phylogenetic
trees. Despite its advantage, AFLP analysis is a technique requiring highly purified DNA and
its use is generally restricted to laboratories specializing in taxonomy.

Microsatellite markers are repeat regions of short nucleotide sequences that can be analyzed
using PCR amplification of unique flanking sequences and separation of the resulting bands
by gel electrophoresis. Microsatellite markers are used extensively in the population analysis
of eukaryotic organisms such as microalgae.

The genomic fingerprinting method for bacteria known as rep-PCR is a simple and reproduc-
ible method for distinguishing closely related strains and deducing phylogenetic relation-
ships. The method is based on the fact that bacterial DNA naturally contains interspersed
repetitive elements, such as the REP, ERIC, and BOX sequences, for which standard PCR
primers are available. The method requires very simple preparation of DNA samples without
the need for extensive purification and is especially useful for the initial screening of cultures
collected from marine samples.

Pulsed-field gel electrophoresis (PFGE) is a method in which the orientation and duration of
an electric field is periodically changed, allowing large-molecular-weight fragments of DNA
(up to 1000 kb) to migrate through the gel. Bacteria can be embedded in the gel and lysed
in situ into large fragments, using a restriction enzyme that recognizes rare restriction sites.
PFGE typing detects small changes in the genome resulting from the insertion or deletion of
gene sequences or mutations that alter the restriction enzyme sites. PFGE is also very useful
for analysis of virus diversity.

Determination of DNA properties is used

in bacterial and archaeal taxonomy
Determination of the ratio of nucleotide base pairs has been used in taxonomy since the
1970s, and currently still forms part of the formal description of species, although this may
soon change due to the development of improved methods (see Chapter 4). The principle of
this method is that two organisms with very similar DNA sequences are likely to have the
same ratio of guanine (G) + cytosine (C) to the total bases. DNA is extracted from the bac-
terium or archaeon and the concentration of the bases is determined by high-performance
liquid chromatography (HPLC) or by determining the “melting point” (Tm) of DNA. This
is the temperature at which double-stranded DNA dissociates; because the hydrogen bonds
connecting G–C pairs are stronger than those connecting adenine (A) and thymine (T), the
Tm increases with higher ratios of G + C. Closely related organisms have very similar GC
ratios, and this can be used to compare species within a genus. However, it is important to
remember that the converse is not true: bacteria and archaea possess a wide range of GC
ratios and completely unrelated organisms can share a similar GC ratio.

A more useful measure of relatedness for the definition of species is DNA–DNA hybridiza-
tion. Purified DNA from two organisms to be compared is denatured by melting and mixing
their DNA. When cooled, DNA with a large number of homologous sequences will reanneal
to form duplexes. The amount of hybridization can be measured if the DNA from one of the
organisms is labeled by prior incorporation of bases containing radioactive isotopes of 14C
or 32P. The amount of radioactivity in the reannealed DNA collected on a membrane filter is
measured and the percentage hybridization is calculated by comparison with suitable con-
trols. Taxonomists often use the benchmark of 70% or greater hybridization (under carefully
standardized conditions) as the definition of a bacterial species. The difficulties associated
with the concept of bacterial and archaeal species, and the use of improved genomic methods
to overcome them, are discussed in Chapter 4.

DNA sequence data are used for identification

and phylogenetic analysis
Sequence data derived both from microbes cultured in the laboratory and from direct commu-
nity analysis of the environment are used to determine the degree of phylogenetic relatedness
 Methods in Marine Microbiology 49

between organisms. Data processing depends on the availability of publicly accessible data-
bases (e.g. GENBANK, Ribosome Database Project, and EMBL) and specialized computer
software. The first step is usually to use BLASTn, which compares a newly generated nucleo-
tide sequence to previously described sequences deposited in databases. Various mathematical
methods and computer algorithms are used for the construction of phylogenetic trees, but details
of the theory of each technique and its merits are beyond the scope of this book. In one common
method (neighbor-joining tree), the different sequences are compared in order to determine the
evolutionary distance between all the permutated pairs of sequences in the dataset by calculat-
ing the percentage of non-identical sequences. Statistical corrections to allow for the possibility
of multiple mutations at a given site are included in the calculation. A distance matrix is then
constructed, and a phylogenetic tree is drawn by grouping the most similar sequences and then
adding less similar groups of sequences. The lengths of the lines in the tree are in proportion
to their evolutionary distances apart (e.g. Figure 4.1). In another approach, called parsimony,
the tree is constructed after calculating the minimum number of mutational events needed for
each pair to transform one sequence into the other. The tree is then drawn using the minimum
number of such parsimonious steps. This method relies on the inherent assumption that evolu-
tion proceeds in one direction by the fastest route; this is undoubtedly an oversimplification.
The validity of trees can be tested by “bootstrapping”—a computational technique in which
multiple iterations are made of trees based on subsamples of the sites in an alignment.

Sequence analysis of rRNA molecules shows that certain nucleotide sequences are highly
conserved in certain regions. The nucleotide bases are numbered in a standard notation and
signature sequences are derived which are characteristic of groups of organisms at different
taxonomic levels. For example, the sequence CACYYG (where Y is any pyrimidine) occurs
at approximate position 315 in the 16S rRNA molecule in almost all Bacteria but does not
occur in Archaea or Eukarya. Sequence CACACACCG at position 1400 is distinctive of
Archaea but does not occur in Bacteria or Eukarya. Signature sequences can therefore be
used to design domain-specific oligonucleotide probes for identification of any member of the
respective group in a marine sample. Other sequences may be diagnostic for phyla, families,
or even genera. An essential feature of modern taxonomy is to target these regions of the
DNA by PCR amplification to produce multiple copies that can be sequenced. If the complete
sequences are obtained from cultured microbes, the approved process to ascribe a taxonomic
name requires additional information about the physiological properties and phylogenetic
relationships of the organism. Therefore, when deposited in the databases, these sequences
serve as a reference source against which sequences from the environment can be compared.
This is a very powerful technique for culture-independent community analysis, especially in
combination with metagenomics and FISH.

DGGE and TRFLP can be used to assess

composition of microbial communities
Different strategies for microbial community analysis are shown in Figure 2.9. Denaturing
gradient gel electrophoresis (DGGE) and terminal restriction fragment length polymorphism
(TRFLP) are methods for separating PCR amplicons. They were the mainstay of community
analysis studies throughout the 1990s–2000s but they have been now largely replaced by
HTS. They are still used by some laboratories, and many important research papers refer to
these methods, so a basic understanding of how they work is important.

In DGGE, PCR products are run in a polyacrylamide gel under a linear gradient of denatur-
ing conditions—typically, 7 M urea and 40% formamide at a temperature of 50–65°C. Small
sequence variations result in different migration of DNA fragments. Double-stranded DNA
migrates through the gel until it reaches denaturing conditions that cause separation of the
double helix (“melting”). DGGE primers must contain a guanine- and cytosine-rich sequence
(GC clamp), which is resistant to complete denaturation under the conditions used, so that the
band is “locked” in place according to the Tm. In analysis of community DNA, a set of bands
will be generated representing variations in the sequence of the target gene (Figure 2.10). A
typical analysis of an environmental sample will generate more than 50 bands, and many dif-
ferent samples can be compared on a single gel. This provides a rapid and efficient method of
“community fingerprinting” for monitoring qualitative changes in community composition at
50 Chapter 2

Figure 2.9 Strategies for cultiva- different locations or under different conditions. Sometimes, the results of DGGE community
tion-independent analysis of micro- analysis are interpreted in a semi-quantitative fashion, using the assumption that the intensity
bial communities using molecular of bands indicates the relative abundance of the species producing them. However, great cau-
biological techniques. BAC, bacterial tion is needed in drawing such conclusions because the techniques used for DNA extraction,
artificial chromosome; DGGE, dena- the presence of inhibitors, the nature of the primers, and the PCR reaction conditions can
turing gradient gel electrophoresis; all have significant effects on the results obtained. Temperature gradient gel electrophore-
EST, expressed sequence tag; SSU, sis (TGGE) works using similar principles, but the concentration of denaturing chemicals
small subunit; TRFLP, terminal restric- remains the same while the temperature of the gel is increased gradually and uniformly.
tion fragment length polymorphism.
In TRFLP analysis, PCR is performed on DNA extracted from the mixed community with
one of the primers labeled with a fluorescent probe. After PCR, the products are cut using
specific restriction enzymes. Each PCR product produces a fluorescent molecule with the
probe at the end. The size of the fragments is determined by differences in the presence or
absence of restriction sites in a particular region of the molecule or by deletions and inser-
tions in the region. Like DGGE, TRFLP can provide a quick semi-quantitative picture of
community diversity but, again, results must be interpreted carefully.

Advances in DNA sequencing enable

improved microbial community analysis
In the well-established technique of chain termination during synthesis of a DNA copy devel-
oped by Frederick Sanger, a sequencing primer is added to a single-stranded DNA (ssDNA)
molecule in four separate reaction mixtures and chain extension begins using the normal
nucleotides (dNTPs). The different reactions contain a small amount of dideoxy nucleotides
(ddNTPs), which are synthetic analogs of the nucleotides lacking an –OH group on the sugar.
Because of this, the DNA chain is extended only up to the point at which one of the ddNTPs
is incorporated. The products from the four reactions are compared by electrophoretic sepa-
ration, and the sequence of the DNA is determined by reading the sequence of nucleotides at
which the chain extension is terminated. The original technique was modified to allow direct
sequencing of PCR products as double-stranded DNA (dsDNA) using dNTPs labeled with
different fluorescent dyes, which can be read by lasers. Automated DNA sequencers allow
robotic handling and high throughput of multiple samples, generating color-coded printouts
of the sequence and data suitable for direct analysis by computer. For community analysis,
it is often necessary to obtain only a partial sequence. A recent improvement of the Sanger
 Methods in Marine Microbiology 51

Figure 2.10 Example of a DGGE

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 gel for microbial community analy-
sis. The gels show 16S rRNA gene
fragments from water, and replicate
tissue slurry samples from three
colonies of the coral Pocillopora
damicornis. Numbers indicate bands
of interest that were excised and
sequenced. Reprinted from Bourne
19 21 22
and Munn (2005) with permission of
2
24
25
John Wiley & Sons.
3 16 20
1 4 11 26
5 9 14
17 27
10 15 18 23 28
6 7 12
8 13
1 2 3 4 5 6 1 2 3 5 6 4 1 2 4 5 6

Water A B C

HIGH THROUGHPUT,
i LOW COST,
HIGH IMPACT
method involves the application of microfluidic separation devices so that all steps in the
process are fabricated onto an electronic wafer to produce a “lab on a chip.” Much of the commercial impetus
for the development of new HTS
Even with advances in automation, a Sanger sequencing run for community analysis typi- technologies is driven by biomedi-
cally produces only ~100 sequences with a length of ~650 base pairs (bp). In the mid-1980s, cal companies seeking to develop
intensive research by biotechnology companies led to new methods that permit rapid and personalized medicine based
inexpensive DNA sequencing. These were called “next-generation sequencing” methods to on individual human genome
sequencing. Sequencing the first
emphasize their novelty. This term persists, although it is clearly a misnomer once a tech-
human genome in 2003 involved
nique has passed the design and development stage; it is more appropriate to use the term multinational consortia and cost $3
HTS, emphasizing the length of sequence reads provided. billion. By 2006, it had decreased
to ~$300,000 and today some
This development continues to occur at a very rapid rate and only a brief summary of some of the companies offer a service cost-
main methods is given here—images and animations of the methods are available on the com- ing less than $1000. Widespread
pany websites. The method of pyrosequencing was developed in 2005 by the company 454 Life genome screening offers many
Sciences and quickly became the main method for metagenomic studies in marine microbiology. potential benefits for advances
The process depends on PCR amplification in minute individual reactions with a one-by-one in healthcare (although there
nucleotide addition cycle, where the pyrophosphate (PPi) released from the DNA polymerization are significant ethical and data
privacy concerns). The spin-off
reaction is transformed into chemiluminescent signals. This method is now obsolete.
for microbial ecologists is that it
is now possible to explore biodi-
The Applied Biosystems SOLiD sequencer is based on sequencing catalyzed by DNA ligase. versity at an unprecedented scale.
DNA fragments are linked by adaptor sequences to complementary oligonucleotides on 1 Individual laboratories can conduct
mm magnetic beads. After amplification by emulsion PCR, the beads are attached to the massive metagenomic analyses,
surface of a specially treated slide inside a fluidics cassette. A universal sequencing primer or sequence multiple strains of
that is complementary to the adapters on the DNA fragments is added, with sequential addi- cultured microbes at relatively
tion of a set of four fluorescently labeled oligonucleotides that hybridize to the DNA frag- low cost within a short period
ment sequence adjacent to the universal primer, so that DNA ligase can join the nucleotides. (although the cost of processing
Multiple cycles of ligation, detection, and cleavage are performed, with the number of cycles the enormous amounts of data
determining the eventual read length. generated can be considerable).
Common bioinformatics protocols
and public databases facilitate a
The Ion Torrent system contains electronic systems combining integrated circuits with metal- rapid rate of discovery, although
oxide semiconductors and transistors that detect the change in potential each time a proton is improvements in standardization
released after a nucleotide is added during DNA synthesis. It is very cost-effective, yielding are needed to improve interpre-
~500 million reads with a length of ~400 bp. tation (Thompson et al., 2017).
Pedros-Alio et al. (2018) review the
The method currently favored by most microbial ecologists is the Illumina sequencing tech- dramatic impact of HTS in marine
nology. This involves the attachment of genomic DNA fragments to a transparent surface of microbial ecology but remind us
a flow cell, where they are amplified to create millions of clusters, each containing ~1000 of the challenge that still remains
copies of the same template. These DNA templates are sequenced by synthesis with DNA to describe the diversity of the
polymerase, employing reversible terminators with fluorescent dyes that can be removed after millions of microbial species that
exist in the oceans (Salazar and
each round of synthesis. The instrument scans the incorporation of each nucleotide using a
Sunagawa, 2017).
52 Chapter 2

laser and an algorithm assigns the base sequence, using a reference sequence for internal
quality control. Although the read length of sequences (~300 bp) is much lower than other
technologies, this is compensated by very high output (millions of sequences per run at very
low cost) and a very low error rate.

The most recent phase in the evolution of HTS is the ability to sequence single DNA mol-
ecules without any necessity for an amplification stage. One of the most successful devel-
opments is the PacBio SMRT (single molecule, real time) platform manufactured by the
company Pacific Biosciences. DNA polymerization occurs in arrays of fabricated nanostruc-
tures—tiny holes in a metallic film covering a chip. Because these apertures are smaller than
the wavelength of laser excitation light, only a tiny area at the bottom of the well is illumi-
nated. DNA polymerase molecules are attached to the bottom of each well—the incorpora-
tion of nucleotides being polymerized produces real-time bursts of fluorescent signal within
the illuminated region and the extension of DNA chains by single fluorescent nucleotides can
be monitored in real time. This process can sequence single DNA molecules rapidly, produc-
ing long sequence reads of over 104 bases, although it currently has high error rates. It can
also detect modification of DNA bases by methylation, which is an important phenomenon
leading to alterations in the control and activity of DNA-mediated processes like transcrip-
tion, or defense against phage infection.

Another exciting development is the MinION sequencer produced by Oxford Nanopore

Technologies. This is a USB device the size of a desk stapler, which can provide rapid results,
even in the field, without the need for highly expensive laboratory sequencers. The dsDNA is
denatured by a processive enzyme which ratchets one of the single strands (ssDNA) through
a biological nanopore embedded in a synthetic membrane. The technology has also been used
successfully to directly sequence mRNA and ss RNA viruses. A voltage is applied across the
membrane and the different nucleotide bases restrict the flow of ions in a distinctive way as
the ssDNA passes through the nanopore. Machine learning is then used to translate current
fluctuations into nucleotide base calls, resulting in a real-time sequence analysis for each
channel. The length of Oxford Nanopore reads is determined by the integrity of the input
DNA and single reads >1 million bases in length have been successfully sequenced. The
DNA sequence can be inferred by monitoring the current at each channel. Another promising
technology has been developed by the BGI company; their sequencer employs combinatorial
chemistry to incorporate a fluorescent probe onto a DNA anchor on a “nanoball” followed by
high-resolution digital mapping.

Each method has advantages for different applications and costs per gigabase vary more
than 100-fold (excluding the cost of the infrastructure). The long and accurate read lengths
obtained with Sanger methods mean that they continue to be used for sequencing PCR ampli-
cons, genomes from cultured organisms, or sequences with repetitive regions. The current
generation of long-read sequencers generate data quickly but have error rates that are rather
high for community analysis. However, with sufficient coverage it is possible to correct
read errors sufficiently to partially predict protein functions. Hybrid approaches are used
increasingly, combining the advantages of long reads offered by some systems with the low
error rates of other technologies such as Illumina. Future technological developments will
undoubtedly lead to further capabilities.

Elucidating the full genome sequence of microbes

provides insights into their functional roles
The whole genome shotgun sequencing method was pioneered in the 1990s by J. Craig
Venter and colleagues to accelerate progress in the Human Genome Project. It has since been
applied to many thousands of marine bacteria and archaea and a few species of eukaryotic
microbes. Tables 3.1 and 6.1 list some important examples illustrating how genomics has
provided insights into the physiology, adaptations, and ecology of marine microbes. DNA
is fragmented into small segments up to 1000 bp and sequenced from both ends. Typically,
when using Sanger or long-read HTS methods, ~10 replicate sequences will be obtained for
different sections of the genome, but 50–100 runs are needed. Overlap consensus algorithms
often fail at these levels due to high memory requirements, and a network analysis method
 Methods in Marine Microbiology 53

(de Bruijn graph) is used. Computer algorithms are used to check the sequences for overlap-
ping, contiguous segments (contigs) and assemble them into larger pieces (scaffolds), which
are then compiled into a draft genome. In the early days, this process often involved multi-
partner teams and took many months. Today, robotic handling of samples and automated
HTS methods mean that it is possible for an individual laboratory to compile a draft genome
within hours or days. Nevertheless, alignment of contiguous nucleotide sequences, genome
mapping, and annotation of the sequences—the prediction of gene sequences and attachment
of predicted protein-coding sequences and other biological information—often relies on rig-
orous collaborative efforts between different laboratories and is most useful when different
strains of the microbe are sequenced. For this reason, most genome sequences published in
databases are at the draft stage. Most bacterial and archaeal species possess single circular
chromosomes of a manageable size, but there are many exceptions to this rule which can lead
to complications in the interpretation of genomic data. The technical difficulties of sequenc-
ing the much larger and more complex genomes of eukaryotic microbes are discussed in
Chapter 6.

Once sequences are complete and gaps are closed, auto-annotation of the genome sequence
is carried out by computer software that predicts the open reading frames (ORFs); these are
sequences of codons beginning with a start codon (usually AUG) and ending with a stop
codon (UAA, UAG, or UGA). These are recognized by searching for characteristic conserved
sequences at the promoter region (where DNA polymerase binds), ribosomal binding sites,
and start and stop codons on the transcribed mRNA. DNA sequences—and the predicted
amino acid sequences of the proteins that they would encode—are compared with those
in DNA and protein databases. More detailed examination using other evidence is used to
refine the annotation of the genome. Often, many of the predicted ORFs cannot be reliably
linked to a function using current knowledge of bioinformatics. Therefore, the next stages
in molecular analysis are functional genomics and proteomics. This can involve inactiva-
tion of specific regions of the genome and high-resolution separation and identification of
proteins, although for many organisms (especially those with streamlined genomes), this
can be problematic and transcriptomics under different conditions will prove more useful.
Such post-genomic analyses of marine microbes are revealing many previously unknown
properties and activities. It is important to bear in mind that each genome sequence is only
a “snapshot” of one individual microbial strain. As reduced sequencing costs and enhanced
bioinformatics enable analysis of genomes from multiple isolates of the same species, very
large variations in genome properties can be revealed. Indeed, such comparative genomics
can yield important information about adaptations to specific environments.

Metabarcoding and metagenomics have led to

major advances in microbial community analysis
In the past two few decades, techniques based on analysis of SSU rRNA gene sequences have
underpinned microbiome research in the marine environment, as well as in plants, soils, and
the human body. Progress has been slower for eukaryotic microbes, but discoveries in this
field are now accelerating, and use of other genomic sequences have been useful, such as
internal transcribed spacer (ITS) regions of the genome encoding the rRNAs or the rbc gene
encoding RubisCO in phototrophic protists (see Chapter 6). The study of viral diversity in
the environment (viromes) remains difficult, because there is no equivalent universal marker
gene that can be identified, even for the most abundant viruses (see Chapter 7).

The concept of metagenomics developed in the late 1990s as a method of analyzing col-
lections of genes sequenced from microbial communities in the environment (Figure 2.9).
In the first studies, methods for handling large DNA fragments up to 300 kb were devel-
oped, so that pieces of DNA could be isolated and recognized by detection of their signature
sequence of SSU rRNA (or other marker gene) using Southern blotting (a process by which
DNA fragments separated by electrophoresis are transferred to a membrane and identified
by hybridization with a probe sequence). The large fragments were cloned directly into the F
plasmids of E. coli to create bacterial artificial chromosomes (BACs), which have the addi-
tional advantage that expression of some of the inserted genes may occur in the E. coli host
harboring the vector.
54 Chapter 2

This is a slow and difficult process, but fortunately it was quickly replaced by the application
of the shotgun sequencing method described above. This allowed large numbers of random
fragments of DNA to be extracted from environmental samples. Mass sequencing was made
possible by the deployment of multiple automated DNA sequencers and the development of
new bioinformatics software to reconstruct the overlapping DNA sequences (contigs) aris-
ing from many small fragments. This led to the first major marine environmental survey
by Venter and colleagues in 2004–2007, (GOS, see p.55). The frequency of specific DNA
sequences gives an indication of the abundance of a particular organism, and even very rare
organisms should be represented if enough sequences are analyzed. Where this approach
is used for community analysis based on a barcode such as the 16S rRNA gene, it is often
referred to as metabarcoding. The advent of the 454 HTS platform enabled further explora-
tion of marine microbial diversity by the International Census of Marine Microbes (ICoMM,
see Box 4.1). Despite its success, it became increasingly obvious that the PCR-amplification
stage used in early sequencing methods led to biases that distorted estimations of richness
and diversity in microbial communities. To avoid these biases, it is preferable to directly iden-
tify DNA fragments containing the code for rRNA. This became possible with the advent of
Illumina sequencing, which facilitated an enormous increase in throughput at comparatively
low cost. An Illumina sequencing run can produce tens of millions of metagenomic reads,
which are enough to recover the thousands of rRNA-encoding DNA fragments obtained from
shotgun sequencing of all the DNA present in a sample. This can provide enough information
to reliably capture the structure of the microbial community. These fragments are named
metagenomic Illumina tags (miTags). Even though Illumina reads are very short, the mul-
tiple overlapping fragments produced can provide accurate estimates of entire rRNA gene
sequences to be used for taxonomic characterization.

Advances in DNA sequencing have only been possible because of parallel advances in com-
puting and bioinformatics to interpret the huge datasets produced by the community analy-
sis sequencing runs. Many algorithms have been developed to filter, cluster, and annotate
the sequences. Only a few of the main approaches are mentioned here. The first step is to
check the quality of the sequencer output. In all platforms, there is a certain probability that
the sequencer software will wrongly identify a base. Various software tools are available to
remove artefacts such as low-quality base calls, primer and adapter sequences, reads that are
too short, chimeras, or obvious contaminating sequences. If we are interested in knowing
what kinds of organisms are present in a sample, the next step in a community analysis based
on SSU rRNA sequencing will be to cluster similar sequences to identify the operational
taxonomic units (OTUs). Because of the problems of defining bacterial and archaeal species,
OTUs are used as a proxy, by grouping organisms using an arbitrary sequence similarity cut-
off. The choice of similarity thresholds for determining similarity depends very much on the
research question being investigated. In earlier studies, sequences which only occur once in
the dataset (singletons) were removed to simplify statistical comparison between samples, but
this is unnecessary with modern algorithms. Recently, the increased quality of DNA sequence
results has led to reporting community analysis data as amplicon sequence variants (ASVs)
rather than OTUs (these issues are discussed in more detail in Chapter 4). Taxon assignment
of OTUs or ASVs depends on comparing sequences with those of reference organisms in
databases such as SILVA, RDP, or GreenGenes. Algorithms such as QIIME2, MOTHUR, or
DADA2 are used to sort and cluster the data to simplify the computational processing. These
calculate a distance matrix set at various threshold levels, leading to a biome table, which
clusters them according to the thresholds selected. The taxonomy assignment of taxonomic
affiliations (genus, family, order, etc.) can then be added to the records of the OTUs or ASVs.

Further downstream analysis of the results of the survey can then be carried out. Data is often
presented in a variety of different graphical formats to facilitate interpretation. For example,
phylogenetic trees accompanied by a stacked bar chart, pie charts, or heat maps using different
colors for different taxa such as orders or classes are often used (e.g. Figure 10.9). Diversity
within a sample (alpha-diversity) is measured using various indices such as Chao1, Simpson’s
index, and Shannon–Wiener index (p.119). For diversity between samples (beta-diversity) cor-
relation coefficients known as the Bray–Curtis dissimilarity index or Spearman rank cor-
relation are commonly reported. Principal components analysis (PCA) and multidimensional
scaling (MDS) plots are often used to highlight groupings within a community, for example
by linking to metadata such as physical conditions, nutrient availability, or other variables.
 Methods in Marine Microbiology 55

It is important to recognize that any community analysis will have unavoidable biases,
SAILING TOWARDS
depending on the combination of sampling and sequencing methods and the algorithms used
to analyze the results. Careful design of experiments and selection of data processing meth- i A BETTER
UNDERSTANDING OF
ods can minimize these biases, but it will always be impossible to be certain that we have
OCEAN MICROBIAL
found the “holy grail” of a description of the true community structure in complex marine
DIVERSITY
communities.
Several sampling expeditions using
sailing vessels have been instru-
Omics technologies provide information about the mental in vastly expanding our
functional gene composition of a microbial community knowledge of microbial diversity
in the oceans (Figure 2.12). The
As well as the narrower metabarcoding approach based on diversity of a subsection of a single first major metagenomic dataset of
gene, the full power of metagenomics is its provision of information about the range of genes 6.3 × 109 bases was published by J.
within a microbiome. This provides insight into linkages between function and phylogeny for Craig Venter in 2004. Further sur-
uncultured organisms and enables the discovery of genes encoding novel enzymes or other veys conducted during the Global
Ocean Sampling (GOS) expeditions
proteins. The endpoints of metagenomic studies get us closer to answering the key questions
with his yacht Sorcerer II resulted
in microbial ecology mentioned earlier—“who’s there, what are they doing, and how are they
in the discovery of six million
doing it?” Advances in DNA isolation methods and sequence analysis have enabled the full, or gene families in 2007. Since then,
almost complete, genome sequence of many microbes to be determined without the need for individual laboratories and inter-
isolation and culture. This has led to spectacular advances in our understanding of the poten- national consortia have conducted
tial functions of microbes in marine environments, including symbiotic and pathogenic inter- many metagenomic surveys. The
actions. There are many examples of ecologically important processes discovered through international Malaspina circum-
metagenomics throughout the book. For example, one of the first successes of this approach navigation expedition (2010–2011)
was the identification of proteorhodopsin genes revealing a new kind of photoheterotrophy in added many microbial sequences
marine bacteria, discussed in Box 3.1. Other notable discoveries include chemosynthetic sym- from the deep ocean. Perhaps the
most productive of these coordi-
bionts of invertebrates (Chapter 10), nitrogen-fixing bacteria, and ammonia-oxidizing archaea
nated research endeavors is the
(Chapter 9), to name just a few. Assembling genomes from uncultivated organisms follows the
international Tara Oceans project
same basic principles as that for cultivated organisms, with the important additional steps of from 2009–2012, which collected
binning the sequences and annotation of gene function (Figures 2.9, 2.11 and 2.12). over 3.5 × 104 samples for diversity
studies from surface and mesope-
A range of software programs are available for assembling the sequences into closed partial lagic waters at 210 globally dis-
or complete genomes via “binning.” The key step of binning involves sorting the assem- tributed sites. An initial analysis of
bled contigs into the best estimate of groups that might belong together. If microdiversity 7.2 × 1012 trillion bases of genomic
is not too high, this gives a starting point for deriving the genome of a single population or data were assembled into an inven-
group of closely related populations. A common approach is to assess the GC content or the tory of 40 million genes—of which
over 80% were novel. Hundreds
of major research papers based on
these samples have already been
published and these are having
major influences on our under-
standing of microbial ecology and
the effects of climate change.

Figure 2.11 Overall workflow

and bioinformatics tools for shot-
gun metagenomic analysis. HMM,
hidden Markov model; KEGG,
Kyoto Encyclopedia of Genes and
Genomes; STRING, Search Tool for
the Retrieval of Interacting Genes/
Protein. Reprinted from Kim et al.
(2013), CC-BY-3.0.
56 Chapter 2

Figure 2.12 A selection of the sci-

entific and technological advances in
the recent history of marine micro-
bial ecology are represented through
time in the upper panel. Reprinted
from Salazar and Sunagawa (2017)
with permission of Elsevier.

abundance distribution of k-mers (all the possible subsequences of a specific length, which
will be characteristic of particular organisms. Many different systems (“pipelines”) are avail-
able for sequence assembly, binning, and downstream analysis. The MEtaGenome ANalyzer
(MEGAN) software is a well-established tool for analyzing HTS output, using BLAST
analysis for taxonomic and functional binning. The faster methods work by co-assembling
sequences with guidance from reference genomes though BLAST analysis for similarity, but
larger computational resources and several days processing time on specialized servers are
needed for de novo assembly. The expansion of metagenomics has resulted in accumulation
of massive amounts of data. Archiving and curating sequences, together with the associated
metadata such as descriptions of the sample source and details of methodology, is carried out
by specialized databases such the Genomes OnLine Database (GOLD). This is essential to
ensure conformity, reproducibility, and as a reference for other studies. MEGAN and other
pipelines are used for comparative analysis of multiple datasets, using databases such as
SEED, which organizes categories of gene functions into a hierarchical classification, and the
Kyoto Encyclopedia for Genes and Genomes (KEGG), a collection of pathways integrating
genes, proteins, RNAs, and metabolic pathways. A flowchart of the major steps in analysis of
metagenomic data is shown in Figure 2.11.

Genomes can now be obtained from single

cells in environmental samples
It is important to recognize that a genome constructed from metagenomic sequences extracted
from the environment (a Metagenome Assembled Genome, MAG) is not the same as a genome
constructed from a clonal population of a cultured organism. In view of the microdiversity
within natural populations of bacteria and archaea, even the highest quality MAGs will likely
represent an agglomeration of very closely related individuals. However, it is now possible to
isolate a single microbial cell from the environment, amplify the whole genome, sequence it
with HTS, and reconstruct the genome. This is known as a Single (cell) Amplified Genome or
SAG. Whole genome amplification (WGA) is carried out using a thermostable mutant form
of the high-fidelity DNA polymerase obtained from φ29 phage. The process is known as
multiple displacement amplification—unlike PCR, it uses random primers—resulting in large
fragments with low frequency of errors. Femtogram amounts of DNA from a single cell can
be amplified over 1000-fold to provide the amount needed for HTS. There are several sources
of error during the process, such as uneven genome coverage and chimera formation, which
must be accounted for and overcome. Many other factors, such as nature of the sample, DNA
extraction and preservation, nature of the cell wall, and packaging of DNA, can also affect the
 Methods in Marine Microbiology 57

quality of the results. Appropriate methods for lysis of the cells under investigation must be
VIRAL
used, with precautions needed to prevent DNA damage, contamination, or residues of chemi-
cals that will interfere with amplification. Despite these difficulties, single-cell genomics has i METAGENOMICS:
THE LONG AND
been successfully used in numerous studies of bacteria and archaea in the marine environ-
THE SHORT OF IT
ment. It has proved especially important in the discovery of new processes in the carbon and
nitrogen cycles (Chapters 8 and 9). It is now being used to obtain whole genome sequences The great advances in understand-
from protists directly isolated from the environment, which has previously been very difficult ing of marine bacterial ecology
(Box 6.1). FACS is the most commonly used method for isolating single cells; picoliter-sized through HTS and metagenomics
droplets can be sorted into tubes or microwell plates. Another method is optofluidics, in which have been possible because of the
universal rRNA marker genes, but
cells are captured by using a microfluidic chip and examined microscopically to select those
there is no equivalent in viruses.
of interest; these are moved into another compartment of the microfluidic device for WGA
HTS with Illumina sequenc-
cells. This is particularly valuable for samples with a low biomass. The gel microencapsulation ing technology produces short
technique described on p.41 has also been adapted for single-cell genomics. sequences, which are very useful
for predicting viral gene functions,
but metagenome assembly is dif-
IN SITU ACTIVITY OF MICROBIAL COMMUNITIES ficult because of nucleotide level
diversity among viral strains. The
Microelectrodes and biosensors measure advent of long-read HTS technol-
microbial processes at the microhabitat scale ogy such as PacBio and MinION
offers the potential to directly cap-
Special electrochemical microelectrodes can be constructed with tip diameters of 10 µm or ture complete genomes of many
less that are inserted using micromanipulators into habitats to measure environmental and marine viruses, but these both
chemical changes. They are made of glass and metal with an ion-selective membrane that require large amounts of input
generates voltage by charge separation. Some have liquid ion-exchange membranes incorpo- DNA, which is challenging for
rating a synthetic ionophore, which is specific for target compounds. These probes can be left natural viral community samples.
in place for long periods, and when linked to recorders, they provide continuous data about Long-read technologies also suffer
from high error rates, which are
environmental changes and the rates of microbial and chemical processes in microhabitats.
particularly problematic in viral
For example, the development and use of oxygen, sulfide, and pH microsensors made it pos-
metagenomics due to the shorter
sible to analyze the gradients of these factors in sediments and biofilms, which were used viral gene length compared to host
to show how sulfur bacteria (p.80) live in opposing gradients of sulfide and oxygen with genes, leading to frameshift errors.
only very small concentrations of both chemical species being present in the oxidation zone. Warwick-Dugdale et al. (2019) of
Microsensor techniques can also be used to investigate oxygenic photosynthesis and respira- Plymouth Marine Laboratory and
tion at a micrometer resolution in biofilms and sediments. Optical microsensors that detect the University of Exeter adapted a
small changes in light are also used. Recent innovations have included the development of method to amplify viral DNA from
microscale biosensors suitable for use in seawater and marine sediments to detect a range of seawater and obtained long reads
compounds. Here, biological components such as immobilized enzymes or antibodies are with MinION. A hybrid approach
that employed short reads for long-
incorporated into the electrodes and coupled to a signal converter so that the results of a
read error correction, which they
reaction are amplified to produce an electronic reading. The development of such biosensors
call VirION, improved the recovery
for various inorganic nitrogen species has been a major advance, permitting detailed study of complete viral genomes and
of the processes of nitrification and denitrification. Biosensors for methane and volatile fatty highly microdiverse virus popu-
acids have also been used to probe microbial transformations in anaerobic sediments and lations from seawater from the
planktonic aggregates such as marine snow particles. Studies using oxygen microelectrodes Western Channel Observatory.
have shown that even the tiniest aggregates (Figure 1.7) contain anaerobic niches.

Radioisotopes can be used to detect

metabolic activity in a community
Specific metabolic transformations can be studied by measuring the rate of incorporation of a
radioactively labeled precursor of a key substrate. This method is widely used in biochemistry to
follow the pathways of metabolic reactions—if one or more atom(s) of a compound is replaced
with a radioisotope, we can track its progress as it is metabolized. One of the earliest applica-
tions of radioactive isotopes in marine microbial ecology was the introduction of 14C-labeled
bicarbonate (H14CO3−) into bottle experiments to measure rates of photosynthesis (primary pro-
duction) by phytoplankton. 14C is a relatively safe isotope, because it has little energy and a long
half-life. Bottles containing phytoplankton can be positioned on a mooring line at various depths
to assess the effects of irradiance and temperature on productivity, with appropriate control bot-
tles from which light is excluded. Phytoplankton cells can be collected by filtration after incuba-
tion, and the amount of labeled carbon fixed can be detected by measuring radioactive decay in
a scintillation counter, or by autoradiography. Bacterial production in seawater can be measured
by the incorporation of [3H]-adenine or [3H]-thymidine into RNA or DNA (respectively), or
58 Chapter 2

by incorporation of labeled amino acids such as [3H]-leucine into proteins. The flux of organic
material in seawater can be measured using [14C]-glucose. Many other microbial processes can
be measured using appropriately labeled compounds. Other commonly used radioisotopes for
this type of study are 32P, 18O, and 35S. Appropriate correction factors for the discrimination of
enzyme processes against the heavier radioisotopes must be included in calculations.

Incorporation of the labeled precursor into individual cells can also be detected by microau-
toradiography (MAR), allowing us to assess the metabolic activities of different cells within
a community. Radiolabeled substrate is added to a sample of a natural community and incu-
bated—it could be a water sample, sediment, or animal tissue, for example—so that cells
that incorporate the labeled compound become radioactive after incubation. Samples are then
placed onto microscope slides or membranes and overlaid with a liquid photographic emulsion
containing silver nitrate. After a few days, the film is examined microscopically. Radioactive
decay will expose the film by deposition of silver salts, so in cells where the radioisotope has
been incorporated, silver grains will become visible as dark spots under the light microscope.
MAR is a long-established technique that has found many new applications in marine ecology
in combination with the gene-based technique of FISH, discussed earlier. This combination of
methods, called MAR-FISH, is extremely sensitive and is a powerful tool for revealing which
organisms in a community are carrying out particular metabolic processes. Unlike some other
methods, it is not subject to the biases that result from DNA extraction or PCR methods. In
MAR-FISH, the incubated sample is treated with a permeabilizing agent and FISH oligonucle-
otides probes are added before adding the MAR photographic emulsion. Duplicate images of
the resulting exposed specimen are examined. Under the light microscope, the isotope-labeled
cells can be seen while the FISH-labeled cells are visible under the fluorescence microscope.
The images can be overlaid digitally so that we can identify the organisms carrying out the
metabolic reaction under study. In most cases, the FISH probe will be an oligonucleotide
recognizing an rRNA target, which means that we can assign a taxonomic affiliation to active
cells. An important example of the use of MAR-FISH is the discovery of chemolithotrophic
fixation of CO2 by deep-sea archaea, described in Box 8.1.

Stable-isotope probing (SIP) tracks fluxes

of nutrients in communities
To follow the flux of major nutrients, particularly carbon and nitrogen, another widely used
approach is to supply the community with a substrate enriched with a heavier stable iso-
tope—13C and 15N are most commonly used—and to couple this with cultivation-independent
methods to identify the organism. SIP provides direct evidence linking substrate assimilation
to specific taxa, thus indicating their functional roles in the community. Usually, rRNA or
DNA is used as the biomarker. For example, to find out which organisms in a mixed commu-
nity are methanotrophs (consuming compounds with a single C atom [C1] such as methane
or methanol, see p.90) we can add C1 compounds containing a high proportion of 13C atoms.
Since 13C has a higher mass than the commoner 12C, those organisms that are incorporating
C1 compounds will contain DNA with a higher density than others. The heavy and light frac-
tions of DNA can be separated by isopycnic centrifugation at high speed in a cesium chloride
gradient. Ethidium bromide or another DNA stain is included so that the bands can be visual-
ized and removed from the centrifuge tube with a needle. It can then be used for community
analysis using PCR with universal primers or subjected to metagenomic sequencing by HTS.
If the taxonomic identity of active groups is the goal, tracking incorporation of the isotope
into rRNA has some advantages, because rates of rRNA synthesis are usually higher than
DNA and are independent of replication. However, an additional step of reverse transcription
to make cDNA will be needed before performing PCR. Isotope incorporation into proteins,
lipid fatty acids, and metabolites can also be used to indicate a specific metabolic activity.

NanoSIMS allows metabolic transfers to

be measured at subcellular levels
Although the SIP methods described above allow us to determine groups of organisms
responsible for metabolic processes, they do not resolve the activities of individual cells, or
compartments within cells. A technique called nanoscale secondary ion mass spectrometry
 Methods in Marine Microbiology 59

(nanoSIMS) is a type of imaging system that enables elemental transformations to be moni-

tored at subcellular resolution. This technique has proved particularly valuable in combina-
tion with SIP for determining interactions and nutrient flow between associated microbes,
or between symbiotic (or parasitic) microbes and their hosts. Several important discoveries
using this technique are described in later chapters, including the assimilation and transfer
of nitrogen and carbon (15NH4+, 15N2, 13CH4) in syntrophic consortia of methane-oxidizing
archaea and sulfate-reducing bacteria in sediments (Figure 3.13C), 15N2 fixation by intracel-
lular cyanobacterial endosymbionts of diatoms (Figure 4.11), and chemolithotrophic symbi-
onts of animals (Chapter 10).

In nanoSIMS, a beam of Cs+ or O − (the primary ion) is focused onto the surface of a sample
with a resolution of 50 nm or less. The ions cause “sputtering” of a few atomic layers from
the surface, and clusters of atoms are ejected; some of them become spontaneously ionized.
The composition of these secondary ions is characteristic of the composition of the analyzed
area. The secondary ion beam is mass filtered by a magnetic device, resulting in separation
of up to seven different types of ion, each of which can be measured independently. The
nanoSIMS-SIP technique can be made even more powerful by combining it with FISH or
CARD-FISH, allowing phylogenetic identification of cells and subcellular structures con-
taining the stable isotope. In a variant called HISH-nanoSIMS (halogen in situ hybridiza-
tion), a rare element such as 19F attached to oligonucleotide probes is introduced directly
into the cells. Fluorescent antibodies can also be introduced to detect the location of specific
proteins.

Microarrays enable assessment of gene

activity in the environment
One of the major methods used in functional genomics is DNA microarray technology. DNA
oligonucleotide sequences identified from microbial genomes are attached using photolithog-
raphy or ink-jet technology on the surface of a silicon chip in a highly ordered array to pro-
duce a device that can act as a probe for hundreds or thousands of genes. For some bacteria,
chips with probes for every expressed gene in the genome have been designed. The target
nucleic acids (mRNA or cDNA) are labeled with fluorescent reporter groups and lasers detect
hybridization of complementary sequences to the oligonucleotide on the microarray chip.
Microarrays are particularly useful for determining the effects of changing conditions on the
patterns of gene expression, for example, following nutrient changes in mesocosm experi-
ments or during different phases of interactions with a host. The results of such experiments
are often presented as color-coded heat maps indicating which genes are upregulated, down-
regulated, or unaltered during the transition. They have also been used to follow community
changes by phylogenetic identification. The PhyloChip assay contains ~1.2 × 106 probes for
variable regions of the 16S rRNA gene, enabling identification and measurement of relative
abundance of over 5 × 104 microbial taxa; notable examples of its use include the assessment
of healthy and disease states in corals (Chapter 11) and the rapid assessment of community
changes during pollution incidents (Chapter 13). Microarrays have also been developed for
high-throughput analysis of functional genes in community analysis. The GeoChip 5.0 micro-
array contains ~1.7 × 105 probes detecting sequences from >1500 gene families related to car-
bon, nitrogen, sulfur and phosphorus cycling, energy metabolism, antibiotic resistance, metal
resistance/reduction, organic remediation, stress responses, bacteriophage, and virulence.

Metatranscriptomics, metaproteomics, and metabolomics

reveal microbial activities in the environment
While knowledge of the metagenome indicates the potential activities within a microbial
community, it does not reveal the expression of genes and hence their actual involvement in
particular processes. Metatranscriptomics allow the analysis of expressed genes by measur-
ing production of messenger RNA (mRNA). Because mRNA has a short half-life, it provides
a near real-time picture of microbial activity. Samples can be frozen or preserved rapidly,
so that they can be easily collected over time and analyzed later. As shown in Figure 2.9,
metatranscriptomics involves the isolation of mRNA and the production of a complemen-
tary DNA (cDNA) copy using the enzyme reverse transcriptase. Specific gene transcripts
60 Chapter 2

can be identified using microarray hybridization. A more commonly applied technique is

to sequence short segments (500–800 bp) of the transcribed cDNA; these are known as
expressed sequence tags (ESTs). Databases are built up to link ESTs with specific gene func-
tions. Metatranscriptomes may be mapped onto metagenomic bins to provide functional
information on metagenome assembled genomes (MAGs).

Even knowing which genes are expressed is one step short of knowing which enzymes and
other proteins are present in a community, because these may all be modified after transla-
Figure 2.13 An example of a tion owing to complex regulatory processes. Proteomics enables separation of individual
mesocosm enclosure in a fjord at the proteins by two-dimensional separation of the proteins present in a mixed sample using a
Espegrend Marine Research Field combination of SDS-PAGE and isoelectric focusing. The protein spots are digested in the
Station at the University of Bergen, electrophoresis gel and analyzed by mass spectrometry. Peptide patterns can be compared
Norway, during an experiment with those in databases, or full protein sequences can be obtained and homology with well-
conducted in the spring of 2017 characterized proteins can be determined. Metabolomics aims to assess the total metabolic
led by Kay D. Bidle and Kimberlee profile of a cell, organism, or community. This usually involves chromatographic separation
Thamatrakoln (Rutgers University, of different classes of chemical compounds (amino acids, carbohydrates, lipids, etc.) and
NJ and Elizabeth Harvey (University identification by mass spectrometry. New developments allow direct identification without
of Georgia). In this study, 12 meso- the need for separation.
cosms consisted of plastic floating
frames (~2 m diameter) attached to
bags constructed of polyethylene, Microfluidics enables study of microscale processes
extending 8 m below the surface;
Microfluidics is a technology for handling and controlling fluids in microliter or picoliter
each contained ~23,000 L of fjord
quantities. It was developed in the 1990s by combining microanalytical and microelectronic
water. Each bag tapered into a set-
tling cone attached to a bottle to circuits and has expanded due to the ability to fabricate microdevices using soft lithography.
collect sinking particles. After the By combining it with microscopic examination, microfluidics allows us to study the behavior
addition of nutrients to the enclo- of individual microbial cells or groups of cells, and to manipulate their environment at a scale
sures, they were mixed with air bub- appropriate to their size. Nano-sized channels and porous membranes can be used to control
blers for several days and subsequent the flow of nutrients and excretion products from cells while keeping the microbial cells in
sampling from various depths was a fixed space. Application of this technology in microbial ecology has been championed by
done using pumps. Bottles contain- Roman Stocker and colleagues at MIT and ETH Zurich, resulting in many important find-
ing sinking material were collected ings related to the behavior of microbes in a “sea of gradients” and the effects of surface and
and replaced by scuba divers boundary effects. Notable examples considered in detail elsewhere in the book involve the
throughout the experiment. Credit: motility and chemotaxis of bacteria in response to nutrient gradients (Box 3.1) and its role in
Kay D. Bidle (Rutgers) and Jozef I. infection by coral pathogenic bacteria (Box 11.1).
Nissimov (Scottish Institute of Marine
Science, Oban).
 Methods in Marine Microbiology 61

Mesocosm experiments attempt to

simulate natural conditions
Carrying out controlled experiments is essential to understand the effects of factors such as
nutrient additions, light, and temperature on community dynamics and microbial processes
in the marine environment. Laboratory investigation of microbial processes in water uses
bottle experiments, in which volumes of water up to a few liters can be handled. Experimental
bottles can also be attached to CTD frames or mooring lines for incubation in situ at differ-
ent depths. Use of these techniques has led to great advances in our understanding of pho-
totrophic and heterotrophic processes, for example, by comparison of metabolic activities of
samples incubated under light and dark conditions or the effects of temperature and nutri-
ent additions. Such small-scale experiments are a vital first stage in the study of microbial
processes in marine samples but extending studies to the natural environment is difficult.
Great care is needed to ensure that vessels used in these experiments are sterile and free of
chemical contaminants. Adsorption effects on the walls of containers can result in the local
concentration of nutrients, which affects the composition and metabolic properties of the
community. In general, the smaller the sample, the more likely it is that significant distortions
due to these “bottle effects” will occur.

One way to overcome these problems is to use mesocosms—enclosures holding thousands

of liters of seawater under semi-controlled conditions in an attempt to simulate natural sea
environments or conduct experimental manipulations. Large tanks containing coastal seawa-
ter pumped in at controlled rates can be used in shore-based experimental stations. However,
a more successful approach is the use of large enclosures constructed of polythene rein-
forced with vinyl and nylon—essentially, these are very big and very strong plastic bags
CAN BACTERIAL
?
immersed in the sea. The materials must be chosen carefully to simulate natural conditions as
closely as possible. One of the longest-running facilities is based at the University of Bergen, LIGHT BE SEEN
Norway, which has maintained mesocosm systems available for international experiments FROM SPACE?
since 1978 (Figure 2.13). Mesocosm bags are suspended in the relatively calm, deep waters Miller et al. (2005) asked this
of a fjord. The bags are filled by pumping in unfiltered fjord water so that natural communi- question after looking at reports
ties of phytoplankton, zooplankton, bacteria, and viruses are introduced. Some success has of the “milky sea” phenomenon
been achieved in the development of very large mesocosm containers (up to 35 m3) for use in well known to sailors in the Arabian
open ocean conditions, although there are major technical difficulties in the construction of Sea and featured in the classic
enclosures strong enough to withstand waves and currents. Jules Verne novel Twenty Thousand
Leagues under the Sea, published
Generally, it is impractical to conduct experiments in the open sea because of the continuous in 1870. Miller and colleagues
obtained the location and time of
turbulence and dispersion of water and the organisms it contains. There are also consider-
a milky sea occurrence reported
able logistical problems in conducting research at sites sufficiently far from the continental by a ship’s captain some years
shelf to avoid land effects. However, a notable exception is the series of experiments carried earlier and noted this record in the
out from 1993–2009 to investigate the effect of iron additions on phytoplankton composition ship’s log: “The bioluminescence
and productivity. These experiments were possible because of the deployment of an inert appeared to cover the entire sea
tracer compound (sulfur hexafluoride, SF6). The dispersion of SF6 can be measured ana- area, from horizon to horizon …
lytically, thus establishing the degree of dilution of iron in the water mass under study. The and it appeared as though the ship
significance of these experiments and plans for further large-scale investigations involving was sailing over a field of snow or
measurement of microbial parameters are discussed in Box 8.1. gliding over the clouds” (Nealson
and Hastings, 2006). Miller et al.
then checked the output from
the meteorological satellites that
Remote sensing permits global monitor global cloud coverage
analysis of microbial activities under daytime and nighttime
conditions. After adjusting spectral
At the opposite extreme from techniques to measure microenvironmental changes, the frequencies of the output, they
use of remote sensing by orbiting satellites has provided valuable information about the produced images (Figure 4.6) of the
activities of planktonic organisms in the surface layers of the oceans. While the ocean milky sea, showing that it covered
appears blue to the human eye, sensitive instruments detect subtle changes in ocean color >15000 km2; they deduced that
due to the presence of plankton, suspended sediments, photosynthetic pigments, and dis- this was bacterial bioluminescence,
solved organic chemicals. The first such satellite was the Coastal Zone Color Scanner probably due to Vibrio harveyi in
(CZCS) launched in 1978, which collected light reflected from the sea surface at differ- association with massive blooms of
Phaeocystis algae. They estimated
ent wavelengths, including the absorption maximum for chlorophyll. Scanning has been
the total bioluminescent bacterial
continued using the Sea-viewing Wide Field-of-View Sensor (SeaWiFS) system and other population of this milky sea to be
satellites. Figure 8.2 shows a typical image from SeaWiFS, revealing chlorophyll levels about 4 × 1022 cells.
62 Chapter 2

for the monitoring of phytoplankton productivity. Satellite-based remote sensing is now

used to measure a wide range of chemical and physical properties of the global ocean and
can distinguish functional groups of microorganisms. Interpretation of the sensor data is
a specialist field, with many different algorithms devised to interpret and display results.
Different spatial resolution can be used to zoom in on particular areas, to identify pro-
cesses and hazards influencing pelagic and coastal waters. For example, it can be used to
monitor natural phytoplankton blooms (Figure 6.9) or track the development of harmful
algal blooms, as discussed in Chapter 11.

Conclusions
This chapter has shown that marine microbiologists use a very wide range of methods to
study the diversity and activity of microbes, and examples of their application in research
will be found throughout the book. We now have techniques enabling us to explore the prop-
erties and activities of marine microbes at every level from single cells to global assemblages.
The most significant advances in marine microbiology have occurred as a result of the devel-
opment of new molecular methods and exciting technological developments, but we are also
seeing a renaissance of longer-established techniques such as culturing and microscopy that
provide essential insights that complement the molecular revolution. The next few years will
undoubtedly see further technological progress and will be dominated by increasing applica-
tion of metagenomics, transcriptomics, and proteomics, plus major advances in single-cell
biology. The challenge for microbiologists will be to manage and interpret the huge datasets
that are now accumulating and to translate them into real understanding of the ecology of
marine microbes and their role in diverse activities.

References and further reading Cultivation

Button, D.K., Schut, F., Quang, P., et al. (1993) Viability and isolation of
Sampling
marine bacteria by dilution culture: Theory, procedures and initial results.
Conathon, M. (2013) Rockets top submarines: Space exploration dollars Appl. Environ. Microbiol. 59: 881–891.
dwarf ocean spending. Center for American Progress. Online at https:// Carini, P. (2019) A “cultural” renaissance: Genomics breathes new life
www.americanprogress.org/issues/green/news/2013/06/18/66956/rock- into an old craft. mSystems 4: e00092-19.
ets-top-submarines-space-exploration-dollars-dwarf-ocean-spending/
(accessed 6 June 2019). Connon, S.A. & Giovannoni, S.J. (2002) High-throughput methods for
Moniz, R. & Coley, K. (2017) Challenges in the deep: The role of manned culturing microorganisms in very low-nutrient media yield diverse new
submersibles in an autonomous age. Online at https://www.oceannews. marine isolates. Appl. Environ. Microbiol. 68: 3878–3885.
com/featured-stories/challenges-in-the-deep-the-role-of-manned-sub- Donachie, S.P., Foster, J.S., & Brown, M. V (2007) Culture clash: Challenging
mersibles-in-an-autonomous-age (accessed 6 June 2019). the dogma of microbial diversity—Commentaries. ISME J. 1: 97–99.
Rahlff, J., Stolle, C., Wurl, O., et al. (2017) SISI: A new device for in situ Giovannoni, S.J., Foster, R.A., Rappé, M.S. & Epstein, S. (2007) New cul-
incubations at the ocean surface. J. Mar. Sci. Eng. 5: 46. tivation strategies bring more microbial plankton species into the labora-
Imaging and flow cytometry tory. Oceanography 20: 62–69.
Giovannoni, S. & Stingl, U. (2007) The importance of culturing bacterio-
Amann, R. & Fuchs, B.M. (2008) Single-cell identification in microbial
plankton in the ‘omics’ age. Nat. Rev. Microbiol. 5: 820–825.
communities by improved fluorescence in situ hybridization techniques.
Nat. Rev. Microbiol. 6: 339–348. Giovannoni, S.J., Tripp, H.J., Givan, S., et al. (2005) Genetics: Genome
streamlining in a cosmopolitan oceanic bacterium. Science 309: 1242–1245.
Chisholm, S.W., Olson, R.J., Zettler, E.R., et al. (1988) A novel free-living
prochlorophyte abundant in the oceanic euphotic zone. Nature 334: 340. Morris, R.M., Rappe, M.S., Connon, S.A., et al. (2002) SAR11 clade domi-
nates ocean surface bacterioplankton communities. Nature 420: 806–810.
Johns, C.T., Nissimov, J.I., Grubb, A., et al. (2019) The mutual interplay
between calcification and coccolithovirus infection. Environ. Microbiol. Rappé, M.S., Connon, S.A., Vergin, K.L. & Giovannoni, S.J. (2002)
21: 1896–1915. Cultivation of the ubiquitous SAR11 marine bacterioplankton clade.
Nature 418: 630–632.
Malfatti, F. & Azam, F. (2009) Atomic force microscopy reveals
microscale networks and possible symbioses among pelagic marine bacte- Schut, F., Gottschal, J.C. & Prins, R.A. (1997) Isolation and characterisa-
ria. Aquat. Microb. Ecol. 58: 1–14. tion of the marine ultramicrobacterium Sphingomonas sp. strain RB2256.
FEMS Microbiol. Rev. 20: 363–369.
Partensky, F. & Garczarek, L. (2010) Prochlorococcus: Advantages and
limits of minimalism. Ann. Rev. Mar. Sci. 2: 305–331. Stingl, U., Tripp, H.J., & Giovannoni, S.J. (2007) Improvements of high-
throughput culturing yielded novel SAR11 strains and other abundant
Pennisi, E. (2017) Meet the obscure microbe that influences climate, ocean
marine bacteria from the Oregon coast and the Bermuda Atlantic Time
ecosystems, and perhaps even evolution. Science. doi: 10.1126/science.
Series study site. ISME J. 1: 361–371.
aal0873.
 Methods in Marine Microbiology 63

Zengler, K. (2009) Central role of the cell in microbial ecology. Microbiol. Sleigh, C. (2013) The secret life of the laboratory: PCR at 30. The
Mol. Biol. Rev. 73: 712–729. Guardian. Online at https://www.theguardian.com/science/2013/
nov/26/the-secret-life-of-the-laboratory-pcr-at-30 (accessed 7 June
Zengler, K., Toledo, G., Rappé, M., et al. (2002) Cultivating the uncul-
2019).
tured. Proc. Natl. Acad. Sci. USA 99: 15681–15686.
Stepanauskas, R. (2012) Single cell genomics: An individual look at
RNA- and DNA-based methods microbes. Curr. Opin. Microbiol. 15: 613–620.

Bourne, D.G. & Munn, C.B. (2005) Diversity of bacteria associated with Thomas, T., Gilbert, J., & Meyer, F. (2012) Metagenomics—A guide from
the coral Pocillopora damicornis from the Great Barrier Reef. Environ. sampling to data analysis. Microb. Inform. Exp. 2: 3.
Microbiol. 7: 1162–1174. Thompson, L.R., Sanders, J.G., McDonald, D., et al. (2017) A commu-
Compeau, P.E., Pevzner, P.A., & Tesler, G. (2011) Why are de Bruijn nal catalogue reveals Earth’s multiscale microbial diversity. Nature,
graphs useful for genome assembly?. Nat. Biotechnol. 29: 987. 551(7681): 457.

Coutinho, F.H., Gregoracci, G.B., Walter, J.M., et al. (2018) Metagenomics Warwick-Dugdale, J., Solonenko, N., Moore, K., et al. (2019) Long-read
sheds light on the ecology of marine microbes and their viruses. Trends viral metagenomics captures abundant and microdiverse viral populations
Microbiol. 26: 955–965. and their niche-defining genomic islands. PeerJ 7: e6800.

Escobar-Zepeda, A., Vera-Ponce de León, A., & Sanchez-Flores, A.

In situ activity
(2015) The road to metagenomics: From microbiology to DNA sequenc-
ing technologies and bioinformatics. Front. Genet. 6: 348. Boughner, L.A. & Singh, P. (2016) Microbial ecology: Where are we now?
Giovannoni, S.J., Britschgi, T.B., Moyer, C.L., & Field, K.G. (1990) Postdoc. J. 4: 3–17.
Genetic diversity in Sargasso Sea bacterioplankton. Nature 345: 60–63. Morando, M. & Capone, D.G. (2016) Intraclade heterogeneity in nitro-
Heather, J.M. & Chain, B. (2016) The sequence of sequencers: The history gen utilization by marine prokaryotes revealed using stable isotope
of sequencing DNA. Genomics 107: 1–8. probing coupled with tag sequencing (Tag-SIP). Front. Microbiol. 7:
1932.
Hiraoka, S., Yang, C. & Iwasaki, W. (2016) Metagenomics and bioin-
formatics in microbial ecology: Current status and beyond. Microbes Musat, N., Musat, F., Weber, P.K., & Pett-Ridge, J. (2016) Tracking micro-
Environ. 31: 204. bial interactions with NanoSIMS. Curr. Opin. Biotechnol. 41: 114–121.

Hongoh, Y. & Toyoda, A. (2011) Whole-genome sequencing of uncultur- Nagy, K., Ábrahám, Á., Keymer, J.E., & Galajda, P. (2018) Application of
able bacterium using whole-genome amplification. Methods Mol. Biol. microfluidics in experimental ecology: The importance of being spatial.
733: 25–33. Front. Microbiol. 9: 496.

Hugerth, L.W. & Andersson, A.F. (2017) Analysing microbial community Radajewski, S., Ineson, P., Parekh, N.R., & Murrell, J.C. (2000) Stable-
composition through amplicon sequencing: From sampling to hypothesis isotope probing as a tool in microbial ecology. Nature 403: 646.
testing. Front. Microbiol. 8: 1561. Rusconi, R., Garren, M., & Stocker, R. (2014) Microfluidics expanding the
Kim, M., Lee, K.-H., Yoon, S.-W., et al. (2013) Analytical tools and data- frontiers of microbial ecology. Annu. Rev. Biophys. 43: 65–91.
bases for metagenomics in the next-generation sequencing era. Genom. Rusconi, R. & Stocker, R. (2015) Microbes in flow. Curr. Opin. Microbiol.
Inform. 11: 102. 25: 1–8.
Kozińska, A., Seweryn, P., & Sitkiewicz, I. (2019) A crash course in Staley, J.T. & Konopka, A. (1985) Measurement of in Situ activities of non-
sequencing for a microbiologist. J. Appl. Genet. 60: 103–111. photosynthetic microorganisms in aquatic and terrestrial habitats. Annu.
Land, M., Hauser, L., Jun, S.-R., et al. (2015) Insights from 20 years of Rev. Microbiol. 39: 321–346.
bacterial genome sequencing. Funct. Integr. Genom. 15: 141–61. Taylor, G.T. (2019) Windows into microbial seascapes: Advances in
Logares, R., Sunagawa, S., Salazar, G., et al. (2014) Metagenomic 16S nanoscale imaging and application to marine sciences. Ann. Rev. Mar.
rDNA Illumina tags are a powerful alternative to amplicon sequencing Sci. 11: 465–490.
to explore diversity and structure of microbial communities. Environ. Tu, Q., Yu, H., He, Z., et al. (2014) GeoChip 4: A functional gene-array-
Microbiol. 16: 2659–71. based high-throughput environmental technology for microbial commu-
Pedros-Alio, C., Acinas, S.G., Logares, R., & Massana, R. (2018) Marine nity analysis. Mol. Ecol. Resour. 14: 914–928.
microbial diversity as seen by high-throughput sequencing. In, Microbial Zark, M., Broda, N.K., Hornick, T., et al. (2017) Ocean acidification
Ecology of the Oceans, Third edition., J.M. Gasol & D.L. Kirchman, eds., experiments in large-scale mesocosms reveal similar dynamics of dis-
pp. 47–97. John Wiley & Sons. solved organic matter production and biotransformation. Front. Mar. Sci.
Rhoads, A. & Au, K.F. (2015) PacBio sequencing and its applications. 4: 271.
Genom. Proteom. Bioinf. 13: 278–289.
Roh, S.W., Abell, G.C.J., Kim, K.-H., et al. (2010) Comparing microar- Remote sensing
rays and next-generation sequencing technologies for microbial ecology Miller, S.D., Haddock, S.H.D., Elvidge, C.D. & Lee, T.F. (2005) Detection
research. Trends Biotechnol. 28: 291–9. of a bioluminescent milky sea from space. Proc. Natl Acad. Sci. USA 102:
Salazar, G. & Sunagawa, S. (2017) Marine microbial diversity. Curr. Biol. 14181–14184.
27: R489–R494. Nealson, K.H. & Hastings, J.W. (2006) Quorum sensing on a global scale:
Singer, E., Wagner, M., & Woyke, T. (2017) Capturing the genetic makeup Massive numbers of bioluminescent bacteria make milky seas. Appl.
of the active microbiome in situ. ISME J. 11: 1949–1963. Environ. Microbiol. 72: 2295–2297.
Chapter 3
Metabolic Diversity
and Ecophysiology

Microbes have been evolving for between three and four billion years, resulting in a huge
diversity of metabolic types. Organisms have evolved to obtain energy from light or vari-
ous inorganic substances, or by breaking down different organic compounds to their basic
constituents. Some species obtain key elements for building cellular material directly from
inorganic minerals, while others require complex “ready-made” organic compounds. The
selective advantage of the ability to utilize particular nutrients under a set of physical condi-
tions has led to the evolution of the enormous metabolic diversity that we see today. Such
diversity results in the occurrence of microbes in every conceivable habitat for life. Microbial
metabolism fuels biogeochemical cycles, transformation of nutrients, and other processes
that maintain the web of life in the ocean, underpinning all aspects of marine ecology. The
first part of this chapter provides a brief overview of cell structure and genome organization.
This is followed by discussion of the diverse metabolic pathways by which microbes obtain
and transform energy and fix carbon and nitrogen for cellular growth. Next, the strategies
for nutrient acquisition and uptake are considered and the physiological factors affecting
microbial growth in plankton and on surfaces are discussed. The main focus of this chapter
is on the metabolism and ecophysiology of the Bacteria and Archaea, although some of the
biochemical pathways and mechanisms (especially photosynthesis and transformations of
organic molecules) are relevant to the eukaryotic fungi and protists, discussed in Chapter 6.

Key concepts
• Members of the Bacteria and Archaea share similar physiology and central metabolic
processes, although there are substantial differences in cell structure.
• Microbes demonstrate huge diversity in the routes by which they obtain nutrients,
extract energy, and produce the precursors needed for synthesis of macromolecules.
• All life depends on the activity of autotrophic organisms using light or chemical energy
to fix CO2 into cellular material.
• Bacteria, archaea, and protists in the photic zone harness energy from light by a range
of mechanisms to support autotrophic and heterotrophic metabolism.
• Chemolithotrophic activity of bacteria and archaea is widespread in the water column,
as well as in sediments, vents, and seeps.
• Diverse mechanisms are responsible for microbial transformations of carbon, sulfur,
and nitrogen.
66 Chapter 3

• Most planktonic microbes are genetically adapted to slow growth with minimum
nutrients, while others can exploit patches of high nutrient concentration.
• Microscale bacterial activities such as chemotaxis result in global scale effects on bio-
geochemical cycles.
• Intercellular communication and antagonism are major features of biofilm communities.
• The most abundant marine microbes are adapted to growth under conditions of low
temperature and high pressure.

A BRIEF OVERVIEW OF CELL

STRUCTURE AND FUNCTION
Bacteria and archaea show a variety of
cell forms and structural features
As introduced in Chapter 1, life may be organized into three phylogenetic domains, the
Bacteria, Archaea, and Eukarya. Members of the Bacteria and Archaea are usually con-
sidered to be unicellular organisms with a simple cell structure. The basic cell forms are
spherical or ovoid cells (cocci), rods (bacilli), and spiral cells (spirilla). Some groups have
distinctive morphologies such as filaments, tightly coiled spirals, or cells with buds, stalks,
or hyphae. As evident from Table 1.2, marine bacteria and archaea show enormous diversity
in cell size, ranging from about 0.1 μm to 750 μm in diameter. Cell size has a great effect on
metabolic and physiological processes in the cell.

The cytoplasmic membrane controls cell processes

via transport of ions and molecules
All cells possess a 6–8 nm-wide cytoplasmic membrane, illustrated in Figure 3.1. In bacterial
and eukaryotic cells, this is composed of a phospholipid bilayer containing hydrophobic fatty
acids linked to glycerol moieties via ester linkages, plus a wide range of proteins. Archaeal
membranes have a different structure and are more variable. The hydrophobic sidechains
are 5-carbon hydrocarbons units (isoprenoids) linked to glycerol via ether linkages. The
lipids can be either 20-carbon units (phytanyl), forming a bilayer, or 40-carbon diglycerol
tetraethers in which the phytanyl sidechains are covalently linked to form a monolayer. The
monolayer structure is more stable at high temperatures and is particularly prevalent in ther-
mophilic archaea, such as those living at hydrothermal vents. Hydrocarbon ring structures
also occur in different groups of Archaea.

phosphate
ether L-glycerol
isoprene sidechain
linkage
(a)

Figure 3.1 Structure of microbial

membranes. (a) Lipid structure
in Archaea, showing sidechains fatty acid sidechain ester
of isoprene units with ether link (b) linkage D-glycerol
phosphate
to ­glycerol phosphate. (b) Lipid
structure in Bacteria and Eukarya,
showing sidechains of fatty-acid units
with ester link to glycerol phosphate.
(c) Lipid bilayer of Bacteria, Eukarya,
and some Archaea (left), and the
lipid bilayer lipid monolayer
monolayer structure found in most
Archaea (right). (c)
 Metabolic Diversity and Ecophysiology 67

Bacterial membranes are stabilized by Mg2+ and Ca2+ and by pentacyclic compounds known
as hopanoids. These are resistant to degradation and large quantities have accumulated in the
environment over the many millennia of bacterial evolution. The total global mass of carbon
deposited in hopanoids is estimated at about 103 Gt (Pg), approximately the same as the mass
of organic carbon in all organisms living today. A very large fraction of fossil fuels such
as petroleum and coal are composed of hopanoids, much of which has accumulated from
the settlement of bacteria in the oceans. Because different bacteria produce characteristic
hopanoid structures, these compounds are frequently used as markers for the analysis of
microbial communities in marine sediments. Hopanoids have not been detected in archaea.

Despite these differences, the function of the cytoplasmic membrane is the same—acting
as a selectively permeable barrier controlling movement of ions and molecules in and out of
the cell. Transport mechanisms are needed for nutrient uptake, excretion, and secretion of
extracellular products. As well as controlling the entry/exit of substances to/from the cell by
its selective permeability and its specific transport proteins, the cytoplasmic membrane is
the most important site of energy generation and conservation. During metabolism, electron
carriers associated with the membrane bring about a charge separation across the membrane.
H+ ions (protons) become concentrated on the exterior surface of the membrane and OH− ions
concentrate on the inner surface. This charge separation creates an electrochemical potential
that can be converted to chemical energy. Bacteria and archaea show enormous variation in
the oxidation-reduction reactions used to produce this gradient, but the common “currency”
of energy exchange is always the molecule ATP used for synthesis and mechanical work by
the cell. When protons cross the membrane through a specific port, the contained energy is
captured in the conversion of ADP to ATP.

Cells may contain organelles,

microcompartments, and inclusion bodies
One of the original defining features of bacteria and archaea as “prokaryotic” cells (p.3) was
that they have a simple cellular organization with no membrane-bound nucleus or organ-
elles such as mitochondria or chloroplasts. However, many bacterial cells do show extensive
internal compartmentalization. Members of the phylum Planctomycetes clearly belong to the
Bacteria, but have a membrane-bound nucleus and some have a membrane-bound organelle
(the anammoxosome). Extensive invaginations or aggregates of membrane vesicles are espe-
cially important in phototrophs and chemolithotrophs with very high respiratory activity.
Phototrophs grown in low light intensity will produce more internal membranes and pig-
ments than those grown under high illumination, in order to harvest available light with the
greatest efficiency.

Gas vesicles are also important in the ecology of marine phototrophs such as Cyanobacteria.
These have a 2 nm thick, single-layered, gas-permeable wall composed of two hydrophobic
proteins surrounding a central space. These form a very strong structure into which gas
diffuses from the cytoplasm, resulting in increased buoyancy of the cell. This enables pho-
totrophic bacteria to maintain themselves in the desired zone of light intensity in the water
column. Organisms from deep habitats have narrow vesicles as they are more resistant to
hydrostatic pressure.

Many types of autotrophic bacteria contain icosahedral structures known as carboxysomes.

These are about 120 nm diameter and surrounded by a thin, selectively permeable envelope,
architecturally similar to the protein coat of viruses. They contain RuBisCO, the key enzyme
used by many autotrophs for the fixation of CO2 (p.83) and serve as concentration mecha-
nisms to enhance the efficiency of this process. Other similar structures act as centers for
certain catabolic reactions, such as aldehyde oxidation.

Many bacteria also contain granules of organic or inorganic material used as a store of energy
or structural components. Sometimes they occur as simple concentrations of material free
within the cytoplasm, but they are often enclosed in a thin, single-layered lipid membrane.
For example, a wide range of marine bacteria produce polyphosphate granules and carbon
and energy storage polymers such as polyhydroxyalkanoates. Large inclusions of elemental
liquid sulfur up to 1.0 μm in diameter occur in sulfur-oxidizing bacteria and can constitute up
68 Chapter 3

to 30% of the cell weight (see Figure 1.4C). Cyanobacteria also contain granules composed

?
CAN BACTERIA BE of large amounts of α-1,4-linked glucan, which is the polymerized carbohydrate product of
MULTICELLULAR? photosynthesis. They also contain granules composed of polypeptides rich in the amino acids
Our blinkered view of bacteria as arginine and aspartic acid, which appear to be produced as a reserve of nitrogen.
simple, independent unicellular
organisms grew from the clas-
sical emphasis on isolating pure
The nature of the cell envelope has a
cultures. But bacteria show many major effect on physiology
properties of differentiation and
coordinated behavior more com- Almost all bacteria and archaea possess some type of cell wall external to the cytoplasmic
monly associated with eukaryotic membrane. The cell wall is responsible for the shape of the cell and provides protection
organisms. For example, cells in an unfavorable osmotic environment. The key component of the cell wall in Bacteria
within some filamentous cyano- is peptidoglycan, a mixed polymer composed of alternating residues of two amino sugars
bacteria communicate to coordi- (N-acetyl glucosamine and N-acetyl muramic acid) with a tetrapeptide sidechain composed
nate formation of heterocysts for of a small range of amino acids. The great mechanical strength of peptidoglycan lies in the
segregation of nitrogen fixation formation of cross-links between the peptide sidechains, forming a mesh-like molecule.
and photosynthesis (p.137). Although variation in the types of amino acids and the nature of cross-linking occurs, pep-
Myxobacteria form fruiting bodies tidoglycan with essentially the same structure occurs in all Bacteria with the exception of
consisting of hundreds of cells,
the Planctomycetes.
some with different functions.
Some magnetotactic bacteria
It is traditional to differentiate bacteria as either Gram-positive or Gram-negative types,
aggregate together in clusters
of up to 40 cells that swim in based on their staining reaction in a method devised by Gram in 1884. Although this basic
coordinated fashion (Figure 4.5). distinction remains a useful first stage in the examination of unknown bacteria in culture
Cable bacteria exist as multicel- and underpins some important physiological differences, it has no phylogenetic signifi-
lular filaments for long-distance cance. As shown in Figure 3.2, Gram-positive bacteria have a relatively thick, simple wall
electron transport (Box 4.1). Many composed of peptidoglycan, together with ribitol or phosphate polymers known as teichoic
bacterial signaling molecules have acids. Gram-negative bacteria have a thinner but more complex wall with a very thin pep-
been identified, with networks for tidoglycan layer that lacks teichoic or teichuronic acids, together with an additional outer
the integration of signals between membrane (OM), anchored tightly to the peptidoglycan layer via a lipoprotein. The OM is
cells in order for the community a lipid-protein bilayer complex but is very different from the cytoplasmic membrane, con-
to make “decisions” about gene
taining unique outer membrane proteins (OMPs) and large amounts of lipopolysaccharide
expression and cellular differen-
tiation. Whether in pure colony
(LPS), both of which have important functions affecting cell physiology and interactions
or a mixed community, bacteria with other organisms. LPS is composed of lipid A covelently bound to a core polysaccharide
communicate via these signals. and strain specific sidechains, which are immunogenic and known as the O-antigen. The
Dunny et al. (2008) argue that “the OMPs have very distinctive properties. The major proteins, called porins, are trimeric struc-
sociobiology of bacteria, largely tures that act as channels for the diffusion of low-molecular-weight solutes across the outer
unappreciated and ignored by the membrane. Other proteins act as specific receptors for key nutrients before transport into the
microbiology research community cell. An important feature of many of these proteins is that they are not produced constitu-
two decades ago is now a major tively. Their synthesis can either be repressed or induced, according to the concentration of
research area … increasingly con- the specific substances. An especially important example of this in marine bacteria is in the
sidered in the context of evolution-
acquisition of the essential nutrient iron, which may be imported into the cell via secreted
ary biology.”
compounds known as siderophores (p.92). Expression of genes encoding siderophores and
OMP receptors for their entry into the cell (as well as many other genes) may only occur
when iron is in short supply. As a consequence, bacteria grown in laboratory culture—
which provides excess iron unless special precautions are taken—may have very different
OM structure and physiological properties from those in natural environments, which are
usually very iron-depleted. Other OMPs are responsible for the detection of environmental
changes such as pressure, temperature, and pH, and communication of this information
via signaling systems. Because the OM is an additional exterior permeability barrier with
different selectivity to the cytoplasmic membrane, some substances cross one membrane,
but not the other. This means that the periplasm—the space between the membranes—is
an important feature of Gram-negative bacteria. Specific periplasmic binding proteins act
as a shuttle to translocate a substrate from an OM receptor to a transport protein in the
cytoplasmic membrane. Proteins such as enzymes and toxins may be transported across the
cytoplasmic membrane but remain in the periplasm until the cell is lysed, unless there are
specific mechanisms to excrete them across the OM. As cells grow, parts of the OM may
balloon out to form small vesicles that are released from the cell, taking with them some of
the periplasmic contents. These can include nucleic acids, enzymes, and toxins. Vesicles can
act as a delivery system for these contents when they fuse with other membranes, which is
especially important in pathogenic and symbiotic interactions of bacteria with their hosts.
The vast majority of marine bacteria have the Gram-negative type of cell envelope. Recent
 Metabolic Diversity and Ecophysiology 69

work shows that vesicles carrying DNA from diverse bacteria are abundant in the oceans,
providing discrete packages that may be involved in exchange of genetic information and
energy sources among marine plankton.

The cell walls of Archaea are very variable in composition. They do not contain peptido-
glycan, but a few types contain a related compound called pseudomurein or thick layers of
other polysaccharides that provide the same osmoprotective function. The main cell walls
in most Archaea are known as S-layers, which are composed of paracrystalline arrays of a
single type of protein or glycoprotein arranged in a tetragonal or hexagonal lattice-like struc-
ture. These structures are built via self-assembly of identical subunits. Many Bacteria—both
Gram-positive and Gram-negative—also possess an S-layer as an additional component as
the outermost part of the envelope (Figure 11.13). Interestingly, S-layers are found in almost
all phylogenetic groups of both Bacteria and Archaea and seem to represent one of the sim-
plest biological membranes, which probably developed very early in evolution. Where the
S-layer is the only component of the cell wall, it provides mechanical strength and osmotic
stabilization, but when it occurs as an additional external surface layer it may have diverse
functions such as protection against phages, predators, phagocytosis, enzymes, and toxins.
Figure 3.2 Schematic representa-
tion of the cell envelopes of Gram-
In addition to the cell envelope, many organisms secrete a slimy or sticky extracellular matrix.
negative (A) and Gram-positive (B)
It is usually composed of polysaccharide, in which case it may form a network extending into
bacteria. Key: 1 cell membrane,
the environment, sometimes termed the glycocalyx. In other cases, there may be a more dis-
2 peptidoglycan, 3 outer membrane,
tinct rigid layer (capsule), which may contain protein. The main importance of the glycocalyx 4 phospholipid, 5 periplasm, 6
in marine systems is in the attachment of bacteria to plant, animal, and inanimate surfaces, lipoprotein, 7 protein, 8 LPS, 9 porin.
leading to the formation of biofilms. The bacterial glycocalyx is critical for the settlement Credit: Francisco2, CC BY 3.0 via
of algal spores and invertebrate larvae; the significance of these processes in biofouling of Wikimedia Commons.
surfaces in the marine environment is considered in Chapter 13. The release of these sticky
exopolymers is also very important in the aggregation of bacteria and organic detritus in
the aggregation of TEPs and marine snow particles discussed in Chapter 1. Furthermore, a
large proportion of the dissolved organic carbon in ocean waters probably derives from the
glycocalyx of bacteria. Slime layers can act as a protective layer, preventing the attachment of
phages and penetration of some toxic chemicals. In pathogens, the presence of a capsule can
inhibit engulfment by host phagocytes, and in free-living forms it is commonly assumed to

i
inhibit ingestion by protists. However, many grazing flagellates seem to feed voraciously on
PROPERTIES OF LPS
capsulated planktonic bacteria, probably because the capsule contains additional nutrients.

Serological typing based on differ-

Genome size and organization determines ences in the O-antigen is impor-
bacterial and archaeal lifestyles tant in identification and variation
in immunological specificity is
Bacterial and archaeal genomes comprise the entire genetic information of the cell, con- important in the epidemiology
taining the DNA code specifying all of the structural proteins, enzymes, regulatory pro- of bacterial diseases (discussed in
teins, and RNA molecules needed for growth and replication. As a general rule, bacterial Chapter 12). LPS also stimulates
and archaeal genomes contain very little non-coding DNA (unlike eukaryotes), so the size the activation of complement
of the genome is directly proportional to the number of genes and hence the lifestyle of the and related host responses, which
organism. The number of genes for the core functions of the cell (energy generation, replica- is very important in the interac-
tion, and protein synthesis) is quite constant and therefore forms a higher proportion of small tions of pathogenic or symbiotic
bacteria with their animal hosts.
genomes. Organisms with larger genomes are likely to have a higher proportion of genes
Lipid A is also an endotoxin, caus-
for functions such as nutrient transport, signaling, regulation of transcription, DNA repair,
ing hemorrhage, fever, shock,
motility, and chemotaxis. This provides them with the greater ability to adapt to factors such and other symptoms caused by
as different nutrient sources and environmental conditions. Organisms with large genomes Gram-negative pathogens of
have the capacity to encode numerous regulatory pathways and specialized metabolic func- vertebrates. Blood collected from
tions. Table 3.1 illustrates this principle, showing the range of genome sizes in representative horseshoe crabs is used to screen
examples of Bacteria and Archaea. for endotoxins in the pharmaceuti-
cal industry because it is sensitive
As seen in the table, some bacterioplankton show very reduced genomes. The genomes found to picomolar concentrations of LPS.
in cultured members of the OM43 and SAR11 clades—the most abundant groups of organ- Fortunately, this use, which threat-
isms in the oceans—are the smallest known for free-living organisms. A theory known as ens Limulus populations on the
US East Coast, may soon be over,
genome streamlining was proposed to explain this phenomenon, in which organisms gain a
because a recombinant protein has
reproductive benefit by evolving to eliminate non-essential genes. As noted above, all bacte-
gained regulatory acceptance as a
rial and archaeal genomes can be considered to be streamlined by the lack of non-coding replacement.
DNA, but some organisms such as the OM43 and SARll representatives have gone further by
70 Chapter 3

Table 3.1 Examples of genome sizes of selected marine members of the Bacteria and Archaea. (Data from NCBI
Genome Database).

Microbea Size ORFsd Comments

(Mb)b,c

Bacteria

“Ca. Ruthia magnifica” 1.16 1099 Obligate intracellular sulfur-oxidizing chemolithoautotrophic symbiont
of giant clam Calyptogena magnifica found at hydrothermal vents.

“Ca. Riegeria santandreae” 1.34 1344 Obligate intracellular sulfur-oxidizing chemolithoautotrophic symbiont
of flatworm Paracatenula.

“Ca. Atelocyanobacterium 1.44 1148 Nitrogen-fixing cyanobacterium lacking many components of

thallassa” ALOHA photosystems and CO2 fixation. Symbiotic, dependent on algal host
for nutrition.

Strain HTCC218 (OM43 clade) 1.30 1354 Methylotroph, metabolizes C1 compounds (but not CH4). Cultured
representative of OM43 clade. The smallest (just!) genome of a
free-living organism.

“Ca. Pelagibacter ubique” 1.31 1354 Chemoorganotrophic heterotroph. Cultured representative of SAR11
HTCC1062 (SAR11 clade) clade. One of the smallest genomes of a free-living organism.

Aquifex aeolicus VF5 1.59 1526 Hyperthermophilic chemoautotroph. Despite its reduced genome, it
possesses flagella and is motile.

Prochlorococcus marinus MED4 1.66 1716 Cyanobacterium. High-light adapted strain of the most abundant
group of oxygenic photosynthetic organisms.

Thermotoga maritima MSB8 1.86 1858 Anaerobic, hyperthermophilic, fermentative, chemoorganotroph.

Thiomicrospira crunogena XCL-2 2.43 2244 Free-living sulfur-oxidizing chemolithoautotroph symbiont.

Prochlorococcus marinus MIT9313 2.14 2275 Strain of P. marinus found at intermediate depths, adapted to exploit
more variable nutrient conditions.

“Ca. Endoriftia persephone” 3.20 n.d. Intracellular symbiont of hydrothermal vent tubeworms but retains
genes for free-living stage. Sulfur-oxidizing chemolithoautotroph.

Nitrosococcus oceani ATCC19707 3.53 3095 Aerobic, ammonia-oxidizing chemolithoautotroph

(Gammaproteobacteria).

Synechocystis sp. PCC6803 3.57 3168 Oxygenic photosynthetic cyanobacterium, abundant in upper
photic zone.

Bdellovibrio bacteriovorus HD100 3.78 3541 Bacterial predator that penetrates other bacteria, replicates, and causes
cell lysis.

Pseudoalteromonas haloplanktis 3.97 3494 Psychrophilic chemoorganotrophic heterotroph. Adaptations for

TAC125 (Chr*2) growth at low temperatures and reactive oxygen species.

Aliivibrio fischeri ES114 4.28 3814 Chemoorganotrophic heterotroph possessing two chromosomes;
(Chr*2) free-living and in bioluminescent symbiosis of squid.

Caulobacter crescentus NA1000 4.04 3886 Chemoorganotrophic heterotroph with a complex lifecycle of
planktonic and stalked sessile stages.

Ruegeria pomeroyi DSS-3 4.60 4306 Member of the Roseobacter clade with a major role in the sulfur cycle
via DMSP metabolism. Many amino acids and carboxylic acid
transporters.

Trichodesmium erythraeum IMS101 7.75 4549 Filamentous cyanobacterium producing differentiated cells (diazocytes)
for temporal regulation of photosynthesis and nitrogen fixation.

Vibrio parahaemolyticus 5.12 4692 Chemoorganotrophic heterotroph. Abundant in marine and estuarine
RMID2210633 (Chr*2) inhabitants. Associated with chitinous surfaces of molluscs and
crustacean shellfish. High replication rate at 37°C; human pathogen.

Rhodopirellula islandica K833 7.43 6851 Chemoorganotrophic, heterotrophic planctomycete. Complex cell
structure with internal cell compartments.

(Continued)
 Metabolic Diversity and Ecophysiology 71

Table 3.1 (Continued)

Microbea Size ORFsd Comments

(Mb)b,c

Archaea

Nanoarchaeum equitans Kin4-M 0.49 536 Hyperthermophilic, obligate symbiont attached to cells of Igniococcus
hospitalis. One of the smallest known cells, minimal genome lacking
genes for most metabolic pathways.

Igniococcus hospitalis KIN4/1 1.30 1442 Hyperthermophilic chemolithoautotroph from undersea volcano,
reducing sulfur with H2. Host for N. equitans.

Methanocaldococcus jannaschii 1.74 1762 Hyperthermophilic obligate anaerobe from hydrothermal vents.
DSM2611 Produces methane form H2 + CO2.

“Ca. Nitrosopumilus maritimus” 1.65 1795 Chemolithoautotroph which oxidizes very low concentrations of NH3
SCM1 in the deep ocean.

Pyrococcus furiosus DSM3638 1.91 1990 Hyperthermophilic obligate anaerobe; chemoorganotrophic

heterotroph from hydrothermal vents. Highly motile.

Cenarchaeum symbiosum A 2.05 2017 Psychrophilic ammonia-oxidizing chemolithoautotroph associated with

the sponge Axinella.

Archaeoglobus fulgidus DSM4304 2.04 2413 Hyperthermophilic obligate anaerobe from hydrothermal vents.
Chemolithoautotroph via sulfate reduction.

Methanosarcina acetivorans C2A 5.75 4542 Metabolically and versatile archaeon from marine and coastal
sediments capable of producing methane from multiple substrates,
including acetate. The largest archaeal genome known.
a Ca. = Candidatus (interim name for uncultured organism); b Mb = megabases; c Chr*2 indicates possession of two chromosomes; d ORFs = open

reading frames (predicted proteins).

losing the ability for metabolic flexibility and specializing in the use of low concentrations
of a few carbon compounds that characterize ocean environments. One explanation for this
is that competition between species will favor organisms with efficient and economic repli-
cation (by requiring less DNA). Large population sizes will increase this selective pressure,
which is why genome streamlining is prevalent in bacteria and archaea, and particularly so
in bacterioplankton. Extreme genome streamlining also minimizes overall size of the cell,
increasing the surface area to volume ratio (see Figure 1.2), leading to more efficient uptake
of scarce nutrients, in proportion to the cell’s needs. The genome sizes of Prochlorococcus
marinus are also extremely streamlined. Again, these have massive population sizes in sur-
face waters, being the most abundant phototrophs. In this case, although classified as the
same species, we can see large differences in genome size and gene content between different
strains (ecotypes) adapted to different levels of light and the availability of nitrogen sources
at different depths. In summary, genome streamlining increases the success of organisms
in nutrient-poor environments, either by winning the competition to acquire their share of
scarce resources, or by using them more efficiently. This is evident by the high gene frequen-
cies of such organisms, as indicators of their abundance and diversity.

As discussed in detail later in this chapter, we can describe organisms adapted to these low
levels of nutrients as oligotrophs, which have a “slow and steady” lifestyle in stable niches
that show little variation in composition over time. The low average size of genomes in
marine metagenomes is evidence that organisms with this lifestyle are most abundant. In
contrast, other organisms described as copiotrophs—often associated with particles—have
a “feast or famine” lifestyle and are able to grow rapidly when exposed to high substrate
levels and respond to higher concentrations of nutrients. This alternative strategy is adopted
by many marine bacteria with higher numbers of genes, conferring the ability to carry out a
wider range of functions according to shifts in prevailing conditions. Variable factors such as
availability of different nutrients, oxygen, light, temperature, pressure, surface attachment,
or circadian rhythms require organisms to have a wider repertoire of enzymes and transport
systems plus efficient systems for the regulation of gene expression.
72 Chapter 3

It is also evident from Table 3.1 that symbiotic and parasitic microbes often possess the
smallest genomes of all. In this case, a different evolutionary mechanism is thought to oper-
ate, in which the organisms become dependent on their host for many of their metabolic
functions. During evolution, reduction in genome size is thought to occur by genetic drift due
to reduction in effective population sizes, with reduced opportunities for recombination and
horizontal gene transfer. In addition, positive selection pressure results in loss of genes for
metabolic functions that are provided by the host. The microbe may retain only the essential
core genes for cell structure and central metabolism. The genome reduction of symbionts and
pathogens of animals is explored in detail in Chapters 10 and 11, but one salient point will be
discussed here to illustrate the principles by comparing the genome sizes of free-living and
symbiotic sulfur-oxidizing bacteria (SOB). Note from Table 3.1 that the bacterial symbiont
of the clam Calyptogena has a highly reduced genome in comparison to free-living SOB,
whereas the symbiont of the tubeworm Riftia has a genome of similar size to its free-living
relatives. Both are intracellular symbionts of their host, but the clam symbiont is obligately
dependent on its host and is passed from one generation to the next via the eggs, whereas the
tubeworm symbiont has a free-living stage and is acquired at each host generation.

The table also shows examples of bacteria and archaea with minimal genomes that are sym-
bionts of parasites of other microbes with which they are intimately associated. The archaeon
Nanoarchaeum depends on its archaeon host Igniococcus hospitalis for most nucleotides,
amino acids, and lipids and has a highly reduced genome. “Ca. Atelocyanobacterium thal-
assa” (UCYN-A) is a cyanobacterium that lacks many components of systems for photo-
synthesis and CO2 fixation and depends on its unicellular algal hosts for energy and carbon
compounds. In return, it fixes nitrogen for its host.

The most common form of bacterial genome organization is a singular circular chromo-
some. The length of the DNA molecule can be over 1000 times the length of the cell, and
cells achieve the remarkable feat of coiling, folding, and packing the DNA in such a way that
the processes of replication, transcription, and translation can occur efficiently. The circu-
lar chromosome exists as a compact structure organized into supercoiled domains by DNA
gyrase. Replication of the DNA involves a number of polymerase and ligase enzymes and
proceeds bidirectionally from an origin on the chromosome and is coordinated with cell divi-
sion processes to ensure that each daughter cell receives a complete chromosome. In rapidly
growing cells, the DNA may be replicating at more than one point on the chromosome. Some
bacteria have linear chromosomes or possess more than one chromosome. Notably, members
of the Vibrionaceae have two chromosomes, which confer important properties for their evo-
lution by gene acquisition, as discussed in Chapter 12.

Although archaea possess small compact genomes like those of bacteria, organization of the
nuclear material is more complex and variable in archaea. In some species, packaging and
stabilization of the DNA structure is achieved by supercoiling, but other species possess posi-
tively charged proteins known as histones, which have amino acid sequences homologous to
those found in eukaryotes. Short sequences of DNA are wound around clusters of histones
to form tetrameric clusters called nucleosomes. Some extremely thermophilic archaea have
an additional type of DNA gyrase that induces positive supercoiling in the DNA to protect it
against denaturation. The process of DNA replication is also different from that in bacteria.
Although most archaea possess single circular chromosomes, there are often multiple origins
of replication, and the structure of the replication complex and polymerase enzymes again
resembles that found in eukaryotes.

In addition to the chromosome(s), many bacteria and archaea contain small, circular, extra-
chromosomal elements of genetic information called plasmids. These replicate indepen-
dently of the main chromosome, although they rely on enzymes encoded on it. Plasmids
have great biological significance, because they encode genes that are useful to the cell under
certain conditions, rather than those connected with mainstream metabolic processes. Often,
plasmids can be transferred between cells and rapidly spread through a population via conju-
gation. This increases the adaptability of a population in various ways. The most important
plasmids are those containing genes for antibiotic resistance, pathogenicity (e.g. toxins or
colonization factors), or the ability to degrade recalcitrant organic compounds (e.g. naturally
occurring or xenobiotic hydrocarbons).
 Metabolic Diversity and Ecophysiology 73

Microbes use a variety of mechanisms

to regulate cellular activities
Even the simplest microbes contain hundreds of genes encoding enzymes and other macro-
molecules required for the hundreds of biochemical reactions in metabolism. To ensure effi-
cient use of resources and to respond to changes in environmental conditions, bacterial and
archaeal cells need to regulate both the activity of enzymes and the level of macromolecular
synthesis. Regulation of the activity of enzymes is a very rapid process, occurring by various
mechanisms such as feedback allosteric inhibition, in which an excess of the end product of
a biosynthetic pathway often inhibits the first enzyme of the pathway. Some other enzymes
are modified by covalent addition of groups, which inhibits catalytic activity. Although some
proteins are required under all conditions and are said to be constitutive, cells employ many
mechanisms of controlling the level of gene expression to regulate the production of enzymes
or other proteins that are only required under certain conditions. Because messenger RNA
(mRNA) is very short-lived, switching transcription on or off provides a very rapid response
to changing circumstances. Also, bacterial and archaeal genes are often grouped together in
operons, so that expression of related genes (e.g. several enzymes in a biosynthetic pathway)
is regulated together under the control of a single promoter and operator region, upstream of
the structural genes. As noted above, copiotrophic planktonic microbes possess many regula-
tory systems, and the same applies to microbes that grow on surfaces or colonize animals or
plants. The expression of a gene involves the binding of RNA polymerase to a specific region
of the DNA (the promoter). Many genes are subject to negative control, when regulatory
compounds bind to specific DNA-binding proteins. This causes an altered conformation of
the protein, allowing it to bind to the DNA at the operator region (between the promoter and
the gene) leading to the blocking of transcription. Genes can also be regulated positively—in
this case, an activator protein acts to help the RNA polymerase bind efficiently to the pro-
moter region of the DNA. Often, several operons will be regulated by the same repressor or
activator, so they are expressed coordinately. Sometimes, regulation of a master gene results
in a cascade of subsequent regulatory steps, so that expression of hundreds of genes is up-
or down-regulated in response to a single stimulus. A set of genes like this is known as a
regulon.

One of the most common methods used by cells to respond to external stimuli is via two-
component regulatory systems. In this case, the signal does not act directly to control
transcription, but is detected by a sensor kinase protein in the cell membrane, which then
transduces the signal to a response regulator protein in the cytoplasm. The kinase enzyme is
composed of two domains, one of which is exposed to the exterior and detects an environ-
mental or chemical signal. When this occurs, the cytoplasmic domain of the protein becomes
phosphorylated. The phosphoryl group is then transferred to a second specific protein called
the response regulator. When phosphorylated, the response regulator interacts with specific
regions of the DNA and can either activate or repress transcription of specific genes, depend-
ing on the system. Chapters 11 and 12 describe important examples of such signaling and
global control networks when considering infection of animals by symbiotic or pathogenic
bacteria from the aquatic environment.

Translation of mRNA and protein synthesis in the Archaea and Bacteria are similar, employ-
ing the 70S ribosome (Figure 2.7). However, many of the component ribosomal proteins of
the Archaea and Eukarya show closer homology than they do with those from the Bacteria.

SOURCES OF ENERGY AND CARBON

Microbes obtain energy from light
or oxidation of compounds
Metabolism, the sum of all the biochemical reactions within a cell, consists of catabolism
and anabolism. In catabolism, cells break down and oxidize larger molecules yielding energy
via exergonic reactions, which is used for the synthesis of macromolecules by endergonic
(energy-requiring) reactions. Adenosine triphosphate (ATP) is regarded as the chemical cur-
rency of the cell, mediating the transfer of energy during these processes (Figure 3.3). It is
convenient to classify microbes in terms of their energy source as either chemotrophs (which
74 Chapter 3

Figure 3.3 The energy metabolism

of all cells depends on hydrolysis
of adenosine triphosphate (ATP) to
adenosine diphosphate (ADP). This
is highly exergonic (negative Gibbs
free energy, -ΔG) and is used to
drive endergonic (+ΔG) biochemical
processes such as biosynthesis, active
transport, or movement.

produce ATP by the oxidation of inorganic or organic compounds) or phototrophs (which use
light as a source of energy to produce ATP). Chemotrophs may be further divided into those
that can obtain energy solely from inorganic compounds (chemolithotrophs) and those that
use organic compounds (chemoorganotrophs). Chemolithotrophy is unique to members of the
Bacteria and Archaea, while chemoorganotrophy is found in all groups of cellular microbes.

Microbes differ in their source of carbon

to make cellular material
Those organisms that can use CO2 as the sole source of carbon are termed autotrophs, while
those that require preformed organic compounds as a source of carbon are termed hetero-
trophs. Autotrophs are primary producers that fix carbon from CO2 into cellular organic
compounds, which are then available to be used by heterotrophs. Most chemolithotrophs and
phototrophs are also autotrophic, but some (e.g. purple non-sulfur photosynthetic bacteria
and some sulfur-oxidizing chemolithotrophs) can switch between heterotrophic and auto-
trophic metabolism. The term mixotroph is used to describe organisms with a mixed mode
of nutrition. For example, some sulfur-oxidizing bacteria are not fully lithotrophic, as they
require certain organic compounds because they lack key biosynthetic enzymes. Many pro-
tists, including numerous types of flagellates and dinoflagellates, are mixotrophic because
they derive nutrition both from photosynthesis and from the engulfment of prey such as bac-
teria. Indeed, among the protists there is a spectrum from absolute heterotrophy to absolute
autotrophy. Table 3.2 shows a summary of these nutritional categories.

PHOTOTROPHY AND CHEMOTROPHY

Phototrophy involves conversion of
light energy to chemical energy
Virtually all life on Earth depends directly or indirectly on solar energy, and it is likely
that simple mechanisms for harvesting light energy developed very early in the evolution of
life. Phototrophs contain light-sensitive pigments, which use energy from sunlight to oxidize
electron donors. Plants, algae, other phototrophic protists, and cyanobacteria use H2O as the
electron donor, resulting in the production of O2 (oxygenic photosynthesis). Other types of
phototrophic bacteria use H2S, S0, H2SO32-, or H2, meaning that no O2 is generated (anoxy-
genic photosynthesis). In the process of photophosphorylation, light energy creates a charge
separation across membranes and “traps” energy from the excited electrons into a chemi-
cally stable molecule (ATP) and also generates NADPH (nicotinamide adenine dinucleotide
phosphate [reduced]), as shown in Figure 3.4A. Many phototrophs are also autotrophic and
use ATP and NADPH for the fixation of CO2 into cellular material, in which case the process
meets the strict definition of photosynthesis.
 Metabolic Diversity and Ecophysiology 75

Table 3.2 Nutritional categories of microorganisms

Energy source Carbon source Hydrogen or electron Representative examples

source

Photolithoautotrophy

Light CO2 Inorganic Cyanobacteria

Purple sulfur bacteria
Phototrophic protists

Photoorganoheterotrophy

Light Organic compounds Organic compounds or H2 Purple non-sulfur bacteria

Aerobic anoxygenic bacteria
Proteorhodopsin-containing
bacteria and archaeaa

Chemolithoautotrophy

Inorganic CO2 Inorganic Sulfur-oxidizing bacteria

Hydrogen bacteria
Methanogens
Nitrifying bacteria and archaea

Chemoorganoheterotrophy

Organic compounds Organic compounds Organic compounds Wide range of bacteria and archaea
Fungi
Phagotrophic protists

Mixotrophy (combination of lithoautotrophy and organoheterotrophy)

Inorganic Organic compounds Inorganic Some sulfur-oxidizing bacteria, e.g.

Beggiatoa

Mixotrophy (combination of photoautotrophy and organoheterotrophy)

Light + organic compounds CO2 + organic Inorganic or organic Phagotrophic photosynthetic

compounds protists (some flagellates and
dinoflagellates)
aThe role of proteorhodopsin in nutrition is uncertain (see Box 3.1).

Light energy
Absorption
of light by
antenna pigments Figure 3.4 (a) Trans-
Excitation energy
formation of light energy
to chemical energy in the
Reaction light reactions of photo-
center synthesis. (b) Structure
Charge separation of chlorophyll a, showing
four pyrrole rings (green)
containing magnesium
Redox Membrane at the center (yellow).
electrochemical potential Bacteriochlorophyll a has
energy energy phyt ol chain
an identical structure,
except for changes in the
shaded regions (purple).
Transfer of Transfer of Other chlorophylls and
electrons H+ and Pi
bacteriochlorophylls
have other substitutions
NADPH ATP at various parts of the
Chemical bond molecule, resulting in
energy
different light absorption
(a) Carbohydrate (b) properties.
76 Chapter 3

Although many microbes possess photochemical mechanisms, true photosynthesis occurs

only in cells containing the magnesium-containing pigments known as chlorophylls (found in
oxygenic cyanobacteria and chloroplasts) or bacteriochlorophylls (found in anoxygenic pho-
totrophic bacteria). Because of their efficiency in both transforming light energy into ATP
through electron flow, the provision of reducing power (NADPH, NADH), and the use of water
as reductant, chlorophyll-based photosynthesis has evolved to become the most prevalent pho-
tosystem. Chlorophyll and bacteriochlorophyll molecules are closely related, apart from a small
region of the porphyrin molecule (Figure 3.4B). There are also small differences in other parts
of the molecule, resulting in structures that absorb different wavelengths of light. A variety
of accessory pigments are also present, notably carotenoids, which function primarily to pro-
tect cells from harmful photooxidation reactions that can occur in bright light, such as the
generation of toxic forms of oxygen. Carotenoids such as fucoxanthin may be important in
light harvesting for photosynthesis in eukaryotic algae. Antenna pigments surround complexes
of 50–300 molecules of chlorophyll or bacteriochlorophyll combined with proteins (reaction
centers). These arrays occupy a large surface area to trap the maximum amount of light of
different wavelengths and transfer its energy to the reaction center. In cyanobacteria, the most
important antenna pigments are the phycobilins phycoerythrin and phycocyanin. The diversity
of different pigments with different light-absorbing properties has great ecological significance,
because it determines the optimum niche within the water column and allows organisms with
different combinations of pigments to coexist in illuminated habitats by absorbing a particular
fraction of the light energy that other members of the community are not absorbing.

The photosynthetic pigments are contained within special membrane systems that create the
structure necessary for generation of a proton-motive force. In photosynthetic bacteria, there
are usually extensive invaginations of the cytoplasmic membrane; these can be complex and
multilayered in cyanobacteria. In chloroplasts, the photosynthetic pigments are attached in
stacks of sheet-like membranes called thylakoids.

Oxygenic photosynthesis involves two

distinct but coupled photosystems
Oxygenic photosynthesis, in which H2O is the electron donor leading to production of O2, is the
most important contributor to primary productivity accounting for the major roles played by
cyanobacteria, diatoms, and dinoflagellates and other protists in ocean processes. Of course,
it is also familiar as the mechanism used by seaweeds and plants. Photosynthesis is character-
ized by the presence of two coupled photosystems operating in series. As shown in Figure 3.5,
photosystem I (PS I) absorbs light at long wavelengths (>680 nm) and transfers the energy to
a specialized reaction-center chlorophyll (P700). Photosystem II (PS II) traps light at shorter
wavelengths and transfers it to chlorophyll P680. Absorption of light energy by PS I leads to
a very excited state, which then donates a high-energy electron via chlorophyll a and iron-
sulfur proteins to ferredoxin. The electron can then pass through an electron transport chain,
returning to P700 and leading to ATP synthesis via generation of a proton-motive force (cyclic
photophosphorylation). ATP formation and reduction of NADP+ to NADPH also occurs via a
noncyclic route involving both photosystems. These processes are known as the light reaction
of photosynthesis. In the light-independent (dark) reaction, ATP and NADPH molecules are
used to reduce CO2, leading to its fixation into carbohydrate. The reaction may be represented
as follows, where (CH2O) represents the basic unit of carbohydrates:

CO2 + 3ATP + 2NADPH + 2H + + H 2 O → (CH 2 O ) + 3ADP + 3Pi + 2NADP +

There are some important differences between the photosynthetic machinery of

Prochlorococcus and other cyanobacteria, which are discussed in Chapter 4.

Anaerobic anoxygenic photosynthesis uses

only one type of reaction center
A group known as the purple phototrophic bacteria contain bacteriochlorophyll a, but lack
PS II and therefore cannot use H2O as an electron donor in noncyclic electron transport. The
 Metabolic Diversity and Ecophysiology 77

3H+ Figure 3.5 Electron flow in oxy-

ADP + Pi ATP
genic photosynthesis. The light-
driven electron flow generates an
NAD reductase ATP electrochemical gradient across the
LIGHT LIGHT
2H+ + 2 synthase membrane, leading to ATP synthesis.
8H+ Cytochrome bf
2 NADP+ NADPH
complex Electrons flow through two photo-
Fd FAD systems, PS1 and PS2. Electrons are
QA QB 4PQH Cyt b 6
FeS provided by water and the oxygen-
evolving complex (OEC) generates
FeS membrane
Pheo A oxygen. Key: Pheo, pheophytin;
P680 4PQ Q, quinone; Cyt, cytochrome; PC,
Cyt f
Mn
plastocyanin; A, chlorophyll a; FeS,
OEC PC –Cu P700
2H2O O2 + 4H+ non-heme iron-sulfur protein; Fd, fer-
8H+ PC –Cu redoxin; FAD, flavin adenine dinucle-
Phot osyst em II Phot osyst em I otide; P680 and P700, the reaction
3H+
centers of PSI and PSII, respectively.

light reaction generates insufficient reducing power to produce NADPH, and these bacteria
therefore use H2S, H2 or S0 because they are better electron donors. For this reason, they
do not generate O2 during photosynthesis and most are strict anaerobes found in shallow
sediments and microbial mats. This mode of photosynthesis was almost certainly the first to
evolve, before the development of PS II by cyanobacteria enabled water to be used as elec-
tron donor. Interestingly, some cyanobacteria growing in sulfide-rich habitats can carry out
anoxygenic photosynthesis because they rely only on PS I.

Aerobic anoxygenic phototrophy is CAN AANP

widespread in planktonic bacteria ? BACTERIA FIX CO2?
Another form of phototrophy that does not result in generation of O2 has recently been dis- Graham et al. (2018) recently
covered to be widespread and important. Up to 20% of the bacteria in the photic zone may be screened genomes assembled from
metagenomic databases obtained
aerobic anoxygenic phototrophs (AAnP). Screening of metagenomic datasets for puf genes,
from the Tara Oceans dataset
which code for the subunits of the light harvesting and reaction-center complexes, reveals and, surprisingly, discovered draft
their presence in a range of globally distributed bacterial groups in several phyla—they also genomes of nine bacteria that pos-
occur in freshwater and the surface of soils. Like the purple bacteria, these bacteria do not sess the potential for both AAnP
generate O2 during photosynthesis. However, unlike them, they do grow in aerobic conditions and CO2 fixation via the CBB cycle.
and sufficient transfer of electrons to drive ATP production by cyclic photophosphorylation These bacterial genomes contained
only occurs in the presence of O2. These organisms are generally photoheterotrophs and lack the genes encoding two subunits
the enzymes for fixation of CO2 into cell material. They use O2 for respiratory metabolism (PufML) of the type II photochemi-
of organic carbon and for the synthesis of bacteriochlorophyll a. However, AAnP bacteria cal reaction center and the large
metabolize more efficiently when light is available, suggesting that in oligotrophic waters and small subunits (RbcLS) of
RuBisCO. These organisms form a
they can use energy from both light and scarce nutrients simultaneously. In these bacte-
clade with no cultured representa-
ria, light inhibits bacteriochlorophyll a synthesis, so its activity diminishes during the day.
tives within the alphaproteobacte-
Marine AAnP populations are complex and dynamic, forming a significant component of rial family Rhodobacteriaceae, for
bacterioplankton in certain oceanic areas. Their high metabolic activity suggests that AAnP which Graham et al. propose the
bacteria may contribute to ocean energy budgets and carbon cycling, but the extent of this is name “Ca. Luxescamonaceae.” This
currently unclear. group is globally distributed but
constitutes only a minor com-
ponent (0.1–1.0%) of free-living
Some phototrophs use rhodopsins as plankton in the photic zone. These
light-harvesting pigments bacteria also possess genes for the
synthesis of bacteriochlorophyll
True photosynthesis is unknown in the Archaea, although some extremely halophilic types and the oxidation of thiosulfate
contain a specialized protein called bacteriorhodopsin, which is conjugated to the carotenoid or sulfite, suggesting that they
pigment retinal. This appears as purple patches in the membrane of cells grown under low can grow photolithoautotrophi-
oxygen concentration and high-light intensity. Light energy is used to form a proton pump, cally. Graham et al. note that their
which generates ATP to provide sufficient energy for slow growth when other energy-yielding metagenomic assembly method
reactions are not possible. Bacteriorhodopsin also functions to maintain the Na+/K+ balance has some limitations, and confirm-
in the cell in highly saline conditions. A similar light-mediated ATP synthesis system based ing that these bacteria do indeed
possess a new method of photo-
on a molecule termed proteorhodopsin is now known to be very widespread amongst marine
synthesis will require proof of gene
members of the Bacteria and Archaea in the photic zone. The discovery of proteorhodopsin
expression.
and its ecological significance is discussed in Box 3.1.
 78 Chapter 3

BOX 3.1 RESEARCH FOCUS

Microbes use solar power to top up energy levels

Discovery of a novel light-harvesting mechanism. The discov- was demonstrated in many of the major bacterial groups and in
ery of proteorhodopsin (PR) was one of the first demonstrations of archaea, diatoms and dinoflagellates, other protists, and viruses
the power of metagenomics to reveal previously unknown types of (Ernst et al., 2014; Sharma et al., 2006). PR is now realized to be
metabolism in ocean processes. As stated by Pinhassi et al. (2016): ubiquitous and present in over half of all heterotrophic bacterio-
“With most techniques and research approaches, what one finds plankton in the surface ocean (Figure 3.6).
in nature is what one is looking for; the rest remains invisible.”
Rhodopsin is a membrane protein that contains a light-absorbing PR-based phototrophy is dominant in picoplankton. So, we now
pigment (retinal) found in the rods of the retina in animal eyes. It know that life on Earth has evolved two efficient—but very differ-
has been known for some time that membranes of the archaean ent—light-harvesting strategies for phototrophy. Finkel et al. (2013)
genus Halobacterium contain an analog of this protein—given hypothesized that the simple, PR-based proton pump mechanisms
the name bacteriorhodopsin—which use light energy to synthe- should be more commonly found in organisms than the complex
size ATP (p.109). This was thought to be a process unique to photochemical reaction centers (RCs) based on chlorophylls or
these extremely halophilic Archaea. However, while carrying out bacteriochlorophylls. Although PRs appear not to provide enough
sequence analysis of DNA from an uncultivated group of bacteria energy for CO2 fixation, they have the advantage of requiring the
known as SAR86, Béjà et al. (2000) found evidence of a genetic expression of only a single membrane protein and are easily spread
sequence with strong sequence homology to the archaeal rho- by HGT. Photochemical RCs, on the other hand, are highly com-
dopsins. Using bioinformatic tools, Béjà and colleagues made a plex structures of numerous proteins and pigments and are “costly”
structural model of the protein believed to be encoded by the gene for the cell to maintain. Finkel et al. used a systematic approach to
and identified the transmembrane domains that are a critical fea- examine 115 metagenomic datasets for the abundance of genes from
ture associated with the rhodopsin function. Phylogenetic analysis oxygenic photosystems I and II and the anoxygenic photosystems
showed that the gene for the proteobacterial protein (which they RC1 and RC2 and compared these with rhodopsin homologs. The
called proteorhodopsin, PR) is phylogenetically distinct from the metagenomes had been prepared based on pre-filtration of different
archaeal protein bacteriorhodopsin (see p. 155 for comments on size fractions (see p.31) and samples filtered with a pore size greater
the naming of rhodopsins). The gene for the PR protein was then than 0.8 µm—which would include most protists—were dominated
cloned in Escherichia coli, in which it was expressed as an active by RCs. By contrast, in samples of the <0.8 µm fraction—con-
protein and acted as a proton pump employed for the synthesis taining most bacteria and archaea—48% of the cells contained
of ATP. Béjà et al. (2001) then analyzed the properties of mem- a rhodopsin gene and only 18% had RC genes of any kind. Some
branes isolated from marine bacterioplankton concentrated from organisms such as cyanobacteria, dinoflagellates, and diatoms can
seawater by filtration, showing that the functional reaction cycle contain both phototrophic mechanisms. Finkel and colleagues con-
is expressed in the natural environment. Subsequently, genes clude that the majority of bacterial and archaeal cells in the photic
encoding PR have been shown to be present in a very diverse zone contain phototrophic potential. They point out that abundance
range of marine microbes, including the first cultured strain of genes does not necessarily indicate the expression or function
(“Ca. Pelagibacter ubique”) of the highly abundant SAR11 clade
(Giovannoni et al., 2005); see p.41. As more surveys were con- +
+ H
ducted, it became clear that proteorhodopsin is not restricted to H

the Proteobacteria, or even to the Bacteria. Frigaard et al. (2006) Light +

+ H
demonstrated PR genes in DNA of Archaea extracted from vari- H
+ +
ous depths in the Pacific Ocean and the GOS metagenomic data- H
Periplasm H
sets (see p.55) provided further evidence of the wide distribution +
H
of PRs in diverse microbes. Phylogenetic analysis of the genes
required for synthesis of the PR protein and the associated reti-
PR Cell membrane
nal pigment provided strong evidence of horizontal gene trans-
fer (HGT) between organisms. McCarren and DeLong (2007)
+
showed that acquisition of a cassette of just six genes confers the H
ability to harvest biochemical energy from light on a microbe and Cytoplasm ATPase P P A

suggested that this could happen easily in a single genetic trans- + Pi

ADP
H
fer. Further studies showed that there is sequence variation in PR + +
+ H H
H +
from bacteria isolated from different regions and depths, leading H P P P A

to altered biophysical properties. Spectral tuning of PR—due to ATP

small changes in amino acid composition at a critical region of
the molecule—provides an ecophysiological adaptation to ensure Figure 3.6 Proteorhodopsin (PR) uses light energy to trans-
maximum absorption of different wavelengths of light, which locate protons (H+) across the cell membrane which create
penetrate water to different depths (Figure 1.5). Over the next few a proton-motive force that drives ATP synthesis via a multi
years, evidence for PR-driven transduction of energy from light subunit membrane-bound ATPase.
 Metabolic Diversity and Ecophysiology 79

BOX 3.1 RESEARCH FOCUS

of these genes, but numerous transcriptomic and proteomic studies have also been acquired via HGT by some “classic” organohet-
have shown that PR is highly expressed in various ocean regions erotrophs like Vibrio spp. Gómez-Consarnau et al. (2010) showed
and depths (Pinhassi et al., 2016). Since many of these studies are increased long-term survival of a starved Vibrio strain in seawater
based on single time points, Sieradzki et al. (2018) sought to deter- exposed to light, rather than held in darkness. By contrast, no dif-
mine whether PR metagenomes and metatranscriptomes varied ferences in survival were observed in a PR-deficient mutant strain.
according to the season at three contrasting locations, with major Wang et al. (2012) showed that cell viability of cells in nutrient-poor
differences in nutrient concentration, chlorophyll levels, bacterial environments was due to light-induced proton pumping via PR,
and viral counts, and heterotrophic production. PR dominated gene leading to increased ATP production. Use of bonus “solar panels”
abundance and expression of phototrophy in this dynamic envi- certainly seems to be an effective strategy of bacteria to supple-
ronment throughout the year, with oxygenic photosynthesis and ment energy when organic nutrients are scarce, conferring a fitness
AAnP at much lower levels. PR expression was dominated by the advantage for bacteria to survive periods of resource deprivation.
SAR11 cluster, followed by other Alpha- and Gammproteobacteria. But not all bacteria possess PR—is it possible to detect genomic
In areas with low chlorophyll a concentrations, genes for the blue- differences between related organisms with and without this mech-
light absorbing PR dominated, as found in other oligotrophic open anism? Kumagai et al. (2018) compared the genomes of 41 PR- and
ocean waters (Dubinsky et al., 2017). 35 PR+ strains of Flavobacteriia. Differences appear to be related
to evolution of fundamentally different lifestyles and ecophysi-
What benefits does PR provide to cells? Given that PRs are so ological strategies, which they dubbed the “solar panel or parasol”
widespread in diverse microbes and that there may be over 20000 hypothesis. Kumagai et al. conclude that PR- Flavobacteriia pos-
copies of the protein, occupying a considerable fraction of the cell sess genes that encode synthesis of UV-absorbing pigments in their
membrane (Béjà et al., 2000), it seems likely that PR must provide a membranes (“parasols”) that shield the cells from UV damage to
significant ecological advantage to planktonic bacteria and archaea their cells. This means that they cannot take advantage of light
that possess it. However, experimental proof of the physiological energy, whereas PR+ cells exploit the light for phototrophy (via PR
benefits that PR provides have proved elusive. Giovannoni et al. “solar panels”) and accept the cost of repairing the damage to their
(2005) found that “Ca. Pelagibacter ubique” (SAR11) grew equally DNA (via synthesis of a photolyase enzyme).
well in seawater cultures in dark and light conditions. However,
measurement of oxygen consumption and gene expression under In view of their presence in such diverse organisms, PR may have
conditions of carbon starvation—leading to reduced respiration various physiological roles ranging from promoting long-term star-
and oxidative phosphorylation—showed that, in the light, the bac- vation survival to short-term adaptation to low-nutrient levels; also
terium switches to use of PR phototrophy to provide ATP (Steindler the presence of PR genes may have hitherto unidentified interac-
et al., 2011). Gómez-Consarnau et al. (2007) showed that Dokdonia tions with other gene systems that provide more subtle ecological
sp. (a PR-containing member of the Flavobacteriia) showed a six- or physiological benefits (Pinhassi et al., 2016). Further research
fold increase in growth yield when grown in light of the appropriate should reveal how much contribution PR makes to the nutrition of
wavelength. The growth-enhancing effect of light was most effec- the picoplankton microbes and to carbon cycling and ecosystem
tive at low concentrations of organic matter. Interestingly. PR genes energy fluxes in the ocean.

Chemolithotrophs use inorganic electron donors

as a source of energy and reducing power
Chemolithotrophs derive energy for the synthesis of ATP using an electron transport chain
for the oxidation of inorganic molecules, and these bacteria play a major role in biogeochemi-
cal cycling in marine habitats. Until quite recently, it was thought that chemolithotrophy was
only of major significance in sediments and around hydrothermal vents, but genomic data
have shown that some of these processes are also highly important in the water column. The
energetics of chemolithotrophy depend on the free-energy yield of coupled reactions between
an electron donor and electron acceptor. Electron donors with sufficient reducing power to
drive ATP synthesis include sulfur, sulfide, hydrogen, ammonium, nitrite, and iron (Fe2+) and
these have many geological or biological origins. A general feature is that the yield of ATP
from oxidation of inorganic substances is very low; hence, the ecological and biogeochemical
importance of chemolithotrophs is magnified because they need to oxidize large quantities
of material in order to grow. Many chemolithotrophs are obligate aerobes, because oxygen
is the most energetically favorable electron acceptor. However, anaerobic processes using
sulfate, nitrate, and nitrite as electron acceptor are very important in anoxic sediments and
deep ocean waters.
80 Chapter 3

S2–
S0 Many bacteria oxidize sulfur compounds
Many types of bacteria use sulfur in elemental (S0) or reduced (S2−, S2O32−) forms as electron
S S2O3 2– donor for chemolithotrophic metabolism, using O2 as electron acceptor. Most are also obligate
electron autotrophs (fixing CO2 via production of NADH) or mixotrophs (using organic compounds
e– transport as a source of carbon). Sulfur-oxidizing bacteria (SOB) include genera such as Thiobacillus,
chain Thiothrix, and Thiovulum—these are strict aerobes found in the top few millimeters of
AMP
SO32– marine sediments that are rich in sulfur. SOB are also very prominent at hydrothermal vents
2e–
and cold seeps, both as free-living forms and as symbionts of invertebrates (Chapter 10),
sulfite 2e–
where autotrophic SOB can form the base of food webs in the absence of input from photo-
oxidase synthesis. A major challenge facing SOB is the opposing gradients of reduced sulfur com-
APS ADP
pounds they need for energy and the oxidants (O2 or NO3-) needed as electron acceptors.
Pi
SOB frequently show chemotactic or aerotactic motility in order to seek out the oxic-anoxic
SO42– substrate level SO42–
phosphorylation interface where O2 and reduced sulfur compounds can exist together. Other SOB have other
physiological or behavioral adaptations to overcome this challenge and symbionts of inverte-
Figure 3.7 Oxidation of reduced brates often depend on their host for a specialized blood supply or movement through sedi-
sulfur compounds by chemolitho- ments to provide the correct balance of HS- and O2. Large sulfur bacteria such as the family
trophs. Sulfite is most commonly Beggiatoaceae are very common in microbial mats and surface sediments. In order to bridge
oxidized with the enzyme sulfite the gap between reductant and oxidant, they store large amounts of elemental S and NO3- as
oxidase (left side of diagram) result- an alternative electron acceptor. Different strains are very diverse in their metabolism, vary-
ing in the direct generation of ATP ing from chemorganotrophy to chemolithoautotrophy. There are numerous variations in the
via the electron transport chain and pathways used for sulfur oxidation; an outline of the main biochemical reactions is shown
a proton-motive force generated in Figure 3.7. The complete oxidation of sulfide to sulfate generates eight electrons, and
across the membrane that leads the intermediate oxidation states provide substrates for numerous other organisms and syn-
to ATP synthesis by ATPase. Some trophic interactions (Figure 3.8). In particular, the production of sulfate in sediments plays a
sulfur-oxidizing bacteria use the key role in the breakdown of organic material in sediments, by providing nutrition for sulfate-
enzyme adenosine phosphosulfate reducing bacteria (SRB) as discussed below. Thioploca and Thiomargarita obtain energy
(APS) reductase (right side). by using sulfide as an electron donor, coupled with NO3- as electron acceptor and can grow
autotrophically or mixotrophically using organic molecules as carbon source. They grow in
nitrate-rich waters above anoxic bottom waters with high levels of organic matter. Each cell
contains a very thin layer of cytoplasm around the periphery and a liquid vacuole that con-
stitutes 80% of the cell volume and stores very high concentrations of NO3−. Some archaea
also oxidize sulfur; these include the hyperthermophilic and acidophilic genus Sulfolobus
found at hydrothermal vents. Some phototrophic bacteria (members of the Chromatiaceae,
Chlorobiaceae, Rhodospirillaceae, and Cyanobacteria) found in microbial mats can grow
anaerobically, using reduced sulfur compounds as electron donors for photosynthesis.

Figure 3.8 The sulfur cycle in

marine sediments. Abbreviations:
OSM = organo-sulfur molecules,
ANME = anaerobic methane oxida-
tion, SCI = sulfur cycle intermediates,
Corg = organic matter, DIET = direct
interspecies electron transport, LDET
= long distance electron transport.
Bacterial and archaeal groups are
shown in green; note that the affilia-
tions of the Deltaproteobacteria and
Epsilonproteobacteria have recently
been amended (see Chapter 4),
Redrawn from Wasmund et al.
(2017), CC BY 4.0.
 Metabolic Diversity and Ecophysiology 81

Mechanisms of sulfur oxidation are clearly very diverse, and SOB have evolved many inter-
WHIRLING BACTERIA
esting mechanisms for solving the challenge of obtaining their energy source and reducing
power. Recently, an exciting new mechanism was discovered whereby some members of i FAN THEMSELVES TO
OBTAIN OXYGEN
the family Desulfobulbaceae form very long, multicellular filaments that transfer electric
currents over centimeter-long distances in sediments, remotely coupling sulfide oxidation Bacteria that oxidize sulfides show
in anoxic sediment layers with O2 reduction at the surface. The properties of these “cable special adaptations to cope with
bacteria” and their ecological role are discussed in Box 4.2. the varying gradient of O2 and
S at the oxic-anoxic interface.
Thar and Kuhil (2002) discovered
unusual microaerophilic bacteria
Many chemolithotrophs use hydrogen growing in highly sulfidic marine
as an electron donor sediments that developed in
Nivå Bay in Denmark, owing to
Hydrogen is a common product of the breakdown of organic compounds and many bac- the burial of large amounts of
teria can use it as the electron donor. In aerobic hydrogen bacteria, electrons from H2 are decaying seaweed in the sedi-
transferred to quinone and proceed via a cytochrome chain, ending with the reduction of ment during winter storms. SRB
O2 to H2O. The enzyme hydrogenase catalyzes the reaction H2 + ½O2 → H2O. The reaction generate large quantities of S2−,
has a high yield of energy, supporting the synthesis of ATP. Cultured examples found in and the whole area can become
marine habitats include Alcaligenes, Pseudomonas, and Ralstonia, most of which can fix anoxic and colonized by other
CO2 to grow autotrophically, although when organic compounds are available the synthesis bacteria. One type uses an unusual
of hydrogenase and CBB cycle enzymes is usually repressed. Hydrogen-oxidizing bacteria sideways rotary movement to
are typically associated with sediments and suspended particles where an oxic-anoxic inter- migrate toward O2 and attaches
to solid surfaces via stalks, form-
face provides the optimum conditions for growth. Some chemosynthetic endosymbionts of
ing conspicuous veils of mucus-
invertebrates in vents, seeps, and sediments can use H2 for chemosynthesis. Some thermo- like polymer. Stalks of different
philic chemolithotrophs found at hydrothermal vents use NO3- to oxidize H2, S0, or S2O32−. bacterial cells stick to each other to
build up the veil, and holes in the
structure are quickly repaired. The
Bacterial and archaeal nitrification is a major bacteria position themselves at the
process in the marine nitrogen cycle preferred concentration of 2 μM
O2, either by chemotaxis to O2 or
The inorganic compounds ammonia (NH3), ammonium ion (NH4+), and nitrite (NO2-) serve by increasing their stalk lengths.
as substrates for chemolithotrophy in the process known as nitrification. The process occurs The attached cells keep rotating
in two steps; there is some variation in the enzymes and intermediates, but the overall reac- to enhance the rate of O2 uptake
tions are: through advection of a flow of
water. Although the bacterium can
+ - + -
1. Ammonium oxidation: NH 4 + 2H 2 O ® NO2 + 8H + 6e be grown in enrichment cultures
- - +
2. Nitrite oxidation: NO2 + 2H 2 O ® NO3 + 2H + 2e - in the laboratory, it has not been
isolated, so classification has been
difficult. Muyzer et al. (2005) used
One group of microbes carry out oxidation of NH3 or NH 4 + using the enzyme ammonia molecular analysis of the enrich-
monooxygenase (AMO, an integral membrane protein), which produces the intermediate ment cultures to show that the
hydroxylamine (NH2OH); this is then oxidized to nitrite (NO2 - ) by hydroxylamine oxido- bacterium is a member of the
reductase in the periplasm. A different group of microbes oxidize NO2 - to nitrate (NO3- ) Epsilonproteobacteria, which they
using the enzyme nitrite oxidoreductase (NXR). Nitrifying bacteria are widely distributed named “Ca. Thioturbo danicus”
and active in suspended particles and in the upper, oxic layers of sediments. The ammonia (“the sulfur whirl of Denmark”).
oxidizers Nitrosomonas and Nitrosococcus fix CO2 and are obligate chemolithoautotrophs,
while the nitrite oxidizers Nitrobacter, Nitrococcus, and Nitrospina can be mixotrophic,
using simple organic compounds as a source of carbon. These reactions yield little energy,
and oxidation of 35 ammonia molecules or 15 nitrite molecules is required to produce fixa-
tion of one molecule of CO2 resulting in low growth yields.

Through these activities, nitrifying bacteria play a major role in nitrogen cycling in the oceans
by degrading organic matter to create nitrate, which is a major source of nitrogen for primary
production by phytoplankton, algae, and plants; it is often the limiting nutrient (see Chapter
9). Nitrification is especially important in shallow coastal sediments and beneath upwell-
ing areas such as the Peruvian coast and the Arabian Sea. Like the phototrophs, nitrifying
bacteria have extensive internal structures to increase the surface area of the membrane. It is
difficult to obtain estimates of the abundance and community structure of nitrifying bacteria.
Although most can be cultivated in the laboratory, the energetics of this mode of chemolithot-
rophy mean that the bacteria grow slowly and are difficult to work with. Immunofluorescence
and genomic methods reveal that Nitrosococcus oceani and similar strains are widespread
in many marine environments, with worldwide distribution at concentrations between 103
82 Chapter 3

and 104 cells mL-1. This organism is thought to be responsible for significant oxidation of
ammonia in the open ocean. Nitrospira also seems to be distributed worldwide. Study of
the activities of nitrifiers and their contribution to nitrogen cycling is usually carried out
using isotopic methods with 15 NO2 - or 15 NH 4 + or by using various inhibitors of nitrification
enzymes (e.g. nitrapyrin inhibits AMO). Nitrification is a strictly aerobic process and suf-
ficient O2 usually only penetrates a few millimeters into sediments. However, the activity
of burrowing worms, molluscs, and other animals (bioturbation) increases O2 availability to
deeper levels of sediments.

Until quite recently, ammonia oxidation was believed to be carried out exclusively by the
specialized aerobic genera of the Bacteria named above. We now know that highly efficient
nitrification is also carried out by the Thaumarchaeota phylum of the Archaea. This is of
particular importance because this group is so abundant, especially in the deep ocean. This
is explored in Box 5.1.

Ammonia can also support anaerobic

chemolithoautotrophy
Anaerobic ammonia oxidation (anammox), using nitrite as the oxidant and yielding nitro-
gen as the end product, is now known to be an important component of the marine nitrogen
cycle, with the overall reaction NH 4 + + NO2 - ® N 2 + 2H 2O. Anammox is only carried out by
members of the Planctomycetes phylum, which possess an unusual cell structure including a
membrane-bound nucleus and organelle-like structures (Figure 4.13). Those species that carry
out anammox possess an organelle known as the anammoxosome, which can occupy up to
half of the cell and has a membrane composed of unusual lipids known as ladderanes, densely
packed in a rigid ladder-like array that is impervious to many agents. As shown in Figure 3.9,
the enzyme nitrite reductase reduces NO2 - to NO, which then reacts with NH 4 + via hydrazine
hydrolase yielding hydrazine (N2H4)—a highly reductive compound used as rocket fuel—so
this special membrane is needed to prevent it leaking into the cytoplasm. Finally, N2H4 is
oxidized to N2 by the enzyme hydrazine dehydrogenase, yielding electrons, some of which are
used by the anammoxosome electron transport chain to generate a proton-motive force for the
synthesis of ATP. Like aerobic ammonia-oxidizing bacteria, anammox bacteria are autotro-
phic, using CO2 as their sole carbon source and NO2 - as electron donor in the overall reaction:

NH 4 + + 1.32NO2 - + 0.066HCO3 - + 0.13H +

® 1.02N 2 + 0.256NO3 - + 0.066(CH 2O0.5H 0.15 ) + 2.0H 2O

The ecological importance of anammox is discussed in Chapter 9.

Figure 3.9 Anaerobic ammonia

oxidation (anammox). a. A simplified
depiction of a planctomycete anam-
mox cell, showing the anammoxo-
some organelle, which has a dense
impermeable membrane composes
ladderane lipids. b. The pathway for
the anammox reaction. Reprinted
from DeLong (2002) with permission
of Springer Nature.
 Metabolic Diversity and Ecophysiology 83

CARBON AND NITROGEN FIXATION MICROBIOLOGISTS

The Calvin–Benson–Bassham (CBB) cycle is the i DOUBTED EXISTENCE
OF ANAMMOX
main method of carbon fixation in autotrophs Anammox was discovered in the
In autotrophs, the generation of ATP by the mechanisms described in the previous sec- late 1980s by the Delft microbiolo-
tions is used to drive the incorporation of carbon into organic compounds for biosynthesis gist Gijs Kuenen, who was working
by oxidation of CO2. The great majority of photoautotrophs and chemolithoautotrophs do with a yeast company to help them
this via the cycle of reactions first discovered in plants by Calvin, Benson, and Bassham in to reduce pollution from the efflu-
ent of their plant. When nitrate was
the 1950s. The key enzyme in this pathway is ribulose bisphosphate carboxylase/oxygen-
added to an anaerobic digester,
ase (RuBisCO), which is unique to autotrophs. RuBisCO is a relatively inefficient enzyme, the concentration of toxic ammo-
and cyanobacteria and some chemolithoautotrophs have evolved CO2 concentrating mecha- nia fell. Fellow microbiologists were
nisms (CCMs) that increase CO2 levels around the active site of the enzyme up to 1000 very skeptical that a microorganism
times. Bicarbonate (HCO3-) accumulates in the cytosol via the action of several high- and was responsible for this process—a
low-affinity transporters in both the plasma membrane and the thylakoid membrane (in reaction predicted on thermody-
cyanobacteria); synthesis of these is regulated in accordance with external levels of CO2 namic grounds by Broda (1977)—
and HCO3-. Large amounts of RuBisCO are contained in inclusion bodies called carboxy- but Kuenen and colleagues
somes, about 100 nm in diameter and surrounded by a thin protein shell or envelope. They persisted and eventually obtained
contain a crystal-like array of up to 250 RuBisCO molecules and another enzyme, carbonic highly enriched cultures of the
bacterium responsible, which
anhydrase (CA). HCO3- diffuses into the carboxysome and CA converts it to CO2. These
they named “Ca. Brocadia anam-
mechanisms greatly increase the efficiency of carbon fixation, and if genetic mutations are
moxidans.” Ten other Candidatus
generated resulting in structural changes or loss of the carboxysomes or CA, cells require species have since been described
much higher concentrations of CO2 than normal for growth. Diversity in the components in wastewater treatment plants, all
and activity of the CCMs contributes to the competitive ability of cyanobacteria to occupy members of the Planctomycetes.
different habitats. Biotechnological applications are
being developed to enhance their
As shown in Figure 3.10A, each turn of the CBB cycle incorporates one molecule of CO2 use for treatment of wastewater
by carboxylation of ribulose bisphosphate (RUBP). The intermediate molecule is unstable and industrial effluent at much
and immediately dissociates to two molecules of 3-phosphoglycerate (PGA). This is then lower costs and reduced output
phosphorylated to form glyceraldehyde-3-phosphate, which is also a key intermediate of the of greenhouse gases than conven-
tional methods (Kuenen, 2008;
glycolytic pathway. Therefore, monomers of hexose sugars are synthesized by a reversal of
Pereira et al., 2017).
glycolysis. The cycle requires 12 molecules of NADPH and ATP to reduce 12 PGAs to 12
glyceraldehyde-3-phosphates, and a further six ATPs to convert six ribulose phosphates to six
ribulose bisphosphates. Thus, the overall reaction mechanism requires 12 NADPH plus 18
ATP, for the incorporation of six molecules of CO2 to generate one molecule of C6 monomer
(hexose sugar). Hexose monomers are converted by various reactions to other metabolites
and the building blocks of macromolecules. If energy and reducing power are in excess sup-
ply, hexoses are converted to storage polymers such as starch, glycogen, or polyhydroxybu-
tyrate, and deposited as cellular inclusions. The final step in the CBB cycle is the regeneration
of one molecule of RUBP by another important enzyme, phosphoribulokinase.

Some Archaea and Bacteria use

alternative pathways to fix CO2
Some Archaea and Bacteria in the deep sea, sediments, and vents, as well as some symbi-
onts of marine invertebrates (see Chapter 10), can fix CO2 via at least two other possible
routes. The reverse or reductive tricarboxylic acid (TCA) cycle (Figure 3.10B) uses ferre-
doxin-linked enzymes leading to the formation of acetate. It was first described in Aquifex
and Chlorobium, but sequence analysis of microbial communities from hydrothermal vents
reveals that genes encoding the key enzyme of this pathway (ATP citrate lyase) are more
common than genes for RuBisCO. This suggests that the reverse TCA cycle may be the main
mechanism of CO2 fixation in chemolithoautotrophs in these habitats.

Anammox bacteria use the acetyl-coA pathway, in which acetate is formed by combining the
methyl group formed from reduction of one molecule of CO2 with the carbonyl group from
reduction of another CO2 molecule with nitrite as the electron acceptor:

CO2 + 2NO3− + H 2O → CH 2O + 2NO3−

84 Chapter 3

CO2
COOH
COSCoA
C O
CH3 acetyl-CoA
CH3
2 Fdred
pyruvate

COOH COOH
Light or C O ATP
12 12 CoA CH2
chemical
3-phosphoglycerate ATP energy CH2 HO C COOH
36 12 COOH CH2
6 ATP + Pi oxaloacetate
COOH
ribulose1,5-bisphosphate 12
1,3-bisphosphoglycerate citrate
30
36 12 NADPH
several steps COOH
COOH
12 NADP+ COSCoA
HC COOH
10 12 CH2
gyceraldehyde- gyceraldehyde- CH2
CH2
3-phosphate 3-phosphate COOH
36 COOH COOH
30 CO2 CO2 isocitrate
succinyl-CoA C O
1
fructose-6-phosphate CH2

6 2 Fdred CH2 NAD(P)H

Biosynthesis CoA COOH
2-oxoglutarate
(a) (b)

Figure 3.10 Major pathways of The acetyl-coA pathway for reduction of CO2 is also used by methanogens and by acetate-
carbon fixation in autotrophs. (a) The producing chemoorganotrophs growing in anaerobic sediments. In this case, H2 serves as the
CBB cycle. One molecule of hexose electron donor in the reaction:
sugar is produced from each six mol-
ecules of CO2 incorporated; figures
in circles show the total number 4H 2 + H + + 2HCO3- ® CH 3COO - + 4H 2O.
of carbon atoms involved. (c) The
reverse (reductive) tricarboxylic acid
cycle used in some chemolithotro- Fixation of nitrogen makes this essential element
phic autotrophs. Each cycle results available for building cellular material in all life
in the fixation of one molecule of
CO2 and production of one pyruvate Although 80% of the Earth’s atmosphere is dinitrogen gas (N2), most organisms cannot
molecule, which is subsequently metabolize it directly. Only about 5–8% of naturally occurring fixation of this N2 occurs
phosphorylated using ATP, leading to abiotically (as a result of lightning activity), so life on Earth depends almost entirely on diaz-
the biosynthesis of cellular material. otrophy, the reduction of N2 into NH3, which is then assimilated into organic compounds for
(Intermediate steps are omitted for biosynthesis of amino acids, nucleotides, and other essential compounds. Diazotrophy is a
clarity). highly energy-demanding process that occurs only in members of the Bacteria and Archaea.
The necessary genes and proteins are highly conserved, but there is great physiological
and phylogenetic diversity among nitrogen-fixing species, which include photolithotrophs,
chemolithotrophs, and chemoorganotrophs. The reduction of N2 to NH3 is extremely energy
demanding, because N2 is a very unreactive molecule due to the stability of the dinitrogen
triple bond. At least 16 ATPs are required for the reaction: N2 + 8H+ + 8e- → 2 NH3 + H2.

As shown in Figure 3.11, incorporation of N2 depends on the nitrogenase complex, consisting

of the enzymes dinitrogenase (which is composed of the two copies of each of the NifD and
NifK subunits and contains Fe + Mo) and dinitrogenase reductase (which is composed of two
copies of the NifH subunit and contains Fe). For reduction of each molecule of N2, four pairs
of electrons are sequentially passed from ferredoxin or flavodoxin to the dinitrogen reductase
subunits (NifH). For each electron transferred, two ATP molecules are required to provide
the energy for a change in conformation of the NifH subunits. This allow them to interact
with the dinitrogenase component and transfer the electrons before dissociating. In the first
step, a pair of electrons combine with two protons to form H2, then three further pairs of
electrons are sequentially routed through the reductase enzyme to complete the next steps in
 Metabolic Diversity and Ecophysiology 85

8e- Figure 3.11 Nitrogen fixation. The

16ATP
Red nitrogenase complex consists of
FeS subunits of dinitrogenase reductase
Ox 16ADP+ 16Pi (NifH, orange) and dinitrogenase,
+ composed of two copies of each of
2H
H2 NºN the NifD (blue) and NifK (green)
+
2H subunits.
+ HN=NH H2
2H
Dinitrogenase(FeMo)
+ H2N-NH2
2H 2xNifD
2xNifK
2 NH3

Dinitrogen reductase(Fe) 2 NH3 Biosynthesis

2xNifH

reduction of N2, which occurs at the Fe-Mo cofactor of the dinitrogenase enzyme. The cycle
of electron transfer and ATP-mediated association and disassociation of the nitrogenase com-
plex is repeated at each step.

There are about 20 nif genes involved in synthesis and regulation of the nitrogen-fixing sys-
tem. Although organisms show variation in some of the nif genes, the key genes nifHDK
that encode the nitrogenase complex are highly conserved, enabling the use of probes and
sequence analysis to detect nitrogen-fixing microbes in environmental samples, as discussed
later. The nif genes are tightly regulated by different processes and occur in operons under
common control (the nif regulon). Transcription of the nif structural genes encoding these
enzymes is prevented by the presence of O2 and fixed nitrogen such as NH3 and NO3-. In addi-
tion, dinitrogenase is very labile in the presence of O2. This poses no problem under anaero-
i
WHAT’S IN A NAME?
bic conditions, but aerobic nitrogen fixers such as photosynthetic cyanobacteria must employ GENES AND PROTEINS
special strategies for protecting the enzyme from O2, discussed in Chapter 4. Furthermore,
Distinguishing between proteins
bacteria and archaea have a rapid and reversible system for shutting down nitrogen-fixing
and the genes that encode them
activity in the presence of excess NH3 in the environment by limiting the flow of electrons is sometimes confusing, so it is
to the nitrogenase complex. The aerobic cyanobacteria and the anaerobic archaeal metha- important to become familiar with
nogens play a particularly significant role in nitrogen cycling in the oceans, as discussed in the conventions for abbreviations.
Chapter 9. Microbial proteins are typically rep-
resented as a three- or four-letter
abbreviation, in normal type with
HETEROTROPHIC METABOLISM the first letter capitalized, while the
genes that encode them are writ-
Many marine microbes obtain energy by the ten in lower case italic type. For
fermentation of organic compounds example, Nif includes all members
of the group of proteins involved
All energy-yielding metabolic processes depend on coupled reduction-oxidation (redox) reac- in nitrogen fixation; the individual
tions, with conservation of the energy in the molecule ATP using the reaction: structural and regulatory proteins
are called NifA, NifB, NifC, etc. The
genes are represented by nif, nifA,
ADP + Pi + energy  ATP + H 2O nifB, etc. An operon contains sev-
eral genes that are transcribed into
Fermentation occurs only under anaerobic conditions and is the process by which substrates a single mRNA under the control
are transformed by sequential redox reactions without the involvement of an external elec- of a single regulatory gene, so we
tron acceptor. Fermentative microbes can be either strict or facultative anaerobes; the latter often need to describe these col-
can often grow using respiration when O2 is present. Fermentation yields much less ATP lectively. For example, nifHDK rep-
resents the three genes nifH, nifD,
than aerobic respiration for each molecule of substrate, so growth of facultative anaerobes
nifK in the order that they occur
is better in the presence of O2. A very wide range of fermentation pathways exist in marine in the operon. Finally, different
bacteria and archaea, especially those associated with degradation of organic material in operons may be under the control
sediments, animal guts, and microbial mats. Carbohydrates, amino acids, purines, and of a single regulatory protein; this
pyrimidines can all serve as substrates. Some individual species are adapted to use only a is called a regulon. In this example,
limited range of substrates, while others can carry out fermentations with various starting we can say that all the Nif proteins
materials. are encoded by the nif regulon.
86 Chapter 3

DO MARINE PLANTS
Anaerobic respiration has major
? AND ANIMALS
DEPEND ON
importance in marine processes
In respiration, electrons are transferred via a sequence of redox reactions through an electron
SYMBIOTIC NITROGEN transport chain located in the cell membrane, which leads to the transfer of protons to the
FIXATION?
exterior of the membrane and generates a proton-motive force. The commonest and most
In terrestrial biology, the symbiotic familiar type of respiration is aerobic, in which O2 is the terminal electron acceptor, but many
partnership of leguminous plants microbes carry out anaerobic respiration using other terminal electron acceptors. Anaerobic
such as peas, beans, and alfalfa respiration uses electron transport chains containing cytochromes, quinones, and iron-sulfur
with symbiotic bacteria belonging proteins like those in aerobic metabolism. Some organisms can carry out both aerobic and
to the genus Rhizobium has been anaerobic respiration, using alternate electropositive electron acceptors such as Mn4+, Fe3+,
extensively investigated, because
NO3-, and NO2- that yield sufficient energy for anaerobic growth. But because O2 is the most
of its importance in soil fertility.
Does something similar happen in
efficient electron acceptor and produces the most energy for oxidation of an electron donor,
the marine environment? The high such facultative organisms will always use O2 preferentially if it is available. The respiration
productivity of seagrass beds and of more electronegative electron acceptors like SO42-, S0, and CO2 is only possible anaerobi-
mangroves is linked to close cou- cally. Several types of anaerobic respiration have major importance in marine biogeochemi-
pling with nitrogen-fixing bacteria cal processes; especially denitrification, sulfate reduction, methanogenesis, and methane
in a biofilm of bacteria growing on oxidation, which are considered below.
the roots (the rhizosphere), mak-
ing fixed nitrogen available to the
plants. In recent years, many exam- Nitrate reduction and denitrification
ples of nitrogen-fixing symbionts
have also been discovered in corals,
release nitrogen and other gases
sponges, molluscs, crustacean Many organisms use nitrate ( NO3− ) as an electron acceptor in anaerobic respiration. The
larvae, and other invertebrates, as complete process of reduction is known as dissimilative reduction of nitrate. The sequence of
well as inter-microbial symbioses in transformations is NO3− → NO2− → NO → N 2O → N 2 and the overall net redox reaction is:
diatoms and other microalgae. The
importance of these associations is
discussed in Chapters 9 and 10. 2NO3 - ® 10e - + 12H + ® N 2 + 6H 2O

This can be broken down into the following sequential reactions, catalyzed by the enzymes
shown below:
- + - -
(1) NO3 + 2H + 2e ® NO3 + H 2O (nitrate reductase)
- + -
(2) NO3 + 2H + e ® NO + H 2O (nitrite reductase)
+ −
(3) 2NO + 2H + 2e → N 2O + H 2O (nitric oxide reductase)
(4) N 2O + 2H + 2e − → N 2 + H 2 O (nitrous oxide reductase)
+

Some organisms only carry out the first reaction, but others carry out the subsequent steps
and this results in the formation of the gases nitric oxide, nitrous oxide, and nitrogen. Because
these gases are released and may return to the atmosphere, resulting in the loss of biologically
useful nitrogen, the process is known as denitrification. Furthermore, the intermediate gases
have environmental consequences in climate warming and production of acid rain, so these
processes are of critical importance in marine nitrogen cycling, especially in anoxic environ-
ments and oxygen minimum zones (Chapter 9). The synthesis of the enzymes involved in
denitrification is repressed by molecular O2 and induced by high levels of NO3-. However,
denitrification is an important beneficial process in the treatment of organic-rich wastewater
because it lowers the levels of oxygen-demanding microbial growth. Most denitrifiers are
phylogenetically and metabolically diverse members of the Proteobacteria, which are facul-
tative anaerobes and can carry out aerobic respiration of many different organic compounds.

Sulfate reduction is a major process in marine sediments

Sulfate is abundant in seawater and sulfate-reducing bacteria (SRB) are very widely distrib-
uted in anoxic marine sediments, as members of microbial mat communities, and at hydro-
thermal vents. Although most marine members of the Bacteria and Archaea can carry out
assimilative sulfate reduction for biosynthesis of cellular compounds, those known as SRB
carry out dissimilative sulfate reduction. The SRB can link the oxidation of substrates to ATP
generation, using SO42- as electron acceptor and leading to the production of H2S. Although
SRB are found in several phyla of the Bacteria, the most important types are currently
assigned to the class Deltaproteobacteria (see Chapter 4 for discussion of their taxonomic
 Metabolic Diversity and Ecophysiology 87

Figure 3.12 Pathways of assimilative

and dissimilative reduction of sulfate.

position). Only one member of the Archaea is known to reduce sulfate: the hydrothermal
vent genus Archaeoglobus. A wide range of organic compounds generated by many different
members of the microbial community can be used by SRB as electron donors (including H2,
organic acids, and alcohols produced by fermentation). Many SRB oxidize acetate in order to
provide electrons for SO2- reduction using the acetyl-coA pathway. The reversible reactions
of this pathway can also be employed for acetate synthesis, enabling the bacteria to grow
autotrophically by incorporation of CO2. A few SRB can also grow autotrophically using H2.

As shown in Figure 3.12, the first step in sulfate reduction is activation of sulfate by ATP to
adenosine-5′-phosphosulfate (APS). In assimilative sulfate reduction, another P is added to
APS from ATP to form phosphoadenosine-5'phosphosulfate, which is then reduced to SO32−
and then to HS− for the synthesis of the amino acids cysteine and methionine, coenzyme
A, and other cell factors that contain sulfur. By contrast, in dissimilative reduction, APS is
reduced directly to SO32− and then H2S. The dissimilative metabolism of SRB results in the
production of large amounts of H2S which is excreted from the cell, where it can serve as a
substrate for SOB and react with metals to form metal sulfides that characterize marine sedi-
ments. Sulfate reduction is usually considered a strictly anaerobic process, but some types are
tolerant of O2. Some SRB can also reduce other oxidized sulfur compounds including sulfite
(SO32–), dithionite (S2O42–), thiosulfate (S2O32-), trithionate (S3O62-), tetrathionate (S2O62-), and
elemental sulfur (S0). This enables SRB to utilize intermediates produced by SOB as well as
those produced from their own metabolism.

Because they are ubiquitous in anoxic sediments, SRB play a major role in the sulfur and
carbon cycles through the oxidation of organic matter reaching the sea floor. An important
aspect is the interaction of these activities with the production and oxidation of methane, dis-
cussed below. Their activities also cause problems of corrosion and toxicity in the oil industry
and they influence the mobilization of toxic mercury, discussed in Chapter 13. SRB are also
implicated in black band disease of corals, discussed in Chapter 11.

MICROBIAL PRODUCTION AND

OXIDATION OF METHANE
Methanogenesis is unique to the Archaea
Many members of the Euryarchaeota produce methane as the final step in the anaerobic bio-
degradation of organic material. Bacterial fermentation produces H2, CO2, and acetate as end
products, which are then used by methanogens. Most methanogens use H2, with CO2 serving
as both the oxidant for energy generation and for incorporation into cellular material. The
overall reaction is: CO2 + 4H2 → CH4 + 2H2O. This reaction depends on the input of eight
88 Chapter 3

electrons, added two at a time, which leads to four steps with intermediate oxidation states.
This depends on coenzymes, unique to this group of Archaea, that transfer electrons from
hydrogen (or other donors, if used) and act as carriers for the C1 unit from the substrate (CO2)
to the end product (CH4). Besides H2 + CO2, other compounds methanogens form methane
from CH3-containing compounds such as methanol, acetate, and dimethyl sulfide. In all the
possible pathways, the final step of reduction leads to the formation of a single molecule of
ATP via generation of a proton-motive force. Sugars and fatty acids are not directly used for
methane generation, but because methanogens exist in syntrophic communities with fermen-
tative bacteria, virtually any organic compound can eventually be converted to methane. In
the presence of high levels of SO42-, methanogens are in competition with SRB because both
groups use H2 or acetate. Therefore, methanogens are usually found in deeper sediments
where sulfate levels are lower.

Methane produced by archaeal activity in sediments has a number of fates. Much of the
methane produced in anoxic sediments diffuses into adjacent oxic regions, where it sus-
tains the growth of methanotrophic bacteria—thus recycling the gaseous end product of
degradation of organic matter—and a large fraction is oxidized anaerobically, as discussed
below. Methane is also trapped in a crystal lattice of water molecules, forming solid methane
hydrate (clathrate). This only forms under appropriate conditions of geology, temperature,
and pressure—the hydrate stability zone—forming widespread deposits on the continental
shelf, typically in sediments between 250 and 550 m deep but sometimes mixed with other
hydrocarbons nearer the surface. The production of methane in sediments and deep-sea seeps
is the basis of many symbioses with invertebrates, as discussed in Chapter 10.

Besides anoxic sediments, methane is found in the oxygenated seawater of oceans. Some of
this is due to the activity of methanogens as members of the intestinal microbiota of cope-
pods, fish, and other marine animals or their activity in anoxic niches inside marine snow
particles. Many protists, especially ciliates, also contain methanogens as endosymbionts.

Methane is produced in the surface ocean

by bacterial cleavage of phosphonates
Oceanographers have long puzzled over the fact that large amounts of methane flux to the
atmosphere occurs in well-oxygenated ocean gyres, whereas methanogenesis is a strictly
anaerobic process. However, it has recently been found that many bacteria are capable of
producing methane in the oxygenated conditions of the surface ocean and that this accounts
for about 4% of the methane reaching the atmosphere. This occurs via the action of C-P lyase
enzymes that, under phosphate-limited conditions, cleave methylphosphonate (MPn) com-
pounds. These are simple, single-carbon compounds with a stable phosphonate (C-P) bond.
Gene sequences for both the biosynthesis of MPn and for the C-P lyase enzyme are present
in marine metagenomes. The reaction creates ethylene (which is consumed by heterotrophs)
and methane, most of which escapes to the atmosphere; so the main benefit to the degrad-
ing organism appears to be the release of phosphate needed for biosynthesis of nucleotides,
lipids, and other key cellular components. MPn-polysaccharides appear to be mostly syn-
thesized by cyanobacteria such as Prochlorococcus, and experimental studies have shown
DOM contains large amounts of these compounds that are degraded aerobically by bacteria
in seawater in sufficient quantity to explain the daily flux of methane. Some cultured strains
of SAR11, the most abundant clade of oceanic microbes, have been shown to cleave MPn
when phosphate limited, although others do not. Intriguingly, some SAR11 have been shown
to possess MPN synthase, suggesting that members of this clade may be either a source or
sink of methane in the ocean.

Anaerobic oxidation of methane (AOM) in

sediments is coupled to sulfate reduction
As described above, methane produced by the anaerobic process in deep sediments diffuses
upwards to be consumed by aerobic methanotrophs in the oxic zone. However, for many
years, geochemists had been puzzled by observations that as much as 80% of this disap-
peared before it reached zones where contact with oxygen could occur. Sulfate is the only
possible electron acceptor that could account for anaerobic methane oxidation in these anoxic
 Metabolic Diversity and Ecophysiology 89

sediments. Stable isotope methods combined with 16S rRNA gene sequencing and FISH
(Figure 3.13B) led to the discovery of a syntrophic consortium of Archaea and SRB carrying
out AOM via the reaction:

CH 4 + SO 4 2 - ® HCO3 - + HS- + H 2O

A model of these interactions is shown in Figure 3.13A. The anaerobic methanotrophic

Archaea (ANME) are closely related to the methanogens and it is thought that the reaction
functions as a reversal of methanogenesis. The syntrophy with SRB may be obligate, because
attempts to culture them have been unsuccessful. The archaeal partners have been shown to
fix nitrogen and transfer it to the SRB (Figure 3.13C). Nitrogen fixation is very expensive in
terms of energy consumption and the growth of these consortia is very slow—the doubling
time is 3–6 months—but it seems to be advantageous for the archaea to transfer fixed nitro-
gen to the SRB, which they need to utilize methane. Different microbial consortia have sub-
sequently been discovered to carry out ANME via coupling to reduction of NO3 - , NO2 - Fe3 +
and Mn4+ and it is likely that organisms capable of using other electron acceptors remain to
be discovered.

The sulfide and bicarbonate products of the reaction are very important. These consortia
contribute to the construction of reefs of precipitated carbonate at methane seeps, especially
in the Black Sea. The sulfide can be used in chemosynthetic symbioses involving tubeworms,
clams, or giant SOB, which form thick carpets on the seabed above the methanotrophic
consortia. Elsewhere, extensive microbial growth above methane seeps may sustain deep-sea
coral communities. This type of methanotrophic metabolism may have evolved very early on
Earth, possibly predating the development of O2 in the atmosphere from oxygenic photosyn-
thesis (about 2.2 BYA).

Figure 3.13 A. Model showing

anaerobic oxidation of methane
(ANME) in marine sediments, based
on the findings of Boetius et al.
(2000). The metabolic consortium
contains methane-oxidizing archaea
surrounded by sulfate-reducing bac-
teria (SRB). Other reaction pathways
using alternative electron accep-
tors are now known. B. Confocal
micrograph visualized by FISH with
probes directed against the archaeal
(red) and bacterial (green) partners.
C. NanoSIMS image showing 15N
incorporation, indicating transfer of
fixed nitrogen from the archaeal cells
to SRB. Credits: A. Reprinted from
DeLong (2000) with permission of
Springer Nature; B, C. Anne Dekas
and Victoria Orphan, California
Institute of Technology.
90 Chapter 3

Many marine microbes oxidize methane

and other C1 compounds
Chemoorganotrophs catabolize a wide range of organic compounds to obtain energy and a
carbon source for biosynthesis. The simplest of these is methane (CH4), and many organ-
isms known as methylotrophs can utilize this or various other methyl- or one-carbon (C1)
compounds. A very wide range of bacteria in different phylogenetic groups can carry out this
process and it is of great important in carbon and sulfur cycling in ocean processes. In par-
ticular, the compound dimethylsulfoniopropionate (DMSP) is produced in great amounts by
phytoplankton and supplies much of the carbon and almost all of the sulfur for the growth of
heterotrophic bacteria in ocean food webs, including the dominant SAR11 and Roseobacter
clades (see Chapter 9). Many methylotrophs can also use more complex organic compounds,
although some bacteria in the Alphaproteobacteria and Gammaproteobacteria are obligate
methylotrophs and can use only C1 compounds in their metabolism.

A subset known as the methanotrophs can grow using methane. These possess a unique
enzyme system, methane monooxygenase (MMO), which occurs in membrane-bound and
cytoplasmic forms whose synthesis depends on the availability of copper. MMO obtains elec-
trons from NADH and incorporates O2 leading to the formation of methanol, with subsequent
conversion to formaldehyde by methanol dehydrogenase in the reaction:

CH 4 → CH 3COOH → CH 2O → HCOO − → CO2

The coenzymes tetrahydrofolate or methanopterin enable conversion of the CH2O intermedi-

ate, producing electrons that enter the electron transport chain to produce a proton-motive
force for ATP synthesis.

Different phylogenetic groups of methanotrophs vary in the route of assimilation of C1 units

into cell material. Some use the unique ribulose monophosphate pathway, which requires
one molecule of ATP and no reducing power to assimilate three molecules of CH2O into
one molecule of glyceraldehyde-3-phosphate. Other types use the serine pathway for carbon
assimilation into acetyl CoA, which is less efficient due to the requirement for two molecules
of both NADH and ATP:

i
MOST MARINE
BACTERIA REQUIRE
SODIUM CH 2O + CO2 + 2NADH + 2ATP + CoA → CH 3 -CO~S-CoA
A particular feature of the majority
of marine bacteria is their abso- Apart from the importance of free-living forms in methane oxidation in sediments and ocean
lute requirement for Na+ (usually processes, methanotrophs also occur as symbionts of mussels found near cold seeps of meth-
in the range 0.5–5.0%) and they ane-rich material on the ocean floor, providing the animals with a direct source of nutrition
fail to grow when K+ is substituted (see Chapter 10). Methanotrophs are also important in bioremediation of low-molecular-
in the culture medium. Sodium is weight halogenated compounds, an important process in contaminated marine sediments
required by transport proteins (per-
meases), which bring substrates
into the cell via Na+-substrate NUTRIENT ACQUISITION AND MICROBIAL GROWTH
symport across the membrane
and also help to stabilize the outer Microbial metabolism depends on nutrient uptake
membrane found in Gram-negative
bacteria. This distinguishes them The remainder of this chapter is mainly concerned with the growth of heterotrophic bacteria
from closely related terrestrial and and archaea and the environmental factors that affect them. A huge diversity of biochemical
freshwater species. As might be pathways enables heterotrophs to obtain their carbon and energy from the breakdown of pre-
expected, most grow optimally formed organic compounds, including carbohydrates, organic acids, amino acids, and pep-
at a concentration of NaCl similar tides by respiration and fermentation. Besides C and N—the major components of cellular
to that found in seawater (about material, which heterotrophs acquire from preformed organic compounds—microbes need a
3.0–3.5% NaCl), although they dif- supply of the macronutrients P, S, K, Fe, Mg, Ca, and Na These are needed as key constitu-
fer greatly in the lower and upper ents of macromolecules like proteins, nucleotides, and lipids or in smaller quantities as cofac-
limits for growth. Gram-positive tors for enzyme function or stabilization of structures such as nucleic acids, membranes, and
bacteria (which are much less com-
ribosomes. The importance of these elements in ocean processes is discussed in Chapters 8
mon in seawater) do not seem to
have this requirement.
and 9. In addition, microbes often depend on the presence of very small amounts of metals
such as Mn, Mg, Co, Zn, and Cu as enzyme cofactors. There is also growing recognition of
 Metabolic Diversity and Ecophysiology 91

the importance for marine microbial growth of growth factors such as vitamins (especially
the B vitamins cobalamin, thiamin, and biotin), or specific amino acids, purines, or pyrimi-
dines. Individual microorganisms are often auxotrophic—lacking the biosynthetic pathways
needed for specific compounds—and they must obtain these compounds from other organ-
isms within the community. This applies particularly to members of the picoplankton, which
have often undergone genome reduction resulting in loss of genes for certain pathways and
therefore depend on exogenous sources. Despite the perception that microalgae like diatoms
and dinoflagellates just need light, CO2, and a few inorganic nutrients, many are auxotrophic
and do not synthesize particular essential vitamins. For example, half of surveyed species
of algae require vitamin B12, cobalamin. The availability of vitamins depends on overall
microbial community composition or tight associations between algae and bacteria such as
Roseobacter, Marinobacter, Sulfitobacter, and their relatives. This has a major effect on the
regional and seasonal distribution and activity of phytoplankton.

Apart from a few very small hydrophobic molecules that can diffuse into the cell, all
nutrients are transported into the cell via transport proteins in the cytoplasmic membrane.
All transport proteins depend on an energy-dependent conformational change when a
substance binds to it, forming a channel that carries the substance across the membrane.
Transporters may be specific for particular compounds or sometimes transport groups of
related compounds. Some molecules like sugars and amino acids can cross the membrane
using simple channels using energy via a proton pump and others are transported by a
phosphotransferase system. The presence of the outer membrane (OM) in Gram-negative
bacteria (the dominant form in bacterioplankton) requires them to use additional OM recep-
tors and periplasmic binding proteins to bridge this additional permeability barrier. ATP-
binding cassette (ABC) transporters consist of a periplasmic binding protein (PBP) that is
highly efficient at binding low concentrations of substrates that are then transferred to a
membrane-spanning transporter that is linked to a cytoplasmic ATP hydrolase, supplying
energy. Many hundreds of high-affinity, multisubstrate ABC transporters and even greater
numbers of the highly-expressed PBPs have been identified in metagenomic and metapro-
teomic analyses of low-nutrient ocean water. For example, the genome of SAR11 bacteria
encodes individual transporters that import various organic compounds (including amino
acids, DMSP, taurine, polyamines, simple sugars, and phosphonates) and inorganic com-
pounds (e.g. iron and complexed phosphate). Proteins of another system, tripartite ATP-
independent transporters (TRAP), which uses an electrochemical gradient to drive nutrient
uptake, are also common. Another system of TonB-dependent transporters (TBDTs) in
the OM is also widespread and particularly important for the transport of iron complexes,
vitamins, and aromatic compounds. The presence of particular transporters can indicate
which nutrients are limiting.

As noted in Chapter 1, apart from phagotrophic protists, microbes can only feed on relatively
small molecules (usually <600 mDa) which are transported into the cell. The origin of these
compounds in the marine environment as a component of dissolved organic matter (DOM)
is discussed in Chapter 8. When cells die or are lysed by viruses, or when fecal pellets from
zooplankton are released, large amounts of organic material are made available to hetero-
trophic microbes. But most of this will be in form of complex macromolecules like nucleic
acids and cytoplasmic and membrane proteins and lipids, which cannot be assimilated by
bacterial or archaeal cells. Therefore, many produce extracellular enzymes (exoenzymes),
including chitinases, amylases, nucleases, and proteases for degradation of biopolymers into
monomers. Chitinase is especially significant in marine bacteria, since chitin (a polymer
of N-acetyl glucosamine, NAG) is such an abundant compound in the sea, due to its pres-
ence as a major structural component of the exoskeleton of many invertebrates (especially
crustaceans), and the cell walls of fungi. Many marine bacteria possess surface-associated
chitinases and the released NAG is an excellent source of both carbon and nitrogen. As in
all cases in which bacteria degrade macromolecules in their environment, mechanisms must
exist for transporting the monomers produced into the cell. In a natural bacterial assemblage,
degradation of complex macromolecules by one type of microbe makes nutrients available to
many others. Such processes are particularly important in particles of marine snow and may
lead to a plume of DOM as the particle falls through the water column, because colonizing
bacteria produce DOM more quickly than they can use it (Figure 1.11). Bacteria inhabiting
this plume grow at much faster rates than those in bulk seawater.
92 Chapter 3

There are two strategies for the deployment of exoenzymes. Some remain closely associated
SELFISH BACTERIA
i GOBBLE LARGER
OLIGOSACCHARIDES
with the cell membrane, which maximizes the opportunity for the cell to benefit from the
breakdown of the molecule. Others are excreted from the cell and diffuse freely into the sur-
rounding water. Production of extracellular hydrolytic enzymes by a cell requires an “expen-
The notion that ocean bacteria sive” commitment because of the energetic costs of synthesizing and exporting the enzyme
always use exoenzymes to hydro- and of importing the degradation products. The cell-associated and free exoenzymes have
lyze high molecular weight DOM different cost–benefit ratios—the cell-associated enzymes “pay off” at nanomolar concentra-
to molecules <600 Da has been tions of substrate, while the free enzymes are only profitable at micromolar concentrations.
challenged in a discovery by Greta However, bacteria producing free exoenzymes in dense populations can benefit from the
Reintjes and colleagues from MPI
enzymes released by neighboring cells. In some species, exoenzyme production is coordi-
Marine Microbiology, Bremen.
Reintjes et al. (2017) incubated
nated via quorum sensing, discussed below.
natural ocean microbial communi-
ties with the fluorescently-labeled Acquisition of iron is a major challenge
polysaccharides (FL-PS) laminarin
and xylan (produced by algae for marine microbes
and present in large amounts in Iron is an essential constituent of cytochromes and iron-sulfur proteins, which have crucial
seawater) and chondroitin sulfate
roles in electron transport processes, especially photosynthesis and nitrogen fixation. Although
(present in animal cartilage and
known to be degraded by marine
iron is one of the most common elements in the Earth’s crust, the amount of free iron in the
bacteria). Using fluorescence open ocean can be virtually undetectable (picograms per liter) without ultraclean assay tech-
microscopy, she observed that 26% niques. In oxygenated seawater, almost all the iron occurs in the ferric Fe(III) state bound
of cells took up FL-PS. To follow this tightly to organic ligands, so it is a major challenge for marine microbes to obtain enough iron
up, Reintjes et al. combined FL-PS for growth. Many bacteria, both autotrophs and heterotrophs, secrete chelating agents known
staining with single-cell identifica- as siderophores that bind to iron, preventing its oxidation and allowing active transport into the
tion by FISH and super-resolution cell via specific receptors on the cell surface. In culture, these compounds are produced under
structured illumination microscopy. conditions of iron restriction and some structures have been characterized. Over 500 chemi-
FL-PS molecules were seen to cal structures have been described; these include phenolics and derivatives of amino acids or
bind rapidly to cells of the phyla
hydroxamic acid. The first siderophores to be characterized from oceanic bacteria are structur-
Bacteroidetes and Planctomycetes
and the gammaproteobacterial
ally quite different from those previously described and behave as self-assembling amphiphilic
genus Catenovulum. FL-PS appear molecules. Aquachelins from Halomonas aquamarina and marinobactins from Marinobacter
to be taken into the periplasm by a both have a water-insoluble fatty-acid region and a water-soluble region of amino acids. In
TonB-dependent OM transporter, laboratory studies in the absence of iron, these compounds form clusters of molecules attached
where they are degraded into together by their fatty-acid tails. Upon binding to Fe(III), the micelles come together to form
mono-, di-, and trisaccharides for vesicles. It is not yet known whether this phenomenon of aggregation occurs in the natural envi-
transport into the cytoplasm. This ronment, but it might be important for bacteria associated with particles containing local high
“selfish” uptake prevents the poly- concentrations of organic matter. The mechanism by which the cell takes up these siderophores
saccharide breakdown products is also unknown. As discussed in Chapter 9, iron cycling has a major role in the primary pro-
from being used by other bacteria.
ductivity of oceans and artificial iron fertilization of the oceans has been considered a potential
method to mitigate the effects of elevated levels of atmospheric CO2.

Iron deprivation is also a problem for pathogens of vertebrate animals, which produce iron-
binding proteins as a defense mechanism. For example, Vibrio vulnificus and V. anguilla-
rum produce siderophores, known as vulnibactin and anguibactin respectively, and these are
important factors in virulence (Figure 11.13).

Siderophores are increasingly recognized as having additional functions besides iron trans-
port, including the ability to transport a variety of other metals, boron, to act as signaling
molecules and antibiotics, and to regulate oxidative stress. Some also sequester toxic heavy
metals to prevent uptake by cells.

Marine bacterioplankton use two trophic strategies

Marine planktonic bacteria show two patterns of growth and multiplication, depending on
their acquisition of organic matter. These lifestyle capabilities can be readily assessed from
metagenome sequences, by identifying clusters of gene groups that characterize each type.

Some bacteria, known as copiotrophs, can grow rapidly when exposed to high substrate levels.
This trophic type is associated with surfaces and particles or nutrient “hotspots” surround-
ing phytoplankton or marine snow particles (Figure 1.11). They show a “feast or famine”
lifestyle, in which rapid growth rates and increased cell size can be induced in response to
 Metabolic Diversity and Ecophysiology 93

increased levels of nutrients. They often produce free exoenzymes and highly diverse nutrient
WHY IS IRON SO
transporters and have relatively large cells (>1 µm3) that often show motility and chemotaxis
in response to nutrient gradients (Box 3.2). Estimating the growth rate of bacteria in natural ? SCARCE, YET SO
IMPORTANT?
seawater is difficult because of the complicating effects of viral lysis and protistan grazing,
but there is no doubt that growth rates are nothing like the rapid doubling times observed Most parts of the open ocean
with those bacteria that can be easily grown in laboratory culture. Perhaps, in nature, copio- contain very little iron to support
trophs have doubling times of a few hours under optimal conditions. Conversely, “famine” marine microbial life, yet it is an
results in reduced synthesis of macromolecules, size reduction during cell division, and other essential element required for
adaptations to starvation, discussed below. many key biological processes,
including photosynthesis and
nitrogen fixation. Why does life
In contrast, most free-living planktonic bacteria can be described as oligotrophs, with less
depend on such a scarce nutrient?
adaptability and regulatory flexibility, a smaller number of specific transporters and less reli- The iron-containing proteins that
ance on exoenzymes for nutrient acquisition. Their small size—often associated with stream- are critical to these reactions are
lined genomes—and slow growth rates (doubling once per day or longer) are an adaptation to highly conserved across diverse
the chronic low levels of nutrients. phylogenetic lineages, suggesting
that they evolved very early in the
history of Earth. The early evolu-
Growth rate and turnover of organic material tion of life occurred under anoxic
conditions, but the development of
depend on nutrient concentrations oxygenic photosynthesis about 2.4
Bacterial growth efficiency (BGE) can be defined as the ratio of biomass or ATP produced to BYA resulted in a sudden transition
the amount of substrate utilized. The yield is always lower than predicted by theoretical calcu- of the atmosphere and oceans to
lations based on the energetics of biochemical pathways because organisms require a certain an oxidizing state. Iron precipi-
amount of nutrient for “housekeeping” functions such as transport, maintenance of internal tated in large quantities and the
amount of soluble iron in seawater
pH and a proton gradient across the membrane, DNA repair, and motility (if present). These
today is lower than the concentra-
functions take priority over biosynthesis for growth or reproduction. At low growth rates due tion needed to sustain microbial
to nutrient limitation, the cell will use proportionately more of the available substrate for main- life. Aquatic microbes, including
tenance energy. Below a certain threshold, cells will divert all the substrate to just “staying the cyanobacterial ancestors that
alive” rather than growing. Uncoupling of catabolic and anabolic processes often occurs at low caused the change in oxygen con-
growth rates when growth is limited by a substrate other than the energy source. Maintaining tent of the atmosphere, faced an
the maximum possible energy state of the cell enables it to resume growth rapidly when con- iron-shortage crisis and this neces-
ditions change. These effects of nutrient levels and BGE are important concepts in assessing sitated the development of efficient
the role of microbes in carbon flux and productivity in marine systems. Because nutrient con- mechanisms for scavenging the
centrations in the sea are generally low, growth rates of oligotrophs are necessarily slow and increasingly low levels in the envi-
ronment. Other transitional metals
a large proportion of energy intake is used for maintenance energy, especially active transport
are occasionally found to replace
and synthesis of extracellular enzymes for acquiring nutrients. Therefore, net BGE will be
iron, but they are equally scarce
low, despite considerable turnover of organic material. This is particularly true when N and P in the oceans, and iron is uniquely
are in limiting concentrations. However, as noted in Chapter 1, there is an increasing recogni- suited to its function.
tion that the ocean environment is very heterogeneous with respect to nutrients and that many
marine microbes are associated with local concentrations of organic matter.

Copiotrophic marine bacteria may

show rapid growth in culture
Although the vast majority of marine bacteria have not yet been cultured, many copiotrophic
bacteria from ocean water, as well as those associated with surfaces and sediments, can be
isolated and grown in the laboratory. During growth and reproduction, a bacterial cell must
synthesize additional cell membranes, cell wall components, hundreds of different RNA mol-
ecules, thousands of proteins, and a complete copy of its genome. The coordination of these
activities is subject to complex regulation of gene expression. With a few exceptions (such as
the budding and stalked bacteria), almost all bacteria reproduce by binary fission into two
equal-sized cells. When growing under optimal or near-optimal conditions, the cell doubles
in mass before division and regulation of DNA replication and cell septation are largely con-
trolled by the rate of increase in cell mass. The growth cycle of culturable heterotrophs in
the laboratory is well known. Upon introduction to the growth medium, unbalanced growth
(without cell division) occurs during a lag phase, during which the cell synthesizes metabo-
lites, enzymes, coenzymes, ribosomes, and transport proteins needed for growth under the
prevailing conditions. Under ideal conditions, bacteria then enter a logarithmic phase and
grow exponentially. Different species vary enormously in the time taken for the population
94 Chapter 3

to double (generation time). Environmental factors such as temperature, pH, and O2 avail-
ability (for aerobes) have a major influence on the rate of growth. Growth rate increases
with temperature. Many marine bacteria isolated from water at 10°C can be grown in the
laboratory with doubling times of about 7–9 h, while those isolated at 25°C typically double
every 0.7–1.5 h. Of course, laboratory culture will select for fast-growing copiotrophic bac-
teria. All organisms have a minimum, optimum, and maximum temperature for growth, and
this often—but not always—reflects their evolution in a particular habitat. Above the mini-
mum temperature, metabolic reactions increase in rate as temperature increases, and so the
doubling time reduces. The maximum temperature is usually only a few degrees above the
optimum because of the thermostability of enzymes and membrane systems. Some marine
bacteria have exceptionally low generation times. For example, Vibrio parahaemolyticus can
divide every 10–12 min at 37°C (although this temperature is not encountered in its normal
estuarine habitat) and the hyperthermophile Pyrococcus furiosus doubles every 37 min at
100°C. At the other extreme, oligotrophic psychrophiles growing just above 0°C may have
generation times of several days or weeks.

Bacteria adapt to starvation by coordinated

changes to cell metabolism
In culture, bacteria continue exponential growth until they enter the stationary phase
because a limiting nutrient is exhausted, toxic by-products of growth accumulate to
become self-limiting, or some density-dependent signal limits growth. Cells entering the
stationary phase do not simply “shut down” but undergo genetically programmed changes
that involve the synthesis of new starvation-specific proteins induced via the RpoS global
regulatory system. When cells of laboratory-cultured copiotrophic marine bacteria are
starved, they become smaller and spherical. In the 1980s, Morita coined the term “ultra-
microbacteria” to describe the very small cells (0.3 μm diameter or less) that develop in
this way. To achieve reduction in cell size, some proteins and ribosomes are degraded
in a controlled fashion, but 30–50 new proteins are induced in response to starvation.
These include proteins that are essential for survival with limited sources of energy and
nutrients as well as proteins that enable cells to acquire scarce substrates at very low
concentrations. Cell surfaces become more hydrophobic, which increases the adhesion of
bacteria to surfaces, and surface association in biofilms appears to be a major adaptation
to nutrient limitation. Changes in the fatty-acid content of membranes promote better
nutrient uptake and provide increased stability, while cell walls become more resistant to
autolysis. Starved cells have a much-reduced ATP content, but maintain a proton-motive
force across the membrane. Starved cells are more resistant to a variety of stress con-
ditions, and mutations in any of the genes in the survival induction pathway may lead
to death in the stationary phase. Most oligotrophic pelagic bacteria are also very small
and have similar morphology to the ultramicrobacteria formed by starvation of cultured
marine isolates in the laboratory. For some time, the prevailing paradigm was that marine
bacteria are “normal” cells that have undergone reduction in size and are in a state of per-
petual near-starvation. This led to the erroneous view that bacterioplankton are virtually
inactive in the natural environment. In fact, we now know that the naturally occurring
small cells found in the sea are more active than larger bacteria on a per volume basis
and a large fraction of supposedly inactive cells can be induced to high metabolic activ-
ity by addition of substrates. However, determination of in situ growth rates of marine
heterotrophic microbes remains one of the most difficult aspects. Recent techniques have
relied on the use of different fluorochromes in flow cytometry (Table 2.1) to determine
cell biomass. Some protocols attempt to distinguish dead and live bacteria by employing
fluorochromes directed against nucleic acids, while others use fluorochromes that detect
respiratory activity. One method combines nucleic acid staining with a permeable dye and
an impermeable dye both attached to gene probes to measure membrane integrity and
DNA content. Cell viability can also be assessed by measuring the uptake of radiolabeled
compounds (such as amino acids) by microautoradiography (p.58) or the use of fluorescent
electron acceptors to monitor the presence of an active electron transport chain. Contrary
to earlier ideas, these methods reveal that planktonic microbes in the open sea are meta-
bolically active and their small size is an evolutionary (genotypic) adaptation to ensure
efficient use of scarce nutrients in their oligotrophic environment.
 Metabolic Diversity and Ecophysiology 95

Some bacteria enter a “viable but

nonculturable” state in the environment
The terms “viable but nonculturable” (VBNC) and “somnicell” (“sleeping” cells) were origi-
nally introduced in the 1980s by Rita Colwell and coworkers, who observed that Vibrio chol-
erae does not die off when introduced into the environment. Since then, the VBNC state has
been studied extensively and shown to occur in a wide range of Gram-negative (and a few
species of Gram-positive) bacteria, including pathogens of humans, fish, and invertebrates.
The phenomenon is thus especially important in studies of the ecology of autochthonous
pathogenic marine bacteria and the survival characteristics of introduced pathogens (e.g.
from wastewater) and indicator organisms. This term has been used by some microbiologists
to explain the discrepancy between the number of bacterial cells that can be visualized by
direct observation (e.g. epifluorescence or immunological detection) and those that can be
enumerated using viable plate counts. The VBNC state could constitute a dangerous reser-
voir of pathogens, which cannot easily be detected, and it is also relevant to evaluation of the
survival of introduced genetically modified organisms into the environment. A variety of fac-
tors, especially nutrient deprivation and changes in pH, temperature, pressure, or salinity can
initiate the cascade of cellular events leading to the VBNC state. The VBNC state remains
a matter of considerable controversy among microbiologists. While many microbiologists
believe these events to be an inducible response to promote survival under adverse condi-
tions, many remain skeptical about the validity of the VBNC state as an adaptive stage. One
of the most controversial aspects of the VBNC state is whether cells can be resuscitated to
a form that grows on agar plates in the laboratory. Many studies have found that addition of
nutrients or temperature changes can cause VBNC bacteria to revert, provided that the cells
have not been in the VBNC state for too long. Others refute these findings and state that the
effect is due to the regrowth of a few remaining culturable cells in the population. An alter-
native explanation of the VBNC state view is that when exponentially growing bacteria face
a sudden insult, such as the deprivation of nutrients, it results in the decoupling of growth
from metabolism. Consequently, cells may suffer a burst of oxidative metabolism, resulting
in lethal free radicals. Bacteria can avoid this if they can induce changes to protect DNA,
proteins, and cell membranes. The shock of sudden transfer of cells into a rich medium when
they are still in the process of adaptation to life in the aquatic environment could result in
rapid death. The inability of cells in the VBNC state to detoxify peroxides and other free
radicals commonly present in culture media or induced within the cells themselves during
exposure to rich media seems to be a key explanation for the phenomenon of non-culturabil-
ity. This may be due to repression of the gene-encoding periplasmic catalase, which breaks
down toxic peroxide.

It is very important to realize the distinction between the VBNC state from the “unculturabil-
ity” of most marine bacteria. As discussed in Box 2.1, these are better described as “not yet
cultured,” since, although many cannot be cultured using conventional methods, considerable
progress is being made using dilution methods and other new approaches. This is completely
different from the VBNC state, which is a transition shown by bacteria that grow readily in
laboratory culture to a starvation or other shock.

Many bacteria use motility to search for

nutrients and optimal conditions
The ability to swim toward high concentrations of favorable nutrients (chemotaxis) is an
important property of many marine bacteria, which swim due to the presence of one or
more flagella. These are long helical filaments attached to the cell by means of a hook-like
structure and basal body embedded in the membrane. The flagellar filament in members of
the Bacteria is about 20 nm in diameter and is made of many subunits of a single protein
(flagellin) synthesized in the cytoplasm and passed up from the basal body through the cen-
tral channel of the filament, to be added at the tip. Much less is known about flagella in the
Archaea; although the basic architecture is similar to that in the Bacteria, the flagellum is
made of several different proteins and is much thinner (about 13 nm in diameter); possibly too
thin for the subunits to pass up the central channel. Genomic analysis shows little homology
between bacterial and archaeal flagellum proteins.
96 Chapter 3

Figure 3.14 Schematic drawing of

the architecture of the basal body
and hook region of bacterial fla-
gella based on electron microscopy
and identification the constituent
proteins. The left side shows the
complex of the Salmonella H+-driven
motor and the right side is that of
the Vibrio Na+-driven motor. The
flagellin subunits are secreted from
the cell and assemble at the hook to
form a helical filament (not shown)
that rotates to propel the cell. IM
inner membrane; OM outer mem-
brane; PG peptidoglycan layer.
Credit: Reprinted from Minamino
and Imada (2015) with permission
from Elsevier.

The basal body of the bacterial flagellum consists of a central rod and a series of rings embed-
ded in the cell wall and membrane, composed of nearly 30 different proteins, as shown in
Figure 3.14. Movement is caused by the rotation of the basal body acting as a tiny rotary
motor, which couples the flow of ions to generation of force. The flagellum is rigid and
behaves like a propeller. For many years, the remarkable mechanics of the motor and the pro-
cess of self-assembly of the fine flagella tubules have attracted the attention of engineers for
possible exploitation in nanotechnology. The mechanism has been extensively studied in thr
human-associated bacteria Escherichia coli and Salmonella. Rotation is caused by a proton-
motive force as H+ ions flow from the outside to the inside of the cell through the Mot protein
subunits surrounding the C and MS rings in the membrane. In motion reminiscent of a tur-
bine, the protons exert electrostatic forces on charges on the C and MS rings as they move
through a channel in the Mot proteins, causing a rotatory movement. About 1000 protons are
needed for each revolution. Genomic analysis shows that the model of flagellar assembly and
function derived from studies of E. coli and Salmonella applies remarkably well to many
bacteria in diverse phyla. Many genes are highly conserved—especially those connected
with the MS ring, C ring, rod, and flagellin export process—although there are small but
important differences between different systems. Most significantly for our discussion here is
the finding that most motile marine bacteria that have been investigated seem able to switch
between the use of either H+ or Na+ ions to create the ion motive force. This is an advantage
in alkaline (~pH 8) seawater. When the flagellar motor of a marine bacterium like Vibrio is
being driven by Na+, it can rotate at up to 1700 hz for short periods, about six times as fast as
the H+-driven motor of Salmonella. In Vibrio spp. the Mot proteins that act as the ion channel
for H+ are replaced by Pom proteins that act as a Na+ channel and additional H and T rings,
as seen in Figure 3.14. Analysis of the polar flagellar motor shows that the Na+-driven motor
has additional components that are responsible for detecting the speed of rotation or the flux
of Na+ ions, discussed in Box 3.2.

The swimming behavior of many marine bacteria is very different to the “classical” system
studied extensively in E. coli and Salmonella, which usually have multiple (4–10) flagella.
When E. coli cells are observed in a neutral environment under the microscope, they move
in an apparently random fashion. Counterclockwise (CCW) rotation of all motors causes
 Metabolic Diversity and Ecophysiology 97

the flagella to form a bundle that propels the bacterium on a relatively straight path (“run”)
at 10–30 µm s-1 before executing a brief “tumble” and swimming off in another direction.
The tumble is caused by a brief change to clockwise (CW) rotation of the flagella, disrupting
the bundle leading to random reorientation of the cell. The presence of attractants or repel-
lents modifies the frequency of tumbling. Cells tumble less frequently and therefore spend
longer swimming up a gradient of increasing concentration of an attractant and show a net
movement toward the source of the substance. The bacterium senses minute changes in con-
centration with time, through a series of chemoreceptors on the cell surface and a chemical
signaling system involving changes in the methylation state of chemotaxis (Che) proteins that
relays information to a tumble generator at the base of the flagellum. This complex process
involves many different proteins encoded by numerous genes. This method of movement is
termed a “biased random walk” and biophysical considerations show that it is an energeti-
cally efficient mechanism for moving in high nutrient concentrations, but it depends on rela-
tively low uniform speed in straight runs, with a random turn angle.

It is estimated that 40–70% of marine bacteria may be motile, although the observation
of motility shows considerable diurnal and seasonal variations and depends strongly on
the water composition, and the smallest cells (<0.4 µm) have the lowest motile fraction.
Obviously, chemotaxis is only relevant for copiotrophic bacteria that have the genomic
potential to respond to localized concentrations of organic carbon, although not all copio-
trophs are motile. Most high-speed bacterial community members use both H+ and Na+ ion
motors simultaneously. Marine bacteria almost always possess a single polar flagellum for
swimming (although cells may become peritrichously flagellated in contact with surfaces,
as discussed below).

Although there are far fewer studies of chemotaxis in marine bacteria, we now know that
they behave quite differently, showing higher chemotactic speed (chemokinesis) and preci-
sion than E. coli. Very high speeds enable small marine bacteria to swim in a relatively
straight line before a 180o reversal of direction or a “flick” mechanism of ~90o (Figure 3.15).
Although energetically more expensive than the random walk, it is necessary if the bacteria
are able to detect and respond to the small point sources of nutrients that occur in the sea due
to turbulence. A few marine bacteria have been observed swimming in short bursts at up to
450 μm s-1—equivalent to several hundred body lengths per second. The significance of these
responses of marine bacteria to gradients in the sea is discussed in Box 3.2.

Flagella also have a mechanosensory function

When Vibrio parahaemolyticus is transferred from suspension to solid surfaces, the bacterial
cells undergo a remarkable change in morphology. They stop dividing normally and elongate
to about 30 μm. At the same time, they start synthesizing large numbers of lateral or perit-
richous flagella as well as the normal polar flagellum (Figure 3.16). The Na+-driven polar
flagellum is effective when the cell is in the free-living, planktonic state, while the H+-driven
lateral flagella enable cells to move better through viscous environments or over surfaces.
This can be observed in the laboratory by the phenomenon of swarming on the surface of
agar plates. In the natural environment, differentiation to the swarmer form enables vibrios
to adopt a sessile lifestyle, which is significant in the formation of biofilms on surfaces or
the colonization of host tissues. This “swim-or-stick” switch is determined by the action of
the polar flagellum acting as a sensor. Increased viscosity as the bacterium nears a surface
slows down flagellar rotation, leading to induction of the lateral phenotype. Experimentally
induced interference of the Na+ ion flux with chemical agents also induces differentiation
to the swarmer state. Genetic analysis of the Na+ motor is leading to an understanding of
this mechanism. However, little is known at present about how the signal from the flagellum
induces changes in the expression of the large number of genes required for altered cell divi-
sion and synthesis of different flagella. It is likely that there is a master regulatory switch. In
vibrios, the polar flagellum is enclosed in a sheath formed from the outer membrane. It is not
known whether the flagellum rotates within the sheath or whether the filament and sheath
rotate as a single unit. The lateral (H+) flagella do not appear to possess a sheath. Motility and
biofilm formation are also critical factors in infection by the cholera pathogen, V. cholerae
(p.334) and in the life cycle of Caulobacter crescentus (p.123).
98 Chapter 3

BOX
BOX 3.23.2 RESEARCH FOCUS
RESEARCH FOCUS

How can microscale bacterial swimming affect global processes?

High-speed swimming—“run and reverse” motility. The idea those produced by a point source such as a lysed algal cell or plumes
that seawater is a non-uniform environment with many small generated by a falling marine snow particle. The bacteria concen-
ephemeral hotspots of nutrients within a dynamic matrix of organic trated in a nutrient “hot spot” within a few seconds and the fastest
polymers was introduced in Chapter 1. Bacteria move over tiny, 20% of the population gained a tenfold higher nutrient concentration
micrometer distances, whereas energy flow into the ocean through than nonmotile cells. In response to a nutrient plume, motile bacteria
convection, wind, and planetary rotation creates turbulence on a gained a fourfold increase in nutrient exposure.
scale many orders of magnitude greater. However, the effects of tur-
bulence translate down to powerful micrometer-scale shear forces. Flagella as a propeller and rudder—“run-reverse-flick” motil-
Therefore, if motility and chemotaxis are to be effective, marine ity. This backwards and forwards movement alone could mean that
bacteria need to swim very fast and show rapid responses to altered the bacterium retraces its steps endlessly, which would not lead to
conditions. Major advances in our understanding of marine bacte- the high-performance chemotaxis observed. This suggested that
rial motility and chemotaxis emerged from the work of Jim Mitchell another aspect to the movement occurs. Xie et al. (2011) tracked the
and colleagues at Flinders University, who used video microscopy trajectories of over 800 individual cells of V. alginolyticus using fast
to record the behavior of natural assemblages of marine bacteria in video imaging and fluorescence microscopy. The strain used has a
enriched seawater samples. They observed tight microswarms of left-handed flagellar filament, meaning that forward direction results
bacteria with cells swimming at a mean speed of 230 µm s-1 and from CCW rotation of the motor, while reversal of the motor pulls the
exhibiting rapid accelerations (Mitchell et al., 1995). Further studies cell backward. They observed that the bacterium alternates between
observed the clustering of bacteria around air bubbles and staying forward and backward runs and reorients by reversing (180° turn) or
in these clusters by an almost instantaneous reversal of direction making a “flick” with an average angle of ~90o to the previous run.
at the end of each run (Mitchell et al., 1996). The authors observed Close inspection of the video images shows that the transition from
different speeds, which they attributed to the different flagellar run to reverse occurs within ~1/30 s, but transitions in the opposite
motors driven by H+ or Na+. By observing the rotation of flagella of direction take longer, ~ 1/10 s. When cells were stained with a fluo-
Vibrio alginolyticus that had stuck to glass slides, Muramoto et al. rescent dye, it could be seen that during the switch from forwards
(1995) had previously shown that the sodium motor rotates rapidly to backward swimming, a small kink forms between the cell body
at a stable speed. Mitchell and Barbara (1999) tested the effects of and the flagellum and this is amplified by the CCW rotation, pushing
compounds that inhibit either H+ or Na+ ion gradients on an enriched the cell body at an angle. After ~0.1 s, the flagellum realigns with
seawater community, as well as isolated cultures of E. coli and the the body axis of the cell and forward swimming resumes in a new
marine bacteria Shewanella putrefaciens and Pseusoalteromonas direction. Xie et al. showed that the frequency of flicking correlates
haloplanktis. Results indicated that marine bacteria use both kinds with swimming speed and flicking diminished when Na+ ions were
of motor simultaneously, with Na+ motors accounting for ~60% of reduced, or the viscosity of the medium was increased. By introduc-
the speed. Other experiments showed that clusters of bacteria occur ing micropipette tips filled with different concentrations of a che-
around particles of organic matter, or even track the release of nutri- moattractant (the amino acid serine) at the edge of the microslide,
ents as motile algal cells move through the water, with up to 12 con- Xie and colleagues demonstrated that V. alginolyticus uses the same
secutive correct turn turns enabling bacteria to remain in the wake three-step motility pattern to efficiently randomize their swimming
of a swimming algal cell (Barbara and Mitchell, 2003a; Barbara and directions, moving towards a point source of nutrient and remaining
Mitchell, 2003b). This run and reverse mechanism was shown to localized near it. The flick allows swift randomization of swimming
be present in 70% of phylogenetically diverse motile marine bac- direction, whereas the backtracking allows the bacterium to explore
teria examined (Johansen et al., 2002). In a generally low-nutrient environments that are structured rather than spatially homogenous.
environment such as the open ocean, it is energetically impossible If necessary, reversal of the motor can quickly bring the cell back to
for bacteria to seek out nutrients by biased random movement. the high concentration of a nutrient. Stocker (2011) considers that the
Therefore, at this stage of the research, marine bacterial chemotaxis hybrid locomotion employing flicks could be commonplace among
was best explained by bacteria relying on turbulent flow bringing marine bacteria, and that earlier descriptions of run and reverse motil-
them within a patch of increased nutrient concentration, where they ity (e.g. in P. haloplanktis) may have overlooked flicks. To investigate
would use the run and reverse movement strategy to maximize the the mechanical origin of the seemingly impossible off-axis flick, Son
opportunity that they remain associated with it. et al. (2013) used ultra-high-speed video microscopy to show that the
flick occurs 10 ms after the cell reverses direction from a backward
Studying motility with microfluidics. Further major advances in run and begins a forward run, revealed by the alignment between the
the study of motility and chemotaxis have been made using micro- direction of swimming and the cell head orientation (Figure 3.15).
fluidic devices and the fusion of observations of microbial behavior The flick takes ~60 ms. During backward swimming, the cell is under
with biophysical principles, pioneered by Roman Stocker and col- tension, because the thrust from the flagellum and the drag on the
leagues (MIT and ETH Zürich). Microfluidics devices enable study head are equal and directed away from each other, whereas it is under
of very small volumes of fluid (µL to pL) in networks of micro chan- compression during forward swimming. This led Son et al. to reason
nels mounted on a microscope slide. Video microscopy coupled that the reason that flicks only occur during forward swimming is
with computational systems allows tracking and analysis of the due to a material instability of the hook region of the flagellum and
movement of individual bacteria in real time (Rusconi et al., 2014). they explain the mechanical basis by which this buckling occurs dur-
Using this approach, Stocker et al. (2008) measured the response ing a flick. They then showed experimentally that the probability of
of Pseudoalteromonas haloplanktis to nutrient patches mimicking flicking during a cycle varies with swimming speed, as predicted.
 Metabolic Diversity and Ecophysiology 99

BOX 3.2 RESEARCH FOCUS

The hook stiffens by twisting during swimming, preventing buckling Vibrio coralliilyticus which is chemotactic towards DMSP produced
until the motor reverses again at the end of a run. Further experi- by coral (see Box 11.1).
ments showed that P. haloplanktis and a mixed seawater community
show the same mechanism of buckling, as does the coral pathogen Marine bacterial chemotaxis is both fast and precise. In follow-up
experiments using microfluidics to track responses of thousands of
individual V. alginolyticus cells to serine gradients, Son et al. (2016)
investigated how swimming speed affects chemotaxis. By varying
the Na+ concentration, they showed that the bacteria display chemo-
kinesis—speeding up as they move up the concentration gradient,
getting closer to the source. Faster cells also show greater chemo-
tactic precision—the strength of accumulation of cells at the peak
of the gradient. In other words, the bacteria reach a nutrient source
more quickly and, once there, remain more closely associated with
the hotspot. Higher speeds alone promote the frequency of flicks
for the reason discussed in the previous section. The chemokinetic
speed enhancement of a population occurs rapidly and is therefore
an advantageous adaptation to exploit transient patches of resources
in the sea.

The ecosystem scale effects of microscale processes. The experi-

mental studies described above show how we have obtained consider-
able understanding of the microscale processes by which patchiness
of nutrients leads to adaptations in bacterial chemotaxis, which
affects nutrient acquisition and utilization. How microbial life in this
“sea of gradients” is translated into a quantitative framework for for-
aging behavior of microbes and its consequences in the ecosystem is
the subject of a major review by Stocker (2012). Adaptations such as
rapid swimming and hybrid chemotactic responses are clearly suc-
cessful adaptations whose benefits outweigh the major energetic costs
of the mechanisms. They enable motile marine bacteria to exploit
ephemeral nutrient patches such as sinking particles or DOM released
from phytoplankton blooms. Mathematical models of the effect of
chemotaxis in “chasing” a plume of marine snow indicate that growth
rates of bacteria swimming into plumes increase 10-fold (Kiørboe
and Jackson, 2001) and these estimates are consistent with microflu-
idics experiments by Stocker et al. (2008). Thus, motile copiotrophs
might channel more DOM than oligotrophs and the remineralization
of organic matter within sinking marine snow particles will stimulate
primary production. Motile bacteria that cluster near phytoplankton
enhance their productivity by feedback of inorganic nutrients, as
well as accelerating the remineralization of phytoplankton DOM.
Figure 3.15 High-speed video microscopy of V. alginolyticus.
Atomic force microscopy reveals that a large proportion of algae or
a: Image sequence captured with high-intensity dark-field
cyanobacteria are intimately connected with heterotrophic bacteria
microscopy at 420 frames s−1, showing the kinematics of the
(Figure 2.3), increasing direct nutrient flux between autotrophs and
cell head and polar flagellum just before and during a flick.
heterotrophs. Stocker (2012) concludes his review by discussing new
b: Schematic of the head orientation at selected times is
tools for exploring microbial behavior at the single-cell level and the
overlaid on the cell trajectory captured at 1,000 frames s−1.
integration of these ideas into ecological frameworks. In an attempt
Cell head positions are shown by circles at 1 ms intervals. The
to provide a unified view of the diverse microbial adaptations and
inset shows the entire trajectory subsampled at 30 frames s−1
their effects on the resource landscape, these ideas have recently been
(open circles). c: Alignment, q (directional cosine) between
developed into a “foraging mandala” (Fernandez et al., 2019). This is
cell head and swimming direction, for the trajectory in b,
a graphical representation that maps microbial behavior onto a space
reveals the elements of the swimming cycle, particularly the
in which a variety of physical and biological factors are distilled into
short forward swimming segment (here, 18 ms long; red)
two fundamental parameters of gathering resources—the time taken
before the 60-ms-long flick (blue). d: Swimming speed during
to find new resources and the yield obtained from one resource patch.
a flick. e: TEM showing the single polar flagellum (mean head
The biophysical processes included in the model encompass autotro-
length 3.2 µm, mean flagellar contour length 4.6 µm. f, g:
phic and heterotrophic bacteria and algae as well as the impacts of
Schematics (not to scale) of the flagellar filament, hook and
virus predation and mortality. Fernandez et al. conclude that their
rotary motor during backward swimming (f), when the hook
model provides “a blueprint for quantitative understanding of micro-
is in tension, and during forward swimming (g), when the
organism–resource interactions at the level of single cells, from
hook is in compression. Reprinted from Son et al. (2013) with
which the vast global flow of carbon and other elements starts.”
permission of Springer Nature.
100 Chapter 3

Figure 3.16 Transmission electron

micrographs of Vibrio parahaemolyti-
cus. (A) Grown in liquid, showing the
swimmer cells with single, sheathed
polar flagellum. (B) Grown on a sur-
face showing the elongated swarmer
cell with numerous lateral flagella.
Bars represent ~1 μm. Image cour-
tesy of Linda McCarter, University of
Iowa; see McCarter (1999).

Microbes also respond to light, magnetic

fields, and other stimuli
Movement toward light (phototaxis) is common in a diverse range of phototrophic bacte-
ria and the genes controlling the response have probably been acquired in diverse phyla
by horizontal gene transfer. Responses often depend on the ability to integrate information
about changes in the intensity of light of different wavelengths and the physiological status
of the cell. Negative phototaxis is also important, because many microbes need to move
away from high-intensity light (especially in the UV range) that causes damage to DNA and
proteins. Depending on the organism, sudden changes in motility in response to light inten-
sity can result from biased run and tumble or run and reverse mechanisms. Scotophobotaxis
can be observed under the microscope when a bacterium moves out of a light source and
reverses flagellar rotation to reenter the light. Positive and negative photokinesis, i.e. greater
swimming speed causing cells to accelerate towards or away from sources of illumination, is
also common in cyanobacteria and purple photosynthetic bacteria. True phototaxis involves
movement towards or away from a light source through direct sensing of the direction of
illumination, rather than just detection of a spatial gradient of intensity. This is widespread
in eukaryotic algae, but less so in bacteria. This mechanism occurs via photosensory pro-
teins containing chromophores that change conformation when they absorb photons. Various
types of photosensory proteins have been described. The archaeon Halobacterium salinarum
possesses rhodopsin molecules, which act as light sensors influencing rotation of the flagella.
Transduction of signals for the control of motility occurs via various mechanisms such as
light-powered proton gradients or electron transport chains; in many cases, these are initi-
ated by two-component signal transduction systems and systems homologous to the bacterial
chemotaxis cascade Che proteins. It is unclear how small unicellular organisms can detect
the direction of light. Pigments that shade part of the cell may be involved, or the cell may act
as a lens that focuses light on the edge of the cell furthest from the light source.

Other tactic movements include those towards or away from high concentrations of oxygen
(aerotaxis) and ionic substances (osmotaxis). Magnetotaxis is a specialized response seen in
certain species of bacteria found in marine and freshwater muds which contain intracellular
inclusion bodies called magnetosomes (Figure 4.4). These bacteria orient themselves in the
Earth’s magnetic field, although their movements are primarily in response to gradients of
chemicals and oxygen.

Gliding and twitching motility occur on surfaces

Gliding motility does not involve flagella and only occurs when bacteria are in contact with
an aqueous film on a solid. It is especially important in the responses of cyanobacteria and
members of the Cytophaga and Flavobacterium genera to gradients of oxygen and nutrients
in microbial mats. It is likely that there are several different mechanisms employed for gliding
motility and these are not yet fully understood. The most common mechanism used by gliding
cyanobacteria appears to be a kind of jet propulsion through the exclusion of slime via minute
pores in the cell surface. However, in Flavobacterium it is thought that rotatory motors, similar
to flagellar motors, transmit force to special outer membrane proteins, which results in a ratchet-
like movement of the cell envelope, analogous to the movement of a tank on caterpillar tracks.
 Metabolic Diversity and Ecophysiology 101

Many bacteria also show twitching motility, seen as jerky and irregular movements under the
microscope. It occurs via the action of short filaments called type IV pili that occur in tufts
or all over the cell surface. These are made of multiple subunits of pilin protein arranged
helically. The pili bind to the surface and the pilin subunits rearrange, causing the filament
to be under tension. This leads to retraction and pulls the cell forwards in a motion similar to
the use of a grappling hook. Twitching motility is especially important in the formation and
development of biofilms.

Microbes colonize surfaces via formation of biofilms

The importance of surface colonization by the formation of biofilms was introduced in
Chapter 1. There are significant differences in the physiological properties of free plank-
tonic cells and those that are in biofilms or associated with particles. Although biofilm
formation has been recognized for more than a century, it is only in recent years that sig-
nificant advances in the study of biofilm physiology have been achieved, largely because
of the application of confocal microscopy and FISH techniques. Bacteria and diatoms both
initiate biofilm formation as a result of environmental cues, particularly nutrient avail-
ability. Bacteria may undergo considerable developmental changes during transition from
the suspended, planktonic form to the sessile, attached form. The events occurring in
biofilm formation by single species of bacteria have been studied extensively in labora-
tory investigations, and the process can be regarded as a form of cellular development
toward a multicellular lifestyle—some researchers liken it to a tissue. Organic molecules BACTERIA STICK
(largely polysaccharides and proteins) will coat any surface with a conditioning film within
minutes of placing it into seawater. The initial stage in colonization by bacteria usually
i TOGETHER TO WARD
OFF PREDATORS
involves motility toward the surface, and changes in microviscosity as the cell approaches
One of the advantages of the bio-
the surface may cause flagellar rotation and motility to slow, as discussed above. This film mode of life for bacteria seems
causes a transient, reversible adsorption to the surface, mediated by electrostatic attrac- to be protection against predation
tion and van der Waals forces. Genetic studies involving the creation of biofilm-deficient (grazing) by phagotrophic protists.
mutants have shown the importance of motility due to peritrichous flagella, gliding, and Matz et al. (2008) compared the
type IV pili in the movement of bacteria across the surface in order to contact other cells grazing efficiency of two flagel-
to form microcolonies. late protists common in coastal
waters—the surface-feeding
During the process of biofilm formation, expression of specific genes is induced—this Rhynchomonas nasuta and the sus-
depends very much on the nature of the substrate. In experimental studies, bacteria will pension feeding Cafeteria roenber-
gensis—against a large number of
quickly colonize inert surfaces such as pieces of plastic, glass, or stainless steel immersed
bacteria isolated from the surface
in seawater, but the genes expressed are quite different from those expressed when they
of a seaweed. C. roenbergensis
colonize organic substrates such as chitin. Some marine heterotrophic bacteria express chitin increased in density in response to
binding and chitinase genes selectively when they encounter chitin, so that they can begin to planktonic cells of the bacteria and
utilize it as a nutrient source. This is especially important in the colonization of the exoskel- predated them with high effi-
etons of crustaceans, as discussed in Chapters 11 and 12. The most important developmental ciency. By contrast, when grown as
change following attachment is the expression of copious quantities of exopolymeric sub- biofilms, most of the bacteria were
stances (EPS). These provide a strong and sticky framework that cements the cells together. resistant or toxic to the surface-
The chemical and physical properties of EPS are very variable, and the nature and amount feeding flagellate R. nasuta. One
produced depends on the bacterial species, concentration of specific substrates, and envi- of the most effective anti-grazing
ronmental conditions. The majority are polyanionic because of the presence of acids (e.g. compounds produced by biofilm
cells was identified as the alkaloid
d-glucuronic, d-galacturonic, or d-mannuronic acids), ketal-linked pyruvate, or the presence
violacein, which inhibits protozoan
of phosphate or sulfate residues. The EPS often form a complex network of long, interlinked feeding at nanomolar concentra-
strands surrounding the cell (glycocalyx). Both rigid and flexible properties can be conferred tions by inducing a conserved
on the biofilm, depending on the secondary structure of the EPS, and interactions with other eukaryotic cell death program. The
molecules such as proteins and lipids may produce a gel-like structure. The mature biofilm authors concluded that biofilm-
often has a complex architecture composed of pillars and channels. Stalked bacteria and specific resistance against preda-
diatoms may increase the length of their stalk when growing in dense biofilms. Once estab- tion contributes to the successful
lished, the dense packing of bacteria and the diversion of metabolism to the production of persistence of biofilm bacteria in
EPS may mean that the cells are metabolically active but divide at very slow rates. various environments. Unraveling
the interactions between bacterial
biofilms and eukaryotic cells may
In nature, biofilms are rarely composed of single species. Bacteria, algae, flagellates, ciliates,
reveal important clues to the evolu-
other protists, and viruses will all interact in the mature biofilm, and there is growing evi-
tion of compounds that affect cell
dence that gene transfer between different microbial species is greatly enhanced within bio- function.
films. This is of great significance in the evolution of organisms with altered characteristics.
102 Chapter 3

Pili are important for bacterial attachment

to surfaces and genetic exchange
Pili (also known as fimbriae) are fine hair-like protein filaments on the surface of many
bacteria. Pili are typically about 3–5 nm in diameter and 1 μm long. They are composed of a
single protein, although a cell may have different types and the amino acid sequence of each
type can show significant strain-to-strain variation. The most common pili are those involved
in the attachment of bacteria to surfaces (some microbiologists restrict the use of the term
fimbriae to these adhesive structures). Their function has been particularly well studied in
the interaction of pathogenic bacteria with mucous membranes of animals. There is often a
specific receptor recognition site on the pili, which can explain why certain strains of bacteria
attach specifically to particular hosts or tissues. Host immune responses to bacterial attach-
ment are often important in resisting infection. Adhesive pili are often encoded by phages
or plasmids and may be exchanged by transduction or conjugation, respectively. The human
pathogen V. cholerae provides a good example: here, pili play a crucial role in infection, and
the acquisition and expression of genes are important factors in explaining the transition
of this bacterium from its estuarine or marine habitat to the human gut. Other examples
discussed later in the book include the attachment of Roseobacter to invertebrate larvae and
the attachment of swimming cells of Caulobacter to surfaces, where pili develop into a form
of holdfast. More generally, pili play a key role in the attachment of bacteria to inanimate
surfaces as the first stage in biofilm formation. As well as the more common adhesive pili,
some cells possess the type IV pili responsible for a twitching motility observed on surfaces,
discussed above.

Sex pili are quite different structures, often several micrometers long, formed only by donor
?
WHAT’S IN A NAME?
QUORUM SENSING (“male”) bacteria involved in the process of conjugation. Sex pili and the machinery needed
for replication and transfer of DNA from donor to recipient cells are encoded by conjugative
The definition of quorum is the
minimum number of members of
plasmids. Sex pili are important in the transfer of genes, such as those encoding antibiotic
a decision-making body (such as a resistance or degradative ability, between cells of bacteria in the marine environment. In
committee or council) required to addition, sex pili are the receptors for certain phages, and an application of this is the use of
conduct the business of that group F+ RNA phages as an indicator of fecal pollution in seawater.
and make decision. Thus, the use
of the term quorum sensing (QS)
to describe density-dependent Antagonistic interactions between microbes
gene regulation implies a coopera- occur on particles or surfaces
tive behavior, which is challenged
by ecologists based on evolution- As noted above, the activity of extracellular enzymes leads to the dissolution of organic
ary theory. Redfield (2002) com- material from marine snow particles as they sink through the water column, and much of
mented that “The appeal of the this DOM becomes available to other members of the plankton. Extracellular enzyme activ-
idea that bacteria act cooperatively ity is likely to provide little “return” for free-living bacteria at low density. Composition of
has caused the postulated benefits the bacterial community inhabiting the particle and the plume of DOM that follows it could
of quorum sensing to be accepted therefore have significant effects on the release and subsequent utilization of nutrients. We
uncritically as the explanation for
know that there are extensive antagonistic interactions between microbes in dense communi-
the role of autoinducers in gene
regulation.” Alternative terms
ties inhabiting the soil or the gut of animals, owing to the production of antibiotics that inhibit
have been proposed to avoid this growth of organisms unrelated to the producer strain, but there have been few such investiga-
controversy: these include “posi- tions of antagonistic interactions between marine bacteria. However, studies indicate that a
tional” (Alberghini et al., 2009), large proportion of marine-particle-associated bacterial isolates do possess antibiotic activity
“diffusion” (Redfield, 2002), and against other pelagic bacteria. Such antibiotic interactions are also likely to be widespread in
“efficiency” (Hense et al., 2007) sediments, microbial mats, and biofilms on the surface of algae, plants, and animals
sensing. The term “autoinduction”
(used in the original description of
LuxI/LuxR regulation) is the most Quorum sensing is an intercellular communication
neutral because it does not imply
a general ecological function.
system for regulation of gene expression
However, Platt and Fuqua (2010) Quorum sensing (QS) is a mechanism of intercellular communication through the controlled
argue that these terms gener- production, release, and detection of threshold concentrations of signal molecules. QS is widely
ate a “semantic quagmire” and used by a variety of bacteria to coordinate communal behavior. Usually, this involves the coor-
that the ecological context of QS
dinated expression of certain genes in response to population density. Study of this phenom-
regulation is complex and affected
enon is leading to new insights into the physiology and ecology of attached marine microbial
by multiple aspects of natural
environments. communities. The signal molecules used in QS are chemically diverse, but the commonest and
best studied are N-acyl homoserine lactones (AHLs), which are widespread in diverse bacteria.
 Metabolic Diversity and Ecophysiology 103

AHLs have different chain lengths, and presence or absence of oxo- and hydroxyl groups. In a
mixed bacterial population, there may be many different structures present.

QS was first studied in detail in the bacterium Vibrio fischeri (since renamed Aliivibrio fisch-
eri), in which it is used to control bioluminescence. When grown in laboratory broth cultures,
these bacteria emit no light until the bacteria enter the late logarithmic or stationary phase
and the population reaches a certain critical density (typically about 107 cells mL-1). This is
because the bacteria synthesize a freely diffusible autoinducer molecule and release it into
the medium. Low-density cultures can be induced to show bioluminescence by the addition
of supernatants from high-density cultures. The autoinducer is an AHL synthesized from
S-adenosylmethionine by the protein LuxI. As illustrated in Figure 3.17, AHL is produced
in the cytoplasm and passively diffuses through the bacterial membrane. Accumulated AHL
diffuses back into the cell, binding to the protein LuxR, a polypeptide of about 250 amino
acids comprising two domains; the N-terminal domain binds the AHL and the C-terminal
domain binds to a palindromic sequence (lux box) upstream of the lux operon promoter.
Thus, when a certain threshold concentration is reached, the bioluminescence genes encod-
ing the various proteins required for the bioluminescence system are expressed by activation
of transcription of the operon (Figure 3.17C).

This AHL-mediated process QS known as the LuxI/LuxR system is now known to be very
widespread and used by many Gram-negative bacteria to regulate the activity of many genes,
including colonization and virulence factors. This description of the LuxI/LuxR QS mecha-
nism is somewhat simplified, as bacteria differ considerably in the details of the regula-
tory circuits and other factors are often involved. The luxI and luxR genes are often located
adjacent to one another on the chromosome. In the A. fischeri bioluminescence system, the Figure 3.17 A. Quorum sens-
transcription of luxI is controlled by the LuxR-AHL complex, which results in autoinduc- ing control of bioluminescence in
Aliivibrio fischeri. The process is con-
tion of further AHL synthesis and amplification of the response. AHL-QS occurs only in
trolled by two regulatory genes luxI
Gram-negative bacteria and homologs of luxI are widespread in proteobacteria. Structural
and luxR in two different operons.
variations in the AHL sidechains, which bind to the cognate LuxR homolog, leads to high
The luxI gene encodes the enzyme
specificity of signaling in different bacterial taxa. responsible for synthesis of the AHL
3-oxo-C6-homoserine lactone. A.
Elucidation of the regulation of bioluminescence in another marine vibrio, V. harveyi, has Representation of a cell at low in a
revealed three separate systems. Like A. fischeri, V. harveyi synthesizes and responds to low-density population, in which
the AHL molecule, in this case termed AI-1 synthesized by LuxLM, and a separate boron-­ there is a low constitutive transcrip-
containing furanosyl diester autoinducer, AI-2 synthesized by LuxS. In this case, the tion of luxI and luxR. LuxI catalyzes
synthesis of AHLs (green dots),
which diffuse from the cell. B. As cell
density increases, AHLs produced by
other A. fischeri accumulate in the
area surrounding the cells and diffuse
into the cell at high concentration.
LuxR is a repressor of the lux operon
promoter (−). When LuxR binds the
AHL autoinducer I, it binds strongly
to the lux box upstream of the pro-
moter and transcription of the right
operon is enhanced (+). Production
of AHL increases exponentially, giv-
ing an autocatalytic feedback loop
as well as initiation of biolumines-
cence. Genes luxA and luxB encode
the α and β subunits of the enzyme
luciferase. Genes luxC, luxD, and luxE
encode enzymes for the synthesis of
the aldehyde substrates from fatty
acids. The LuxR-AHL complex also
binds at the luxR promoter, but in
this case, it represses transcription,
resulting in a compensatory negative
feedback. C. Overview of reactions in
bacterial bioluminescence.
104 Chapter 3

autoinducer molecules do not diffuse back into the cell and interact with a regulatory protein,
DO ALGAE AND
? BARNACLES
INTERPRET BACTERIAL
but are recognized by sensor kinase proteins LuxN and LuxQ, which have a histidine kinase
domain and a response regulator domain. The signal is then transduced by a phosphorylation
mechanism to a cascade of other proteins that determines the expression of the structural
QS SIGNALS?
genes for bioluminescence. Highly conserved luxS homologs occur in many Gram-negative
Many mechanisms for enhanc- and Gram-positive bacteria. These AI-2 synthase genes have been found in about half of all
ing and inhibiting QS have sequenced bacterial genomes and their wide distribution suggested that AI-2 might be a uni-
been identified, suggesting that versal signaling molecule, which bacteria use for communication both within and between
manipulation of QS-controlled species. However, this idea has lost favor because many bacteria that produce AI-2 do not
processes is important in many
possess a cognate receptor for it to function in signaling. Also, recent work indicates that in
bacterial-bacterial and bacterial-
many bacteria, LuxS may have a primary role in methyl metabolism rather than QS, which
eukaryotic associations. Karen Tait
and colleagues at Plymouth Marine may explain its wide distribution.
Laboratory found that biofilms
composed of wildtype Vibrio Gram-positive bacteria do not use AHLs but rely on peptide signals via two-component
anguillarum strongly enhanced response regulator proteins resembling those of V. harveyi. In this case, the autoinducers are
settlement of zoospores of the alga usually short peptides and are highly species specific. They are actively exported from the
Enteromorpha, but vanM mutants cells by ATP-binding transporters.
(defective in AHL synthesis) did not
(Tait et al., 2005). Surprisingly, this Following its discovery and detailed investigation in marine vibrios, QS has emerged as
is not due to chemotactic attrac- one of the most important mechanisms of gene regulation in bacteria, and application of
tion up concentration gradients
this knowledge extends across the whole field of microbiology. QS is especially important
of the bacterial AHL, but due to
in surface-associated communities, with many properties such as motility, adhesion, EPS
negative chemokinesis—swim-
ming speed of zoospores decreases synthesis, production of allelopathic chemicals, and other aspects of biofilm formation being
as they near the wildtype biofilm subject to density-dependent gene regulation. Experimental studies of single-species biofilms
(Wheeler et al., 2006). A similar have shown that the production of AHLs is important in determining the three-dimensional
role for AHLs was shown in the structure of mature biofilms, and AHL mutants produce densely packed biofilms that are
attraction of larvae of the barnacle more easily dislodged from the surface. In multispecies marine biofilms bacteria may utilize
Balanus improvisus to biofilms of V. signaling molecules produced by other species or actively inhibit or degrade them by produc-
anguillarum, Aeromonas hydrophila, tion of specific enzymes (a phenomenon known as quorum quenching).
and Sulfitobacter sp. (Tait and
Havenhand, 2013). These studies The critical role of QS in symbiotic or pathogenic interactions with plant and animal hosts
provided the first evidence of such
has been particularly well studied. In the marine context, the regulation of bioluminescence
communication across the bacterial
and eukaryotic domains, but it is
in the symbiosis of A. fischeri with Euprymna scolopes squid (Chapter 10) is an impor-
a cue rather than a signal, because tant example. Besides the Lux/LuxR system. an additional QS system (AinS/AinR) is also
“signal” has a specific meaning in important; this system regulates early colonization of the squid light organ using a different
evolutionary biology—indicating autoinducer. Recent studies indicate that QS may be even more complex, with a high diversity
something produced explicitly of AHLs being recognized. Other examples include the production of disease by Vibrio spp.
to invoke a response from other in humans, fish and corals (Chapters 11 and 12). Bacteria associated with other organisms
organisms to coordinate activities including corals, sponges, and algae—as well as marine snow particles—have been shown to
between the signal producer and produce AHLs, suggesting that QS is important in the production of extracellular enzymes
the responder. and interspecies interactions in these densely populated habitats. The role of QS in bacterial
associations with corals is discussed in Box 11.2.

Despite strong evidence from cultured isolates and gene surveys that many marine bacteria
can produce the components of QS systems, there is limited insight into the role of QS in situ.
Sensitive detection methods for AHLs and other autoinducers have shown them to be present
in marine snow, algae, and a range of invertebrates, where they are likely involved in regula-
tion of synthesis of exoenzymes involved in the cycling of carbon, phosphorus, nitrogen, and
other key nutrients. The ocean bioluminescence due to association of Vibrio bacteria with
algal blooms is a demonstration that QS is active in the ocean on a massive scale (Figure 4.6).
The settlement and metamorphosis of algal spores and invertebrate larvae is strongly depen-
dent on bacterial biofilms and QS may be directly involved in these processes via recognition
of the signaling molecules. Most QS signals break down rapidly due to basic hydrolysis in
seawater. In addition, many bacteria and algae have been shown to interfere with QS by the
production of inhibitors of AHL synthesis or receptor binding and diverse bacteria produce
acylase or lactonase enzymes that degrade AHLs. This “quorum quenching” (QQ) has the
obvious consequence of conferring a competitive advantage to the organisms producing QQ
molecules because of the removal of the signals from the QS producers and the degraded
AHLs also provide valuable nutrition. Thus, in multispecies community, the competitive
interactions are very complex. QQ is a promising method for the control of biofouling.
 Metabolic Diversity and Ecophysiology 105

PHYSICAL EFFECTS ON MICROBIAL WHAT HAPPENS

GROWTH AND SURVIVAL ? WHEN MESOPHILES
GET A COLD SHOCK?
Most marine microbes grow at low temperatures While many mesophilic bacte-
Over 90% of the ocean has a temperature of 5°C or colder. The temperature of seawater in ria can protect themselves from
the deep sea and in polar regions ranges from −1 to 4°C, while internal fluids in sea ice can sudden exposure to low tem-
be as low as −35°C in winter. Except for sea ice, these temperatures are very stable and little peratures by initiating a cold-shock
response—rapid expression of
affected by seasonal changes. The overwhelming majority of marine microorganisms are
a suite of cold-shock proteins
adapted to this cold environment and conditions can only be considered “extreme” from a
(Csp) that mainly function at the
human perspective. Psychrophilic (cold-loving) microbes are defined as those with an opti- post-translational levels as RNA
mum growth temperature of less than 15°C, a maximum growth temperature of 20°C, and a chaperones. These are helper
minimum growth temperature of 0°C or less. In fact, many deep sea and polar bacteria have molecules that protect RNA from
quite a narrow temperature growth range and may lose viability after brief exposure to typi- misfolding, or unwinding and
cal laboratory temperatures. Therefore, special precautions are needed in their collection, degrading damaged RNA, whereas
transport, and culture. Psychrotolerant bacteria are those that can grow at temperatures as cold acclimation mechanisms in
low as 0°C but have optima of 20–35°C; many organisms from shallow seawater or coastal psychrophiles—evolved for living
temperate regions fall into this category. Apart from understanding their considerable impor- where cold temperatures are the
tance in nutrient cycling and ocean processes, there has also been recent interest by astrobi- norm—tend to be involved in the
regulation of transcription and
ologists in psychrophiles because of the planned future exploration and search for possible
translation. Organisms may sense
life on Europa, the ice-covered moon of Jupiter. Biotechnologists also study psychrophiles the temperature downshift due
because of the industrial potential of their fatty acids and extracellular polymer-degrading to changes in the supercoiling of
enzymes such as chitinase, chitobiose, and xylanase (Chapter 14). DNA, which affects the recognition
of the promoter regions of certain
Low temperatures particularly affect the folding capacity of structural proteins, enzymes, and genes and leads to induction of csp
flexibility of membranes, inhibiting catalysis and transport functions. RNA and DNA become genes (Phadtare, 2004).
more stable, inhibiting the processes of replication, translation, and RNA. Psychrophiles have
many adaptations to overcome these effects. Comparison of the genomes of diverse bacte-
rial and archaeal psychrophiles with those of mesophiles confirms the importance of amino
acid composition and the prevalence of certain residues at critical sites for protein function.
Proteins from psychrophiles are more flexible at low temperatures because they have greater
amounts of α-helix and lesser amounts of β-sheet than those from other organisms. Higher
protein flexibility around the active site of enzymes enhances catalysis by reducing the acti-
vation energy needed for substrate binding. Adaptations of some DNA-associated proteins
are critical to relaxation of DNA under cold stress, and genetic differences leading to altered
secondary structure permits replication, transcription, and translation to occur effectively.
Adaptations to ensure efficient active transport across membranes at low temperatures include
the incorporation of large amounts of unsaturated fatty acids into the membrane, which helps
to maintain membrane fluidity. Omega-3-polyunsaturated fatty acids (PUFAs), once thought
to be nonexistent in bacteria, have been found in Antarctic and deep-sea isolates and have con-
siderable biotechnological potential (see Chapter 14). Transcriptomic studies show that cells
adapt to low temperatures by upregulating genes for the synthesis of transport proteins to over-
come the problem of lower diffusion rates across the membrane at low temperatures, as well as
enzymes for the synthesis of structural components of the cytoplasmic and outer membranes.

Microbes growing in hydrothermal systems

are adapted to very high temperatures
Some bacteria and archaea can grow at temperatures above 60°C. Such thermophilic organ-
isms are found in the marine environment in areas of geothermal activity including shal-
low submarine hydrothermal systems, abyssal hot vents (black smokers), and active volcanic
seamounts (Figure 1.13). In deep-sea vent systems, the temperature of seawater can exceed
350°C. As this superheated water mixes with cold seawater, a temperature gradient is estab-
lished and diverse communities of thermophilic organisms with different temperature optima
occur. Those organisms that can grow above temperatures of 80°C are termed hyperthermo-
philes. Table 5.1 shows the temperature growth ranges of representative marine species. Most
hyperthermophiles described to date belong to two major groups of the Archaea (Chapter 5),
many of which have been isolated because of their biotechnological potential. Only a few
major genera of bacteria, including Thermodesulfobacterium, Aquifex, and Thermotoga are
106 Chapter 3

hyperthermophilic. In both domains, hyperthermophiles occupy very deep branches of the

phylogenetic tree, consistent with the idea that life evolved in high-temperature environments.
Hyperthermophiles show a range of physiological types and can be aerobic or anaerobic, che-
moorganotrophic, or chemolithotrophic. Their enzymes and structural proteins are adapted
to show optimum activity and stability at high temperatures. The overall structure of proteins
from hyperthermophiles often shows relatively little difference from that of homologous pro-
teins in mesophiles. However, variation in a small number of amino acids at critical locations
in the protein seems to affect the three-dimensional conformation, permitting greater stabil-
ity and function of the active site of enzymes. Intracellular proteins from hyperthermophiles
also contain a high proportion of hydrophobic regions and disulfide bonds, which improve
thermostability. Enzymes from hyperthermophiles are useful in high-temperature industrial
processes (Chapter 14). Adaptations of the cell membrane also occur to ensure stability and
effective nutrient transport at high temperatures. The membranes of archaea contain ether-
linked isoprene units and hyper-thermophiles usually possess monolayer membranes, which
appear to be more stable at high temperatures.

Microbes that inhabit the deep ocean must

withstand a very high hydrostatic pressure
Pressure increases by one atmosphere (atm = 0.101 megapascal, Mpa) for every 10 m water
depth and over 75% of the ocean’s volume is more than 1000 m deep. As we now know that
bacteria and archaea are distributed in great numbers throughout the water column, as well
as in marine sediments many hundreds of meters deep, growth under conditions of very high
pressure is the normal state for the vast majority of marine microbes. Zobell and Morita pio-
neered the study of deep-sea bacteria in the 1940s and 1950s. Recent advances in sampling
and cultivation methods together with the application of molecular techniques to the study of
diversity and physiology are leading to some significant new insights. Bacteria and archaea
that have been isolated from depths down to ~3000 m are usually found to be piezotolerant
and can grow over a wide range of pressures up to 30–40 mPa when cultured in the laboratory,
although increasing pressure above ~20 mPa results in reduced metabolic activity and growth

i
HIGH PRESSURE rates. In contrast, many organisms isolated from the deep sea grow optimally at pressures
PRESERVED A >30 mPa; these are known as piezophiles. (The alternative terms barotolerant and barophilic
DEEP-SEA SNACK may also be used, although the piezo- prefix is preferred, because it indicates pressure rather
High pressure decreases the than baro- (weight)). Hyperpiezophiles display optimal growth rates at >60 mPa and many of
binding of substrates to enzymes, these are unable to grow at lower pressures and may die if decompressed for more than a short
which explains why the metabo- period. By using special isolation techniques involving collection in pressurized chambers
lism of shallow-water or terrestrial and cultivation in solid silica gel media, an increasing number of species of obligate piezo-
psychrotolerant organisms is philes have been cultured in recent years, including some from the deepest habitats such as
much slower when incubated in the Marianas Trench in the Pacific Ocean (10500 m). Genetic studies of extreme piezophiles
the laboratory under pressure. A indicate that many have a close resemblance to common psychrotolerant types (e.g. the gam-
remarkable demonstration of this
maproteobacterial genera Shewanella, Photobacterium, Colwellia, and Moritella), although
effect occurred in 1968, when
the submersible vessel Alvin sunk
some unique taxa have also been discovered. At abyssal depths (>4000 m), psychrophiles
accidentally to a depth of 1500 appear to be ecologically dominant over bacteria from shallow waters carried there by sink-
m. While Alvin was being lowered ing organic matter. The most abundant sources of piezophiles are nutrient-rich niches such as
into the sea for a dive, the cable decaying animal carcasses or the gut of deep-sea animals. However, oligotrophic piezophiles
snapped. As water poured through adapted to very low-nutrient concentrations also occur in seawater. Some hyperpiezophilic and
the hatch, the crew managed to hyperthermophilic Archaea have been found near hydrothermal vents, including Pyrococcus
escape but left their lunchboxes— yayanossi, which has optimal reproduction rates at 52 mPa and 98oC.
containing meat sandwiches and
apples—on board the vessel. When A range of adaptations seems to be present in deep-sea organisms. It is important to note that
Alvin was recovered nearly a year very low temperatures (apart from near hydrothermal vents) and very low-nutrient conditions
later, these were found in almost
(apart from the localized occurrence of concentrations of organic matter) characterize the
perfect condition—looking fresh
enough to tempt a couple of scien-
deep sea, as well as high pressure. We must therefore consider the adaptations of piezophilic
tists to take a bite—because of the deep-sea microorganisms as a response to the combined effects of these factors. Protection of
high pressure and temperature of enzymes from the effects of pressure seems to be due mainly to changes in the conformation of
~2–4°C. At atmospheric pressure, proteins. Proteins in piezophiles seem less flexible and less subject to compression under pres-
growth of contaminating bacteria sure, owing to a decreased content of the amino acids proline and glycine. Cells grown under
would have spoiled the food within high pressure also contain high levels of osmotically active substances, which are thought to
a few days. protect proteins from hydration effects of high pressure. The most well-studied adaptation
 Metabolic Diversity and Ecophysiology 107

to high pressure and low temperature is a change in membrane composition. Membranes of

piezophiles contain a higher proportion of polyunsaturated fatty acids and have a more tightly
packed distribution of fatty-acyl chains. Pressure also affects DNA secondary structure.

Application of molecular genetics to the study of two piezophilic bacteria, Photobacterium

profundum and Shewanella sp., has revealed some insight into the mechanisms of regula-
tion of the pressure response. When P. profundum is shifted from atmospheric to high pres-
sure, the relative abundance of two outer membrane proteins (OmpH and OmpL) is altered.
These proteins act as porins for the transport of substances across the outer membrane. An
increased production of OmpH probably provides a larger channel, suggesting that the pres-
sure response enables the bacteria to take up scarce nutrients more easily (as would occur in
the deep-sea environment). A pair of cytoplasmic membrane proteins regulate transcription
of the ompL and ompH genes. Interestingly, these have a high-sequence homology to the
genes encoding ToxR and ToxS proteins, which were first discovered in V. cholerae, where
they detect changes in temperature, pH, and salinity during the transition from the aquatic
environment to the host, leading to the expression of virulence factors (Figure 12.1). The
homology indicates some common ancestry of this environmental sensing system that has
evolved to perform different functions according to habitat. Some other genes have also been
shown to be important in the response to pressure in a deep-sea Shewanella sp. and these may
be grouped into pressure-regulated operons. Use of gene probes directed against these genes
could aid in the identification of new species of piezophiles that we cannot currently culture.

Obligate aerobes always require the presence of O2 and use it as the terminal electron accep-
tor in aerobic respiration. Facultative aerobes can carry out anaerobic respiration or fermenta-
tion in the absence of O2 or aerobic respiration in its presence. Even though O2 is not required,
the growth of facultative organisms is better in its presence on account of the greater yield of
ATP from aerobic respiration. Microaerophiles carry out aerobic respiration but require an
O2 level lower than that found in the atmosphere. Obligate anaerobes carry out fermentation
or anaerobic respiration and many are killed by exposure to O2, although some are aerotoler-
ant and survive (but do not grow) in its presence. Examples of all categories occur in marine
bacteria and archaea.

Oxygen can exist in various forms that are highly reactive and toxic to all cells unless they
possess mechanisms to destroy them (reactive oxygen species, ROS). Singlet oxygen (1O2) is
a high-energy state that causes spontaneous oxidation of cellular materials that forms dur-
ing photochemical reactions; phototrophs usually contain carotenoid pigments, which con-
vert singlet oxygen to harmless forms (quenching). For this reason, non-phototrophic marine
organisms exposed to bright light (such as those inhabiting clear surface waters) are also
often pigmented. Various other ROS form during respiration (Figure 3.18). Superoxide (O2−)
and hydroxyl (OH+) radicals are particularly destructive and react rapidly with cellular com-
pounds. The evolution of mechanisms for the removal of ROS was a major step in the transi-
tion of the biosphere from anaerobic to aerobic following the development of oxygen-evolving
photosynthesis. Organisms capable of aerobic growth usually contain the enzymes catalase
(2H2O2 → H2O + O2), superoxide dismutase (2O2− + 2H+ → H2O2 + O2), and peroxidase (H2O2
+ NADH + H+ → 2H2O + NAD+). Superoxide reductase is an enzyme originally found in
Pyrococcus furiosus and thought to be unique to the Archaea, but genome sequence analysis
has shown that it may be widely distributed in obligate anaerobes in place of superoxide dis-
mutase. This enzyme reduces superoxide to H2O2 without the formation of O2 (O2− +2H+ + cyt
cred → H2O2 + cyt cox).

Ultraviolet irradiation has lethal and mutagenic effects

Research into the effects of ultraviolet (UV) radiation on marine microbes is needed because
of growing evidence that UV radiation is increasing at certain locations on Earth, as a result
particularly of ozone depletion in the upper atmosphere and the formation of an ozone “hole”
over Antarctica and the Southern Ocean. Restrictions in the use of chlorofluorocarbons
(CFCs) is leading to partial closure of the hole over Antarctica, but the layer still appears
to be thinning near the equator and middle latitudes. The lethal and mutagenic effects of
UV radiation result from damage to DNA. UV-B causes direct damage to DNA through the
108 Chapter 3

O2 oxygen formation of pyrimidine dimers, while the main effects of UV-A are due to formation of toxic
e– oxygen and hydroxyl radicals. Various mechanisms for the repair of UV-induced damage
exist, including nucleotide excision repair and light-activated enzyme repair (photoreactiva-
O2 superoxide tion). Studies of DNA damage in bacteria in surface waters show that there is a pronounced
effect over the course of the day, with maximal damage evident in the late afternoon and
e– 2H+
repair occurring during the night. We do not yet fully understand the ecological significance
of these processes. Bacteria produce a range of UV-screening products such as mycosporine-
H2O2 hydrogen peroxide
like amino acids and scytonemin, a complex aromatic compound formed in the sheath of
e– H+ some cyanobacteria. Some bacteria isolated from corals in very clear surface waters show
extreme resistance to the effects of UV radiation by enhancing the activity of NAD(P)H
OH* hydroxyl radical quinine oxidoreductase, a powerful antioxidative enzyme. These mechanisms could have
e– H+ significant biotechnological potential in human health, as products for skin-protection treat-
ments and for overcoming the effects of oxidative stress during aging.
H2O water

Bacterial bioluminescence may protect

Figure 3.18 Formation of toxic
intermediates during reduction of bacteria from ROS and UV damage
oxygen. Some members of the gammaproteobacterial order Vibrionales exhibit the distinctive fea-
ture of bioluminescence. Bacterial bioluminescence occurs due to the action of the enzyme
luciferase, which is a mixed-function oxidase that simultaneously catalyzes the oxidation of
reduced flavin mononucleotide (FMNH2) and a long-chain aliphatic aldehyde (RCHO) such
as tetradecanal (Figure 3.17C). Blue-green light with a wavelength of about 490 nm is emit-
ted because of the generation of an intermediate molecule in an electronically excited state.
Luciferases from all bioluminescent bacteria are dimers of α (~40 kDa) and β (~35 kDa) sub-
units, encoded by the luxA and luxB genes that occur adjacently in the lux operon, along with
other gene-encoding enzymes, leading to the synthesis of the aldehyde substrate via fatty-
acid precursors. The overall process is dependent on ATP and NADPH. Bioluminescence is
regulated by quorum sensing (Figure 3.17A–B).

Bioluminescent bacteria occur as free-living forms in seawater and on organic debris, as com-
mensals in the gut of many marine animals, and as symbionts of the light organs of some squid
and fish. The ecological benefits for animal hosts of harboring bioluminescent symbionts are
obvious (Box 10.3), and selection pressure over many millennia must also have been a major
evolutionary force in the development of such a complex process. However, it can consume
up to 20% of the cell’s energy and benefits for the bacteria are unclear. The discovery of
complex mechanisms of regulation of bioluminescence by density-dependent quorum sensing
adds a further dimension to this enigma. Numerous Vibrio species in coastal water contain
lux genes but are not visibly luminescent in culture. Some experiments suggest that the cell’s
own light emission may function in the repair of DNA at night. UV damage causes formation
of pyrimidine dimers that prevent DNA replication. In blue light, a process called photoreac-
tivation occurs. The enzyme DNA photolyase binds to the dimers and excises them, and other
enzymes then restore the damaged segment of DNA. Induction of the DNA repair process
is initiated by a complex regulatory system called SOS. Mutants of Vibrio harveyi which do
not emit light due to mutations in the luxA or luxB have lower survival when incubated in the
dark after UV-irradiation. Thus, dark repair initiated by bioluminescence might promote sur-
vival of bacteria in surface waters. Other research indicates that bioluminescence may protect
bacteria against the toxic effects of oxygen. As well as bacteria, bioluminescence is present
in some fungi and diverse groups of animals and has multiple evolutionary origins—there is
a very wide diversity of both enzymes and substrates for the bioluminescent reaction mecha-
nism. Apart from light emission, the only common feature is the requirement for oxygen, and
bioluminescent reactions may have evolved primarily as a mechanism for detoxification of the
highly toxic derivatives of molecular oxygen (Figure 3.18), with the initial evolutionary driver
being the nature of the substrates rather than the luciferase enzymes.

Microbes use various mechanisms to

prevent osmotic damage
Several genera of the Archaea are extreme halophiles that grow at very high NaCl con-
centrations (15–35%) found in salterns, brine pockets within sea ice, and submarine brine
pools. Extreme halophilicity is rare in the Bacteria, but Salinibacter ruber is an exception.
 Metabolic Diversity and Ecophysiology 109

To protect themselves from dehydration due to loss of water from the cell to the external
environment, marine microbes must maintain the concentration of intracellular solutes at a
high level. One way of achieving this is by accumulating non-inhibitory substances known
as compatible solutes or osmoprotectants. Usually, these types of sugars, alcohols, or amino
acids are extremely soluble in water. For example, many Gram-negative bacteria synthesize
compounds such as glycine-betaine, ectoine, glutamate, or a-glucosylglycerol. These sub-
stances are released when cells lyse, and some bacteria accumulate glycine-betaine from the
environment rather than synthesizing it themselves. Most Gram-positive bacteria accumulate
the amino acid proline as an osmoprotectant. In algae, DMSP is the main osmoprotectant and
its production and release has major impacts on the ocean sulfur cycle (Chapter 9).

The extremely halophilic archaea use a different method to prevent water loss. They have
an active mechanism for pumping K+ ions into the cell until the internal concentration
balances the high concentration of Na+ outside. In some species, a large proportion of the
proton-motive force for the ion pump derives from light-mediated generation of ATP via
the pigment-containing protein bacteriorhodopsin. Extreme halophiles also have other
adaptations for growth at high NaCl concentrations. Their enzymes and structural pro-
teins have a high proportion of acidic amino acids, which protects the conformation from
disruption by high salt concentrations. Internal cellular components, such as the ribo-
somes and DNA-replication enzymes, require high K+ concentrations for their integrity
and activity.

Conclusions
This chapter has explored the very wide range of metabolic activities in marine microbes,
with a particular focus on bacterial and archaeal processes. We have seen a large diversity of
mechanisms by which they obtain energy from light or chemical sources and various ways
in which they use this to fuel their biosynthetic processes. Some types of metabolism, with
major significance in global ecology, have been discovered only in the last few years and
there are undoubtedly many more surprises to be revealed. Investigation of phenomena such
as antagonism and intercellular communication is revealing information about the ways in
which different microbial species interact with their environment and the interdependence
of their metabolic activities. We are beginning to understand how the microscale effects of
microbial processes can affect global processes in the oceans. In subsequent chapters, numer-
ous examples of the activities of such communities in marine habitats will become apparent.

Giovannoni, S.J., Bibbs, L., Cho, J.C. et al. (2005) Proteorhodopsin in the
References and further reading ubiquitous marine bacterium SAR11. Nature 438: 82–85.
Phototrophy Gómez-Consarnau, L., Akram, N., Lindell, K., et al. (2010) Proteorhodopsin
Béjà, O., Aravind, L., Koonin, E. V. et al. (2000) Bacterial rhodopsin: evi- phototrophy promotes survival of marine bacteria during starvation. PLoS
dence for a new type of phototrophy in the sea. Science 289: 1902–1906. Biol. 8: e1000358.

Beja, O., Spudich, E.N., Spudich, J.L. et al. (2001) Proteorhodopsin pho- Gomez-Consarnau, L., Gonzalez, J.M., Coll-Llado, M. et al. (2007) Light
totrophy in the ocean. Nature 411: 786–789. stimulates growth of proteorhodopsin-containing marine Flavobacteria.
Nature 445: 210–213.
Broda, E. (1977) Two kinds of lithotrophs missing in nature. Z. Allg.
Mikrobiol. 17: 491–493. Graham, E.D., Heidelberg, J.F., and Tully, B.J. (2018) Potential for pri-
mary productivity in a globally-distributed bacterial phototroph. ISME J.
Bryant, D.A., and Frigaard, N.U. (2006) Prokaryotic photosynthesis and 12: 1861–1866.
phototrophy illuminated. Trends Microbiol. 14: 488–496.
Inoue, K., Kato, Y., and Kandori, H. (2015) Light-driven ion-translocating
Dubinsky, V., Haber, M., Burgsdorf, I. et al. (2017) Metagenomic analysis rhodopsins in marine bacteria. Trends Microbiol. 23: 91–98.
reveals unusually high incidence of proteorhodopsin genes in the ultraoli-
gotrophic Eastern Mediterranean Sea. Environ. Microbiol. 19: 1077–1090. Kumagai, Y., Yoshizawa, S., Nakajima, Y. et al. (2018) Solar-panel and
parasol strategies shape the proteorhodopsin distribution pattern in marine
Ernst, O.P., Lodowski, D.T., Elstner, M. et al. (2014) Microbial and animal Flavobacteriia. ISME J. 12: 1329–1343.
rhodopsins: structures, functions, and molecular mechanisms. Chem. Rev.
114: 126–163. McCarren, J., and DeLong, E.F. (2007) Proteorhodopsin photosys-
tem gene clusters exhibit co-evolutionary trends and shared ances-
Finkel, O.M., Béjà, O., and Belkin, S. (2013) Global abundance of micro- try among diverse marine microbial phyla. Environ. Microbiol. 9:
bial rhodopsins. ISME J. 7: 448–51. 846–858.
Frigaard, N.U., Martinez, A., Mincer, T.J. et al. (2006) Proteorhodopsin Moran, M.A., and Miller, W.L. (2007) Resourceful heterotrophs make the
lateral gene transfer between marine planktonic Bacteria and Archaea. most of light in the coastal ocean. Nat. Rev. Microbiol. 5: 792–800.
Nature 439: 847–850.
110 Chapter 3

Pinhassi, J., DeLong, E.F., Béjà, O. et al. (2016) Marine bacterial and Born, D.A., Ulrich, E.C., Ju, K.-S. et al. (2017) Structural basis for meth-
archaeal ion-pumping rhodopsins: genetic diversity, physiology, and ecol- ylphosphonate biosynthesis. Science 358: 1336–1339.
ogy. Microbiol. Mol. Biol. Rev. 80: 929–954. Carini, P., White, A.E., Campbell, E.O. et al. (2014) Methane production
Sharma, A.K., Spudich, J.L., and Doolittle, W.F. (2006) Microbial rho- by phosphate-starved SAR11 chemoheterotrophic marine bacteria. Nat.
dopsins: functional versatility and genetic mobility. Trends Microbiol. 14: Commun. 5: 4346.
463–469. Chistoserdova, L., and Kalyuzhnaya, M.G. (2018) Current trends in meth-
Sieradzki, E.T., Fuhrman, J.A., Rivero-Calle, S. et al. (2018) ylotrophy. Trends Microbiol. 26: 703–714.
Proteorhodopsins dominate the expression of phototrophic mechanisms in Cui, M., Ma, A., Qi, H. et al. (2015) Anaerobic oxidation of methane: an
seasonal and dynamic marine picoplankton communities. PeerJ 6: e5798. “active” microbial process. Microbiol. Open 4: 1–11.
Steindler, L., Schwalbach, M.S., Smith, D.P. et al. (2011) Energy starved DeLong, E.F. (2000) Resolving a methane mystery. Nature 407: 577–579.
Candidatus Pelagibacter ubique substitutes light-mediated ATP produc-
tion for endogenous carbon respiration. PLoS One 6: e19725. Karl, D.M., Beversdorf, L., Björkman, K.M. et al. (2008) Aerobic produc-
tion of methane in the sea. Nat. Geosci. 1: 473–478.
Wang, Z., O’Shaughnessy, T.J., Soto, C.M. et al. (2012) Function and regu-
lation of Vibrio campbellii proteorhodopsin: acquired phototrophy in a Lever, M.A. (2016) A new era of methanogenesis research. Trends
classical organoheterotroph. PLoS One 7: e38749. Microbiol. 24: 84–86.
Liu, Y., Beer, L.L., and Whitman, W.B. (2012a) Methanogens: a window
Nitrogen metabolism into ancient sulfur metabolism. Trends Microbiol. 20: 251–258.
Casciotti, K.L., and Buchwald, C. (2012) Insights on the marine microbial Repeta, D.J., Ferrón, S., Sosa, O.A. et al. (2016) Marine methane para-
nitrogen cycle from isotopic approaches to nitrification. Front. Microbiol. dox explained by bacterial degradation of dissolved organic matter. Nat.
3: 00356 Geosci. 9: 884–887.
Dekas, A.E., Fike, D.A., Chadwick, G.L. et al. (2018) Widespread nitrogen Sun, J., Steindler, L., Thrash, J.C. et al. (2011) One carbon metabolism in
fixation in sediments from diverse deep‐sea sites of elevated carbon load- SAR11 pelagic marine bacteria. PLoS One 6: e23973.
ing. Environ. Microbiol. 20: 4281–4296. Valentine, D.L., and Reeburgh, W.S. (2000) New perspectives on anaero-
Dekas, A.E., Poretsky, R.S., and Orphan, V.J. (2009) Deep-sea archaea fix bic methane oxidation. Environ. Microbiol. 2: 477–484.
and share nitrogen in methane-consuming microbial consortia. Science Yan, Z., Joshi, P., Gorski, C.A. et al. (2018) A biochemical framework for
326: 422–426. anaerobic oxidation of methane driven by Fe(III)-dependent respiration.
DeLong, E.F. (2002) All in the packaging. Nature 419: 676–677. Nat. Commun. 9: 1642.
Kuenen, J.G. (2008) Anammox bacteria: from discovery to application.
Nat. Rev. Microbiol. 6: 320–326. Sources and acquisition of nutrients
Pereira, A.D., Cabezas, A., Etchebehere, C. et al. (2017) Microbial Adam, N., and Perner, M. (2018) Microbially mediated hydrogen cycling
communities in anammox reactors: a review. Environ. Technol. Rev. 6: in deep-sea hydrothermal vents. Front. Microbiol. 9: 02873.
74–93. Bergauer, K., Fernandez-Guerra, A., Garcia, J.A.L. et al. (2018) Organic mat-
Stahl, D.A., and de la Torre, J.R. (2012) Physiology and diversity of ammo- ter processing by microbial communities throughout the Atlantic water col-
nia-oxidizing Archaea. Annu. Rev. Microbiol. 66: 83–101. umn as revealed by metaproteomics. Proc. Natl. Acad. Sci. 115: E400–E408.
Croft, M.T., Warren, M.J., and Smith, A.G. (2006) Algae need their vita-
Sulfur metabolism mins. Eukaryot. Cell 5: 1175–1183.
Barton, L.L., and Fauque, G.D. (2009) Biochemistry, physiology and bio- Decho, A.W., and Gutierrez, T. (2017) Microbial extracellular polymeric
technology of sulfate-reducing bacteria. Adv. Appl. Microbiol. 68: 41–98. substances (EPSs) in ocean systems. Front. Microbiol. 8: 922.
Liu, Y., Beer, L.L., and Whitman, W.B. (2012b) Sulfur metabolism in Johnstone, T.C., and Nolan, E.M. (2015) Beyond iron: non-classical bio-
archaea reveals novel processes. Environ. Microbiol. 14: 2632–2644. logical functions of bacterial siderophores. Dalt. Trans. 44: 6320–6339.
Lovley, D.R. (2017) Happy together: microbial communities that hook up Kiørboe, T. (2003) Marine snow microbial communities: scaling of abun-
to swap electrons. ISME J. 11: 327–336. dances with aggregate size. Aquat. Microb. Ecol. 33: 67–75.
Muyzer, G., and Stams, A.J.M. (2008) The ecology and biotechnology of Kiørboe, T., and Jackson, G.A. (2001) Marine snow, organic solute plumes,
sulphate-reducing bacteria. Nat. Rev. Microbiol. 6: 441–454. and optimal chemosensory behavior of bacteria. Limnol. Oceanogr. 46:
1309–1318.
Muyzer, G., Yildirim, E., van Dongen, U. et al. (2005) Identification of
“Candidatus Thioturbo danicus,” a microaerophilic bacterium that builds Koch, A.L. (2001) Oligotrophs versus copiotrophs. BioEssays 23: 657–661.
conspicuous veils on sulfidic sediments. Appl. Environ. Microbiol. 71: Lauro, F.M., McDougald, D., Thomas, T. et al. (2009) The genomic
8929–8933. basis of trophic strategy in marine bacteria. Proc. Natl. Acad. Sci. 106:
Rückert, C. (2016) Sulfate reduction in microorganisms—recent advances 15527–15533.
and biotechnological applications. Curr. Opin. Microbiol. 33: 140–146. Lipson, D.A. (2015) The complex relationship between microbial growth
Thar, R., and Kuhil, M. (2002) Conspicuous veils formed by vibrioid bacte- rate and yield and its implications for ecosystem processes. Front.
ria on sulfidic marine sediment. Appl. Environ. Microbiol. 68: 6310–6320. Microbiol. 6: 615.
Wasmund, K., Mußmann, M., and Loy, A. (2017) The life sulfuric: micro- Long, R.A., and Azam, F. (2001) Antagonistic interactions among marine
bial ecology of sulfur cycling in marine sediments. Environ. Microbiol. pelagic bacteria. Appl. Environ. Microbiol. 67: 4975–4983.
Rep. 9: 323–344. Mühlenbruch, M., Grossart, H.-P., Eigemann, F. et al. (2018) Mini-review:
Phytoplankton-derived polysaccharides in the marine environment and their
Methanogenesis and methylotrophy interactions with heterotrophic bacteria. Environ. Microbiol. 20: 2671–2685.
Boetius, A., Ravenschlag, K., Schubert, C.J. et al. (2000) A marine micro- Paerl, R.W., Bouget, F.-Y., Lozano, J.-C. et al. (2017) Use of plankton-
bial consortium apparently mediating anaerobic oxidation of methane. derived vitamin B1 precursors, especially thiazole-related precursor, by
Nature 407: 623–626. key marine picoeukaryotic phytoplankton. ISME J. 11: 753–765.
 Metabolic Diversity and Ecophysiology 111

Reintjes, G., Arnosti, C., Fuchs, B.M. et al. (2017) An alternative polysac- Stocker, R. (2011) Reverse and flick: Hybrid locomotion in bacteria. Proc.
charide uptake mechanism of marine bacteria. ISME J. 11: 1640–1650. Natl. Acad. Sci. USA 108: 2635–2636.
Schönheit, P., Buckel, W., and Martin, W.F. (2016) On the origin of heter- Stocker, R. (2012) Marine microbes see a sea of gradients. Science 338:
otrophy. Trends Microbiol. 24: 12–25. 628–633.
Sowell, S.M., Wilhelm, L.J., Norbeck, A.D. et al. (2009) Transport func- Stocker, R., Seymour, J.R., Samadani, A. et al. (2008) Rapid chemotactic
tions dominate the SAR11 metaproteome at low-nutrient extremes in the response enables marine bacteria to exploit ephemeral microscale nutrient
Sargasso Sea. ISME J. 3: 93–105. patches. Proc. Natl. Acad. Sci. USA 105: 4209–4214.
Stoecker, D.K., Hansen, P.J., Caron, D.A. et al. (2017) Mixotrophy in the Wilde, A., and Mullineaux, C.W. (2017) Light-controlled motility in pro-
marine plankton. Ann. Rev. Mar. Sci. 9: 311–335. karyotes and the problem of directional light perception. FEMS Microbiol.
Suffridge, C.P., Gómez-Consarnau, L., Monteverde, D.R. et al. (2018) B Rev. 41: 900–922.
Vitamins and their congeners as potential drivers of microbial commu- Xie, L., Altindal, T., Chattopadhyay, S. et al. (2011) Bacterial flagellum as
nity composition in an oligotrophic marine ecosystem. J. Geophys. Res. a propeller and as a rudder for efficient chemotaxis. Proc. Natl. Acad. Sci.
Biogeosciences 123: 2890–2907. USA 108: 2246–2251.
Tang, K., Jiao, N., Liu, K. et al. (2012) Distribution and functions of TonB-
dependent transporters in marine bacteria and environments: implications Quorum sensing, communication
for dissolved organic matter utilization. PLoS One 7: e41204. and multicellularity
Traving, S.J., Thygesen, U.H., Riemann, L. et al. (2015) A model of extra- Alberghini, S., Polone, E., Corich, V. et al. (2009) Consequences of rela-
cellular enzymes in free-living microbes: which strategy pays off? Appl. tive cellular positioning on quorum sensing and bacterial cell-to-cell com-
Environ. Microbiol. 81: 7385–7393. munication. FEMS Microbiol. Lett. 292: 149–161.

Motility and taxis Antunes, J., Leão, P., and Vasconcelos, V. (2018) Marine biofilms: diver-
sity of communities and of chemical cues. Environ. Microbiol. Rep. doi:
Barbara, G.M., and Mitchell, J.G. (2003) Bacterial tracking of motile 10.1111/1758-2229.12694.
algae. FEMS Microbiol. Ecol. 44: 79–87.
Dunny, G.M., Brickman, T.J., and Dworkin, M. (2008) Multicellular
Barbara, G.M., and Mitchell, J.G. (2003) Marine bacterial organisation behavior in bacteria: communication, cooperation, competition and cheat-
around point-like sources of amino acids. FEMS Microbiol. Ecol. 43: ing. Bioessays 30: 296–298.
99–109.
Hense, B.A., Kuttler, C., Müller, J. et al. (2007) Does efficiency sensing
Belas, R. (2014) Biofilms, flagella, and mechanosensing of surfaces by unify diffusion and quorum sensing? Nat. Rev. Microbiol. 5: 230–239.
bacteria. Trends Microbiol. 22: 517–527.
Hmelo, L.R. (2017) Quorum sensing in marine microbial environments.
Fernandez, V.I., Yawata, Y., and Stocker, R. (2019) A foraging mandala for Ann. Rev. Mar. Sci. 9: 257–281.
aquatic microorganisms. ISME J. 13: 563–575.
Hmelo, L.R., Mincer, T.J., and Van Mooy, B.A.S. (2011) Possible influ-
Guttenplan, S.B., and Kearns, D.B. (2013) Regulation of flagellar motility ence of bacterial quorum sensing on the hydrolysis of sinking particu-
during biofilm formation. FEMS Microbiol. Rev. 37: 849–871. late organic carbon in marine environments. Environ. Microbiol. Rep. 3:
Johansen, J., Pinhassi, J., Blackburn, N. et al. (2002) Variability in motil- 682–688.
ity characteristics among marine bacteria. Aquat. Microb. Ecol. 28: Krupke, A., Hmelo, L.R., Ossolinski, J.E. et al. (2016) Quorum sensing plays
229–237. a complex role in regulating the enzyme hydrolysis activity of microbes
McCarter, L. (1999) The multiple identities of Vibrio parahaemolyticus. J. associated with sinking particles in the ocean. Front. Mar. Sci. 1: 55.
Mol. Microbiol. Biotechnol. 1: 51–57. Matz, C., Webb, J.S., Schupp, P.J. et al. (2008) Marine biofilm bacteria
Minamino, T., and Imada, K. (2015). The bacterial flagellar motor and its evade eukaryotic predation by targeted chemical defense. PLoS One 3:
structural diversity. TrendsMicrobiol 23: 267–274. e2744.

Mitchell, J.G., and Barbara, G.M. (1999) High speed marine bacteria use Platt, T.G., and Fuqua, C. (2010) What’s in a name? The semantics of quo-
sodium-ion and proton driven motors. Aquat. Microb. Ecol. 18: 227–233. rum sensing. Trends Microbiol. 18: 383–387.

Mitchell, J.G., Pearson, L., and Dillon, S. (1996) Clustering of marine bac- Redfield, R.J. (2002) Is quorum sensing a side effect of diffusion sensing?
teria in seawater enrichments. Appl. Environ. Microbiol. 62: 3716–3721. Trends Microbiol. 10: 365–370.

Mitchell, J.G., Pearson, L., Dillon, S. et al. (1995) Natural assemblages Tait, K., and Havenhand, J. (2013) Investigating a possible role for the bac-
of marine-bacteria exhibiting high-speed motility and large accelerations. terial signal molecules N-acylhomoserine lactones in Balanus improvisus
Appl. Environ. Microbiol. 61: 4436–4440. cyprid settlement. Mol. Ecol. 22: 2588–2602.

Muramoto, K., Kawagishi, I., Kudo, S. et al. (1995) High-speed rotation Tait, K., Joint, I., Daykin, M. et al. (2005) Disruption of quorum sensing in
and speed stability of the sodium-driven flagellar motor in Vibrio algino- seawater abolishes attraction of zoospores of the green alga Ulva to bacte-
lyticus. J. Mol. Biol. 251: 50–58. rial biofilms. Environ. Microbiol. 7: 229–240.

Pallen, M.J., Penn, C.W., and Chaudhuri, R.R. (2005) Bacterial flagellar Ventura, R. (2019) Multicellular individuality: the case of bacteria. Biol.
diversity in the post-genomic era. Trends Microbiol. 13: 143–149. Theory: doi.org/10.1007/s13752-019-00317-7.

Rusconi, R., Garren, M., and Stocker, R. (2014) Microfluidics expanding Wheeler, G.L., Tait, K., Taylor, A. et al. (2006) Acyl-homoserine lactones
the frontiers of microbial ecology. Annu. Rev. Biophys. 43: 65–91. modulate the settlement rate of zoospores of the marine alga Ulva intestina-
lis via a novel chemokinetic mechanism. Plant, Cell Environ. 29: 608–618.
Schuergers, N., Lenn, T., Kampmann, R. et al. (2016) Cyanobacteria use
micro-optics to sense light direction. Elife 5: e12620.
VBNC bacteria
Son, K., Guasto, J.S., and Stocker, R. (2013) Bacteria can exploit a flagellar
buckling instability to change direction. Nat. Phys. 9: 494–498. Bloomfield, S.F., Stewart, G.S.A.B., Dodd, C.E.R. et al. (1998) The viable
but non-culturable phenomenon explained? Microbiology 144: 1–3.
Son, K., Menolascina, F., and Stocker, R. (2016) Speed-dependent che-
motactic precision in marine bacteria. Proc. Natl. Acad. Sci. U. S. A. 113: Oliver, J.D. (2005) The viable but nonculturable state in bacteria. J.
8624–8629. Microbiol. 43: 93–100.
112 Chapter 3

Effects of pressure, temperature, Kunzig, R. (2000) Mapping the Deep: The Extraordinary Story of Ocean
ROS and UV radiation Science. W.W. Norton & Co.

Chénard, C., and Lauro, F.M. eds. (2017) Microbial Ecology of Extreme Martin, A., and McMinn, A. (2018) Sea ice, extremophiles and life on
Environments. Springer International Publishing, Cham. extra-terrestrial ocean worlds. Int. J. Astrobiol. 17: 1–16.

Counts, J.A., Zeldes, B.M., Lee, L.L. et al. (2017) Physiological, metabolic Marx, J.-C., Collins, T., D’Amico, S. et al. (2007) Cold-adapted enzymes
and biotechnological features of extremely thermophilic microorganisms. from marine Antarctic microorganisms. Mar. Biotechnol. 9: 293–304.
Wiley Interdiscip. Rev. Syst. Biol. Med. 9: e1377. Oger, P.M., and Jebbar, M. (2010) The many ways of coping with pressure.
Czyz, A., and Wegrzyn, G. (2001) On the function and evolution of bacte- Res. Microbiol. 161: 799–809.
rial luminescence. In: Bioluminescence and Chemiluminescence (ed. J. F. Phadtare, S. (2004) Recent developments in bacterial cold-shock response.
Case, P. J. Herring, B. H. Robinson, S. D. H. Haddock, L. J. Kricka and Curr. Issues Mol. Biol. 6: 125–136.
P. E. Stanley). World Science Publishing Company, Singapore, pp. 31–34.
Poli, A., Finore, I., Romano, I. et al. (2017) Microbial diversity in extreme
Czyz, A., Wrobel, B., and Wegrzyn, G. (2000) Vibrio harveyi biolumi- marine habitats and their biomolecules. Microorganisms 5: 25.
nescence plays a role in stimulation of DNA repair. Microbiology 146:
Rees, J. F., De Wergifosse, B., Noiset, O. et al. (1998) The origins of
283–288.
marine bioluminescence: turning oxygen defence mechanisms into deep-
Durvasula, R., and Subba Rao, D. V. (2018) Extremophiles: From Biology sea communication tools. J. Exp. Biol. 201: 1211–1221.
to Biotechnology. CRC Press.
Siezen, R.J. (2011) Microbial sunscreens. Microb. Biotechnol. 4: 1–7.
Fang, J., Zhang, L., and Bazylinski, D.A. (2010) Deep-sea piezosphere and
Tribelli, P., López, N., Tribelli, P.M. et al. (2018) Reporting key features in
piezophiles: geomicrobiology and biogeochemistry. Trends Microbiol. 18:
cold-adapted bacteria. Life 8: 8.
413–422.
Vezzi, A., Campanaro, S., D’Angelo, M. et al. (2005) Life at depth:
Haddock, S.d., Moline, M.A., and Case, J.F. (2010) Bioluminescence in the
Photobacterium profundum genome sequence and expression analysis.
sea. Ann. Rev. Mar. Sci. 2: 443–449.
Science 307: 1459–1461.
Harrison, J.P., Gheeraert, N., Tsigelnitskiy, D. et al. (2013) The limits for
Zeng, X., Birrien, J.-L., Fouquet, Y. et al. (2009) Pyrococcus CH1, an obli-
life under multiple extremes. Trends Microbiol. 21: 204–212.
gate piezophilic hyperthermophile: extending the upper pressure-temper-
Huber, H., and Stetter, K.O. (1998) Hyperthermophiles and their possible ature limits for life. ISME J. 3: 873–876.
potential in biotechnology. J. Biotechnol. 64: 39–52.
Zhang, Y., Li, X., Bartlett, D.H., and Xiao, X. (2015) Current develop-
Kashefi, K., and Lovley, D.R. (2003) Extending the upper temperature ments in marine microbiology: high-pressure biotechnology and the
limit for life. Science 301: 934. genetic engineering of piezophiles. Curr. Opin. Biotechnol. 33: 157–164.
Chapter 4
Diversity of Marine
Bacteria

Although many organisms in the domain Bacteria can be grown and studied in the labora-
tory, the overwhelming majority are known only by genetic information obtained by direct
analysis of DNA in environmental samples. Thus, there is a large discrepancy in the extent
of our knowledge of the properties of the various members of the Bacteria. For those that
cannot yet be cultured, we can infer many of their likely properties by considering their rela-
tionship to well-studied reference species, their habitats, and geochemical evidence relevant
to their activities. Major advances in sequencing technology and bioinformatics mean that we
can now predict the metabolic pathways and likely biogeochemical role through analysis of
genomes reconstructed from environmental sequences. The first section of this chapter con-
tains a synopsis of diversity and discusses the rapidly changing picture of the major phyloge-
netic groups of the Bacteria. This is followed by discussion of the properties of representative
examples of major bacterial taxa, chosen with reference to activities of particular marine
ecological or applied importance.

Key Concepts
• Phylogenetic and genomic methods have led to major changes in the systematics of the
Bacteria and understanding of their diversity.
• Bacteria that are phylogenetically closely related can differ greatly in their metabolic
and physiological characteristics.
• A relatively small number of major bacterial clades dominate most marine habitats, but
extensive microdiversity enables colonization of specific niches.
• The problems of defining bacterial species are being resolved by application of the core
genome and pangenome concept.
• There are probably millions of marine bacterial species, with many of them constitut-
ing the rare biosphere.
• Genomic analysis of cultured bacteria and metagenomic analysis of environmental
samples leads to new insights into bacterial diversity, ecology, and biogeochemical
roles.
114 Chapter 4

OVERVIEW OF BACTERIAL DIVERSITY

Understanding of diversity has been revolutionized
by phylogenetic and genomic techniques
Our understanding of the diversity of bacteria has undergone gradual evolution over the
past 150+ years, since the first attempts to classify them by scientists such as Haeckel and
Cohn. As classification schemes were developed in the 20th century, bacteria that could
be isolated and grown in the laboratory were assigned into groups based on aspects of
phenotype such as morphology (e.g. shape and grouping); structural features (e.g. Gram
reaction); metabolism (e.g. nature of carbon and energy sources, enzymes, chemical
products); behavior (e.g. motility, interactions); and ecology (e.g. habitats, pathogenic-
ity). Biochemistry and genetics of bacteria gained increasing importance as knowledge
of these aspects grew from the 1960s on, resulting in developments such as characteriza-
tion of specific cell components (e.g. lipids, quinones, and DNA base composition). The
methods underlying these traditional phenotypic approaches were described in Chapter 2;
they are still widely used in culture-based microbiology—especially for bacterial identi-
fication—and they remain part of the recognized system by which bacteria are formally
classified and named. However, these approaches are completely inadequate to provide a
true picture of diversity.

As discussed in Chapter 1, the ideas developed by Carl Woese in the late 1970s heralded
a revolution in our thinking about their relationships between different forms of life.
Modern methods of classification attempt to group organisms by their presumed evolu-
tionary relationships. Such phylogenetic systems of classification depend on comparisons
of the genetic information contained in their macromolecules, especially nucleic acids
and proteins. If two organisms are very closely related, we expect the sequence of the
individual units (nucleotides or amino acids) in a macromolecule to be more similar than
they would be in two unrelated organisms. Sequencing of the genes encoding the 16S
rRNA molecules in the small ribosomal subunit (SSU) has been the most widely used
tool in studies of microbial diversity, and the principles of this approach were discussed
in Chapter 2. The development of direct sequencing of 16S rRNA genes in environmen-
tal samples in use since the late 1990s has produced huge databases of genetic informa-
tion that are analyzed to identify groups of phylogenetically related sequences. With this
technique, it soon became apparent that the marine environment contains organisms that
are phylogenetically distinct from many previously known groups obtained by culture.
By far the largest amount of bacterial diversity is represented by organisms that have
not been cultivated but are known only by their genetic “signatures” observed in envi-
ronmental samples. Microbiologists have long realized that there is a large discrepancy
between the numbers of organisms counted using direct microscopic observation and
those recovered on culture media (the “great plate count anomaly”, p.34). The sequencing
of 16S rRNA genes led to a complete reevaluation of the importance of marine members
of the Bacteria (and Archaea, as discussed in the next chapter), which are both more
abundant and more diverse than we could possibly have imagined before the advent of
these techniques.

Our view of bacterial and archaeal diversity is now undergoing another major revolution as
a result of analysis made possible by widespread use of high-throughput sequencing (HTS,
p.51), leading to the ability to obtain many thousands of genetic sequences from individual
fragments. We can now compare sets of multiple genes and obtain entire genome sequences
from individual microbes. Researchers have developed vast databases of global genetic infor-
mation (trillions of nucleotide bases) from different locations, depths, and habitats obtained
via numerous individual and coordinated ocean sampling projects (Figure 2.12). These data-
bases can be continually analyzed and reanalyzed—“mined” for information—leading to a
new frontier in our understanding of diversity of marine microbes and their importance in
ecological and biogeochemical processes. We are getting closer to answering some of the
key questions: how many kinds of microbes are there and what are they doing? How do they
interact with one another? How might they respond to future environmental changes? Some
recent developments are discussed in Box 4.1.
 Diversity of Marine Bacteria 115

Figure 4.1 A view of the tree of

life, encompassing the total diversity
represented by sequenced genomes
and illustrating the diversity of major
bacterial lineages (upper section).
The Candidate Phyla Radiation (CPR)
is composed entirely of organisms
without isolated representatives and
are still in the process of definition
at lower taxonomic levels. The lower
section shows the eukaryotes as a
single cluster and the Archaea as
branches representing the phyla and
newly designated super-phyla (see
Chapter 5). This is an updated ver-
sion of the tree devised by Hug et al.
(2016) shown in Figure 1.1C. Credit:
reprinted from Castelle and Banfield
(2018) with permission from Elsevier.

Bacterial systematics is in transition due

to application of genomic methods
Microbiologists have adopted the binomial system of Latin names (Genus species) first
developed by Linnaeus for classification of plants and animals, but the concept of a bacte-
rial species is very different. In plants and animals, the presence of distinct morphological
differences, sexual reproduction, and geographic separation can all be used to underpin
a theoretical explanation of the concept of species as a group of individuals that can pro-
duce fertile offspring and have evolved to be reproductively isolated from other species.
This definition is biologically meaningless for bacteria, which are haploid and do not have
sexual reproduction; therefore, the concept of a bacterial species is purely operational.
Microbiologists use a polyphasic approach to classify bacteria into different taxa, based on
observing a combination of numerous phenotypic, chemotaxonomic, and genotypic data.
Members of a species should share a high degree of similarity in many independent charac-
teristics and should be characterized by a distinctive phenotypic property (such as morphol-
ogy, biochemical properties, or physiology) that distinguishes them from all other species.
For a cluster of organisms to be considered members of the same species, they need to be
genomically coherent, so phylogenetic information about the relatedness of properties of the
strains must also be considered. This requirement led to the use of DNA-DNA hybridization
(DDH) as a standard in the designation of a species: if two organisms show DDH of ≥70%
116 Chapter 4

BOX 4.1 RESEARCH FOCUS

High-throughput DNA sequencing provides new

insights into marine microbial diversity

The rare biosphere. The concept of the rare biosphere developed of insects, plants, fish, and other organisms. Ecological theory predicts
from the first use of HTS as part of the International Census of Marine that bacteria would show a log-normal species abundance distribution
Microbes Project (ICoMM, Amaral-Zettler et al., 2010). The method curve. Curtis et al. developed a model by which the number of individu-
of tag 454-pyrosequencing (p.51) used enabled analysis of many als in a number of taxa can be derived by measuring the total number of
thousands of sequences of a hypervariable region of 16S rRNA genes individual bacteria in the community (N) and the number of individuals
from a single sample. This overcame the bias inherent in previous (Nmax) of the most abundant species. Based on the limited data available
studies of molecular diversity, in which dominant populations mask at the time, and by speculating on larger scale diversity, they concluded
the detection of low-abundance types. Sogin et al. (2008) concluded that the entire bacterial diversity of the sea would likely be ~2 × 106 spe-
that the number of different kinds of bacteria (operational taxonomic cies. Subsequently, the use of HTS in the ICoMM survey mentioned
units, OTUs) in the oceans could exceed five to ten million and that above led Sogin et al. (2008) to raise that value to 5–10 × 106.
the “rare biosphere is very ancient and may represent a nearly inex-
haustible source of genomic innovation.” It was later realized that By 2016, the volume of microbial gene sequences in the databases
the method used at this time can be prone to multiple errors in a few had skyrocketed, prompting Locey and Lennon (2016) to analyze
reads, which can lead to apparently unique sequences; Kunin et al. them again using these ecological scaling laws. For the first time,
(2010) cautioned that careful analysis and interpretation of pyrose- they combined data from biodiversity studies of microbes, plants,
quencing data is needed to prevent overestimation of species rich- and animals. They concluded that they were successful in develop-
ness. The Illumina technology that superseded 454-pyrosequencing ing a scaling law that predicts the abundance of dominant species
produced shorter reads but had fewer errors. Researchers have con- across 30 orders of magnitude (matching the estimated abundance of
tinued to apply new DNA sequencing technology and develop new microorganisms on Earth). Using this scaling law, they predicted up
bioinformatic methods for data analysis to improve the reliability of to a trillion (1012) species of microbes on Earth. Based on known dis-
estimates of richness and community structure (Pedros-Alio et al., tributions (Whitman et al., 1998) we might expect ~90% of these spe-
2018). Jousset et al. (2017) discuss the factors that drive the presence cies to be marine. However, Willis (2016) criticized the methods used
and persistence of rare microbes, which include competition and neg- in the papers by Curtis et al. (2002) and Locey and Lennon (2016),
ative interactions between species. Also, specialization in the use of arguing that extrapolating SAD curves has no predictive power for
particular nutrients or preference for environmental conditions leads estimating true microbial diversity because it is statistically invalid.
organisms to be rare in some environments and abundant in others. In response, Locey and Lennon argued that their method did not rely
Grazing protists, bacterial predators, and phages are major factors in on extrapolation and was therefore valid, because they used values of
shaping community composition, because they tend to overconsume N and Nmax that matched the scale of predictions.
the most abundant prey, leading to trade-offs between organisms that
can be explained by a theoretical framework known as the “kill the Yarza et al. (2014) analyzed the curated SILVA database for near
winner” hypothesis (Winter et al., 2010; also see p.202). Jousset et al. full-length 16S rRNA sequences of Bacteria (1.4 × 106) and Archaea
note that we may be overlooking a substantial part of the biosphere (5.3 × 104) and concluded that only near complete 16S rRNA sequences
because (at that time) a standard procedure in processing the results give accurate measures of taxonomic diversity. Schloss et al. (2016)
of HTS was to eliminate sequences of singletons (sequences repre- showed that considerable biodiversity has been discovered since they
sented only once). They argue that low-abundance species should not conducted a similar census in 2004, but much of this had been biased
be considered as “analytical annoyances” but treated as full members towards particular phyla and environments. Figure 4.2 shows the fre-
of microbial communities. They provide many examples of ways in quency of distribution of sequences from the various phyla. Schloss
which rare species can have a disproportionate role in biogeochemi- et al. also found that ~95% of new full-length bacterial and archaeal
cal cycles—“less is more”—with particular importance in response sequences belong to OTUs that have been observed more than once
to nutrient cycling, pollutant degradation, and processes that affect already. The rate of discovery of new species is slowing, as indicated
levels of greenhouse gases (e.g. methane consumption and denitrifi- by rarefaction curves (the number of species plotted against the number
cation). Rare microbes can be more active than abundant types and of samples) approaching saturation. Based on these results, the rich-
may enhance community metabolic functions by providing a vast ness values of 1012 proposed by Locey and Lennon seem unrealistically
gene pool for unique or complementary metabolic pathways. high. Schloss et al. conclude that research to date has been effective
at sampling the most abundant organisms, but it struggles to sample
How many bacterial and archaeal species really exist? Surveys the rarer organisms. They emphasize the need to sample environments
based on amplification and Sanger sequencing of 16S rRNA genes con- with high biodiversity using newer sequencing methods that will yield
ducted since the 1990s soon revealed the diversity of ocean microbes. full-length sequences from these rare organisms. In discussion of this
But can we put a numerical value on this diversity, rather than just paper, Amann and Rosselló-Móra (2016) estimate that the census of
concluding … “it’s enormous”? Curtis et al. (2002) argued that it is bacterial and archaeal species might ultimately stop in the lower mil-
not necessary to count every single species in a sample; instead, they lions rather than the billions predicted earlier. They encourage micro-
showed how this can be achieved by measuring the area under the spe- biologists—rather than being dismayed at having “only” millions of
cies abundance distribution (SAD) curve for that environment. This is species—to apply our creative minds to access the currently under-
a method frequently used in macroecology to measure the biodiversity explored rare biosphere with strategies for sampling and analysis that
 Diversity of Marine Bacteria 117

BOX 4.1 RESEARCH FOCUS

Figure 4.2 Frequency of

sequences for bacterial phyla from
(a) marine water samples and
(b) marine sediments and hydro-
thermal vents. Data obtained from
Schloss et al. (2016).

focuses on the unknown to seek new organisms. Without doubt, much system and to ensure that the GTDB taxa are monophyletic, Parks et
new biology and biochemistry is still to be discovered. al. tested their system with a variety of phylogenetic tools.

New phylogenetic trees permit a new approach to bacterial tax- A new taxonomic system. This new approach has resulted in major
onomy. The existing classification system for the Bacteria contains reclassification of taxa above the rank of species—73% of the
many inconsistencies with our goal of grouping organisms according genomes currently assigned to a taxon within the NCBI database
to their evolutionary relationships. Many current taxa are polyphy- had one or more changes. The most changes occurred at the level
letic; the application of 16S rRNA gene relationships has helped enor- of orders. Some of the most important changes at the higher ranks
mously but has not entirely resolved the problem. As discussed above, affect important and well-studied groups with culture-based clas-
there is a huge diversity of unclassified rare taxa that are not currently sifications. These changes include: (a) reclassification of the class
affiliated with known clades. Some (many?) of these will undoubtedly Betaproteobacteria as an order of the class Gammaproteobacteria;
possess hitherto unknown roles in marine processes. In order to focus (b) removal of the Epsilonproteobacteria from the Proteobacteria
on these unknown clades, Yilmaz et al. (2016) analyzed the SILVA and reclassification as a new phylum Epsilonbacteraeota; (c) reclas-
database of bacterial and archaeal 16S rRNA gene sequences, using sification of the order Desulfurellales (Deltaproteobacteria) to a
a new approach to construct phylogenetic trees that included aspects new class within the Epsilonbacteraeota; (d) internal reclassifica-
such as size, phylogenetic depth, and standard nomenclature. This tion of the Firmicutes; and (e) transfer of the Chlorobi to the phy-
involved the use of the candidate taxonomic unit (CTU) proposed lum Bacteroidetes. The GTDB system also allows the uncultivated
by Yarza et al. (2014). Because the major known clades in SILVA microbes to be assigned to named candidate taxa. In particular,
have been annotated manually in a guide tree, Yilmaz et al. were the large Candidate Phyla Radiation appears to be a monophy-
able to reveal previously unknown clades. With this new phylogenetic letic group for which Parks and colleagues propose the name
and taxonomic framework, they performed a meta-analysis of the Patescibacteria. It is hoped that maintenance of an up-to-date
ICoMM marine water samples and other datasets. Because most of normalized genome-based classification will improve the analysis
the new clades were interspersed by known taxa containing cultivated of microbial genome data and consistent reporting of diversity in
species whose genome has been sequenced, use of the bioinformat- scientific communication. But there will undoubtedly be continued
ics software package PICRUSt could be used to predict the possi- debate about the merits of such a major overhaul of long-estab-
ble metabolic functions of the new clades. Thus, up to 92 unknown lished taxonomic system.
clades were characterized and classified. A different recent approach
to incorporating uncultivated organisms into the tree of life was used Towards a new concept of bacterial species. It is now possible to
by Hug et al. (2016). They argued that, despite its widespread use and envisage a genomic basis for defining species (and higher taxa) that
enormous value to date, 16S rRNA gene analysis has some inherent takes account of the huge microdiversity observed and the difficulty
limitations, so that many unknown groups escape our attention. To of defining the boundaries of taxa. By comparing multiple genome
obtain higher resolution, they used an aligned and concatenated set sequences of related organisms, we can define the pangenome—the
of 16 ribosomal protein sequences deduced from available genomes collection of all genes across the set of strains examined. We can then
for each organism (Bacteria, Archaea, and Eukarya were included) to ask which genes are present in every strain—this is the core genome,
construct a new view of the tree of life (Figure 4.1). The main con- which could be used as the proxy for defining a species. Similarity
clusion of this study is that the domain Bacteria includes many more of genomes can then be quantified by “digital hybridization” of con-
major lineages of organisms than either the Archaea or Eukarya, served regions of the query and reference sequences, or by compar-
including the large branch termed the Candidate Phylum Radiation. ing the average nucleotide identity (ANI) of the genome of the query
Parks et al. (2018) took this approach further by developing a phylog- strain with that of a refence strain (Richter and Rosselló-Móra, 2009).
eny inferred from concatenation of 120 ubiquitous, single-copy pro- These methods could replace the current “gold standard” of labora-
teins covering ~95,000 genomes—of which 14% are assembled from tory determination of DNA-DNA hybridization, which is technically
metagenomes, or single-cell genomes, from uncultured organisms. too demanding (Chun et al., 2019). Currently, the nomenclature rules
This produced a new system and database (the Genome Taxonomy require type cultures to be deposited in order for a species names to
Database, GTDB), publicly available to guide reclassification and be validly published, but in future, genome type material may be per-
future classification of newly discovered bacteria. To validate their mitted for uncultured organisms (Konstantinidis et al., 2017).
118 Chapter 4

they are considered to be of the same species. DDH is a difficult technique and prone to
errors, therefore comparison of 16S rRNA gene sequences became widely accepted as the
mainstay of bacterial taxonomy. For many years, two strains were considered as belonging
to distinct species if they shared 16S rRNA similarities <97% and different genera at <95%.
The species threshold was raised to 98.6% in 2006. A variety of other indirect methods of
genome comparison are shown in Table 2.2, and the currently preferred method for compar-
ing genome sequences of different organisms is to calculate the overall genome relatedness,
most commonly by calculating the ANI, where the cutoff value for the species boundary is
95–96%. The rapid development of relatively inexpensive and technically straightforward
whole genome sequencing (WGS), plus a wealth of associated bioinformatics tools, means
that taxonomy is shifting from its reliance on these indirect methods. WGS can provide a
shortcut to direct evidence about the biochemical, physiological or structural features that
are traditionally tested in polyphasic taxonomy. However, new classifications must be com-
patible with existing systems and there are many different opinions about how these novel
approaches should be integrated.

While the designation of species is difficult and results in differences of opinion among
experts, agreement about grouping species into higher taxa is even more difficult. Boundaries
for defining taxa using 16S rRNA gene sequence identities are proposed at 94.5% for genus,
86.5% for family, 82.0% for order, 78.5% for class, and 75% for phylum. As discussed in
Box 4.1, new genome-based criteria for the definition of species and higher taxa are now
being implemented.

? WHAT’S IN A NAME? Although there is still no formal system of classifying bacteria, there are strict rules about
BACTERIAL TAXA the designation and naming of new taxa under a Code of Nomenclature regulated by the
International Committee of Systematics of Prokaryotes. The main aims of the code are to
A few guidelines may help to navi-
ensure that names are stable and necessary and to avoid errors or confusion that may hamper
gate the plethora of Latin bacte-
rial and archaeal names. Orders scientific progress. Expert sub-committees keep the status of taxa in particular groups under
and families take their name from review. A variety of characteristics must be considered before a proposed name for a new
the stem of the name of the type species, genus, or family can be defined as valid. Detailed rules about the published descrip-
genus, and the suffix indicates tions, etymology of the name, and diagnostic features are required. A key requirement is
the rank; families end in -aceae, the designation of a “type strain” of a species and its deposit in internationally recognized
orders in -ales, and classes in -ia or culture collections, where it is archived and maintained in a viable form as a reference point
-ea. For example, Vibrio cholerae for assigning other strains to the species. It is important to note that the Code only specifies
is the type species of the family rules about naming microbes and, like all biologists who specialize in taxonomy, microbiolo-
Vibrionaceae, order Vibrionales, gists often have many differences of opinion about the assignment of organisms to taxa (tax-
class Gammaproteobacteria. Genus
onomists are often categorized as “lumpers” or “splitters”). There are currently over 15600
and species names must always be
written in italics (or underlined).
validly named species of Bacteria and Archaea.
The genus name can be abbrevi-
ated to single letter after first use, Uncultured bacteria cannot be assigned to a valid species, because we do not have full
or a three-letter abbreviation if information about their phenotype and there is no type strain, as required by the Code of
it would cause confusion. For Nomenclature. However, in some cases, it is possible to assign a provisional species name
example, Aeromonas salmonicida with Candidatus designation if genomic sequences point to significant differences from
and Aliivibrio salmonicida would other recognized species in structure, metabolism, and other key properties. Names are writ-
appear as Aer. salmonicida and Ali. ten in the format “Candidatus Genus species” (note the use of quotation marks and italics);
salmonicida in a list. After a genus Candidatus may be abbreviated to Ca. This interim status has been extended to allow the
name, sp. and spp. (not italicized) designation of candidate higher taxa for uncultivated organisms.
indicate a single or multiple
undesignated species, respectively.
Names of all formal taxa start with
a capital letter. According to the
OTUs and ASVs are used to represent
nomenclature codes, names of diversity in community analyses
validly published higher bacterial,
The difficulties of defining bacterial species means that we need a convenient way to iden-
archaeal, and viral taxa should also
be italicized but names of eukary-
tify groups of closely related organisms in gene-based surveys. Operational taxonomic units
otic taxa are not. Publishers vary in (OTUs) have been the traditional way of doing this, as they serve as a pragmatic proxy for
application of this convention and species. Once 16S rRNA gene sequences have been obtained, computer algorithms can clus-
you will see both forms in journal ter them into OTUs on the basis of similarity—typically ≥98.6% to match the threshold for
articles and books. In this book, species. Sequences for other marker genes can be grouped in a similar way, using appropri-
names of all taxa above genus are ate thresholds set by the researcher. This approach has been used in thousands of ecological
not italicized. investigations since the advent of environmental gene sequencing in the 1990s.
 Diversity of Marine Bacteria 119

A more recent approach is to compare all of the individual sequences present in a commu-
nity, generating amplicon sequence variants (ASVs). (The term “exact sequence variants,”
ESVs, is sometimes used.) Until recently, this has been problematic because of small errors in
amplification and sequencing of the marker gene, but recent improvements in control of these
errors means that it is now possible to reliably distinguish sequences differing by as little
as one nucleotide. The main advantage of ASVs is their improved resolution, because the
species similarity threshold can mask considerable differences between microbes that have
distinct phenotypic or ecological attributes. (As an example, see the discussion of genome
differences in ecotypes of the species Prochlorococcus marinus on p.135.) A second advan-
tage is that ASVs in different datasets can be directly compared against one another, which
makes comparison between different studies more efficient. As with conventional taxonomy,
there are differences of opinion—lumpers and splitters—and some consider that the lower
resolution of OTUs can be advantageous for some purposes in analyzing diversity. There is
a growing trend to encourage all ecological survey data to be reported using ASVs—if all
datasets are reported in ASVs, the advantage of easy comparison is retained, but investigators
can always recompute sequences as OTUs if they wish. Because most microbial community
analyses to date have been reported in OTUs, this system is used in this chapter.

When considering microbial diversity, three general ecological concepts are used. Alpha
diversity refers to the average diversity of OTUs found in a habitat or particular sample, i.e.
at a local scale. It is usually expressed as richness, i.e. the number of different OTU types.
Beta diversity is the ratio between diversity at a local scale (alpha) and a regional scale; in
other words, the amount of change in the OTUs between the two habitats. Gamma diversity
is a measure of the overall diversity of an ecosystem or large region. Richness is the sim-
plest measure of diversity, but it does not tell us about evenness of distribution—the relative
abundance of different OTUs. Ecologists have devised many ways to quantify these aspects
of diversity into a single index, and discussion of the methods of calculation and merits of
different approaches is beyond the scope of this book. Besides the simple measure of rich-
ness, two of the most important measures encountered in microbial ecology are the Shannon-
Wiener index (a measure of evenness of populations within a community) and the Simpson
index (a measure of dominance, weighted towards the abundance of the most common OTUs)
and then used to measure proportional abundance.

Marine microbial communities show high alpha diversity

Using the new method of direct sequencing of nucleic acids from the marine environment,
the first studies conducted in the early 1990s soon revealed that the most abundant clones of
16S rRNA bacterial gene sequences did not correspond to cultured species at all. When phy-
logenetic trees were constructed based on genetic similarity, more than three-quarters of the
marine Bacteria revealed by this approach mapped onto about 20 phylogenetic groups with
worldwide distribution. These formed clusters of related types, rather than single lineages,
and genetic variability within these populations was observed to be very high. In some clus-
ters, differences in 16S rRNA sequences were seen to be due to genetic variability associ-
ated with adaptation to different depths of water. Subsequent analyses showed that bacterial
diversity in suspended particles and free picoplankton is very different and varies greatly
with depth and environmental conditions.

Many of the initial studies were conducted in the Sargasso Sea, and new sequences that
could not be linked to cultured bacterial groups were assigned to clades with the prefix
SAR. Among the early discoveries was the abundance and wide distribution of the SAR11,
SAR116, and Roseobacter clades, whose phylogeny indicated them to be members of the
Alphaproteobacteria. The SAR 11 cluster of 16S rRNA sequences was found in almost every
pelagic environment ranging from shallow coastal waters to depths of over 3000 m. It is
probably the most abundant group of microorganisms in the sea and on average accounts for
nearly a third of the cells present in surface waters and a fifth of the cells in the mesopelagic
zone. SAR11 is a deeply branching member of the class Alphaproteobacteria and is phylo-
genetically distinct from all cultured members of the group. Some members of the SAR11
cluster have now been cultured and this has allowed study of genomic and physiological
properties, as discussed in Box 2.1.
120 Chapter 4

In these studies, members of the SAR86 clade (class Gammaproteobacteria) also emerged as
ubiquitous inhabitants of the surface layers of the oceans and shallower coastal waters and
were soon linked to a previously unknown type of photoheterotrophic metabolism in oligo-
trophic conditions (Box 3.1). These studies also showed that sequences matched to the phy-
lum Cyanobacteria accounted for a large segment of the diversity in the photic zone. This was
less surprising, since many can be cultured and some—notably the genera Prochlorococcus,
Synechococcus, and Trichodesmium—had already been extensively studied in view of their
importance in contributing to productivity through photosynthesis. By the turn of the cen-
tury, the picture emerging of marine bacterial diversity was one in which over three-quarters
of the sequences could be grouped into about 12 dominant clades.

Since these early investigations, many research groups have conducted studies in diverse
marine habitats throughout the world and established databases containing many thousands
of sequences from marine samples. Different investigators, using variations of the basic
methods and sampling different geographic regions and depths have repeatedly found simi-
lar phylogenetic groups, which can be clustered into a number of major clades. The vertical
structure of the water column emerged as being the most important factor in overall variation
of diversity in the oceans. There are also high levels of beta diversity, with the composition
of microbial communities greatly affected in time and space, with seasonal and geographic
variations.

However, this “big picture” view hides a staggering amount of alpha diversity at the
microscale. This finding was surprising, because ecological theory predicts that competi-
tion for the limited nutrient resources found in most of the ocean should lead to low levels
of diversity. We can now explain this paradox through our realization of the importance of
factors that are explored in other chapters. Firstly, patchy distribution and gradients of nutri-
ents at the microscale, together with syntrophic and symbiotic interactions, results in het-
erogeneity of the environment, providing different niches that can support diverse microbes
(Chapter 3). Secondly, top-down control of populations by protistan grazing and phage lysis
have major influences on community structure and promote diversity by attacking the most
abundant or active members of the community (Chapters 6 and 7).

As discussed in Box 4.1, the International Census of Marine Microbes (ICoMM) was estab-
lished in 2004 to coordinate and catalog our knowledge of microbial diversity in the oceans,
facilitated by high-throughput sequencing (HTS), emerging at this time through the develop-
ment of the 454 Life Science sequencing technology (p.51). One of the most significant and
surprising early discoveries of ICoMM was the realization that the abundance distributions
of different OTUs are highly uneven, leading to the concept of the “rare biosphere.” Despite
their low abundance, rare microbes can have a disproportionally large effect on ecosystem
functioning.

A TOUR OF THE BACTERIAL AQUARIUM

Even among the tiny fraction of all bacteria that we can culture and study in the labora-
tory, there are over 15000 recognized species with published names, together with several
hundred organisms sufficiently well characterized to be given Candidatus species status.
These are organized into hundreds of higher taxa, many of which contain organisms that
cannot be grouped by phenotype. This makes it impossible for a book like this to provide
a systematic description of all types of marine bacteria. Instead, I will imagine that we are
visiting a virtual aquarium containing representatives of some of the most noteworthy and
best-understood groups (a concept adapted from James W. Brown). Just as a real aquarium
with a complete collection of all the world’s fish would overwhelm the visitor, our bacte-
rial aquarium contains selected examples that have particularly interesting characteristics
or because they are important in marine ecology, biogeochemical cycles, or biotechnology.
Mostly, the “tanks” are arranged by phyla, but in some places interesting properties from
other phyla are mentioned in order to keep similar features together.

The discussion that follows includes information about the properties of some of these cul-
tured, named species as well as information about related types that is gained from genomes
 Diversity of Marine Bacteria 121

constructed by metagenomic assembly or amplification from single cells extracted from the
environment (see p.56 for a summary of methods). Note that the description of taxa given
here mostly follows the current LPSN/NCBI organization, but some significant proposed
changes to higher taxa (especially phyla) arising from the new genome-based GTDB clas-
sification (see Box 4.1) are included.

The Proteobacteria account for about

half of all bacterial ocean diversity
Figure 4.2 shows the frequency of the most abundant sequences in marine water and sedi-
ment samples grouped by phyla. These show that the phylum Proteobacteria is the most
abundant in 16S rRNA gene surveys and it is widely distributed in all major habitats. Almost
all the main metabolic types (autotrophy, heterotrophy, anoxygenic phototrophy, methylotro-
phy, sulfate reduction, nitrogen fixation), requirements for oxygen and mechanisms of CO2
fixation are represented, and many types of cell morphology are found. Cells have Gram-
negative structure. The group has conventionally divided into six classes; namely the Alpha-
Beta-, Gamma-, Delta-, Epsilon-, and Zetaproteobacteria (often abbreviated using the Greek
letters α, β, γ, δ, ϵ, and ζ). Using the GTDB classification, only the Alpha-, Gamma-, and
Zetaproteobacteria remain as classes within the phylum. The Betaproteobacteria is now an
order within the Gammaproteobacteria class. Members of the Delta- and Epsilonproteobacteria
are assigned to new phyla. Despite this reorganization, the biggest tank in our virtual aquar-
ium is undoubtedly reserved for the phylum Proteobacteria and we begin our tour with this
group (Table 4.1).

Members of the class Alphaproteobacteria

are the most abundant marine bacteria
This class includes many types with important functions in marine habitats. Many have
been isolated and cultured and Table 4.2 contains a brief summary of representatives of
the 434 cultivated genera in 15 families and 14 orders. This is a tiny fraction of the total
diversity revealed by 16S rRNA gene surveys of euphotic and mesopelagic zones. Members
display a wide range of metabolic types, evident in cultivated isolates as well as by evidence
from metagenomes. The most abundant types, which include the dominant clades SAR11,
SAR116, OCS116, and OM75, are slow-growing oligotrophs with small cells and stream-
lined genomes. SAR11 is often described as the most abundant type of bacteria on Earth,
although this hides the fact that the clade can be divided into multiple subclades which form
distinct ecotypes that show clear differences in dynamics due to environmental conditions.
Many of the Alphaproteobacteria are metabolic generalists containing genes for utilizing
light via proteorhodopsin and for metabolizing C1 compounds. Representatives of SAR11
and SAR116 have been cultured and given the Candidatus names “Pelagibacter ubique” and
“Puniceispirillum marinum” respectively. These are closely related to the Rickettsiales while
OM75 is related to the Rhodospirillaceae. The Roseobacter clade (family Rhodobacteriaceae)
is also very abundant, sometimes representing up to 20% of bacterial cells in coastal ecosys-
tems and 3–5% in the open ocean, blooming seasonally in association with algal growth. The
genomes of many members of the Roseobacter clade have been sequenced, revealing enor-
mous diversity of metabolic potential and regulatory systems that explain their ubiquity and
success in many marine habitats. There is considerable variation in genome size—some are
highly streamlined, while others are quite large, with over 4000 protein genes. “Ca. Riegeria
santtandreae” is an intracellular symbiont with a genome of only 1.34 Mb, yet it retains a
highly energy-efficient metabolism and serves as the primary energy source for its host, the
flatworm Paracatenula (p.267).

The order Caulobacterales contains prosthecate bacteria

Members of the Caulobacterales are distinguished by the presence of extrusions or append-
ages of the cytoplasm called prosthecae. Unlike almost all other bacteria, members of this
group show a life cycle with unequal cell division, in which the “mother cell” retains its
shape and morphological features while budding off a smaller “daughter” cell. This occurs
122 Chapter 4

Table 4.1 Representative members of the class Alphaproteobacteria. There are 14 named orders

Representative familya,b Distinctive features Representative marine

genera or speciesb

ORDER Caulobacterales (2)

Caulobacteriaceae (4) Mostly stalked cells that produce a flagellated cell. Chemoheterotrophs Caulobacter crescentus
attached to surfaces in soil, freshwater, and marine habitats.

Hyphomonadaceae (15) Some divide by binary fission with motile offspring, others have stalked cells. Algimonas, Hellea,
Oligotrophic chemoheterotrophs in a wide range of marine habitats. Woodsholea,
Oceanicaulis

ORDER Rhizobiales (17)

Aurantimonadaceae (4) Small family of marine chemoheterotrophs, including the coral pathogen Aurantimonas,
Aurantimonas coralicida. Fulvimarina, Martelella

Bradyrhizobiaceae (10) Mostly plant-associated nitrogen-fixers; Rhodopseudomonas is a marine Nitrobacter,

phototroph. Nitrobacter is a widely distributed chemolithoautotrophic Rhodopseudomonas
nitrifying bacterium.

Methylobacteriaceae (3) Pink-pigmented facultative methylotrophs growing on C1 compounds, as Methylobacterium

well as larger carbon compounds. Sometimes associated with human
infections and contamination of fuel oils.

Methylocystaceae (10) Type II methanotrophs utilizing methane and C1 compounds as source of Methylocystis,
carbon and energy; Distinguished by internal membranes containing Methylosinus
methane monooxygenase around the periphery of the cell and assimilating
C1 intermediates via the serine pathway. Widespread in sediments at the
interface between CH4 and O2.

Rhizobiaceae (13) Rhizobium is mostly known for symbiotic nitrogen fixation in plants. Some Rhizobium,
marine species have been isolated from marine sediments and coral mucus. Methylobrevis
Methylobrevis are facultative methylotrophs.

ORDER “Pelagibacterales”

“Pelagibacteriaceae” Comprises at least five related groups of bacterioplankton found in marine “Ca. Pelagibacter
and fresh waters that constitute up to half of all bacterial and archaeal cells ubique”
in the open ocean. Includes the founder group SAR11. Oligotrophs with
small cells and streamlined genomes, phylogenetically closely related to the
Rickettsiales and the ancestor of the mitochondrion.

ORDER Rhodobacterales (1)

Rhodobacteriaceae (148) Very large family with extensive metabolic and ecological diversity. Highly Dinoroseobacter,
abundant in marine environments, with particular importance as aerobic Jannaschia, Pelagimonas,
photo- and chemoheterotrophs and in the carbon and sulfur cycles. Many Phaeobacter,
form close associations or symbiotic interactions with other organisms Rhodobacter,
(especially phytoplankton). Roseobacter,
Roseovarius, Ruegeria,
Silicibacter

ORDER Rhodospirillales (2)

Rhodospirillaceae (37) Diverse metabolism. including purple non-sulfur bacteria found on Rhodospirillum,
sediments or anaerobic niches of microbial mats; these can grow Magnetospirillum, “Ca.
photoheterotrophically or anaerobically via fermentation or aerobically via Riegeria santtandreae”
respiration. Includes magnetotactic species found in salt marshes and
fresh water.

ORDER Sphingomonadales (2)

Sphingomonadaceae (19) Contain distinctive membrane sphingolipids. Widely distributed aerobic Sphingomonas
chemoorganotrophs, usually associated with rich sources of organic matter.
Some species degrade polyaromatic hydrocarbons and toxic compounds,
which may be important in bioremediation. Some species produce
industrially useful polymers.

a Figures in parentheses indicate number of validly published taxa in the next lower rank (families in orders, genera in orders) from the List of

Prokaryotic Names with Standing in Nomenclature (LPSN, www.bacterio.net). b Names in non-italic font and quotation marks are not validly
published under the Bacteriological Code. Genera may have Candidatus status, but affiliation to higher taxa is not confirmed.
 Diversity of Marine Bacteria 123

Figure 4.3 The dimorphic cell cycle

of Caulobacter crescentus. Precise
spatiotemporal patterning of the key
regulatory proteins CtrA, PleC, and
DivJ coordinate cell polarity, DNA
replication, cell division, and flagellar
biosynthesis. Cells are false colored
red to indicate presence and distribu-
tion of active CtrA. Credit: reprinted
from Hughes et al. (2012) with per-
mission from Elsevier.

owing to polarization of the cell, whereby new cell wall material is grown from a single
point rather than by intercalation as occurs in other bacteria. Some, such as Hyphomicrobium
and Rhodomicrobium, bud off progeny cells from hypha-like extensions. Others, such as
Caulobacter, have distinctive stalks by which they attach firmly to algae, stones, or other
surfaces in aquatic environments. The progeny cells are motile and swim away to colonize
fresh surfaces (Figure 4.3). Caulobacter crescentus has been intensively studied as a model
of cellular differentiation. The cell cycle is controlled by a set of three master regulator pro-
teins that determine the coupling of gene expression to cell-cycle events with extraordinary
precision. The increased surface area: volume ratio of stalked bacteria probably enables them
to thrive in nutrient-poor waters and the prosthecae also enable these aerobic bacteria to
remain in well-oxygenated environments by avoiding sinking into sediments. These bacteria
are often the first to colonize bare surfaces and are therefore important in the formation of
biofilms, with significant consequences for larval settlement and biofouling (Chapter 13).

Several alphaproteobacterial genera show magnetotaxis

The genera Magnetospira, Magnetospirillum, and Magnetovibrio (Order Rhodospirillales)
and Magnetococcus (Order Magnetococcales) show magnetotactic behavior, orienting them-
selves in the Earth’s magnetic field. The rod-shaped or spiral cells contain chains of mag-
netic particles comprised of magnetite (Fe3O4) and greigite (Fe3S4) anchored within the cell
(Figure 4.4). A magnetic field imparts torque on the chain of magnetosomes, so that the cell
becomes aligned with magnetic field lines. The bacteria swim in this direction using polar
flagella. The bacteria use this behavior in conjunction with aerotactic responses to locate a
favorable zone in sediments of optimal O2 and sulfide concentrations, which they require
for growth. Although the best-studied examples are found in freshwater mud, magnetotac-
tic bacteria are widespread in salt marshes and other marine sediments. Recent molecular
studies indicate that magnetotaxis is not restricted to a small, specialized lineage as previ-
ously thought, and there are examples in the Deltaproteobacteria (see below). Magnetotactic
bacteria can be isolated quite easily by applying a magnetic field to mixed environmental
samples, although they are very difficult to cultivate. Analysis of the genome of these bacteria
reveals that several genes are required for the acquisition of iron via siderophores and energy-
dependent transport into the cell, followed by its conversion to magnetite or greigite within
the magnetosomes. Microfossils of magnetite crystals found in deep-sea marine sediments
are thought to originate from magnetotactic bacteria at least 50 MYA.
Figure 4.4 Transmission elec-
tron microscopy images of
Magnetospirillum sp. AMB-1. A.
Grown under iron-rich conditions,
showing a chain of magnetite
crystals. B. Grown in the absence
of iron, showing empty magneto-
some vesicles. Credit: reprinted from
Komeili et al. (2004) with permission;
copyright 2004 National Academy of
Sciences, USA.
124 Chapter 4

Figure 4.5 Electron microscopy

images of “Ca. Magnetoglobus
multicellularis”. A. TEM showing the
multicellular consortium consisting
of about 40 bacterial cells containing
lipid inclusions (asterisks) and mag-
netosomes surrounding an acellular
space (Ac). The cells and the mag-
netosomes are specially arranged
so that the total magnetic moment
aligns the whole structure so it can
swim along field lines. Bar = 1 μm.
B. SEM showing the spherical mor-
phology and the tightly bound cells.
Bar = 2 μm. Credits: A. reprinted from
Martins et al. (2007) with permission
Magnetotaxis is also found in other classes and phyla
of John Wiley and Sons. B. reprinted There is growing evidence that the phenomenon of magnetotaxis is much more widespread
from Silva et al. (2008). With permis- than previously thought. In the Proteobacteria phylum, examples are now known from the
sion of Elsevier. Gamma-, Delta-, and Zetaproteobacteria classes. The phyla Nitrospirae and Planctomycetes,
and the candidate phyla “Latescibacteria” and “Omnitrophica” also contain magnetotactic
bacteria.
HOW DO BACTERIA
? MAKE NANO-
SIZED MAGNETS?
A unique bacterium named “Ca. Magnetoglobus multicellularis” forms a compact spherical
aggregate of highly organized flagellated bacterial cells that swim in either helical or straight
The individual magnetosomes in trajectories (Figure 4.5). The bacteria behave as a truly multicellular organism, as the aggre-
magnetotactic bacteria are too gate grows by enlarging the size of its cells and doubling the volume of the whole structure.
small to behave as magnets on Cells divide synchronously and separate into two identical spherical aggregates. The clas-
their own and must be aligned into sification of this bacterium is uncertain; it is currently in the Deltaproteobacteria and has not
a linear chain so that they func- yet been assigned in the GTDB taxonomy.
tion like a compass needle. The
physical tendency of magnetic
dipoles would be to agglomerate The order Betaproteobacteriales includes many rare OTUs
in order to lower the magnetostatic
energy, so the magnetosomes are This group has recently reclassified from a class to an order within the Gammproteobacteria
stabilized with an organic struc- in the GTDB system. It is metabolically diverse, containing 8 orders, 15 families, and ~192
ture to anchor the chain in place. cultivated genera. Many marine members are copiotrophic heterotrophs and they are fre-
Scheffel et al. (2006) identified quently found in association with animal and plant surfaces. Others are chemolithoauto-
a key gene in Magnetospirillum trophic, of which the most important types are sulfur-oxidizing bacteria (SOB) which use
gryphiswaldense that encodes the oxygen (or sometimes nitrate or nitrite) as terminal electron acceptors; some oxidize iron
protein MamJ, which forms a novel (Fe2+→Fe3+). They occur at the interface between oxic and anoxic zones of sediments and
filamentous structure. A MamJ
mats (p.80). Among the best-known genera is Thiobacillus, although many of its species have
deletion mutant showed reduced
magnetic orientation; it produces
recently been reclassified to other genera. OM43 is a significant uncultured clade of methy-
normal magnetite crystals, but lotrophs, which degrades C1 compounds (but not methane). Its genome has been sequenced
these assemble into clusters instead and shown to be the smallest known for a free-living organism, but its taxonomic position
of chains. MamJ is laid down as a in the class is not clear. Betaproteobacteria account for ~5–10% of retrieved 16S rRNA gene
cytoskeleton-like structure from sequences in marine samples, many of which occur in very low abundance, as part of the
pole to pole of the cell, guiding the rare biosphere.
formation of the magnetosomes
and magnetite crystals. Recently,
Taoka et al. (2017) showed that The Gammaproteobacteria is a very large and diverse class
daughter cells receive a functioning
magnetic assembly at cell division, Cultivated members of this class comprise over 21 orders, 56 families, and 351 genera. It
and the static chain-like arrange- contains some of the most familiar bacteria encountered in laboratory microbiology as they
ment of magnetosomes is needed are ubiquitous in aquatic and terrestrial habitats and found almost everywhere apart from
to ensure this. The magnetosome extremely hot or alkaline environments; many are easily cultured with rapid growth; some
gene clusters are highly conserved are important as pathogens of plants, animals, and humans. All major types of metabolism
(Kolinko et al., 2016) and recent are represented. In several large orders, notably the Alteromonadales, Oceanospirillales,
phylogenetic analysis based on and the Vibrionales, the species are almost entirely marine. Examples of cultivated rep-
metagenomic studies suggests
resentative marine genera are shown in Table 4.2. In marine environmental gene surveys,
that magnetotaxis evolved very
Gammaproteobacteria account for 25–30% of retrieved sequences and bacteria in many of
soon after life began ~3 BYA (Lin
et al., 2018). the clades have not yet been cultured. A number of clades have been recognized in low to
medium productivity ocean waters, of which the SAR86 clade is prevalent. This coexists with
 Diversity of Marine Bacteria 125

Table 4.2 Representative members of the class Gammaproteobacteria. There are 21 named orders

Representative familya, b Distinctive features Representative

marine generab

ORDER Enterobacterales (1)

Enterobacteriaceae (58) Mainly occur as commensals or pathogens in the gut of warm-blooded Enterobacter,
animals; important as contaminants in sewage-polluted coastal waters Escherichia,
which may infect marine mammals. Serratia marcescens from this Salmonella, Serratia
source causes a coral disease. Facultatively anaerobic, usually motile by
peritrichous flagella.

ORDER Vibrionales (1)

Vibrionaceae (13) Collectively known as “vibrios;” facultatively anaerobic, typically Aliivibrio,

curved cells, usually motile by polar flagella. Copiotrophic Enterovibrio,
heterotrophs, associated with surfaces, with a major role in biofilm Grimontia,
formation. Includes bioluminescent symbionts of squid and fish. Many Photobacterium,
species are pathogenic (see text). Salinivibrio, Vibrio

ORDER Aeromonadales (2)

Aeromonadaceae (6) Aerobic or facultatively anaerobic heterotrophs. Some are denitrifiers Aeromonas,
with fermentative metabolism. Halotolerant, typically in freshwater and Oceanimonas,
estuarine sediments. A. salmonicida is a pathogen in mariculture and A. Zobellella
hydrophila is an opportunistic human pathogen.

ORDER Alteromonadales (10)

Alteromonadaceae (16) Strictly marine, obligately aerobic fast-growing heterotrophs readily Agarivorans,
isolated from marine particles or surfaces of algae and animals. Motile Alteromonas,
with polar flagellum. Large genomes encode multiple pathways for Glaciecola,
degradation of complex macromolecules (some degrade agar). Marinobacter

Colwelliaceae (3) Strictly marine, aerobic or facultative. Colwellia includes obligate Colwellia,
psychropiezophiles with biotechnological applications. Thalassomonas

Pseudoalteromonadaceae (3) Frequently isolated from the surface of algae and marine snow. Pseudoalteromonas
Versatile metabolism with a wide range of hydrolytic enzymes and
secondary metabolites with antimicrobial influences on biofilm and
biofouling development.

Shewanellaceae (2) Facultative anaerobes. Mostly psychrophilic, some are extreme Psychrobium,
piezophiles from the deep sea. Psychrotolerant species are associated Shewanella
with fish surface microbiota and important in spoilage. Some species
show dissimilatory reduction of metals with applications for microbial
fuel cells.

ORDER Cellvibrionales (5)

Cellvibrionaceae (12) Mostly found in sediments or in association with marine invertebrates. “Ca. Endobugula”,
“Ca. Endobugula sertula” is an endosymbiont of the bryozoan Bugula. Teredinibacter
Teredinibacter turneae is a diazotrophic symbiont of wood-boring
shipworms.

Spongiibacteraceae (5) Originally thought to be specific to marine sponges but appears to be Dasania,
a widely distributed, rare inhabitant of seawater. Motile, aerobic Spongiibacter
heterotrophs. Some contain proteorhodopsin.

ORDER Chromatiales

Chromatiaceae (31) Mostly anaerobic photoautotrotrophs known as the purple sulfur Chromatium,
bacteria, using H2S as electron donor for photosynthesis and depositing Nitrosococcus,
sulfur granules inside the cell. A major group found in many aquatic Phaeobacterium,
habitats, including the upper layers of sulfidic marine sediments and Thiococcus,
mats and as intracellular symbionts of nematodes and gutless worms. Thiospirillum,
Some are photoheterotrophic. Nitrosococcus oceani is a widely “Ca. Thiosymbion”
distributed planktonic chemolithoautotroph oxidizing ammonia.

(Continued)
126 Chapter 4

Table 4.2 (Continued) Representative members of the class Gammaproteobacteria. There are 21 named orders

Representative familya, b Distinctive features Representative

marine generab

Ectothiorhodospiraceae (19) Mostly marine purple sulfur bacteria with spiral cells; sulfur is deposited Aquisalimonas,
outside the cells. Some species are extremely halophilic and Ectothiorhodospira,
alkaliphilic, found in salterns, coastal lagoons, and salt lakes. Halorhodospira

ORDER Methylococcales (3)

Methylococcaceae (17) Includes type I methanotrophs, able to use methane and other C1 Methyloprofundus
compounds as sole carbon and energy sources. Contain methane
monooxygenase in bundles of membrane vesicles in the cytoplasm
and assimilate carbon via the ribulose monophosphate pathway.
Membranes contain sterols. Play a critical role in oxidation of methane
generated in sediments.

ORDER Oceanospirillales (7)

Alcanivoraceae (4) Aerobic, hydrocarbon-degrading bacteria especially common in Alcanivorax,

oil-contaminated environments (see text). Fundibacter,
Kangiella

Hahellaceae (6) Aerobic, rod-shaped cells mainly associated with marine invertebrates Allohahella,
including sponges, corals, molluscs, and other invertebrates as possible Endozoicomonas,
facultative symbionts. Endozoicomonas may be a member of the core Hahella.
microbiomes of corals but its functional role is unclear (p.283)

Oceanospirillaceae (19) Spiral, motile rods Halophilic and widely distributed in seawater, Marinospirillum,
marine organisms, and sediment. Unique capabilities for hydrocarbon Neptunomomas,
degradation (see text). Oceanospirillum

ORDER Thiotrichales (4)

Piscirickettsiaceae (8) Highly diverse family of aerobic heterotrophs with varied Cycloclasticus,
characteristics. Cycloclasticus spp. degrade polycyclic aromatic Methylophaga,
hydrocarbons and are widely distributed in marine sediments, Piscirickettsia
especially contaminated coastal sites and harbors. Methylophaga is a
halophiic methylotroph; some are chemolithoheterotrophs.
Piscirickettsia salmonis is a major pathogen in salmon aquaculture.

Thiotrichaceae (9) Cells are typically large and arranged in filaments arranged in large Beggiatoa,
filaments rods; SOB or chemoheterotrophs prominent in sulfur-rich Leucothrix,
sediments (see text). Thioploca,
Thiomargarita,
Thiothrix
a Figure in parentheses indicate number of validly published taxa in the next lower rank (families in orders, genera in orders) from List of

Prokaryotic Names with Standing in Nomenclature (LPSN, www.bacterio.net). b Names in non-italic font and quotation marks are not validly
published under the Bacteriological Code. Genera may have Candidatus status, but affiliation to higher taxa is not confirmed.

SAR11 and the other alphaproteobacterial clades noted above, with similar metabolic poten-
tial and streamlined genome, although genome analysis—obtained by single-cell genome
amplification—reveals it differs in the use of certain carbon substrates which lessens com-
petition for nutrients. SAR86 can be divided into several ecotypes that show seasonal varia-
tions. The role of proteorhodopsin genes in ATP generation was first identified from SAR86
sequences (Box 3.1).

The Gammaproteobacteria includes many uncultivated

species of sulfide-oxidizing bacteria (SOB)
The chemolithoautotrophic SOB are of major importance in sulfur cycling in sediments and
mats, and as intracellular symbionts of invertebrates. The latter are mostly uncultivated and
unclassified. They include the tubeworm intracellular symbiont “Ca. Endoriftia pachyptila,”
whose 3.4 Mbp genome possesses a wide range of genes permitting a free-living stage out-
side its host, while “Ca. Ruthia magnifica” has a highly reduced genome of 1.16 Mbp and is
 Diversity of Marine Bacteria 127

an obligate clam symbiont. Other unclassified SOB symbionts include “Ca. Thiodiazotropha
spp.” found in lucinid clams. These symbionts are discussed in Chapter 10.

The family Vibrionaceae includes many

important pathogens and symbionts
Because of their importance, members of this family (commonly referred to as the vib-
rios) have been extensively studied and their classification has been frequently revised. The
family is named after the type species Vibrio cholerae, which occurs in association with
zooplankton in estuarine and coastal habitats but colonizes humans, causing epidemics
of cholera. Other important human pathogens include V. parahaemolyticus, V. vulnificus,
and V. alginolyticus, associated with seafood-borne or wound infections, as discussed in
Chapter 12. A wide range of Vibrio spp. also cause economically important diseases of fish
(e.g. V. ­anguillarum, Aliivibrio salmonicida, Photobacterium piscicida), molluscs (e.g. V.
splendidus, V. tubiashii), and crustaceans (V. harveyi, V owensii, V. nigropulchritudo). The
species V. coralliilyticus, V. shilonii, and V. splendidus are also opportunistic pathogens of
corals. These diseases are explored in detail in Chapter 11.

The development of genomic methods such as MLST (p.45) led to a major overhaul of the
phylogeny and classification of the Vibrionaceae and the advent of rapid genome sequencing
led to further insights into the remarkable diversity within this family. A number of fac-
tors contribute to this diversity. Genome analysis of the vibrios shows that substantial gene
acquisition and loss has occurred during their evolution, leading to enhancement of their
physiological and ecological capabilities, facilitated by the presence of two chromosomes.
The overall mean genome size is ~ 40% bigger than that of comparable marine bacteria. All
of the genes recognized in complete genomes can be grouped into ~ 22000 gene families
which constitute the pangenome of the Vibrionaceae, of which ~1630 probably constitute the
core genome—genes that are present in all strains of all species examined. This represents
huge genetic diversity; by comparison, the human genome contains about 25000 genes. Even
within the same species as currently defined, the genomes of different strains can differ
by as much as 20–30%. Thus, the pangenome represents the full genetic potential of the
community of related organisms, but individual strains might have only a small fraction of
these genes that equip it to occupy a particular niche. For example, isolates of V. splendidus
obtained from a single coastal site can differ by as much as 1 Mbp in genome size and may
contain at least a thousand distinct genotypes, each occurring at extremely low environmen-
tal concentrations.

Some members of the Vibrionales exhibit the distinctive feature of bioluminescence. These
occur as free-living forms in seawater, on organic debris, as commensals in the gut of many
marine animals, and as symbionts of the light organs of some squid and fish (Chapter 10).
Most of the marine bioluminescent bacteria that can be isolated and cultured are members
of the family Vibrionaceae—the commonest types being Photobacterium phosphoreum, P.
leiognathi, Aliivibrio fischeri, Vibrio campbellii, and V. harveyi. An image of a massive
ocean bioluminescent bloom—the “milky sea” phenomenon, probably caused by V. harveyi
in association with Phaeocystis algae—is shown in Figure 4.6. The bioluminescent symbi-
onts of flashlight and angler fish (“Ca. Enterovibrio luxaltus” and “Ca. E. escalota”) have not
been cultured; they have reduced genomes compared with other members of the genus and
are probably obligate symbionts (Box 10.4). The reaction mechanisms (Figure 3.17) and pos-
sible functions of bacterial bioluminescence are discussed in Chapter 3.

Members of the order Oceanospirillales break

down complex organic compounds
This group consists of uniquely marine types, as reflected in the names given to the vari-
ous genera shown in Table 4.2. As the name suggests, bacteria in this order were originally
characterized by their spiral cell shape, but they are very diverse in their physiology and
ecology and the spirilla designation is not a reliable distinguishing feature, as several gen-
era are rod-shaped rather than helical. New members of this group have been discovered in
symbiotic association with Osedax worms and in sediments in the vicinity of whale falls (see
128 Chapter 4

Figure 4.6 Modified rendition of a

satellite image of a bioluminescent
bloom “milky sea” observed in the
northwest Indian Ocean in January
1995 and visualized by retrospective
analysis of satellite data. The bloom
was visible on three consecutive
nights, covering an area of 15400
km2. Credits: Main image, Gary Vora,
US Naval Research Laboratory. Inset
reprinted from Miller et al. (2005).
with permission; copyright 2005,
National Academy of Sciences, USA.

Chapter 10). Members of the type family Oceanospirillaceae, including Oceanospirillum

spp. are aerobic and motile and play a major role in the heterotrophic cycling of nutrients in
seawater. Some members in this group are important in the sulfur cycle, especially through
degradation of dimethylpropionosulfonate (DMSP, p.90). Some, such as Alcanivorax spp.,
use alkanes exclusively as carbon and energy sources and are active in biodegradation of
hydrocarbons at natural seeps and oil spills (Box 13.2). Genome analysis reveals multiple
sequences coding for enzymes degrading different alkanes, as well as surfactants that emul-
sify the oil to aid its degradation. There is wide variation in physiological properties such as
optimum growth temperature, halophilicity, and utilization of substrates. The taxonomy of
these genera is in a state of considerable flux, and molecular methods indicate that, although
related, they represent a number of deeper phylogenetic groups.

The family Thiotrichaceae includes some important SOB

Beggiatoa is very common in the top few millimeters of sulfur-rich marine sediments. They
frequently show chemotaxis to seek out the desired gradient of oxygen and sulfur com-
pounds and are also very prominent at hydrothermal vents and cold seeps. Beggiatoa and
other filamentous forms commonly show gliding motility and become intertwined to form
dense microbial mats, often with a complex community structure containing sulfate-reducers
and phototrophs. Although Beggiatoa obtains energy from the oxidation of inorganic sulfur
compounds, it does not possess the enzymes needed for autotrophic fixation of carbon diox-
ide and therefore uses a wide range of organic compounds as a carbon source. Thioploca,
Thiothrix, and Thiomargarita are filamentous sulfur-oxidizing chemolithotrophs important
in the oxidation of sulfide in anaerobic sediments.

Thioploca spp. are multicellular filamentous bacteria that occur in bundles surrounded by a
common sheath (Figure 4.7) and contain granules of elemental sulfur. Thioploca is one of the
largest bacteria known, with cell diameters from 15 to 40 μm and filaments many centimeters
long, containing thousands of cells. Several species, including T. chileae, T. araucae, and T.
Figure 4.7 (a) Filaments of
marina have been described. Huge mats of Thioploca spp. were discovered along the Pacific
Thioploca spp. extending from the
coast of South America, where upwelling creates areas of nitrate-rich water, with bottom
sediment surface into the anoxic
water column at 50–100 m depth at
the shelf off the coast of Chile. The
frame is 15 mm wide. (b) A bundle
of T. araucae filaments extending out
of their sheath. The appearance of
a braid is seen where the filaments
cross over, hence the name Thioploca
(“sulfur braid”). Each filament is
~35 mm wide. Credit: reprinted
from Jørgensen and Gallardo (1999)
with permission from Oxford
University Press.
 Diversity of Marine Bacteria 129

waters becoming anoxic. Blooms of Thioploca can be very dense, up to 1 kg m−2 (wet weight).
Anoxic reduction of hydrogen sulfide is coupled to the reduction of nitrate. Each cell contains
a very thin layer of cytoplasm around the periphery and a liquid vacuole that constitutes 80%
of the cell volume. The vacuole stores very high concentrations of nitrate, which is used as
an electron acceptor for sulfide oxidation. The bacteria can grow autotrophically or mixo-
trophically using organic molecules as a carbon source. In order to access the nitrate in the
overlying water column, the filaments stretch up into the overlying seawater, from which they
take up nitrate, and then glide down 5–15 cm deep into the sediment through their sheaths to
oxidize sulfide formed by intensive sulfate reduction.

Thiomargarita namibiensis holds the record as the largest known bacterium. The spherical
cells are normally 100–300 μm wide, but some reach diameters of 750 μm (Figure 1.4C).
They occur in large numbers in coastal sediments off Namibia and occur in filaments with a
common mucous sheath. Microscopic granules of sulfur reflect incident light, and the name
derives from their resemblance to a string of pearls. The hydrographical conditions off this
coast bring large quantities of nutrients to the surface and massive phytoplankton growth
results in settlement of organic material to the seabed, where it is degraded by bacteria form-
ing large amounts of hydrogen sulfide. Thiomargarita oxidizes sulfide using nitrate and,
like Thioploca, the interior of the cell is filled with a large vacuole. The nitrate stored in the
vacuole and the sulfur stored in the peripheral cytoplasm act as nutrient reserves that allow
these bacteria to grow for several months in the absence of external nutrients.

The proposed phylum Desulfobacterota contains

anaerobic sulfate- or sulfur-reducing bacteria (SRB)
The SRB play a major role in the sulfur cycle and are abundant in anoxic marine sedi-
ments where sulfate from seawater can diffuse several meters down. Eleven families and
63 genera (mostly sharing the prefix Desulfo-) have been validly described as members of
the Deltaproteobacteria, but there has been substantial reclassification under the GTDB sys-
tem. Together with methanogenic archaea, SRB are responsible for the mineralization of
organic material in coastal and shelf sediments. As noted in Chapter 3, the activity of SRB
and archaea changes with depth, due to competition for nutrients and penetration of sulfate,
but the two groups can grow syntrophically when sulfate is depleted. SRB acquire energy
for metabolism and growth by utilizing organic compounds or hydrogen as electron donors
coupled to dissimilative sulfate reduction (Figure 3.12). SRB produce H2S, resulting in the
characteristic stench and blackening (due to deposits of FeS) of decomposition in anoxic
mud, sediments, and decaying seaweed. H2S is highly toxic and adversely affects many forms
of marine life, but it can be used by a wide range of chemotrophic and phototrophic bacteria,
as described in earlier sections, completing the cycling of sulfur. Numerous types of SRB
have been isolated from marine sediments and these are classified on the basis of differing
morphological and physiological properties, as well as 16S rRNA typing. Some genera, such
as Desulfobacter and Desulfonema oxidize acetate and other fatty acids. Another group,
which includes Desulfovibrio and Desulfomonas, can utilize lactate, pyruvate, and certain
fatty acids, but not acetate. Sequences affiliated with this group can also be recovered from
the water column, including clade SAR324 found in waters >500 m deep. These are thought
to be particle-associated and genome analysis indicates metabolic flexibility, including oxi-
dation of carbon monoxide and methylotrophy. It should be noted that some members of the
Firmicutes and Archaea also carry out sulfate reduction.

The recently discovered cable bacteria, which couple oxygen reduction to sulfide oxidation
via long-distance electron transport in sediments, have been identified as members of the
family Desulfobulbaceae within this group (Box 4.2).

The proposed phylum Epsilonbactereota contains major

contributors to productivity at hydrothermal vents
As noted in Box 4.1, new multi-gene analysis does not support the phylogenetic status of
the Epsilonproteobacteria class as members of the Proteobacteria phylum, and they have
been reassigned to the new proposed phylum Epsilonbactereota. There are a relatively
130 Chapter 4

BOX 4.2 RESEARCH FOCUS

Cable bacteria unite to conduct electricity

A network of electrogenic bacterial filaments. Geoscientists at is unknown. The bacteria show positive chemotaxis to increasing
Aarhus University, Denmark noted unusual changes in the distribu- O2 concentration in order to locate the oxic-anoxic interface.
tion of pH in marine sediments, which raised the notion that electric
currents were running through the sea floor (Nielsen et al., 2010). Proof of electrical conduction in cable bacteria. Evidence that
The nature of this process was a mystery until the team, with col- LDET occurs in the sediments came from high-resolution profil-
laborators from scientists at the University of Southern California, ing of pore water chemistry and mapping of electron sources using
showed that gentle washing of the top 20 mm of sediment from sul- different types of micro-electrodes (Nielsen et al., 2010; Risgaard-
fidic sediment revealed a tangled mass of filamentous multicellular Petersen et al., 2014). Some ingenious micro-manipulation experi-
bacteria responsible for long-distance electron transport (LDET, ments by Pfeffer et al. (2012) provided further proof that the
Pfeffer et al., 2012). Subsequently, Schauer et al. (2014) exposed bacterial filaments act as solid conductors for LDET. Metabolic
sulfidic sediment to oxygen and witnessed the rapid growth and activity dropped rapidly when the filament network was cut by
development of a network of filaments, reaching ~2 km per cm 2 passing a thin tungsten wire through the sediment below the oxic-
after 21 days. Examination of the filaments showed that they can anoxic interface. When membrane filters were placed into sedi-
contain over 104 cells and be up to 70 mm long and 0.4–1.7 µm diam- ments, electrical currents only occurred when pore sizes were large
eter, orienting themselves vertically in the upper sediment. Electron enough (>0.8 µm) for the bacteria to grow down vertically through
microscopic examination showed that the surface has 15–17 ridges the membranes to reach the underlying sediment, thus ruling out
400–700 nm wide, with a filled 70–100 nm wide channel between the possibility that electron transfer was due to dissolved or col-
the cytoplasmic and outer membrane (Pfeffer et al., 2012). There loidal compounds. Also, Pfeffer et al. showed that when a layer of
is a gap of 200 nm between adjacent cells in the filament, which glass microspheres was inserted to replace a 5 mm thick layer of the
is bridged by the ridge filling, creating a continuous channel along sediment, LDET only occurred when filaments were able to bridge
the filament. This is assumed to form the conductive “wires” trans- the non-conductive gap. In later experiments, Bjerg et al. (2018)
porting electrons. The whole structure is wrapped in a collective placed filaments in microscope chambers with sulfide and O2 at
outer membrane that acts as insulation. Further detailed analysis opposite ends. They used a laser microdissection beam to cut off
of the architecture of the cell envelope using a range of advanced the electron transport within individual filaments and measured the
imaging techniques was recently described by Cornelissen et al. redox states of cytochromes using resonance Raman microscopy.
(2018). The cell envelope is built by a parallel concatenation of A gradient of cytochrome redox potential was detected along the
ridge compartments, each of which contains a ~50 nm diameter filaments, which broke down if O2 was removed or the filaments
fiber in the periplasmic space (Figure 4.8A). These fibers continue were cut. The filaments showed a loss of electric potential (voltage)
across the junctions between cells in the filament, which have a of 12–14 mV per mm of filament. Bjerg et al. calculated that the
distinctive cartwheel structure formed by invaginations of the outer maximum distance for effective electron transport would be about
membrane. As well as acting as conductive wires, the periplasmic 300 mm, beyond which filaments would have to use their gliding
fibers provide tensile strength to the filaments, which can withstand motility to remain in contact with the O2- or S-rich zones.
substantial stress before breaking.
Bacterial cells cooperate to perform different half reactions.
Motility. Bjerg et al. (2016) used time-lapse microscopy to show The most remarkable feature of LDET is that different cells in the
that cable bacteria are motile, moving by gliding at a mean (highly multicellular filament carry out the oxidative and reductive half-
variable) speed of 0.5 µm s−1. The bacteria frequently moved in reactions separately, as shown in Figure 4.8B (Meysman, 2018).
loops and sometimes twisted around themselves, indicating a heli- Separation of the half reactions over long distances occurs because
cal rotation of the filament, but the mechanism by which this occurs of the electrical connection—like an electric cable—coupling

Figure 4.8 (a) Ultrastructure of cable

bacterium. Upper panel shows a com-
posite TEM image of the multicellular
filament. Lower panel is a model of
the filament, showing ridges contain-
ing periplasmic conductive fibers and
internal structure at cell junctions.
(b) The process of sulfur oxidation in
cable bacteria, showing separation of
the half reactions by LDET. (Credits: A.
Rob Cornellisen, Hasselt University. B.
Reprinted from Meysman (2018) with
permission from Elsevier.
 Diversity of Marine Bacteria 131

BOX 4.2 RESEARCH FOCUS

the supply of electrons from sulfide oxidation by the anodic cells Sea (Netherlands). They were also found, albeit at somewhat lower
to electron removal from oxygen reduction by the cathodic cells. levels, in the surrounding bioperturbated sandy sediment. Despite
Unlike typical SOB, which can only oxidize sulfide in the oxic- frequent burial by the dynamic nature of these sediments, the bac-
anoxic/O2-H2S interface or commute to collect oxidants from teria continue to occupy the uppermost sediment layers, due to
the surface sediment or water column, cable bacteria can “mine” motility enabling them to keep the optimal connections between
sulfide from deeper layers. Cable bacteria can also use nitrate as their separated electron donors and acceptors, as previously shown
an electron acceptor, meaning that oxidation of sulfide can occur in the laboratory by Bjerg et al. (2016). Using microsensors, Malkin
by LDET to anoxic nitrate-rich water overlying the sediment et al. showed that the cable bacteria are responsible for the genera-
(Marzocchi et al., 2014). tion of acid as a result of sulfate reduction. This has marked effects
on the pore water chemistry by increasing the concentrations of dis-
We have grown used to the dogma that all living cells—from uni- solved calcium, manganese and iron. Burdorf et al. (2017) provide
cellular microbes to those within complex multicellular organ- evidence that cable bacteria are globally distributed in different
isms—are responsible for individually securing their own energy coastal habitats and are an important component of biogeochemi-
supply, which occurs by processing of electron donors and accep- cal cycles. Recently, Kessler et al. (2018) showed that they have
tors over very short distances (intracellularly, or within the immedi- a marked effect on dissimilatory reduction of nitrate to ammo-
ate vicinity of the cell). To explain LDET, Meysman (2018) invokes nia (DNRA), which affects loss of bioavailable nitrogen from the
a “thought experiment” imagining two twin brothers separated by system.
20 km—one of whom has access only to carbohydrates as food
and the other who only has access to O2 for respiration—who live Protecting marine organisms from sulfide poisoning. The high
by LDET through a connecting wire. The surprising discovery of levels of highly toxic H2S produced in marine sediments by SRB
physical separation of reactions between different, distant cells has are a potential threat to marine life, but as long as O2 is present it
created a paradigm shift in our understanding of biological electron is promptly oxidized by SOB and does not enter the water column.
transport. However, coastal waters often become hypoxic due to microbial
respiration of decaying organic matter, eutrophication, stratifica-
Phylogeny of cable bacteria. To date, all attempts to isolate and tion, and increased temperature. Nielsen (2016) notes that, during
cultivate cable bacteria have been unsuccessful. All specimens O2 depletion, marine animals survive longer than predicted. Seitaj
so far described have a similar morphology and 16S rRNA gene et al. (2015) conducted a four-year study of a seasonally hypoxic
sequencing indicates that they belong to family Desulfobulaceae coastal marine lake with restricted water exchange with the North
(Pfeffer et al., 2012). Using capillary glass hooks and a microma- Sea. They observed that cable bacteria dominate the sediment in the
nipulator Trojan et al. (2016) isolated individual cable filaments winter, whereas mats of Beggiatoaceae colonize it during the sum-
from marine, saltmarsh, and freshwater sediments in Denmark, mer. The electrogenic activity of the cable bacteria creates a store of
Japan, and the United States. Single-cell genomes were amplified iron oxides before summer hypoxia sets in, and these re-precipitate
and gene sequences encoding 16S rRNA and Dsr proteins (involved as a surface layer of reactive iron oxide that captures free H2S in the
in oxidation of stored sulfur) were identified in the assemblies. By surface sediment, preventing its escape to the water column (eux-
amplifying the genomes of individually picked filaments, Trojan inia). Because of the wide distribution of cable bacteria, Seitaj et
et al. were able to directly correlate sequence information with al. suggest that they may act as a “firewall” that explains why mass
phenotypic properties of the cable bacteria and their environ- mortality in coastal waters due to euxinia is infrequently observed.
mental origin, enabling them to identify the bacteria from all 16
samples as belonging to a monophyletic sister lineage of the genus The activity of cable bacteria may partly explain why seagrasses
Desulfobulus. Two major branches within this clade were identi- are able to grow in anoxic sediments that have phytotoxic levels of
fied and Trojan et al. proposed these as the novel genera “Ca. sulfide. Martin et al. (2019) showed that young seagrass root tips
Electrothrix” (containing four novel candidate species from the leak O2 into the surrounding environment. As well as oxidizing sul-
marine and saltmarsh specimens) and “Ca. Electronema” (contain- fide abiotically, this phenomenon may facilitate sulfide oxidation by
ing two novel candidate species from the freshwater specimens). cable bacteria, as these are found in higher abundance adjacent to
Although genomic data is currently limited, it appears that the sys- the roots leaking O2.
tem of LDET evolved only once, and the responsible genes do not
seem to have spread to other microbial groups. Future research directions. Considerable progress has been made
in understanding this novel mechanism, but many intriguing ques-
Distribution and contributions to ecosystem function. Gene tions remain. How are the energy and products of metabolism
surveys show that cable bacteria occupy organic-rich coastal sedi- shared between cells? How are cell division and construction of the
ments with steep sulfide gradients, but do not feature strongly in filaments controlled, and how do the filaments orient themselves
sandy sediments with mechanical disruption by bioperturbating in the sediment? What is the composition and electrical properties
infauna. However, they are found in the sediment beneath intertidal of the periplasmic fibers? What is the evolutionary history of cable
beds of bivalve molluscs, which trap organic material and form bacteria and what is their role in the ecology of different sediments?
a reef that lifts structures above the surrounding sediment level. How do they interact with other members of the microbial commu-
Malkin et al. (2017) found high densities of cable bacteria in sedi- nity? This newly discovered and fascinating lifestyle of bacteria is
ments accumulating within mussel and oyster reefs in the Wadden ripe for extensive investigation.
132 Chapter 4

small number of cultivated representatives in this group, but environmental gene sequenc-
ing reveals much greater diversity. They carry out anaerobic respiration using organic or
inorganic electron donors and acceptors to support chemoorganotrophic or chemolitho-
autotrophic growth. In marine habitats, Epsilonbacteraeota are particularly associated
with hydrothermal vents and may be the dominant bacteria contribute the major fraction
of the primary production via autotrophy in the waters between 25–60°C surrounding
the geothermal emissions. They appear to play a key role in the initial colonization of
newly formed vents. Here, the most common metabolism is oxidation of reduced sulfur
compounds, hydrogen, or formate, using either oxygen or nitrate as a terminal electron
acceptor and fixing CO2 via the reverse TCA cycle. Examples of cultured genera include
Sulfurimonas, Nautilia, and Caminibacter. As well as free-living forms, members of the
Epsilonbacteraeota are involved in symbiotic associations with invertebrates at hydrother-
mal vents, notably Rimicaris shrimps, the Pompeii worm Alvinella, and the scaly-foot snail
Chrysomallon, discussed in Chapter 10.

Myxobacteria have a complex life cycle

The group of related orders and families known as the myxobacteria are characterized
by cells that have some of the largest bacterial genomes (typically ~9 Mb or more) which
accounts for their complex life cycle with gliding motility and social behavior leading to the
formation of reproductive fruiting bodies with production of myxospores. They are primar-
ily associated with terrestrial soils, but they can be recovered from seawater. Many isolates
are thought to have resulted from spores washed into the sea from coastal runoff, but some
halotolerant and halophilic strains have been isolated and cultured from marine muds and
sands and the surface of algae, seagrass, and invertebrates. These marine types form a dis-
tinct phylogenetic group with representative genera including Plesiocystis, Nannocystis, and
Haliangium. The social behavior characteristic of soil myxobacteria probably does not occur
in marine species. These were classified with the Deltaproteobacteria until recently but have
been reassigned to the new GTDB phylum Myxococcota.

The Bdellovibrionales contains predatory bacteria

This order contains genera (e.g. Bdellovibrio, Halobacteriovorax, and Vampirovibrio) of
small spiral cells with the unusual property of preying on other Gram-negative bacteria. The
most studied type is Bdellovibrio bacteriovorus, whose life cycle is shown in Figure 4.9.
The cells swim at high speed and use a chemosensory system to detect high concentrations

Figure 4.9 Composite of false-

colored SEM images showing the
life cycle of Bdellovibrio bacterivorus
(yellow-colored cell) infecting a
Gram-negative host bacterium
(blue-colored cell). Infection begins
by attachment of a flagellated
Bdellovibrio cell (1, 2). During active
penetration and entry to the host
cell, the flagellum is shed (3–5).
Bdellovibrio replicates in the peri-
plasmic space, producing extracel-
lular enzymes to liberate nutrients
for growth and elongating into
a large filamentous cell (6–7).
When nutrients are exhausted, the
Bdellovibrio differentiates into several
motile cells (8–9), which then swim
to seek out new hosts (10). Credit:
Snejzan Rendulic, Juergen Berger,
and Stephan Schuster; Max Planck
Institute for Developmental Biology.
 Diversity of Marine Bacteria 133

of prey. Different strains are highly specific in targeting particular hosts. It attaches to its
COULD BALOs BE
prey and retractable fibers bring it into close contact with the bacterial cell. A cocktail of
enzymes degrades proteins, lipids, and carbohydrates to create an opening in the cell wall. ? USEFUL IN DISEASE
CONTROL?
Bdellovibrio then enters the periplasmic space between the outer and inner membranes,
where it replicates, producing extracellular enzymes to liberate nutrients for growth. The In the quest to overcome the threat
cytoplasm of the prey is consumed and the Bdellovibrio cell elongates into a large filamen- posed by antibiotic resistance
tous cell. When all the nutrients are exhausted, the filament differentiates into about 15 and the risks to human health,
motile cells, which then seek out new hosts after digestion of the host cell. Bdellovibrio- investigations are underway to
and-like organisms (BALOs) are very widespread in aquatic environments and may exert investigate whether the preda-
tory BALOs could be used as a live
top-down control of populations of other bacteria similar to phages, although their full
therapy. It has been shown to be
ecological role is unknown. effective as a potential alternative
to antibiotics in local and sys-
temic infections in experimental
Members of the Zetaproteobacteria are animals. It has also recently been
suggested that introducing preda-
microaerophilic iron-oxidizers tors into the human gut could be
The original isolate of bacteria in this class of Proteobacteria designated in 2010 was isolated a possible method of restoring a
from a deep-sea hydrothermal vent field off Hawaii and named Mariprofundus ferrooxydans. balanced, diverse microbiome in
Subsequently, isolates assigned to this class have been found at a variety of Fe-rich brack- conditions such as Crohn’s disease
(Mosca et al., 2016), reminiscent
ish and marine sources, including in deep-sea microbial mats and brine seawater interfaces
of the successful “rewilding” of
(DHABs, p.22). Experiments in certain coastal sediments have shown that they can colonize habitats to promote biodiversity
and corrode mild steel. All isolates so far described are microaerophilic Fe(II)-oxidizing in macroecology. Further develop-
chemolithoautotrophs found at opposing gradients of Fe(II) and O2; requiring neutral pH and ments depend on understanding
low O2 concentrations. They are microaerophilic because they require just enough O2 for use how Bdellovibrio interacts with
as an electron acceptor but not so much as to cause abiotic iron oxidation. Their growth on host phagocytes and the immune
surfaces produces characteristic filamentous twisted stalks composed of Fe oxyhydroxide. system (Raghunathan et al., 2019).
Several other species have now been proposed, and related OTUs are abundant in 16S rRNA BALOs are also under investigation
gene datasets from deep-sea samples, indicating that this group may have wider significance for use in prevention of disease in
in biogeochemical cycling. aquaculture (also, see the discus-
sion of phage therapy in Box 14.1).

Members of the Cyanobacteria carry

out oxygenic photosynthesis
The Cyanobacteria is a large and diverse group, and members are characterized by their
ability to carry out photosynthesis in which oxygen is evolved, although some anoxygenic
members have been described. Cyanobacteria contain chlorophyll a, together with acces-
sory photosynthetic pigments called phycobilins. This group was formerly known as the
blue-green algae because of the presence of the blue pigment phycocyanin together with the
green chlorophyll. Designation as algae is inappropriate although the term persists in com-
mon usage and the organisms are still treated as the Cyanophyta division of the algae under
the Botanical Code, even though the genetic and structural features of the Cyanobacteria
mean that they clearly form one of the major phyla of the domain Bacteria. Many genera
contain phycoerythrin pigments, which give the cells a red-orange color, rather than blue-
green color.

Fossil evidence, in the form of morphological structures and distinctive biomarkers

(hopanoids) typical of the group indicates that organisms resembling modern representatives
of cyanobacteria may have evolved about 2.5 BYA. The evolution of oxygenic photosynthesis
in ancestral cyanobacteria was responsible for transformation of the Earth’s early atmosphere
and the subsequent evolution of life on Earth, as discussed in Chapter 1.

Today, cyanobacteria play a major role in global biogeochemical cycles. They occupy very
diverse habitats in terrestrial and aquatic environments, including extreme temperatures and
hypersaline conditions. In the marine environment, cyanobacteria are major contributors to
primary production as members of the phytoplankton, and occupy other important habitats
including sea ice, shallow sediments, and microbial mats on the surface of inanimate objects,
algae, or animal tissue. The Cyanobacteria show a range of metabolic properties besides oxy-
genic photoautotrophy—some can grow anaerobically using H2, H2S. or organic compounds
as electron donors and some may be mixotrophic.
134 Chapter 4

A genome-based classification of the

Cyanobacteria is under development
Until the use of 16S rRNA gene analysis established them as group within the Bacteria, “blue-
green algae” were classified by botanists into about 150 genera and 1000 species based on
morphological features. The Cyanobacteria are morphologically very diverse, ranging from
small undifferentiated rods to large branching filaments showing cellular differentiation.
Unicellular cyanobacteria divide by binary fission, while some filamentous forms multiply
by fragmentation or release of chains of cells. Many types are surrounded by mucilaginous
sheaths that bind cells together. The chlorophyll a is contained within lamellae called phy-
cobilosomes, which are often complex and multilayered. Cyanobacteria show remarkable
ability to adapt the arrangement of their photosynthetic membranes and the proportion of
phycobilin proteins to maximize their ability to utilize light of different wavelengths.

Subsequent phylogenetic analysis shows that groupings based on morphology are very unre-
liable, and many current genera and higher taxa are polyphyletic. As with other groups of
bacteria, the availability of large amounts of genomic data is propelling the development
of a genome-based classification that will better reflect the evolutionary history, ecological,
and physiological diversity of different cyanobacteria. Genomic analysis reveals a number
of important factors that affect the diversity of cyanobacteria and their ability to occupy
specific ecological niches. As with other bacteria, the types of transport systems for limiting
nutrients such as P, N, and Fe are especially important. The components and mechanisms
of the photosynthetic machinery are also highly variable. The structure and arrangement of
photosynthetic membrane systems, the structure of carboxysomes, and the spectral tuning of
pigments to different wavelengths of light have a major effect on the distribution and activity
of different types of cyanobacteria. These factors are leading to major revision of existing
schemes for classification and taxonomy of the group. Therefore, the remainder of this sec-
tion will focus on a few selected genera without attempting to reflect their classification into
higher taxa.

Prochlorococcus is the most abundant

photosynthetic organism
Prochlorococcus is a very small cyanobacterium about 0.6 μm diameter (Figure 4.10), which
is the dominant phototroph in oligotrophic ocean waters. It inhabits large parts of the oceans
at depths of 25–200 m at a density between 105 and 106 mL−1. The global population of 1027
cells means it is the most abundant photosynthetic organism on Earth. It fixes ~ 4Gt (Pg) of

Figure 4.10 TEM of ultrathin sec-

tion of Prochlorococcus marinus MED4
(artificially colored). Credit: Chisholm
Lab, MIT.
 Diversity of Marine Bacteria 135

carbon per year and contributes ~20% of O2 to the atmosphere. Despite this, it was not dis-
covered until 1985, using methods described on p.85. Prochlorococcus is most abundant in
the subtropical gyres, occupying the region from 40°S to 40°N. It contains modified forms of
chlorophyll (divinyl chlorophylls a and b), but lacks phycobilins.

Hundreds of strains have been isolated, cultured, and cataloged as the single species
Prochlorococcus marinus, but this designation hides exceptional diversity. When samples
of Prochlorococcus are isolated from different depths in the water column, it is clear that
there are distinctly different ecotypes adapted for growth in different ecological niches in the
photic zone. Beyond the upper mixed layer of the ocean, nutrient availability increases with
depth, while light intensity and temperature decrease. Prochlorococcus ecotypes are adapted
to life across all parts of the photic zone in the oligotrophic oceans with very low nutrient
levels. Like many of the dominant heterotrophic bacteria discussed earlier, they have stream-
lined genomes and very small cells enabling efficient nutrient uptake and photosynthesis.
Requirements for phosphorus, which is extremely limiting in the open ocean, are reduced
by the lower DNA content and modified cell membranes that contain sulfolipids rather than
phospholipids. Strains isolated from different depths have many differences, including the
ratios of divinyl chlorophylls a2 to b2 and optimal irradiances for photosynthesis. Divinyl
chlorophylls harvest longer wavelengths of blue light which penetrates deeper waters, allow-
ing some ecotypes of Prochlorococcus to grow at depths where the amount of light is less
than 1% of that at the surface.

Methods for culturing many strains of Prochlorococcus were developed allowing studies of
their biochemical and physiological properties, showing that strains could be grouped into
two broad categories—those adapted for high light (HL) and those adapted for low light
(LL). Molecular phylogenetic analysis shows that this distinction appears to reflect an early
evolutionary split into the HL group containing six clades and the LL group containing seven
clades. These show differences in their acquisition of iron and other nutrients, and in their
ability to adapt to different light and redox conditions.

Complete genome sequences have now been obtained for multiple strains of Prochlorococcus
obtained from different depths varying in nutrient composition and light intensity/spectral
quality. The genomes of HL-adapted strains are very much smaller (mean ~1.7 Mb) than
those of other cyanobacteria. Genomes of LL-adapted strains are generally larger, but more
variable in size, and one clade (LL IV) has genomes of ~2.7 Mb. For example, in the first
study of Prochlorococcus genomes, the HL-adapted strain MED4 was found to have the
smallest minimal genome (1716 genes) of any phototroph. Genes for systems such as chaper-
ones, transport systems, and many other systems for signal transduction and environmental
stress responses are absent. A small genome is less expensive for the cell to maintain but
reduces the organism’s ability to respond to changes in its environment. The LL-adapted
strain SS120 found deep in the photic zone has 2275 genes with many genes conferring flex-
ibility to utilize higher and more variable nutrient levels. These two representative strains
share 1352 genes. As more genome sequences have been added, it has become apparent that
the core genome containing the essential genes for cell structure, growth, and replication
shared by all isolates is ~1000 genes. The remaining “flexible” genes are found in only a few
genomes and account for the adaptation of each strain to its local environment. These flexible
genes are clustered in highly variable genomic islands, indicating acquisition by horizontal
gene transfer (HGT). The role of phages in this transfer is discussed in Chapter 7. The size of
the pangenome—the total number of genes carried by all strains of P. marinus—has grown
to over 85000. This represent a huge amount of diversity that clearly explains the ecological
success of subpopulations of these organisms and their ability to vary in abundance accord-
ing to seasonal and geographical light and nutrient gradients, although the function of three-
quarters of the genes is still unknown.

Since they are responsible for so much carbon fixation in the oligotrophic ocean,
Prochlorococcus has a major role in fueling the growth of heterotrophic bacteria with which
they co-exist. Assessment of abundance of Prochlorococcus and SAR11 in the wild (based
on metagenomic sampling) indicates that they co-exist at a median (highly variable) ratio of
about three Prochlorococcus for every four SAR11. In vitro experiments show that SAR11
grows faster in co-culture with Prochlorococcus, but the growth of the latter is unaffected.
136 Chapter 4

Some strains of Prochlorococcus meet SAR11’s requirement for carbon in the form of gly-
THE GLOBAL GENE
i FEDERATION OF
PROCHLOROCOCCUS
cine, but the two organisms probably compete for uptake of DMSP. In turn, it seems that close
association with certain heterotrophs provides benefits in return. Prochlorococcus does not
possess genes for catalase, which is needed to remove reactive oxygen species (ROS, p.108).
The pangenome of P. marinus It seems that this function is met by closely associated heterotrophs, although this has not yet
revealed by analysis of genomes been demonstrated experimentally. The significance of these cooperative versus competitive
from cultured cells or metage- interactions is complicated by the discovery that Prochlorococcus may be mixotrophic. It
nomic assemblies contains about has been known for some time that some strains of Prochlorococcus can grow heterotrophi-
four times the number of human cally in culture and that natural populations can assimilate amino acids and DMSP to meet
genes. That’s impressive enough
their nitrogen and sulfur requirements. Recently, Prochlorococcus has also been shown to
… but examining the cell by cell
diversity of wild populations adds
assimilate glucose and other sugars, which may be sufficient to provide energy and carbon for
an even greater dimension to this growth and cell maintenance. Also, Prochlorococcus has recently been found to release large
genetic diversity. By amplifying the numbers of outer membrane vesicles which probably have important functions in food webs
genomes of single cells (see p.56 and the biological carbon pump. This, together with their role as prey for grazing protists, is
for methods) in a single sample discussed in Chapter 8.
of seawater, Kashtan et al. (2014)
showed that there are hundreds of
subpopulations. Each of these has
a distinctive “genetic backbone” of
Synechococcus spp. dominate the upper photic zone
core gene alleles linked to a set of Synechococcus spp. have slightly larger cells (0.6–1.6 µm) than most Prochlorococcus strains
flexible genes. There seems to be a and have a wider geographic distribution. Marine species tend to predominate in the well-lit
fine-scale co-variation between the top mixed layer (~25 m) of the water column and are more abundant (up to 106 cells mL−1) in
core and flexible gene content, and nutrient-rich coastal waters than in the ocean gyres (~103 cells mL−1). Different species also
Kashtan et al. suggest that these
occur in freshwater and are particularly abundant in the coastal plumes of major rivers. They
subpopulations diverged millions
of years ago resulting in niche
contain chlorophyll a and ancillary pigments known as phycobiliproteins. The nature of these,
partitioning that is stable over together with habitat, morphology, and some genetic information have been used to desig-
long periods. Nucleotide varia- nate ~50 species under the Botanical Code. Genome analysis reveals close similarities of the
tions in genes associated with cell marine strains to Prochlorococcus, while freshwater strains are more distantly related. Such
surface appear to be particularly analysis also reveals that the genus as currently defined appears to be polyphyletic and a new
important. This study was done genus Parasynechococcus has been proposed. Key differences between the genera include
in a particular region (the BATS the nature of the carboxysome proteins and regulatory systems. Like Prochlorococcus, a core
in the Sargasso Sea) and it will be and flexible genome occurs and HGT again plays a large part in genome evolution.
interesting to see if different geno-
types are spread globally via ocean
About one-third of open ocean Synechococcus isolates exhibit a very unusual motility, mov-
circulation.
ing through seawater at speeds of 5–25 µm s−1 (up to ten body lengths) while rotating at ~1 Hz
although no flagella or changes in shape have been observed. Strains of Synechococcus only
appear to show swimming motility if they produce a glycoprotein S-layer on their surface
(p.69). One possibility is that proton motive force powers the movement of cargo-carrying
motors along a continuous looped helical track anchored to the peptidoglycan layer of the
cell wall. This is envisaged as driving rotation of the track which creates helical waves along
the S-layer, causing the cell to rotate and move through the water. The role of swimming in
ecology of these strains is unknown.

Another interesting feature of Synechococcus is its use as for investigation of circadian

rhythms—the biological clock that controls the expression of genes that affect the physi-
ological responses to daily environmental changes in light and temperature. Extensive stud-
ies of the circadian clock have been made using mutants of the model organism S. elongatus,
revealing that a set of core oscillator proteins encoded by kai genes regulate global pat-
terns of gene regulation, the timing of cell division, and compaction of the chromosome. As
a simple approximation, about two-thirds of the genes are up-regulated around dawn and
down-regulated at dusk, while the other third is switched off at dawn and on at dusk. The
genes of the cyanobacterial clock appear to have little homology to those in other organisms
and it seems that systems for circadian clocks have evolved independently in major groups of
organisms, rather than having an ancient origin in ancestral bacteria, as once believed.

Some free-living and symbiotic cyanobacteria fix nitrogen

Nitrogen fixation is of fundamental significance in primary production in the oceans and
many cyanobacteria carry out this process in addition to their important role in carbon
fixation. As discussed in Chapter 3, the reduction of nitrogen to ammonia is an extremely
 Diversity of Marine Bacteria 137

energy-demanding process catalyzed by the key enzyme nitrogenase, which is strongly

inhibited by oxygen. Nitrogen fixation is assumed to have evolved in the anoxic conditions
of early Earth over 3 BYA. The cyanobacterial diazotrophs created a problem for themselves
by generating oxygen by photosynthesis. As oxygen accumulated in the atmosphere, they
evolved various mechanisms whereby the processes of nitrogen fixation and photosynthesis
became segregated, either temporally or spatially.

Some filamentous cyanobacteria achieve the spatial separation of photosynthesis and nitro-
gen fixation by differentiation of specialized cells called heterocysts formed at intervals
along the filament. These cells do not develop photosystem II, so they do not generate oxygen
and are unable to fix CO2; therefore, they cannot make the pyruvate needed as an electron
donor for nitrogen fixation. As a result, they depend on neighboring cells in the filament to
supply them with carbohydrate precursors. In return, they supply adjacent cells with fixed
nitrogen in the form of the amino acid glutamine. The nitrogenase is also protected from
oxygen because the heterocyst becomes surrounded by a gas-impermeable wall. Heterocyst-
forming cyanobacteria are common in freshwater, estuaries, and the Baltic Sea, but are less
commonly observed in the open ocean. However, two widely distributed cyanobacterial spe-
cies, Richelia intracellularis and Calothrix rhizosoleniae form short chains with a termi-
nal heterocyst and are widespread as symbionts inside the cells of diatom genera such as
Rhizosolenia and Hemiaulus. The diatom Chaetoceros has nitrogen-fixing cyanobacterial
partners that are closely associated with the cell, but outside the diatom frustule. The mutu-
alistic nature of some of these associations has been clearly established—the diatom partners
provide nutrients to increase the growth and metabolism of their cyanobacterial partners,
which in turn provide the diatom host with fixed nitrogen (Figure 4.11). Symbiont-containing
diatoms often form substantial blooms that sink rapidly, providing an important mechanism
for rapid transport of fixed carbon and nitrogen to deep waters. It is not known if the intracel-
lular cyanobacteria have a free-living stage, or how they are acquired or maintained during
reproduction of the diatom hosts.

A common mechanism of separating the process of O2 generation and N2 fixation is to

restrict the latter to the night when no photosynthesis occurs. This is the strategy employed
by Crocosphaera watsonii, a large (2.5–6.0 µm) unicellular cyanobacterium present in tropi-
cal and subtropical waters, where it can be detected by the fluorescence of its pigment phyco-
erythrin or by demonstration of the nif nitrogenase genes in association with phytoplankton
in the 3–10 µm size range. Surprisingly, metagenomic analysis indicates rather conserved
genomes, unlike those of Prochlorococcus and Synechococcus. C. watsonii can be cultured
and shown to possess a large genome, with a mean size of 5.9 Mb. Studies of gene expression

Figure 4.11 Nitrogen fixation and

transfer in cyanobacteria revealed
by NanoSIMS. A. Cytophaga bacteria
attached to filamentous cyanobac-
teria, identified by HISH-SIMS. B.
Fixed N is transferred directly to the
bacteria. C. Symbiosis between the
diatom Hemiaulas and the intracel-
lular cyanobacterium Richelia. Inset
is the epifluorescent image of cells
prior to NanoSIMS analysis. N uptake
is localized to the symbionts and also
to host cells. D. A cluster of unicel-
lular Crocosphaera, and a filament
of Trichodesmium incubated in 15N2.
Note the differential uptake of N into
only a few Crocosphaera cells. Credit:
Rachel A. Foster, Birgit Adams, Tomas
Vagner, Niculina Musat, and Marcel
Kuypers; Max Planck Institute for
Marine Microbiology, Bremen.
138 Chapter 4

Figure 4.12 A. Image of a reddish-

brown bloom of Trichodesmium
between the Great Barrier Reef and
the Queensland shore, viewed with
the MODIS satellite. B. Aggregated
filaments of Trichodesmium. Credits:
A. NASA Earth Observatory. B. S.
Kranz, Alfred Wegener Institute,
Bremerhaven.

show that about one-third of the genome is subject to diel variation in expression. Many of the
genes associated with nitrogen fixation are upregulated at the beginning of the dark period.
During the transition from light to dark periods, another method ensures protection of the
nitrogenase complex from the effects of oxygen. Photosystem II shuts down due to reduction
in the pool of quinone intermediates, the rate of electron transfer, and deactivation of key
proteins (see Figure 3.3).

UCYN-A—ARE WE The filamentous cyanobacterium Trichodesmium is probably the most abundant diazotroph
? WITNESSING THE
EVOLUTION OF A
in the open ocean and forms dense masses that are responsible for massive blooms that can
cover many thousands of km2 visible from space, especially in tropical seas (Figure 4.12a).
“NITROPLAST?” Reddish-brown streamers of Trichodesmium are often described as scum or “sea sawdust”
and were described in the journals of James Cook and Charles Darwin—“The whole sur-
Studies conducted by Jonathan
face of the water, as it appeared under a weak lens, seemed as if covered by chopped bits of
Zehr and colleagues using cell
sorting and visualization by FISH hay, with their ends jagged.” The colonies also provide a substrate for association of many
(Figure 4.13A) showed that ‘Ca. other microbes and small invertebrates. Several species with different distributions can
Atelocyanobacterium thalassa’ be distinguished morphologically. Genetically, these fall into two major clades: one con-
(UCYN-A) forms a close asso- taining T. erythraeum and another containing the species T. tenue, T. hildebrandtii, and
ciation on the exterior of its algal T. contortum; also, the related genus Katagnymene should probably be reclassified into
host, although the nature of the this group. Trichodesmium forms dense colonies in which the individual filaments appear
physical interactions and transfer of braided together following self-assembly into aggregates (Figure 4.12b). The cells contain
nutrients is unknown (review, Zehr gas vacuoles that are used to control buoyancy, allowing the filamentous colonies to move up
et al., 2016). In separate studies, and down the water column to find optimum light conditions or nutrient levels, respectively.
Hagino et al. (2013) discovered
Analysis of the large (7.8 Mb) genome of Trichodesmium reveals that it contains several
inclusion bodies in the calcified
stage of B. bigelow (Figures 4.13B,
genes associated with heterocyst formation in other filamentous cyanobacteria, including the
C) and these appear to be intra- key regulatory gene hetR. However, these genes are not fully expressed and Trichodesmium
cellular UCYN-A, based on PCR does not produce heterocysts; instead, the filaments contain groups of specialized cells called
amplification. These differences are diazocytes that contain the nitrogenase complex. These form at intervals along the filament
currently enigmatic, but UCYN-A and make up ~15% of the cells. Unlike heterocysts, these cells are not terminally differenti-
has some genomic similarities to ated and are able to divide, suggesting that their development may follow the same initial
intracellular symbionts of some stages as heterocysts. In addition to this spatial separation, Trichodesmium also separates
freshwater diatoms. Could these nitrogen fixation and oxygen production temporally, but unlike Crocosphaera, this is not by
associations be true endosymbio- day-night separation—Trichodesmium fixes nitrogen during the daytime. Around midday,
sis that might eventually lead to
oxygen production from photosynthesis is lowered by enhancement of respiration (increased
the evolution of a fully integrated
activity of cytochrome c oxidase) and the Mehler reaction (reduction of oxygen to hydrogen
nitrogen-fixing organelle? (Figure
4.13D). This occurred in the past peroxide) in the diazocytes. Interestingly, the diel clock of Trichodesmium works to keep the
with the evolution of the chlo- energy-demanding processes of fixing carbon dioxide and nitrogen active during the day,
roplast from endosymbiosis of a while restricting cell division and diazocyte development to the night.
cyanobacterial ancestor but, as
far as we know, “nitroplasts” have The final example of a nitrogen-fixing cyanobacterium is an obligate symbiont of picoeu-
not evolved. Further progress karyotic algae. Gene surveys for the nif nitrogenase genes revealed widespread distribution
will depend on cultivating the of abundant small cyanobacteria termed UCYN-A that has defied all attempts at cultiva-
bacterium and its host in order to tion. Reconstruction of the genome showed that it is extremely reduced and lacks many
examine membrane systems and gene systems normally found in cyanobacteria. The cyanobacterium cannot live indepen-
study the physiology of exchange
dently and is dependent on its algal host to provide fixed carbon compounds and in return
of carbohydrate and fixed nitrogen
between the partners.
it provides fixed nitrogen. The uncultivated cyanobacterium has been given the name “Ca.
Atelocyanobacterium thalassa” (the name means “incomplete marine cyanobacterium”).
 Diversity of Marine Bacteria 139

Figure 4.13 Association of

nitrogen-fixing cyanobacterium
UCYN-A with its prymnesiophyte
host Braarudosphaera bigelowii. A.
Epifluorescence micrograph visual-
ized using CARD-FISH with probes
directed against UCYN-A1 (red) and
host cells (green) suggest attach-
ment of the bacterium to the surface
of the host cell. B. TEM of B. bigelowii
specimen showing nucleus (N), chlo-
roplasts (C), pentaliths (calcareous
scales) (P), mitochondria (M), and a
spheroid body (arrow), identified as
UCYN-A by PCR amplification of 16S
rRNA genes. C. TEM of intracellular
SB showing internal lamellae (arrow)
and envelope consisting of three lay-
ers possibly corresponding to outer
membrane, peptidoglycan wall, and
plasma membrane of gram-negative
bacteria. D. Possible models of
symbiotic interactions depending on
whether there is full internalization
of UCYN-A, leading to enclosure by
Transcriptional studies show that there are distinct daily cycles for the expression of many a host membrane as a true endosym-
genes, although it lacks two of the three circadian clock genes found in most cyanobacteria. biont (left), or surface attachment
It appears to use host-supplied carbohydrates and fix nitrogen in the daytime, with close (right), in which case mechanisms
coupling to the energy provided by host photosynthesis, but many aspects of the interaction for transfer of metabolites must
remain enigmatic. The importance of the association is further discussed in Chapter 9. exist. CB, cyanobacterium; N, host
nucleus; PL, plastids; M, mitochon-
drion; OM, outer membrane; IM,
Filamentous cyanobacteria are important inner membrane. (Credits: A. Ana
in the formation of microbial mats M. Cabello, University of California,
Santa Cruz. B, C. Reprinted from
Complex stratified communities of microorganisms develop at interfaces between sediments Hagino et al. (2013), CC-BY-4.0. D.
and the overlying water. Filamentous cyanobacteria such as Phormidium, Oscillatoria, and Reprinted from Zehr et al. (2016)
Lyngbya are often dominant members of the biofilm in association with unicellular types with permission from Springer
such as Synechococcus and Synechocystis. Steep concentration gradients of light, oxygen, Nature.)
sulfide, and other chemicals develop across the biofilm. The mat becomes anoxic at night and
H2S concentrations rise. Anoxygenic phototrophs as well as aerobic and anaerobic chemohet-
erotrophs are also present. Cyanobacteria (and other motile bacteria in the biofilm) migrate
through the mat to find optimal conditions. Gliding movement, up to 10 μm s−1, occurs paral-
lel to the cell’s long axis and involves the production of mucilaginous polysaccharide slime.
There are two possible mechanisms by which gliding occurs. One is the propagation of waves
moving from one end of the filament to the other, created by the contraction of protein fibrils
in the cell wall. The other mechanism is secretion of mucus by a row of pores around the
septum of the cell. Some types, such as Nostoc, are only motile during certain stages of their
life cycle, when they produce a gliding dispersal stage known as hormogonia.

Members of the Planctomycetes have atypical cell structure

The Planctomycetes is a deeply branching phylum of the Bacteria, with many unusual fea-
tures of cell structure and morphology, dividing by polar budding and often attached to sur-
faces in rosette-like structures (Figure 4.14). The cytoplasm is divided into compartments,
and some types contain organelle-like structures. When first discovered, they were originally
classified as eukaryotic fungi (Planctomycetes means “floating fungi”), and it was not until
the advent of 16S rRNA gene analysis that they were recognized as members of the Bacteria.
They appear to form a monophyletic group with other phyla to form the Planctomycetes-
Verrucomicrobia-Chlamydiae (PVC) superphylum. Besides those named, this also includes
the phylum Lentisphaerae and the candidate phyla “Poribacteria” and “OP3.” Although
140 Chapter 4

Figure 4.14 Phase contrast

photomicrograph showing the
Rhodopirellula sp. strain P1 isolated
from biofilm on the surface of kelp
Laminaria hyperborean, displaying
ovoid cells, budding and rosette
formation. Reprinted from Bengtsson
and Øvreås (2010), CC-BY-4.0.

monophyletic, the PVC superphylum contains bacteria with extremely diverse lifestyles and
habitats, many of which are of biomedical or biotechnological importance.

The Planctomycetes (Planctomycetota in GTDB) phylum contains one class and three orders
of cultivated types, which can be isolated from marine sediments, surfaces, and seawater.

?
WHERE ARE THE These are aerobic chemoheterotrophs and can be grown on simple organic media, although
PLANCTOMYCETES IN many grow very slowly. Representatives from fresh and brackish water are also known.
THE TREE OF LIFE?
The smallest order Phycisphaerales contains just three species isolated from marine algae.
Use of 16S rRNA gene phylogeny The major order Planctomycetales contains three families, of which the Planctomycetaceae
clearly links the Planctomycetes includes the well-studied genera Blastopirellula, Gemmata, Pirellula, and Rhodopirellula,
with major groups of Gram- currently plus 13 other genera with many marine examples. In 16S rRNA gene surveys from
negative bacteria. However, the coastal waters, OTUs affiliated with the Planctomycetes represent about 6% of the bacterio-
unusual structure of planctomy-
plankton and less than 1% in the open ocean. Although they are not one of the major groups of
cetes cells has encouraged the
planktonic bacteria in pelagic waters, planctomycetes are especially associated with marine
idea that they blur the distinction
between the Eukarya, Bacteria, and snow and often increase markedly in abundance in association with blooms of diatoms or
Archaea. They appeared to lack a other algae. Use of 16S rRNA gene-based methods suggests much greater diversity than the
peptidoglycan cell wall and possess currently accepted classification based on culture; there may be at least 10 classes, 16 orders,
membrane-bound cell compart- and 43 families, which appear to show different distributions on medium and high productiv-
ments. Some types have an anam- ity waters. Only 0.6% of the known diversity of the phylum at OTU level can be assigned
moxosome organelle or nuclear to cultivated strains. Other molecular markers such as the rpoB, carB, and acsA genes give
membrane and some studies sug- increased resolution for classification of new strains and biogeographical studies.
gested that planctomycetes might
carry out endocytosis—a key fea- Genome analysis of cultivated isolates indicates that they possess rather large genomes in com-
ture of eukaryotes—supported by
parison with most marine heterotrophs. The average genome size is 7.6 Mb (range 3.60–12.4
the presence of genes for eukary-
ote-like membrane coat proteins.
Mb). They possess numerous genes encoding sulfatases, enzymes that could be responsible for
There now seems to be enough the breakdown of complex sulfated heteropolysaccharides, which are produced in large quan-
evidence to dispel the idea of tities in marine environments (e.g. by fish and algae) and it is likely that they play an important
eukaryote-like properties. Multiple role in global carbon cycling by turnover of these complex carbohydrates in marine snow and
bioinformatic and biochemical sediments, Remarkably, 57% of genes in the sequenced isolates are of unknown function.
analyses show that planctomycetes
do possess a cell wall (Jeske et al., A separate monophyletic clade within the Planctomycetes contains five uncultivated
2015). Boedeker et al. (2017) used Candidatus genera identified as anammox bacteria. These bacteria possess a unique cell
advanced imaging techniques and organelle, the anammoxosome (Figure 3.9), enabling anaerobic oxidation of ammonia. This
new genetic methods to show process is of major importance in the marine nitrogen cycle, as discussed in Chapter 9.
that Planctopirus limnophila has a
The “Ca. Scalindula” genus contains four species whose sequences have been identified in
Gram-negative cell plan, in which
the cytoplasmic membrane can
marine habitats.
show massive invaginations lead-
ing to an expandable periplasmic
space. They also found that vesicle
The phylum Bacteroidetes has a major role in
formation does not occur, although nutrient cycling via degradation of polymers
macromolecules are taken up by an
This collection of diverse, aerobic, facultative, or anaerobic chemoheterotrophs is now recog-
unusual connection between the
outer and cytoplasmic membranes.
nized as one of the major branches of the Bacteria. Members of the classes Sphingobacteria,
Flavobacteriia, and Cytophaga contains many familiar cultivated species that are easily
 Diversity of Marine Bacteria 141

isolated when marine samples are plated on laboratory media. As usual, this represents only
a small fraction of the diversity shown in environmental gene surveys; members of the phy-
lum can represent up to 20% of 16S rRNA sequences in coastal and open ocean waters, where
they are mainly attached to phytoplankton particulate aggregates of organic material. They
are also abundant in sediments, hydrothermal vents, and polar regions, and associated with
marine animals.

Many of the key genera (Cytophaga, Flavobacterium, Bacteroides, Flexibacter, and

Cellulophaga) are polyphyletic and their taxonomy is confused. Based on analysis of
the gyrB gene, the new genus Tenacibaculum has been created to accommodate marine
types formerly known as Flexibacter. Many marine isolates of the genera Cytophaga,
Cellulophaga, and Flavobacterium have unusual flexirubin and carotenoid pigments and
can be easily isolated as colored colonies on agar media inoculated from sediments, marine
snow, and the surfaces of animals and plants and incubated aerobically at ambient tem-
peratures (Figure 2.6). Agar is normally resistant to bacterial degradation, which is why
it is an ideal gelling agent for culture plates but marine isolates belonging to this group
often cause softening or the formation of craters on agar plates due to degradation. They
produce a wide range of hydrolytic extracellular enzymes, which are responsible for deg-
radation of carbohydrate polymers, chitin, and proteins. As noted on p.92, members of the
Bacteroidetes use surface-associated enzymes to bind and partially degrade polysaccha-
rides, ensuring efficient, “selfish” assimilation of nutrients. This is of major ecological sig-
nificance in marine nutrient cycling, due to the degradation of complex organic materials,
such as the cell walls of phytoplankton and exoskeletons of crustaceans. Some species are
pathogenic for fish and invertebrates. Many are psychrophilic, being commonly isolated
from cold-water marine habitats and sea ice. The normal habitat of the genus Bacteroides
is the gut of mammals and they may be present in sewage-polluted waters and persist for
quite long periods in the sea, prompting investigation of their use as indicators of water
quality (p.366).

Members of the phylum Chloroflexi are

widespread but poorly characterized
Bacteria in the phylum Chloroflexi are found in a wide range of habitats and members show
diverse metabolism, including anoxygenic photosynthesis and aerobic or anaerobic heterotro-
phy. Some are thermophiles. They are also phylogenetically diverse, with nine classes, each
containing only a few genera. Overall, only 26 genera have been cultivated from terrestrial
hot springs, soils, and freshwater. However, in early marine 16S rRNA surveys the SAR202
clade, which is abundant and globally distributed in the aphotic zone, was shown to affiliate
with the Chloroflexi. We now know that they account for about 9% of the total sequences
and in the deep ocean they can comprise >40% of the total bacterial community. Because of
the absence of cultured marine representatives, their metabolic activity and role in marine
processes has been unclear. Recent analysis of MAGs from SAR202 combined with FISH
measurement indicate that they appear to be free-living, large bacteria with genomes of
~3–4 Mbp, probably existing as at least two sub-clades. One group contains pathways for
the metabolism of organosulfur compounds as a source of carbon, nitrogen, and sulfur, sug-
gesting that they play a major role in deep ocean sulfur cycling. A group identified in other
studies has genes encoding multiple families of oxidative enzymes that may be active in the
degradation of cyclic hydrocarbons. This activity is postulated to be involved in the conver-
sion of DOM to refractory DOM. If this proves correct, it means that the Chloroflexi play a
critical role in the microbial carbon pump discussed in Chapter 8.

The phyla Aquificae and Thermotogae are

deeply branching primitive thermophiles
In 16S rRNA phylogenetic trees, these phyla form branches closer to the root than other
groups of bacteria. The Aquificae contains a small number of genera which are all thermo-
philic and mostly microaerophilic oxidizing hydrogen that use hydrogen, thiosulfate, or sul-
fur as the electron donor and oxygen as the electron acceptor, fixing carbon via the reductive
TCA cycle.
142 Chapter 4

Examples include Aquifex aeolicus, A. pyrophilus, Hydrogenothermus marinus, and

Persephonella marinus, which are extremely thermophilic (maximum growth temperature
can be as high as 95°C) and have a major role in primary production at marine hydrothermal
vents. A. aeolicus was one of the first marine bacteria for which a full genome sequence was
obtained in 1998; it was found to have a very small genome (1.6 Mbp). Other species have
also been found to have small genomes, typically <2.0 Mbp.

The Thermotogae also forms a deeply branching, phylogenetically distant group of Bacteria.
As well as evidence from gene sequences, the function of the ribosome is very different
from other Bacteria and is not affected by rifampicin and other antibiotics that affect protein
synthesis. The name of the genus derives from a unique outer membrane (“toga”), which bal-
loons out from the rod-shaped cells. The cells are Gram-negative, but the amino acid compo-
sition of the peptidoglycan is unlike that of other Bacteria, and there are unusual long-chain
fatty acids in the lipids. Thermotoga is widespread in geothermal areas and occurs in shal-
low and deep-sea hydrothermal vents. Different species vary in their temperature optima,
with a range from 55°C up to 80–95°C for the hyperthermophilic species T. maritima and T.
neapolitana. These are fermentative, anaerobic chemoorganotrophs and use a wide range of
carbohydrates. They can also fix N2 and reduce sulfur to H2S. Like Aquifex, these organisms
have considerable biotechnological potential, and genome sequences have been published for
several species. A number of genomes have been sequenced (size 1.7–1.9 Mbp). Many of the
genes are involved in the transport and utilization of nutrients, in keeping with their ability
to use a wide range of substances for growth. The genomes of both Aquifex and Thermotoga
contain a high proportion of genes with homology to archaeal genes, indicating that large-
scale lateral gene transfer has occurred in the evolution of these bacteria.

The Firmicutes are a major branch

of Gram-positive Bacteria
The large and diverse phylum Firmicutes contains unicellular or filamentous bacteria dis-
tinguished by a thick cell wall composed mainly of peptidoglycan, and genomes with a low
G+C ratio, which distinguishes them from the other phylum of Gram-positive bacteria, the
Actinobacteria (low G+C). The phylum contains seven classes, of which the most important
are the Bacilli (two orders, 15 families, 212 genera) and Clostridia (four orders, 18 families,
146 genera). These are likely to be reclassified under the GTDB system.

Within the Bacilli, the family Bacillaceae contains over 62 genera and 460 species and are
best known for their role in ecology of terrestrial soils and plant health, but they are also a
major component of coastal, saltmarsh, and deep-sea sediments. Although relatively little
information exists on their abundance and distribution in marine environments, some spe-
cies have been named because of their initial isolation from marine sediments (e.g. Bacillus
marinus, Marinococcus, Oceanobacillus, Thalassobacillus) and it is likely that great diver-
sity in marine representatives remains to be discovered in these habitats. Members of the
Firmicutes comprise ~4% of 16S rRNA sequences in the water column and ~8% in sedi-
ments, although it is possible that many of these sequences result from bacteria that enter the
sea as dormant endospores with terrestrial soil runoff, especially near large river mouths.
They are mostly aerobic or facultatively anaerobic chemo-organotrophs. They produce a
range of extracellular hydrolytic enzymes that degrade polysaccharides, proteins, lipids,
and nucleic acids which play an important role in organic decomposition in sediments.
Their most distinctive feature is the production of extremely resistant endospores by all
but a few taxa, which allow them to resist high temperatures, irradiation, and desiccation;
endospores may persist for thousands of years. Thermophilic and halophilic species occur
at hydrothermal vents and salterns, respectively. Many have biotechnological applications.
For example, some Bacillus spp. bind and oxidize metals such as manganese and copper
and the spores may have applications in remediation of polluted marine sediments. Some
members of the related genus Paenibacillus produce powerful chitinolytic activity, which
may be useful for the breakdown of shrimp and crab shell waste. Both genera are also used
as probiotics in aquaculture (see p.403), as are Lactobacillus and Pediococcus in the family
Lactobacillaceae. Other genera produce secondary metabolites including antibiotics with
potential applications in biomedicine.
 Diversity of Marine Bacteria 143

Other important genera of the order Bacillales include Staphylococcus and Listeria, which
are occasionally isolated in marine samples, but are probably terrestrial contaminants or very
minor members of the marine bacterial community. However, they can be important as agents
of fish spoilage and food-borne intoxication following processing (p.360). Streptococcus
iniae and some other species are pathogens in warm-water fish and dolphins.

The Clostridia are anaerobic and employ a range of pathways for fermentation of different
types of carbohydrates, amino acids, and purines. As seen with Bacilli, they are best known
for their activity in anaerobic soils, but many species also occur in anoxic marine sediments
and are important in degradation of organic material. Some species reduce sulfate, sulfite, or
thiosulfate. The toxin-producing Clostridium botulinum (type E) occurs in marine mud and
fish guts and can cause the disease botulism in humans, occasionally associated with seafood
(see p.341), or in waterfowl in estuaries or coastal lagoons.

One unusual group of clostridia are the uncultivated giant bacteria “Ca. Epulopiscium fish-
elsoni” and related species. These occur in large numbers in the intestinal tracts of certain
species of herbivorous surgeonfish on the Great Barrier Reef and the Red Sea. They appear to
be mutualistic symbionts, which aid in the breakdown of complex carbohydrate polysaccha-
rides in the red and brown algae ingested by the fish. Different fish species appear to associ-
ate with particular subclades of Epulopiscium, leading to unique gut microbiota that permit
specialization in feeding on certain algal types. Juvenile surgeonfish have been observed to
be coprophagous (consuming the feces of adults) and presumably acquire symbionts through
this behavior. Epulopiscium strains are some of the largest bacteria known; although mor-
phology and cell size are variable, cells up to 400 × 80 μm are common and some rare cells
can reach 600 μm. When originally discovered, they were thought to be a eukaryotic protist
because of their large size and intracellular structure. They have a unique reproductive pro-
cess; they are “viviparous,” meaning that new cells are formed inside the parent cell, which
undergoes a localized cell lysis to release the active progeny. One of the most remarkable
features of the life cycle is that it follows a circadian rhythm. Samples of surgeonfish gut
contents taken early in the morning show small internal offspring appearing near the tips of
the mother cell. The offspring cells grow during the day and are released at night through
partial lysis of the mother cell, which then dies. This process is speculated to have evolved
from the process of endospore formation, which involves the partition of the spore from the
parent cell (Figure 4.15).

Members of the Actinobacteria are a rich source

of secondary metabolites, including antibiotics
The phylum Actinobacteria forms a second major group of Gram-positive bacteria, character-
ized by its high G + C content. It is highly diverse, containing 65 cultivated families and over
500 species. They are mostly free-living rod-shaped or filamentous organisms and many spe-
cies have been cultivated from terrestrial soils, freshwater, and marine sediments. However,
this represents only a small fraction of the diversity revealed by 16S rRNA sequence surveys,

Figure 4.15 “Ca. Epulopiscium

Epulopiscium E. coli fishelsoni”, one of the largest
known bacteria. (a) Light micro-
graph showing comparative sizes of
Epulopiscium, Paramecium (a protist),
and Escherichia coli (a “typical”
bacterium). (b) “Viviparous” release
Paramecium of daughter cell during reproduc-
tion. Credit: Esther R. Angert, Cornell
(a) (b) University..
144 Chapter 4

especially those from coastal and estuarine waters, where they can represent about 10% of

? DO YOU LIKE SUSHI?

ZOBELLIA GENES
MAY HELP YOU
the bacterioplankton

The order Actinomycetales is a large and very diverse group of bacteria with various
TO DIGEST IT.
cell morphologies, ranging from coryneform (club-shaped, e.g. Corynebacterium and
Humans do not produce the Arthrobacter) to branching filaments with reproductive conidiospores (e.g. Streptomyces and
carbohydrate active enzymes Micromonospora). Actinomycetes are widely distributed in marine sediments and particu-
(CAzymes) needed to digest late matter in the water column. Because of their high abundance in soil, it is possible that
polysaccharides—we depend many actinomycetes in coastal sediments are derived from terrestrial runoff, but there are
on members of our gut microbi- also halophilic species and they are also found in deep-sea samples. Their main ecological
ome to do this. During human
importance is in decomposition and heterotrophic nutrient cycling, owing to their produc-
evolution, we have acquired a
diverse array of microbes, includ-
tion of diverse extracellular enzymes, which break down polysaccharides, proteins, and fats.
ing many that degrade complex Many of the actinomycetes have large genomes (up to 9 Mb), showing cellular differentiation
polysaccharides from terrestrial that makes them an exceptionally rich source of secondary metabolites. Many of the most
plants. Some bacteria, especially widely used antibiotics (e.g. aminoglycosides and tetracyclines) come from actinomycetes
members of the Bacteroidetes found in soil. Extensive surveys of the diversity of marine actinomycetes have been under-
can produce hundreds of specific taken in the search for unique compounds with potential uses as antimicrobial or anticancer
glycoside hydrolases and polysac- drugs (e.g. compounds from Salinispora). Marine Micromonospora are used in some aqua-
charide lyases. Hehemann et al. culture probiotics. These biotechnological aspects are considered in Chapter 14.
(2010) identified a new CAzyme,
porphyrinase, that specifically
Mycobacteria are slow-growing, aerobic, rod-shaped organisms distinguished by unusual cell
degrades porphyrin, a complex
wall components, which render them acid-fast in a staining procedure. They are widely dis-
polysaccharide in the cell walls
of red algae. They identified the tributed as saprophytes on surfaces such as sediments, corals, fish, and algae.
substrate recognition site on the
enzyme and showed that the gene The great majority of species in the Actinobacteria are harmless saprophytes, but a few
sequence for this can be found in species are pathogenic. Those with a marine connection include Mycobacterium mari-
marine bacteria, including Zobellia num (a pathogen of fish and marine mammals that can be transmitted to humans) and
galactivorans (Bacteroidetes), that Renibacterium salmoninarum (an obligate pathogen of salmonid fish).
grows on Porphyra red algae. It was
also found in Bacteroides plebeius in
the gut microbiome of a group of
Japanese people investigated, but
not in North Americans. Genomic Conclusions
analysis shows that acquisition
of this gene likely occurred by This chapter has illustrated the great diversity of morphology, metabolism, physiology, and
HGT, because raw seaweed (nori) genetic makeup of members of the domain Bacteria. This diversity has enabled bacteria to
forms a major component of the
colonize every conceivable habitat in the marine realm, and their activities affect all aspects
Japanese diet.
of marine ecology and biogeochemistry. The evolutionary history and classification of the
Bacteria remains problematic. Only a small fraction of the organisms known from molecu-
lar-based studies have been cultured and studied in detail, but genomic methods mean that
completely new types of bacteria, biochemical pathways, ecological interactions, and biogeo-
chemical effects are being discovered at ever-increasing rates. The oceans undoubtedly hold
many more surprises yet to be revealed.

Callahan, B.J., McMurdie, P.J., and Holmes, S.P. (2017) Exact sequence
References and further reading variants should replace operational taxonomic units in marker-gene data
Principles of diversity, classification, and taxonomy analysis. ISME J. 11: 2639.

Achtman, M., and Wagner, M. (2008) Microbial diversity and the genetic Castelle, C.J., and Banfield, J.F. (2018) Major new microbial groups
nature of microbial species. Nat. Rev. Microbiol. 6: 431–440. expand diversity and alter our understanding of the tree of life. Cell 172:
1181–1197.
Amann, R., and Rosselló-Móra, R. (2016) After all, only millions? MBio
7: e00999–e00916. Chun, J., Oren, A., Ventosa, A. et al. (2019) Proposed minimal standards
for the use of genome data for the taxonomy of prokaryotes. Int. J. Syst.
Amaral-Zettler, L., Artigas, L.F., Baross, J., et al. (2010). A global Evol. Microbiol 68: 461–466.
census of marine microbes, In: Life in the World’s Oceans. Diversity,
Distribution, and Abundance. McIntyre, A.D. (ed.) Wiley-Blackwell, pp. Curtis, T.P., Sloan, W.T., and Scannell, J.W. (2002) Estimating prokaryotic
223–246. diversity and its limits. Proc. Natl. Acad. Sci. U. S. A. 99: 10494–10499.

Bergey’s Manual Trust (N.D) Bergey’s Manual of Systematics of Garrity, G.M. (2016) A new genomics-driven taxonomy of Bacteria and
Archaea and Bacteria. Wiley Online Library; available at https:// Archaea: are We There Yet? J. Clin. Microbiol. 54: 1956–1963.
onlinelibrary.wiley.com/doi/book/10.1002/9781118960608. Accessed Hug, L.A., Baker, B.J., Anantharaman, K. et al. (2016) A new view of the
23 April 2019. tree of life. Nat. Microbiol. 1: 16048.
 Diversity of Marine Bacteria 145

Jousset, A., Bienhold, C., Chatzinotas, A. et al. (2017) Where less may Greene, S.E., Brilli, M., Biondi, E.G. et al. (2012) Analysis of the CtrA
be more: how the rare biosphere pulls ecosystems strings. ISME J. 11: pathway in Magnetospirillum reveals an ancestral role in motility in alp-
853–862. haproteobacteria. J. Bacteriol. 194: 2973–2986.
Konstantinidis KT, Rosselló-Móra R, and Amann R. (2017) Uncultivated Hughes, V., Jiang, C., and Brun, Y. (2012) Caulobacter crescentus. Curr.
microbes in need of their own taxonomy. ISME J. 11: 2399–2406. Biol. 22: R507–R509.
Kunin, V., Engelbrektson, A., Ochman, H. et al. (2010) Wrinkles in the Komeili, A., Vali, H., Beveridge, T.J. et al. (2004) Magnetosome vesicles
rare biosphere: pyrosequencing errors can lead to artificial inflation of are present before magnetite formation, and MamA is required for their
diversity estimates. Environ. Microbiol. 12: 118–123. activation. Proc. Natl. Acad. Sci. USA 101: 3839–3844.
Locey, K.J., and Lennon, J.T. (2016) Scaling laws predict global microbial Kolinko, S., Richter, M., Glöckner, F.O. et al. (2016). Single‐cell genom-
diversity. Proc. Natl. Acad. Sci. USA 113: 5970–5975. ics of uncultivated deep‐branching magnetotactic bacteria reveals a con-
Louca, S., Mazel, F., Doebeli, M. et al. (2019) A census-based estimate of served set of magnetosome genes. Environ. Microbiol.18: 21–37.
Earth’s bacterial and archaeal diversity. PLOS Biol. 17: e3000106. Lin, W., Zhang, W., Zhao, X. et al. (2018). Genomic expansion of mag-
Parks, D.H., Chuvochina, M., Waite, D.W. et al. (2018) A standardized netotactic bacteria reveals an early common origin of magnetotaxis with
bacterial taxonomy based on genome phylogeny substantially revises the lineage-specific evolution. ISME J. 12: 1508.
tree of life. Nat. Biotechnol. 36: 996. Luo, H., and Moran, M.A. (2014) Evolutionary ecology of the marine
Pedros-Alio, C., Acinas, S.G., Logares, R. et al. (2018). Marine micro- Roseobacter clade. Microbiol. Mol. Biol. Rev. 78: 573–587.
bial diversity as seen by hight-throughput sequencing. In: Gasol, J.M. and Martins, J.L., Silveira, T.S., Abreu, F. et al. (2007). Grazing protozoa and
Kirchman, D.L. (eds.) Microbial Ecology of the Oceans. John Wiley & magnetosome dissolution in magnetotactic bacteria. Environ. Microbiol.
Sons, pp. 47–98. 9: 2775–2781.
Richter, M., and Rosselló-Móra, R. (2009) Shifting the genomic gold Rosenberg, E., DeLong, E.F., Lory, S. et al. (eds.) (2014) The Prokaryotes.
standard for the prokaryotic species definition. Proc. Natl. Acad. Sci. Alphaproteobacteria and Betaproteobacteria. Springer Berlin,
USA 106: 19126–19131. Heidelberg.
Salazar, G., and Sunagawa, S. (2017a) Marine microbial diversity. Curr. Scheffel, A., Gruska, M., Faivre, D. et al. (2006) An acidic protein aligns
Biol. 27: R489–R494. magnetosomes along a filamentous structure in magnetotactic bacteria.
Schleifer, K.H. (2009) Classification of Bacteria and Archaea: Past, pres- Nature 440: 110–114.
ent and future. Syst. Appl. Microbiol. 32: 533–542. Silva, K.T., Abreu, F., Keim, C.N. et al. (2008) Ultrastructure and cyto-
Schloss, P.D., Girard, R.A., Martin, T. et al. (2016) Status of the archaeal chemistry of lipid granules in the many-celled magnetotactic prokaryote,
and bacterial census: an update. MBio 7: e00201–e00216. ‘Candidatus Magnetoglobus multicellularis.’ Micron 39: 1387–1392.

Schulz, F., Eloe-Fadrosh, E.A., Bowers, R.M. et al. (2017) Towards a bal- Taoka, A., Kiyokawa, A., Uesugi, C. et al. (2017) Tethered magnets are
anced view of the bacterial tree of life. Microbiome 5: 140. the key to magnetotaxis: direct observations of Magnetospirillum mag-
neticum AMB-1 show that MamK distributes magnetosome organelles
Sogin, M.L., Morrison, H.G., Huber, J.A. et al. (2008) Microbial diversity equally to daughter cells. MBio 8: e00679–e00617.
in the deep sea and the underexplored “rare biosphere.” Proc. Natl. Acad.
Sci. 103: 12115–12120. Gammaproteobacteria
Stackebrandt, E., and Ebers, J. (2006) Taxonomic parameters revisited: Frischkorn, K.R., Stojanovski, A., and Paranjpye, R. (2013) Vibrio para-
tarnished gold standards. Microbiol. Today 33:152–155. haemolyticus type IV pili mediate interactions with diatom-derived chi-
Waite, D.W., Vanwonterghem, I., Rinke, C. et al. (2017) Comparative tin and point to an unexplored mechanism of environmental persistence.
genomic analysis of the class Epsilonproteobacteria and proposed reclas- Environ. Microbiol. 15: 1416–1427.
sification to Epsilonbacteraeota (phyl. nov.). Front. Microbiol. 8: 682. Jørgensen, B.B., and Gallardo, V.A. (1999) Thioploca spp.: filamen-
Whitman, W.B., Coleman, D.C., Wiebe, W.J. et al. (1998) Prokaryotes: tous sulfur bacteria with nitrate vacuoles. FEMS Microbiol. Ecol. 28:
the unseen majority. Proc. Natl. Acad. Sci. USA 95: 6578–6583. 301–313.
Willis, A. (2016) Extrapolating abundance curves has no predictive power for Lin, H., Yu, M., Wang, X. et al. (2018) Comparative genomic analysis
estimating microbial biodiversity. Proc. Natl. Acad. Sci. USA 113: E5096. reveals the evolution and environmental adaptation strategies of vibrios.
BMC Genomics 19: 135.
Winter, C., Bouvier, T., Weinbauer, M.G. et al. (2010) Trade-offs between
competition and defense specialists among unicellular planktonic organ- Miller, S.D., Haddock, S.H.D., Elvidge, C.D. and Lee, T.F. (2005).
isms: the “killing the winner” hypothesis revisited. Microbiol. Mol. Biol. Detection of a bioluminescent milky sea from space. Proc. Nat. Acad.Sci
Rev. 74: 42–57. USA 102: 14181–14184.
Yarza, P., Yilmaz, P., Pruesse, E. et al. (2014) Uniting the classification Thompson, C.C., Vicente, A., Souza, R.C. et al. (2009) Genomic taxon-
of cultured and uncultured bacteria and archaea using 16S rRNA gene omy of vibrios. BMC Evol. Biol. 9: 258.
sequences. Nat. Rev. Microbiol. 12: 635–645. Thompson, F.L., Gevers, D., Thompson, C.C. et al. (2005) Phylogeny and
Yilmaz, P., Yarza, P., Rapp, J.Z. et al. (2016) Expanding the world of molecular identification of vibrios on the basis of multilocus sequence
marine bacterial and archaeal clades. Front. Microbiol. 6: 1524. analysis. Appl. Environ. Microbiol. 71: 5107–5115.

Alphaproteobacteria Cable bacteria, deltaproteobacteria

and epsilonbacteriota
Ferla, M.P., Thrash, J.C., Giovannoni, S.J. et al. (2013) New rRNA gene-based
phylogenies of the Alphaproteobacteria provide perspective on major groups, Bjerg, J.T., Boschker, H.T.S., Larsen, S. et al. (2018) Long-distance elec-
mitochondrial ancestry and phylogenetic instability. PLoS One 8: e83383. tron transport in individual, living cable bacteria. Proc. Natl. Acad. Sci.
Giovannoni, S.J., Hayakawa, D.H., Tripp, H.J. et al. (2008) The small U. S. A. 115: 5786–5791.
genome of an abundant coastal ocean methylotroph. Environ. Microbiol. Bjerg, J.T., Damgaard, L.R., Holm, S.A. et al. (2016) Motility of electric
10: 1771–1782. cable bacteria. Appl. Environ. Microbiol. 82: 3816–3821.
146 Chapter 4

Burdorf, L.D.W., Tramper, A., Seitaj, D. et al. (2017) Long-distance Raghunathan, D., Radford, P.M., Gell, C. et al. (2019) Engulfment, per-
electron transport occurs globally in marine sediments. Biogeosci. 14: sistence and fate of Bdellovibrio bacteriovorus predators inside human
683–701. phagocytic cells informs their future therapeutic potential. Sci. Rep. 9:
Cornelissen, R., and Bøggild, A., Thiruvallur Eachambadi, R. et al. (2018) 4293.
The cell envelope structure of cable bacteria. Front. Microbiol. 9: 3044. Rotem, O., Pasternak, Z., and Jurkevitch, E. (2014) Bdellovibrio
Kessler, A.J., Wawryk, M., Marzocchi, U. et al. (2018) Cable bacteria and like organisms. In: The Prokaryotes. Deltaproteobacteria and
promote DNRA through iron sulfide dissolution. Limnol. Oceanogr. 64: Epsilonproteobacteria. Rosenberg, E., DeLong, E.F., Lory, S. et al. (eds.)
1228–1238. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 3–17.

Kuever, J. (2014a) The family Desulfobulbaceae. In: The Prokaryotes. Zetaproteobacteria

Deltaproteobacteria and Epsilonproteobacteria. Rosenberg, E., DeLong,
E.F., Lory, S. et al. (eds.) Springer Berlin, Heidelberg, pp. 75–86. Laufer, K., Nordhoff, M., Halama, M. et al. (2017) Microaerophilic
Fe(II)-oxidizing zetaproteobacteria isolated from low-Fe marine coastal
Kuever, J. (2014b) The family Desulfovibrionaceae. In: The Prokaryotes.
sediments: physiology and composition of their twisted stalks. Appl.
Deltaproteobacteria and Epsilonproteobacteria. Rosenberg, E., DeLong,
Environ. Microbiol. 83: e03118–e03116.
E.F., Lory, S. et al. (eds.) Springer Berlin, Heidelberg, pp. 107–133.
Moreira, A.P.B., Meirelles, P.M., and Thompson, F. (2014) The fam-
Malkin, S.Y., Seitaj, D., Burdorf, L.D.W. et al. (2017) Electrogenic sulfur
ily Mariprofundaceae. In: The Prokaryotes. Deltaproteobacteria and
oxidation by cable bacteria in bivalve reef sediments. Front. Mar. Sci. 4: 28.
Epsilonproteobacteria. Rosenberg, E., DeLong, E.F., Lory, S. et al. (eds.)
Martin, B.C., Bougoure, J., Ryan, M.H. et al. (2019) Oxygen loss from Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 403–413.
seagrass roots coincides with colonisation of sulphide-oxidising cable
McAllister, S.M., Davis, R.E., McBeth, J.M. et al. (2011) Biodiversity
bacteria and reduces sulphide stress. ISME J. 13: 707–719.
and emerging biogeography of the neutrophilic iron-oxidizing
Marzocchi, U., Trojan, D., Larsen, S. et al. (2014) Electric coupling Zetaproteobacteria. Appl. Environ. Microbiol. 77: 5445–5457.
between distant nitrate reduction and sulfide oxidation in marine sedi-
ment. ISME J. 8: 1682–1690. Cyanobacteria
Meysman, F.J.R. (2018) Cable bacteria take a new breath using long-dis- Becker, J.W., Hogle, S.L., Rosendo, K. et al. (2019) Co-culture and bioge-
tance electricity. Trends Microbiol. 26: 411–422. ography of Prochlorococcus and SAR11. ISME J. 13: 1506–1519.
Nielsen, L.P. (2016) Ecology: Electrical cable bacteria save marine life. Bergman, B., Sandh, G., Lin, S. et al. (2013a) Trichodesmium – a wide-
Curr. Biol. 26: R32–R33. spread marine cyanobacterium with unusual nitrogen fixation properties.
Nielsen, L.P., Risgaard-Petersen, N., Fossing, H. et al. (2010) Electric FEMS Microbiol. Rev. 37: 286–302.
currents couple spatially separated biogeochemical processes in marine Biller, S.J., Berube, P.M., Lindell, D. et al. (2015) Prochlorococcus: the
sediment. Nature 463: 1071–1074. structure and function of collective diversity. Nat. Rev. Microbiol. 13:
Pfeffer, C., Larsen, S., Song, J. et al. (2012) Filamentous bacteria trans- 13–27.
port electrons over centimetre distances. Nature 491: 218–221. Biller, S.J., Schubotz, F., Roggensack, S.E. et al. (2014) Bacterial vesicles
Plugge, C.M., Zhang, W., Scholten, J.C.M.. et al. (2011) Metabolic flex- in marine ecosystems. Science 343: 183–186.
ibility of sulfate-reducing bacteria. Front. Microbiol. 2: 81. Cohen, S.E., and Golden, S.S. (2015) Circadian rhythms in Cyanobacteria.
Risgaard-Petersen, N., Damgaard, L.R., Revil, A. et al. (2014) Mapping Microbiol. Mol. Biol. Rev. 79: 373–385.
electron sources and sinks in a marine biogeobattery. J. Geophys. Res. Coutinho, F., Tschoeke, D.A., Thompson, F. et al. (2016) Comparative
Biogeosci. 119: 1475–1486. genomics of Synechococcus and proposal of the new genus
Schauer, R., Risgaard-Petersen, N., Kjeldsen, K.U. et al. (2014) Succession Parasynechococcus. PeerJ 4: e1522.
of cable bacteria and electric currents in marine sediment. ISME J. 8: Dubinsky, Z., and Berman-Frank, I. (2001) Uncoupling primary produc-
1314–1322. tion from population growth in photosynthesizing organisms in aquatic
Seitaj, D., Schauer, R., Sulu-Gambari, F. et al. (2015) Cable bacteria gen- ecosystems. Aquat. Sci. 63: 4–17.
erate a firewall against euxinia in seasonally hypoxic basins. Proc. Natl. Ehlers, K., and Oster, G. (2012) On the mysterious propulsion of
Acad. Sci. 112: 13278–13283. Synechococcus. PLoS One 7: e36081.
Trojan, D., Schreiber, L., Bjerg, J.T. et al. (2016) A taxonomic framework Farnelid, H., Turk-Kubo, K., Ploug, H. et al. (2019) Diverse diazotrophs
for cable bacteria and proposal of the candidate genera Electrothrix and are present on sinking particles in the North Pacific Subtropical Gyre.
Electronema. Syst. Appl. Microbiol. 39: 297–306. ISME J. 13: 170–182.
Vetriani, C., Voordeckers, J.W., Crespo-Medina, M. et al. (2014) Deep- Farnelid, H.M., Turk-Kubo, K.A., and Zehr, J.P. (2016) Identification of
sea hydrothermal vent Epsilonproteobacteria encode a conserved and associations between bacterioplankton and photosynthetic picoeukary-
widespread nitrate reduction pathway (Nap). ISME J. 8: 1510–1521. otes in coastal waters. Front. Microbiol. 7: 339.

Myxobacteria Flombaum, P., Gallegos, J.L., Gordillo, R.A. et al. (2013) Present and
future global distributions of the marine Cyanobacteria Prochlorococcus
Zhang, Y.-Q., Li, Y.-Z., Wang, B. et al. (2005) Characteristics and living and Synechococcus. Proc. Natl. Acad. Sci. U. S. A. 110: 9824–9829.
patterns of marine myxobacterial isolates. Appl. Environ. Microbiol. 71:
Foster, R.A., Kuypers, M.M.M., Vagner, T. et al. (2011) Nitrogen fixation
3331–3336.
and transfer in open ocean diatom-cyanobacterial symbioses. ISME J. 5:
1484–1493.
Bdellovibrio
Hagino, K., Onuma, R., Kawachi, M. et al. (2013) Discovery of an endo-
Mosca, A., Leclerc, M., and Hugot, J.P. (2016) Gut microbiota diversity symbiotic nitrogen-fixing cyanobacterium UCYN-A in Braarudosphaera
and human diseases: Should we reintroduce key predators in our ecosys- bigelowii (Prymnesiophyceae). PLoS One 8: e81749.
tem? Front. Microbiol. 7: 455.
 Diversity of Marine Bacteria 147

Johnson, Z.I., Zinser, E.R., Coe, A. et al. (2006) Niche partitioning among Wagner, M., and Horn, M. (2006) The Planctomycetes, Verrucomicrobia,
Prochlorococcus ecotypes along ocean-scale environmental gradients. Chlamydiae and sister phyla comprise a superphylum with biotechnologi-
Science (80-.). 311: 1737–1740. cal and medical relevance. Curr. Opin. Biotechnol. 17: 241–249.
Kashtan, N., Roggensack, S.E., Rodrigue, S. et al. (2014) Single- Wiegand, S., Jogler, M., and Jogler, C. (2018) On the maverick
cell genomics reveals hundreds of coexisting subpopulations in wild Planctomycetes. FEMS Microbiol. Rev. 42: 739–760.
Prochlorococcus. Science 344: 416–420. Žure, M., Fernandez-Guerra, A., Munn, C.B. et al. (2017). Geographic
Kent, A.G., Dupont, C.L., Yooseph, S. et al. (2016) Global biogeography distribution at subspecies resolution level: closely related Rhodopirellula
of Prochlorococcus genome diversity in the surface ocean. ISME J. 10: species in European coastal sediments. ISME J. 11: 478.
1856–1865. Žure, M., Munn, C.B., and Harder, J. (2015) Diversity of Rhodopirellula
Mishra, A.K., Tiwari, D.N., and Rai, A.N. (2019) Cyanobacteria: From and related planctomycetes in a North Sea coastal sediment employing
Basic Science to Applications. Academic Press. carB as molecular marker. FEMS Microbiol. Lett. 362: fnv127.
Moore, L.R. (2013). More mixotrophy in the marine microbial mix. Proc. Bengtsson, M.M., and Øvreås, L. (2010). Planctomycetes dominate biofilms
Nat. Acad. Sci. USA 110: 8323–8324. on surfaces of the kelp Laminaria hyperborea. BMC Microbiol.10: 261.
Muñoz-Marín, M.D.C., Shilova, I.N., Shi, T. et al. (2019) The transcrip-
tional cycle is suited to daytime N2 fixation in the unicellular cyanobacte- Epulopiscium
rium “Candidatus Atelocyanobacterium thalassa” (UCYN-A). MBio 10: Angert, E. R., and Clements, K. D. (2004) Initiation of intracellular off-
e02495–e02418. spring in Epulopiscium. Molec. Microbiol. 51: 827–835.
Paerl, H. (2017) The cyanobacterial nitrogen fixation paradox in natural Kamanda Ngugi, D., Miyake, S., Cahill, M. et al. (2017) Genomic diver-
waters. F1000Research 6: 244. sification of giant enteric symbionts reflects host dietary lifestyles. Proc.
Rocap, G., Larimer, F.W., Lamerdin, J. et al. (2003) Genome divergence Natl. Acad. USA 1114: E7592.
in two Prochlorococcus ecotypes reflects oceanic niche differentiation.
Nature 424: 1042–1047. Chloroflexi
Walter, J.M., Coutinho, F.H., Dutilh, B.E. et al. (2017) Ecogenomics and Hanada, S. (2014) The phylum Chloroflexi, the family Chloroflexaceae,
taxonomy of Cyanobacteria phylum. Front. Microbiol. 8: 2132. and the related phototrophic families Oscillochloridaceae and
Zehr, J.P. (2011) Nitrogen fixation by marine cyanobacteria. Trends Roseiflexaceae. In: The Prokaryotes. Other Major Lineages of Bacteria
Microbiol. 19: 162–173. and the Archaea. Rosenberg, E., DeLong, E.F., Lory, S. et al. (eds.)
Springer Berlin, Heidelberg, pp. 515–532.
Zehr, J.P. (2015) How single cells work together. Science 349: 1163–1164.
Landry, Z., Swan, B.K., Herndl, G.J. et al. (2017) SAR202 Genomes from
Zehr, J.P., Shilova, I.N., Farnelid, H.M. et al. (2016) Unusual marine uni- the dark ocean predict pathways for the oxidation of recalcitrant dissolved
cellular symbiosis with the nitrogen-fixing cyanobacterium UCYN-A. organic matter. MBio 8: e00413–e00417.
Nat. Microbiol. 2: 16214.
Mehrshad, M., Rodriguez-Valera, F., Amoozegar, M.A. et al. (2018) The
Planctomycetes and bacteroidetes enigmatic SAR202 cluster up close: shedding light on a globally distrib-
uted dark ocean lineage involved in sulfur cycling. ISME J. 12: 655–668.
Boedeker, C., Schüler, M., Reintjes, G. et al. (2017) Determining the bac-
terial cell biology of Planctomycetes. Nat. Commun. 8: 14853. Other taxa
Fernández-Gomez, B., Richter, M., Schüler, M. et al. (2013). Ecology Barka, E.A., Vatsa, P., Sanchez, L. et al. (2016) Taxonomy, physiology, and
of marine Bacteroidetes: a comparative genomics approach. ISME J. 7: natural products of Actinobacteria. Microbiol. Mol. Biol. Rev. 80: 1–43.
1026.
Garcia, R., and Müller, R. (2014) The family Myxococcaceae. In:
Hehemann, J.H., Correc, G., Barbeyron, T. et al. (2010). Transfer of car- The Prokaryotes. Deltaproteobacteria and Epsilonproteobacteria.
bohydrate-active enzymes from marine bacteria to Japanese gut micro- Rosenberg, E., DeLong, E.F., Lory, S. et al. (eds.) Springer Berlin,
biota. Nature 464: 908. Heidelberg, pp. 191–212.
Jeske, O., Schüler, M., Schumann, P. et al. (2015) Planctomycetes do pos- McBride, M.J., Liu, W., Lu, X. et al. (2014) The family Cytophagaceae.
sess a peptidoglycan cell wall. Nat. Commun. 6: 7116. In: The Prokaryotes. Other Major Lineages of Bacteria and the Archaea.
Kartal, B., de Almeida, N.M., Maalcke, W.J. et al. (2013) How to make a Rosenberg, E., DeLong, E.F., Lory, S. et al. (eds.) Springer Berlin
living from anaerobic ammonium oxidation. FEMS Microbiol. Rev. 37: Heidelberg, Berlin, Heidelberg, pp. 577–593.
428–461. Gupta, R.S. (2014) The phylum Aquificae. In: The Prokaryotes. Other
Reintjes, G., Arnosti, C., Fuchs, B.M. et al. (2017). An alternative poly- Major Lineages of Bacteria and the Archaea. Rosenberg, E., DeLong,
saccharide uptake mechanism of marine bacteria. ISME J. 11:1640. E.F., Lory, S. et al. (eds.) Springer Berlin, Heidelberg, pp. 417–445.
Chapter 5
Marine Archaea

This chapter explores the diversity of the Archaea and their role in marine processes. As
discussed in Chapter 1, the application of 16S rRNA gene sequencing methods first devel-
oped in the 1970s led to the establishment of the Archaea as a distinct domain of life, over-
turning the description of them at the time as an uncommon, specialized subset of bacteria
(“archaebacteria”) found only in extreme habitats. Other information, especially the nature
of their ribosomes and mechanisms of transcription and translation has supported this view.
Although many of the most thermophilic and halophilic microbes are indeed members of
the Archaea, species belonging to this domain are very abundant and diverse, and can be
found everywhere in the marine environment, especially the sub-photic ocean. Here, they
play critical roles in the carbon and nitrogen cycles. Recent advances in DNA sequencing and
genomic methods are revealing an unexpected diversity of novel archaeal lineages, leading
to new ideas about the position of Archaea in the tree of life and the origins of eukaryotes.

Key Concepts
• The basic metabolism and organization of cells of Archaea resembles those of Bacteria,
but there are important differences in membrane structure, replication of DNA, and
protein synthesis.
• Numerous new phyla of the Archaea have been described following discovery in envi-
ronmental gene surveys and this has led to reorganization of phylogenetic trees, pro-
viding support for the evolution of the Eukarya from within the Archaea.
• Production of methane by members of the Euryarchaeota is the final step in the anaero-
bic biodegradation of organic material and leads to formation of massive reserves of
methane in deep-sea sediments.
• Anoxic methane oxidation is carried out by members of the Euryarchaeota by reverse
methanogenesis, in syntrophic association with sulfate-reducing bacteria.
• Numerous types of chemolithotrophic and organotrophic Euryarchaeota and
Crenarchaeota show adaptations for growth at very high temperatures and high salinity.
• Thaumarchaeota are abundant in the deep sea, where their main function is oxidation
of ammonia coupled to chemolithoautotrophy, making a major contribution to dark
carbon fixation.
150 Chapter 5

Several aspects of cell structure and function

i
ARCHAEA, ARCHAEA,
EVERYWHERE distinguish the Archaea and Bacteria
The discovery that Archaea are While the basic organization of the cells of bacteria and archaea is similar, there are some
a major component of ocean important structural and physiological differences. Indeed, some aspects of archaeal biol-
microbiota ranks as one of the ogy show closer similarities to those of the Eukarya than they do to the Bacteria, the
most significant surprises to significance of which will become clear when we consider their possible evolutionary rela-
emerge from the application of tionships. As described in Chapter 3, there are major differences between bacterial and
methods to directly sequence archaeal cell membranes, which contain fatty acid-based lipids and isoprenoid-based lipids,
16S rRNA isolated from plank-
respectively. However, analysis of gene sequences has blurred this distinction; some of the
tonic biomass (see p.42). The first
recently discovered uncultivated bacterial taxa in the Candidate Phyla Radiation (Figure
clues came in 1992, when Jed
Fuhrman (University of Southern 4.1) appear to possess genes encoding the isoprenoid synthesis pathway that is typical of
California) and Ed De Long (then Archaea, while the newly discovered DPANN Archaea (see below) possess genes like those
at Woods Hole Oceanographic of Bacteria. Secondly, although archaeal cells possess small compact genomes like those
Institution) independently dis- in bacteria, organization of the nuclear material is more complex and variable in archaeal
covered archaeal sequences in cells. In some species, packaging and stabilization of the DNA structure is achieved either
seawater samples. Fuhrman et al. by supercoiling (using the enzyme DNA gyrase), as in the Bacteria. However, other species
(1992) found sequences in pelagic possess positively charged proteins known as histones, which have amino-acid sequences
water samples from the Pacific homologous to those found in eukaryotes. Short sequences of DNA are wound around
Ocean (100 and 500 m depth)
clusters of histones to form tetrameric clusters called nucleosomes. Some of the extremely
that were only distantly related to
thermophilic Archaea have an additional type of DNA gyrase that induces positive super-
those of any organisms previously
characterized. The closest match coiling in the DNA to protect it against denaturation. The process of DNA replication
for some of these sequences was is also different from that in the Bacteria. Although most Archaea possess single circu-
to extreme thermophiles, which lar chromosomes, there are often multiple origins of replication, and the structure of the
had been assigned to the new replication complex and polymerase enzymes again resembles that found in eukaryotes.
domain Archaea (at that time, Conversely, regulation of gene expression in the Archaea is very like that in Bacteria,
often still referred to as archae- involving the use of repressor or activator proteins to control the level of transcription.
bacteria). DeLong (1992) also Translation of mRNA and protein synthesis in the Archaea shares many features with the
discovered widespread archaeal Bacteria, but many of the component proteins of Archaea and Eukarya show closer homol-
rRNA sequences in coastal surface ogy than they do with those from Bacteria.
waters off the east and west coasts
of North America. Before these
discoveries, it was assumed that New phylogenomic methods have led to
Archaea are only found in extreme recognition of multiple phyla of the Archaea
environments, inhospitable to
other life forms. Within a few years, As mentioned, the group we now know as the Archaea were first discovered in extreme
numerous studies had established environments, including hot springs, hydrothermal vents, high salinity, and low pH habi-
these currently uncultivated tats. With the right conditions, it has proved relatively straightforward to bring these into
Archaea are highly abundant in culture, but almost none of the diverse archaeal taxa that occur in the oceans have been
surface and deep surface waters in cultured. The application of 16S rRNA gene sequencing in the 1980s led to the early recogni-
all the major ocean basins. tion of two major branches (phyla) of the phylogenetic tree, namely the Euryarchaeota and
Crenarchaeota. Later, the new phylum Thaumarchaeota was designated, and many former
members of the Crenarchaeota have now been assigned to this. Despite the recognition of
their abundance in the marine environment, we still know much less about the diversity
of the Archaea than Bacteria. The number of researchers working with archaea is much
smaller, there is a much shorter history of attempts to cultivate them, and environmental
16S rRNA gene surveys have been heavily biased towards bacterial sequences. Based on
16S rRNA sequences from marine habitats, the dominant phyla are the Thaumarchaeota and
Euryarchaeota (Figure 5.1). As noted in previous chapters, despite the undoubted usefulness
of 16S rRNA gene sequencing, reliance on a single gene to infer phylogenetic relationships
can lead to problems. As we saw in the discussion of bacterial diversity and evolutionary
relationships in Chapter 4, there have been major advances in phylogenetic and phyloge-
nomic analysis, which has been extended by the study of other conserved key genes, such
as those encoding proteins involved with transcription and translation, as well as compari-
son of amplified genome sequences with genomes of cultivated representatives of different
taxa. Over 2000 complete or draft archaeal genomes are now available—either from cultured
species or metagenome assemblies—and the rate of discovery continues to increase. Some
examples are shown in Table 3.1. This has led to sequential changes in the main branch of
the tree of life that includes the Archaea and Eukarya, as shown in Figures 4.1 and 5.2. The
current view (Figure 5.2d) shows four major clades of Archaea—designated as the phy-
lum Euryarchaeota, plus the TACK, Asgard, and DPANN groupings (“superphyla”). Each of
 Marine Archaea 151

Figure 5.1 Frequency of sequences

for archaeal phyla from samples of
marine water, sediments, and hydro-
thermal vents. Data obtained from
Schloss et al. (2016).

these contains several distinct phyla and the number of these—and their constituent classes
and orders—is increasing rapidly with the recovery of more sequences from the marine envi-
ronment and other habitats, such as deep sediments, hot and acidic springs, soil, and animal
digestive tracts. These discoveries have important implications for theories on the origin of
eukaryotes, discussed in Box 5.1.
COULD GLOBAL
It is suspected that many of these new species may exist in close associations with other
bacteria or archaea and may form syntrophic partnerships. As our knowledge of archaeal
? WARMING TRIGGER
RELEASE OF ANCIENT
diversity grows, it is likely that a wide range of novel metabolic functions will be uncovered, METHANE?
but the overwhelming importance of archaea in marine environments lies in their activities in About 500–600 Mt of methane
the production and oxidation of methane and in autotrophy linked to nitrification, discussed enters the atmosphere each year.
below. Most comes from the “burps” of
ruminant animals, natural wet-
lands, peat bogs, paddy fields, coal
PHYLUM EURYARCHAEOTA mining, and extraction of oil and
gas. Emissions from ocean Archaea
Many members of the Euryarchaeota produce methane are only 1–2% of the total, because
most methane is consumed by
Production of methane (methanogenesis) is the final step in the anaerobic biodegradation methanotrophs or trapped under
of organic material. Methanogens show high physiological diversity in morphology, cell- pressure in sediments over mil-
wall constituents, and physiology, and are grouped into a number of orders and genera using lennia as solid methane hydrate
these criteria (Table 5.1). Methanogens are mesophilic or thermophilic obligate anaerobes (clathrate). There may be about
and can be isolated using strict anaerobic techniques from a wide range of habitats including ten times as much methane in
anoxic sediments, decomposing material, and the gut of animals. They also occur as endo- clathrate as current natural gas
symbionts of anaerobic protozoa found especially in the hindgut of termites. It is likely that reserves, making it an attractive
shipworms and other marine invertebrates that digest wood and cellulose also harbor ciliates energy source, but the logistics of
extraction from deep marine sedi-
with archaeal endosymbionts. Thermophilic methanogens are also important members of
ments are formidable. Methane is
the microbial community at hydrothermal vents. The key features of methanogenesis as the 25 times more active than CO2 in
final step in decomposition of organic material in anoxic environments were introduced in its greenhouse gas effects. There
Chapter 3. Most genera utilize H2 with CO2 serving as both the oxidant for energy generation is strong geological evidence that
and for incorporation into cellular material using the reactions described on p.87. the mass extinction events of 250
and 56 MYA were due to methane
Although they are strict anaerobes, methanogens can also be found in surface microbial mats release and sudden global warm-
and ocean waters, which can contain high levels of dissolved methane. Presumably, they exist ing. There has been concern that
in the anoxic interior of particles in which oxygen has been depleted by respiratory activity rising sea temperatures in the
of other organisms. Methanogenesis is also significant in deeper waters with upwelling of Arctic could trigger catastrophic
nutrients, where intense heterotrophic oxidation of sinking organic matter leads to oxygen release of ancient methane from
clathrate, but some recent research
depletion.
indicates that this might be less
likely than feared (Ruppel and
The extremely thermophilic archaeon Methanococcus jannaschii has been recognized as Kessler, 2017; Sparrow et al., 2018).
one of the most important members of hydrothermal vent communities and has been studied
152 Chapter 5

BOX 5.1 RESEARCH FOCUS

Evolution of an evolutionary pathway—changing

views of Archaea in the tree of life

Woese and Fox (1977) recognized three distinct primary group- from gene surveys of marine sediments near hydrothermal vents
ings of life on Earth. These ideas were consolidated in the highly in the Arctic Ocean between Norway and Greenland. Based on the
influential paper by Woese et al. (1990)—it has >6500 citations phylogeny of highly conserved protein genes, this soon emerged
to date—proposing the three domains: Bacteria, Archaea, and as the closest link to the Eukarya (Spang et al., 2015), supported
Eukarya (see Chapter 1). There was already considerable accep- by the fact that members of the Lokiarchaeota also contain genes
tance of the endosymbiosis theory, in which the eukaryotic cell encoding a group of eukaryotic signature proteins (ESPs). These
type is believed to have been formed from a chimera of ancestral ideas have not been universally accepted; for example, Nasir et
archaeal cells with ancestral bacterial cells, with strong genetic al. (2016) and Da Cunha et al. (2017) used a different approach to
evidence that mitochondria are descended from alphaproteobac- interpret the phylogeny of Lokiarchaeota proteins and suggested
terial endosymbionts. However, the nature of the archaeal part- that published genome data could be contaminated from distantly
ner has proved much more difficult to establish and promoted related organisms, concluding that they do not “bridge the gap”
much debate. between Archaea and Eukarya. The phylum name was given after
the sampling site, Loki’s Castle black smoker vent, which in turn
Eme et al. (2017) review the changing picture of diversity in the was named after the shape-shifting Norse god, Loki. In the study
domain Archaea since it was first proposed, and how this has led to of Norse mythology, Loki has been the subject of many academic
new ideas about the evolution of the Eukarya. Figure 5.2a shows controversies, so Spang et al. (2015) chose a fitting name for
the three-domain tree of life in which the Archaea and Eukarya an organism at the heart of ongoing debates about the origin of
each represent a monophyletic group sharing a unique common eukaryotes. Spang et al. (2018) gave several reasons why they con-
ancestor to the exclusion of Bacteria. An alternative view of a sidered the claims of Da Cunha et al. to be unfounded. The subse-
two-domain (Bacteria plus “Eocyta”) tree of life (Figure 5.2b) quent discovery and analysis of three related, but separate, phyla,
was proposed early on (Lake et al., 1984; Rivera and Lake, 1992), the Odinarchaeota, Thorarchaeota, and Heimdallarchaeota—
based on evidence from comparison of the structural patterns of they are all named after Norse gods—provided further strong
the ribosomes and the properties of genes encoding elongation support for the origin of eukaryotes within the Asgard group or
factors—these are key cofactors during translation of mRNA to are a sister clade to them (Figure 5.2d). Genes for a wide range
proteins. These authors favored a close relationship between the of ESPs involved in membrane trafficking and production of
Eukarya and the Crenarchaeota. By 2010, the branch containing vesicles—defining features of eukaryotic cells, and previously
the Crenarchaeota was expanded to include three new phyla, fol- thought to be unique to them—led Zaremba-Niedzwiedzka et al.
lowing the discovery of new sequences from metagenomic data, (2017) to conclude that the ancestral host cells involved in initial
forming the TACK superphylum. With this new phylogeny, Guy endosymbiosis with the proteobacterial precursors of the mito-
and Ettema (2011) proposed that the Eukarya branched within, or chondrion already contained many of the key components that
as a sister branch to, this clade (Figure 5.2c). Most recently, another govern eukaryotic cellular complexity. Many unanswered ques-
branch of the Archaea, the phylum Lokiarchaota was discovered tions remain about the sequence of evolutionary events before and

Figure 5.2 A schematic representa-

tion of changes in our understanding
of the relationships between eukary-
otes and archaea over the past 40
years. The DPANN group contains
the Diapherotrites, Parvarchaeota,
Aenigmarchaeota, Nanoarchaeota,
and Nanohaloarchaeota phyla. It is
shown as a monophyletic lineage,
although this is a topic of debate.
Reprinted from Eme et al. (2017) with
permission from Springer Nature.
 Marine Archaea 153

BOX 5.1 RESEARCH FOCUS

after acquisition of the bacterial symbionts that would become work over 12 years, this extremely slow-growing organism—it
the mitochondria and several hypotheses are reviewed by Eme has a lag phase lasting months and only doubles once every 20
et al. (2017). They conclude that further study of Asgard archaea days—was successfully cultivated in a bioreactor and recognized
and deep-branching eukaryotic microbes might reveal even closer through genome sequencing as a member of the Lokiarchaeota.
relatives and younger ancestors of eukaryotes, or perhaps reveal Physiological studies indicate it metabolizes organic acids
intermediate stages in the development of eukaryotic cell compo- through a syntrophic relationship with bacteria that consume the
nents. A recent major breakthrough has emerged from the suc- hydrogen generated. The cells have a remarkable structure, with
cessful cultivation of “Ca. Promethoarchaeum syntrophicum” long tendril-like protrusions. Based on these findings, Imachi et
by Imachi et al. (2019). This organism was recovered from deep al. propose a new “Entangle–Engulf–Enslave” model to explain
marine sediments in 2006, long before the description of DNA the archaeon-alphaproteobacterium symbiosis that led to evolu-
sequences assigned to the Asgard archaea. After painstaking tion of the mitochondria in eukaryotic cells.

extensively. It is a primary consumer of H2 and CO2 produced by geochemical activity at the

vents. Analysis of the complete genome sequence—the first from an archaeon, obtained in
1996—reveals a 1.7-Mb circular chromosome containing about 1700 genes. Sequence analy-
sis of the genes encoding key metabolic pathways and cellular processes shows that these
are similar to those found in Bacteria, whereas those encoding protein synthesis and DNA
replication show more similarity to eukaryotic genes. However, a large number of genes in
M. jannaschii have little homology to bacterial or eukaryotic genes, suggesting that many
new cellular processes remain to be discovered. About 20 other genomes of methanogens
have been sequenced; most are about the same size as M. jannaschii, but the genomes of
Methanosarcina spp. are about four times as big, presumably reflecting more metabolic ver-
satility. The function of many of the putative genes identified in these species is not known.

Methanopyrus is one of the most thermophilic organisms known, with rapid growth (genera-
tion time of 1 h) at a maximum temperature of 110°C. It is found in the walls of black smoker
chimneys in hydrothermal vents. Although it is a methanogen, utilizing H2 and CO2 only, it
is phylogenetically distant from the rest of this group and has very unusual membrane lipids
and thermostabilizing compounds in the cytoplasm.

Anaerobic oxidation of methane (AOM) in

sediments is carried out by syntrophic archaea
As described in Chapter 3, the enigmatic fate of methane produced in marine sediments
was solved by the discovery of anaerobic methanotrophic Archaea (ANME) that carry out
AOM—probably by reverse methanogenesis—in syntrophic consortia with sulfate-reduc-
ing bacteria (SRB), providing fixed nitrogen in return (Figure 3.13). Three clades of SRB-
associated ANME archaea have been identified. The ANME-1 and ANME-2 are most widely
distributed in anoxic sediments, while ANME-3 is mainly associated with mud volcanoes.
They are thought to have very low growth rates, and none have been cultivated, but phy-
logenetic analysis shows they are related to the Euryarchaeota orders Methanomicrobiales
(ANME-1) and Methanosarcinales (ANME-2 and ANME-3), and they show morphologi-
cal similarities to the respective groups. In clade ANME-1, there is experimental evidence
showing that syntrophic coupling occurs via direct electron transfer from the ANME archaea
to the SRB, mediated by the production of extracellular cytochromes and/or conductive
nanowires forming cell–cell connections. Genome analysis also shows that the SRB partners
possess genes for multiheme cytochromes that are not present in free-living SRB, showing
adaptations for syntrophy by both partners. Other similarities to methanogens include the
presence of genes for the seven-step methanogenic pathway (in ANME-1 and ANME-2) and
the same CO2 fixation pathway as Methanomicrobiales (ANME-1). Furthermore, ANME-2
fixes N2, as do members of Methanosarcinales. Close examination of the ANME now reveals
that the associations with SRB are more diverse than originally envisaged. Different ANME
archaea associate in a variety of ways with SRB—sometimes in loose associations, some-
times as structured consortia. ANME-1 archaea have been observed to exist as single cells,
154 Chapter 5

Table 5.1 Properties of some representative methanogens found in marine

sediments and hydrothermal vents

Order and genus Morphology Major substrates Optimum

temp. (°C)

Methanococcales

Methanococcus Irregular cocci H2, formate 35–40

Methanothermococcus Cocci H2, formate 60–65

Methanocaldococcus Cocci H2 80–85

Methanotorris Cocci H2 88

Methanomicrobiales

Methanoculleus Irregular cocci H2, formate 20–55

Methanogenium Irregular cocci H2, formate 15–57

Methanolacinia Irregular rods H2 40

Methanospirillum Spirilla H2, formate 30–37

Methanosarcinales

Methanosarcina Irregular cocci H2, methylamines, 35–60

in groups acetate

Methanococcoides Irregular cocci methylamines 23–35

Methanopyrales

Methanopyrus Rods in chains H2 98

without an obvious SRB partner and ANME-2 and ANME-3 types occur with different kinds
of bacteria, sometimes in mixed-species aggregates.

Since the initial discovery, other processes of AOM in marine methane-seep sediments have
been described, in which Mn4+ or Fe3+ can serve as electron acceptors. The mechanisms of
this reaction and the organisms responsible are unclear. AOM linked to reduction of NO3- or
NO3-, has been observed in freshwater, but not marine, sediments.

The class Thermococci contains hyperthermophiles

found at hydrothermal vents
This class contains three genera of hyperthermophiles: Thermococcus, Pyrococcus, and
Paleococcus, each containing a small number of species. As shown in Table 3.2, these genera
of Euryarchaeota have optimum growth temperatures in excess of 80°C. Their phylogenetic
position as a deeply branching group resembles that of the hyperthermophilic genera Aquifex
and Thermotoga in the Bacteria. It is interesting that extreme thermophiles in both domains
branch very near the root of the tree, and this is taken as evidence that these organisms most
closely resemble those that evolved first in the hotter conditions of the early Earth. Indeed, a
common theory is that life may have evolved in submarine hydrothermal vents or subsurface
rocks. The adaptation of hyperthermophiles to life at high temperatures is discussed on p.105.

Thermococcus celer forms highly motile spherical cells about 0.8 µm in diameter. It is an
obligately anaerobic chemoorganotroph, using complex substrates such as proteins and carbo-
hydrates, with sulfur as the electron acceptor with an optimum growth temperature of 80°C.
Pyrococcus furiosus has similar properties to Thermococcus, but has an optimum growth
temperature of 100°C and a maximum of 106°C. Both organisms have been investigated
extensively because of their biotechnological potential, and their thermostable enzymes are
used in the PCR (p.390). The genomes of several Pyrococcus species have been compared,
revealing evidence of significant gene rearrangements in their recent evolution.
 Marine Archaea 155

Archaeoglobus and Ferroglobus are hyperthermophilic

sulfate-reducers and iron-oxidizers
The genus Archaeoglobus contains several species that are also extreme thermophiles (opti-
mum temperature 83°C) found in sediments of shallow hydrothermal vents and around
undersea volcanoes. They are strictly anaerobic organisms, which couple the reduction of
sulfate to the oxidation of H2 and certain organic compounds, resulting in the production
of H2S. Although Archaeoglobus forms a distinct phylogenetic group in the Euryarchaeota,
it makes some of the key coenzymes used in methanogenesis, and analysis of its genome
sequence shows that it shares some genes with the methanogens. However, it lacks the gene
for one of the key enzymes, methyl-CoM reductase, so the origin of the small amount of
methane produced is unknown. Archaeoglobus also occurs in oil reservoirs and has caused
problems with sulfide “souring” of crude oil extracted from the North Sea and Arctic oil-
fields. Members of the related genus Ferroglobus also found in hydrothermal vents, do not
reduce sulfate, but are iron-oxidizing/nitrate-reducing chemolithoautotrophs.

The Euryarchaeota contains extreme halophiles

Extreme halophiles (family Halobacteriaceae) grow in concentrations of NaCl greater than
9% and many can grow in saturated NaCl solutions (35%). They are found in salt lakes
such as the Great Salt Lake (Utah, USA) and the Dead Sea (Jordan Rift Valley); as well as
in hypersaline anoxic basins, such as those in the Mediterranean and Red Seas (see p.22).
Extreme halophiles also occur in coastal regions in solar salterns, which are lagoons where
seawater evaporates so that sea salt can be harvested. These operate as semi-continuous
systems and maintain a fairly constant range of salinity throughout the year. Both conven-
tional microbiological and 16S rRNA methods show that the overall diversity of micro-
organisms decreases as salinity increases. Up to about 11% NaCl, the range of bacteria is
like that found in coastal seawater (most marine bacteria are moderately halophilic) while
archaea are scarce. However, above 15% salinity, culturable members of the Archaea such as
Halorubrum, Halobacterium, Halococcus, Haloarcula, and Haloquadratum become domi-
i
WHAT’S IN A NAME?
nant, as well as archaea with novel gene sequences. The halophilic archaea occur as rods or RHODOPSINS
cocci and in a variety of unusual shapes including flattened squares Haloquadratum walsbyi.
They may contain very large plasmids constituting up to 30% of the genome. They are che- The extremely halophilic archaeon
Halobacterium was given this
moorganotrophs that can use a wide range of compounds as sources of carbon and energy.
name because at the time of its
The requirement for very high Na+ concentrations is achieved by pumping large amounts of discovery it was considered to be
K+ across the membrane, so that the internal osmotic pressure remains high and protects the a bacterium—because we were
cell from dehydration. The intracellular machinery of these organisms is adapted to high lev- ignorant then of the true differ-
els of salt, and so they are obligate extreme halophiles. Cells lyse in the absence of sufficient ences between the domains that
concentrations of Na+, possibly because Na+ stabilizes the high levels of negatively charged we now know as the Archaea and
acidic amino acids in the cell wall. Bacteria, they were known as the
Archaebacteria and Eubacteria
Halobacterium salinarum and some other species use a membrane protein called bacteri- respectively. Therefore, when
orhodopsin to synthesize ATP using light energy. This compound is so called because it is the rhodopsin-like pigment of
structurally similar to the rhodopsin pigment in animal eyes. As noted in Box 3.1, rhodopsins Halobacterium was discovered, it
was called bacteriorhodopsin, to
consist of a protein covalently bound to a carotenoid pigment, retinal, which absorbs light and
distinguish it from the rhodopsin
generates a proton motive force for the production of ATP. Bacteriorhodopsin absorbs light found in animal eyes. Another
to provide energy for the proton pump required to pump Na+ out and K+ into the cell. Like type of rhodopsin was discovered
proteorhodopsin, bacteriorhodopsin can provide enough energy for a low metabolic activ- later, associated with sequences
ity when organic nutrients are scarce. H. salinarum also contains other types of rhodopsin. of Proteobacteria, so it was given
Halorhodopsin captures light energy used for pumping Cl− into the cell to counterbalance K+ the name proteorhodopsin (see
transport, while two other rhodopsin molecules act as light sensors that affect flagellar rota- Box 3.1). However, subsequent
tion, enabling chemotaxis towards light. studies showed that this molecule
is also present in many marine
planktonic archaea as well as
Uncultivated members of the Euryarchaeota bacteria from other phyla! So, with
hindsight, the molecules have been
are abundant in the plankton given inappropriate names. It’s
In 16S rRNA gene surveys, four major groups of Archaea were identified in coastal, confusing, but unfortunately these
open ocean, and Arctic waters. The dominant Marine Group I (MG-I) comprises the names are now too well established
in the literature to change.
phylum Thaumarchaeota (see below), while MG-II, MG-III, and MG-IV affiliate with
156 Chapter 5

the Euryarchaeota. Of the euryarchaeotal groups, all increase in abundance with depth
(Figure 5.3). MG-II is most abundant in surface waters and may make up 4–20% of the
total bacterial and archaeal community. Although no members of these groups have been
cultivated, genome analysis of MG-II shows that some contain proteorhodopsin and other
evidence of a motile, photoheterotrophic lifestyle, capable of degrading polymers in the sur-
face ocean. A number of sub-clades have been identified, associated with different habitats.
Some of the MG-II appear to be particle-associated and these show evidence of genomic
adaptations typical of the copiotrophic lifestyle (p.93). In deeper waters, MG-II (and the rarer
MG-III and MG-IV) archaea consume fresh organic matter and appear to bloom in response
to pulses of organic matter.

PHYLUM CRENARCHAEOTA
Members of the Crenarchaeota are thermophiles
occurring in hydrothermal vents
The phylum Crenarchaeota—part of the TACK superphylum—is phylogenetically very dis-
tinct from the Euryarchaeota described above, although many have similar physiological
properties, including the ability to grow at extremely high temperatures. All cultured repre-
sentatives are thermophilic, and most are hyperthermophilic, growing at >80oC (examples of
minimum, optimum, and maximum temperature ranges are shown in Table 5.2). They were
initially discovered from extensive studies of terrestrial hot springs, but several other spe-
cies occur in submarine hydrothermal vents. They use a wide range of electron donors and
IS THERE AN UPPER acceptors in metabolism and can be either chemoorganotrophic or chemoheterotrophic. Most
? TEMPERATURE
LIMIT FOR LIFE?
are obligate anaerobes. Members of the phylum all relatively closely related; there are just 31
cultivated, named genera belonging to six orders in the single class Thermoprotei.
Kashefi and Lovley (2003) discov-
ered a new archaeon, related to
Several species isolated from shallow and deep-sea hydrothermal areas belong to the order
Pyrolobus and Pyrodictium, from a Desulfurococcales, of which the type genus, Desulfurococcus, is an obligate anaerobe with
hydrothermal vent at the Juan de coccoid cells found in the vicinity of black smoker chimneys at hydrothermal vents. It oxi-
Fuca Ridge in the Pacific, which dizes elemental sulfur using molecular hydrogen. Pyrodictium cells are disc-shaped and
they nicknamed “strain 121” are connected into a mycelium-like layer attached to crystals of sulfur by very fine hol-
because of its ability to grow up to low tubules. Most Pyrodictium spp. are chemolithoautotrophs which gain energy by sulfide
121°C. This is the standard temper- reduction, but P. abyssi grows chemoorganotrophically by fermenting peptides to CO2, H2,
ature that is used in an autoclave and fatty acids. Pyrolobus fumarii is found in the walls of black smoker chimneys in hydro-
for sterilization at atmospheric thermal vents and has an optimum growth temperature of 106°C. In this very extreme envi-
pressure. Strain 121 grows at this
ronment, its production of organic compounds is a significant source of primary productivity.
temperature and remains viable at
Cells are lobed cocci and the cell wall is composed of protein. It is a facultatively aerobic
130°C for over 2 h. In reviewing
the discovery, Cowan (2003) cau- obligate chemolithotroph, using H2 to reduce NO3- (to NH4+) or S2O32- (to H2S).
tions that exact measurement of
growth and survival at these tem- Staphylothermus marinus forms aggregates of cocci and is a chemoorganotroph, which, like
peratures is difficult. Microscopic Pyrodictium abyssi, ferments peptides to CO2, H2, and fatty acids. It is widely distributed in
examination of cross-sections of shallow and deep-sea hydrothermal systems and is a major decomposer of organic material.
the black smoker chimney mate- It is the largest known member of the Archaea—although normally about 1 µm in diameter,
rial from which the archaeon was it can form very large cells up to 15 µm in high-nutrient concentrations. Unusually, cells also
isolated reveals diverse popula- aggregate in clusters of 50–100 cells.
tions of microbes, indicating that
microbes with even higher growth
Igniococcus is an obligate hyperthermophile which has been detected in gene surveys of
temperatures might exist. Such
extreme thermophily depends on
hydrothermal systems with global distribution. Three recognized species have been culti-
protein and membrane stability vated and characterized: Igniococcus hospitalis, I. islandicus, and I. pacificus. Growth of
(see p.105), but Cowan suggests Igniococcus is anaerobic and occurs through a novel CO2 fixation pathway, using sulfur
that the stability of small molec- reduction with molecular hydrogen as an electron donor. They can be cultured anaerobically
ular-weight cofactors probably using specialized high-temperature fermenters at 90°C. Igniococcus is one of only a few
determines the upper limit for archaea known to have a double membrane and to lack an S-layer. The outer membrane is
life. It is very likely that microbes separated from the cytoplasmic membrane by a wide inter-membrane or periplasmic space
that grow and survive even higher (Figure 5.5B). This surrounds the cell as a loose sac enclosing a very large periplasmic space
temperatures remain to be discov- containing vesicles and tubes budding from the cytoplasmic membrane. These migrate and
ered, but some essential biological
fuse to the outer membrane, forming a dynamic endomembrane system that appears to have
molecules such as ATP cannot exist
secretory functions. As discussed below, these features are important aspects in the parasit-
above about 150°C.
ism of I. hospitalis by Nanoarchaeum equitans.
 Marine Archaea 157

In the order Thermoproteales, Pyrobaculum shows various modes of nutrition. Some species
link sulfur reduction to anaerobic respiration of organic compounds, while others are anaero-
bic chemolithoautotrophs using H2 to reduce NO3- or Fe3+.

PHYLUM THAUMARCHAEOTA
A single clade of ammonia-oxidizing archaea
comprises 20% of the picoplankton
One of the most unexpected discoveries arising from the application of ocean sampling of
16S rRNA genes in the 1990s was the discovery of widespread archaeal gene sequences
in the marine water column, including polar oceans and very deep, cold waters. Initially,

(a)

Surface

Archaea Bacteria

1
Photic Figure 5.3 Distribution of Bacteria and
zone Archaea in the ocean, showing mean
annual depth profiles in the North Pacific
Depth (m)

10
subtropical gyre. Total cell abundance
was measured using the DAPI nucleic acid
stain and Bacteria and Archaea measured
100 using whole-cell rRNA-targeted FISH with
fluorescein-labeled polynucleotide probes.
The counts of Archaea dominated the
picoplankton below 1000 m and were
1000
almost entirely due to the clade now
known as Thaumarchaeota; counts due to
5000 the Euryarchaeota were a few percent of
the total throughout the water column.
0 20 40 60 80 100
Redrawn from Karner et al. (2001) with
(b) % of total microbial count
permission from Springer Nature.
158 Chapter 5

Table 5.2 Growth conditions of hyperthermophilic marine Archaea and

Bacteria

Species Growth temperature (°C) Nutrition Anaerobic/

Aerobic
Min. Opt. Max.

Archaea

Archeoglobus fulgidus 60 83 95 CL An

Ferroglobus placidus 65 85 95 CL An

Igniococcus sp. 65 90 103 CL An

Methanocaldococcus 46 86 91 CL An
jannaschii

Methanococcus igneus 45 88 91 CL An

Methanopyrus kandleri 84 98 110 CL An

Pyrobaculum aerophilum 75 100 104 CL Ae/An

Pyrococcus furiosus 70 100 105 CO An

Pyrodictium occultum 82 105 110 CL An

Pyrolobus fumarii 90 106 113 CL An

TINY CELLS…TINY
i GENOME…ANCIENT
PARASITE?
Staphylothermus marinus 65 92 98 CO An

Thermococcus celer 75 87 93 CO An
The minute cell size and minimal
genome of N. equitans suggest that “Strain 121” 85 106 121 CL An
it is close to the limits for cellular life.
Bacteria
Waters et al. (2003) showed that
it contains only 536 genes, about Aquifex pyrophilus 67 85 95 CL Ae
one-third of which have no known
homologs. Genes for components Thermotoga maritima 55 80 90 CO An
of many metabolic pathways and
Abbreviations: Min., minimum; Opt., optimum; Max., maximum; CL, chemolithotrophic; CO,
energy generation are missing. chemoorganotrophic; An, anaerobic; Ae, aerobic. Data from Huber and Stetter, 1998; Kashefi and
Loss of genes during evolution is Lovley, 2003.
common in obligate endosymbi-
onts and parasites (see Chapter 10),
but Waters et al. concluded that analysis of these psychrophilic archaea (MG-1) identified them as members of the phylum
Nanoarchaeum is a genomically sta- Crenarchaeota, and they were classified as such until 2008. At this time, crenarchaeotes
ble parasite that diverged anciently were known only as extreme thermophiles, and the discovery that these as yet uncultivated
from an archaeal lineage. As an
microbes are ubiquitous and so abundant in the water column was a great surprise.
ectoparasite, Nanoarchaeum is not
genetically isolated from the envi-
ronment and it can acquire genes by One of the early changes in the phylogeny and classification of the archaea resulting from phy-
HGT, so Nicks and Rahn-Lee (2017) logenomic methods was the realization that this group of organisms formed a monophyletic
argue that a different model of group distinct from the Crenarchaeota. This followed the analysis of the genome sequence
genome reduction may apply. They of an uncultivated archaeon “Ca. Cenarchaeum symbiosum” found in association with the
propose that the high-temperature cold-water sponge Axinella mexicana discovered off the coast of California. It accounts for
lifestyle promoted genome stream- 65% of the microorganisms associated with the tissues of the sponge and genomic analysis
lining—the Igniococcus genome indicates that it oxidizes ammonia—it may have a symbiotic function by removing nitrog-
is also small, only 1.30 Mb—and enous wastes from the sponge. When trees were calculated using both small and large sub-
reduction in cell size, as observed in unit rRNA sequences and concatenated ribosomal protein sequences, it became clear that
other thermophiles. Concurrently,
this organism was phylogenetically separate from the hyperthermophilic Crenarchaeota. In
the direct membrane association of
the two thermophilic archaea led
2008, “Cenarchaeum” and the uncultivated mesophilic/psychrophilic ammonia-oxidizing
to sharing of metabolites, result- archaea (AOA) identified in marine waters were therefore moved to the new sister phylum
ing in loss of biosynthetic genes Thaumarchaeota (the name means “wonder” archaea), to be joined later by the Algarchaeota
in Nanoarchaeum. The evolution and Korarchaota to form the TACK superphylum (Figure 5.1). Recently, a further revision
of smaller parasite cells would be has been implemented under the genome-based taxonomy (GTDB, see p.117) and in this
favored because they place lower system, the AOA are now classified again as members of the phylum Crenarchaeota and
metabolic demands on the host that assigned to a new class Nitrosophaeria. (At the time of writing, this change has not been
they inhabit. formally accepted under the Code of Nomenclature).
 Marine Archaea 159

Table 5.3 Properties of genomes from cultivated ammonia-oxidizing Thaumarchaeota

Species1 Source GC (%) Size (Mb) Proteins Growth Ref.3

(enrichment culture) temp.
(o C)2

Nitrosopumilus

N. maritimus Tropical aquarium tank 34.2 1.65 1795 15–35 (32) (1)

N. cobalaminigenes Surface coastal water (50 m) 33.0 1.56 1817 10–30 (25) (1)

N. oxyclinae Surface coastal water (17 m oxycline) 33.1 1.59 1794 4–30 (33) (1)

N. ureiphilus Near-shore coastal sediment 33.4 2.17 2479 15–40 (25) (1)

“N. koreensis” Arctic marine sediment 34.2 1.64 1890 n.d. (2)

“N. sediminis” Arctic marine sediment 33.6 1.69 1949 n.d. (3)

N. adriaticus Surface coastal water 33.4 1.80 2037 15–34 (30) (4)

N. piranensis Surface coastal water 33.8 1.70 2161 15–34 (32) (4)

“N. salaria” Estuarine sediment 36.8 1.27 1665 n.d. (5)

“Nitrosopelagicus”

“N. brevis” Surface ocean (oligotrophic) 33.2 1.23 1419 15–34 (22) (6)

“Nitrosocaldus”

“N. islandicus” 41.5 1.61 1641 50–70 (65) (7)

1 Genus and species names in quote marks indicate Candidatus status. 2 Growth temperature range, optimum in parentheses. 3 References:
(1) Qin et al. (2017); (2) Park et al. (2012a); (3) Park et al. (2012b); (4) Bayer et al. (2019); (5) Mosier et al. (2012); (6) Santoro et al. (2015);
(7) Daebeler et al. (2018). Additional data courtesy of Wei Qin, University of Washington.

As shown in Figure 5.3, archaeal cells (mostly members of the phylum now known as
Thaumarchaeota) comprise a large fraction of the picoplankton, especially in deeper
waters. The total number of bacteria and archaea decreases with depth, from 105 –10 6 ml−1
near the surface to 103 –105 cells mL −1 below 1000 m. Bacteria are most prevalent in the
upper 150 m of the ocean, but below this depth the fraction of archaea equals or exceeds
that of bacteria. The pattern is consistent throughout the year. Combining the figures for
cell density with oceanographic data for the volume of water at different depths indicates
that the world’s oceans contain approximately 1.3 × 028 archaea and 3.1 × 1028 bacteria.
About 1.0 × 1028cells—20% of all the picoplankton—is dominated by a single clade of
the pelagic Thaumarchaeota, suggesting that a common adaptive strategy has allowed
them to radiate through the entire water column of the oceans. The physiology of these
uncultured psychrophilic archaea was completely unknown until a combination of meta-
bolic experiments and metagenomic evidence, together with study of the physiology of
cultured isolates (Table 5.3) showed that they are mostly chemolithotrophic autotrophs
that obtain energy from ammonia oxidation, therefore being the major contributor to
nitrification and carbon fixation in the deep, cold, and dark ocean. These discoveries are
discussed in Box 5.2.
160 Chapter 5

BOX 5.2 RESEARCH FOCUS

Deep-sea FISHing—discovery of the role of archaea

in carbon and nitrogen cycling

What are deep-water archaea living on? The MG-1 group three subunits, encoded by amoA, amoB, and amoC, and archaeal
of archaea, originally identified as members of the phylum homologs of all three genes have subsequently been discovered in
Crenarchaeota and subsequently reassigned to the Thaumarchaeota, numerous metagenomic datasets, implicating the ocean archaea in
is the most abundant microbial group in the mesopelagic water col- nitrification by ammonia oxidation. But further proof was difficult,
umn (Figure 5.3). The first clues that they might be autotrophs were because no mesophilic/psychrophilic oxidizing archaea had been
indicated by stable isotope analysis (p.58) of sedimentary depos- cultivated. The breakthrough came when Könneke et al. (2005)
its laid down during the mid-Cretaceous period about 112 MYA succeeded in cultivating an ammonia-oxidizing archaeon (AOA)
(Kuypers et al., 2001). “Fossilized” cells of archaea could be identi- from enrichment cultures prepared from seawater supplemented
fied by the presence of specific biphytanyl membrane lipids charac- with ammonium chloride and gravel from a tropical aquarium
teristic of the Crenarchaeota. The organic matter in these archaeal tank. Laboratory experiments showed that cultures of the organ-
cells contained a higher ratio of 12C to 13C, indicating that they had ism, which they named “Ca. Nitrosopumilus maritimus” strain
fixed CO2 by enzymic processes. Wuchter et al. (2003) detected SCM1, were exceptionally efficient at utilizing low levels of ammo-
these membrane lipids and archaeal 16S rRNA sequences in sea nia. Further evidence for the role of AOA-based autotrophy in
water. Using in situ radioactive tracer experiments in dark condi- marine systems was obtained in an enrichment experiment con-
tions, they showed that archaeal cells incorporate large amounts ducted by Wuchter et al. (2006), in which coastal sea water from
of 13C from bicarbonate into the distinctive membrane lipids, thus the North Sea was kept in a mesocosm in darkness for six months
confirming that they are chemolithoautotrophic. Teira et al. (2004) without addition of any nutrients. During this time, the accumula-
investigated this process further by developing a highly sensitive tion of characteristic archaeal membrane lipids was accompanied
CARD-FISH method in which oligonucleotide probes directed by the consumption of ammonia and production of nitrite. CARD-
against specific archaeal groups, and Herndl et al. (2005) coupled FISH showed that the water became enriched with archaea with
this with microautoradiography (MAR-FISH, p.58) to measure 16S rRNA sequences that were very similar to the cultured “Ca.
metabolic activity by incubating seawater samples with radioac- N. maritimus.” In addition, when the amoA gene was isolated
tive [3H]-leucine or [14C]-bicarbonate. After collecting the cells on from the archaeal cells, one dominant sequence closely matched
a membrane filter, the incorporation of the radioactive tracer into those of “Ca. N. maritimus” and the archaeal amoA gene from the
individual cells (tagged by the FISH probe) was visualized micro- Sargasso Sea metagenome. Following analysis of seawater samples
scopically after overlaying with photographic emulsion. Herndl throughout the year and use of Q-PCR (p.47) to detect levels of
et al. concluded that a large fraction of the Crenarchaeota (now expression of archaeal amoA gene—this was 1–3 orders of mag-
reclassified in the Thaumarchaeota) at depths from 200–3000 m in nitude higher than bacterial amoA—the authors concluded that
the North Atlantic Ocean are metabolically active and that carbon nitrification by AOA in the upper 1000 m of the Atlantic system
fixation by their autotrophic growth makes a major contribution to is more important than that carried out by the ammonia-oxidiz-
productivity. ing bacteria (AOB). Besides higher abundance of AOA than AOB,
and higher expression of amoA of AOA, Martens-Habbena et al.
What is the energy source that fuels autotrophy? In the first (2009) showed that “Ca. N. maritimus” is also much more efficient
metagenomic study conducted in the Sargasso Sea by Venter et at nitrification than AOB. They measured the removal of ammo-
al. (2004), a gene encoding ammonia mono-oxygenase (AMO), nia by cultures of this organism, which grew with a doubling time
the key enzyme in ammonia oxidation, was associated with of 26 h, converting ammonia to nitrate until the ammonia levels
archaeal genome fragments. The enzyme AMO is composed of were depleted below the detection limit of 10 nM. This is about 100

Figure 5.4 Efficiency of NH4+

oxidation by Nitrosopumilus
maritimus SCM1. A. Growth in
NH4+-limited seawater batch
culture. B. Specific affinity of
SCM1 (red), AOB (green), and the
highest values for diatoms and
heterotrophic bacteria (grey). For
comparison, the open bar shows
the highest recorded affinity of
an organotrophic organism for its
carbon substrate. Reprinted from
Martens-Habbena et al. (2009)
with permission from Springer
Nature.
 Marine Archaea 161

BOX 5.2 RESEARCH FOCUS

times lower than the concentration needed by ammonia-oxidizing of organic compounds has no effect on growth rates in vitro, while
bacteria. By measuring the kinetics of substrate binding, Martens- others show obligate mixotrophy, requiring organic compounds—
Habbena et al. showed that SCM1 sustains high specific oxidation including urea, a form of reduced nitrogen whose metabolism had
rates at the exceptionally low ammonium concentrations found in previously been predicted from metagenomic datasets. Another
the open oceans. SCM1 has a remarkably high specific affinity for candidate species “Ca. Nitrosopelagicus brevis” isolated from the
NH3/NH4+, suggesting that organisms like this would successfully open ocean has an extremely small genome and small cell size, sug-
compete with heterotrophic bacterioplankton and phytoplankton. gesting that it has undergone similar streamlining to that seen in
This has been borne out by in situ open ocean estimates of commu- oligotrophic bacterioplankton like SAR11 (Santoro et al., 2015: see
nity substrate affinity (Peng et al., 2016). Given the abundance of p.69). Despite this, it has the genetic machinery to make a range
thaumarchaea in deep ocean waters, it is essential to know whether of vitamins and all amino acids. Like N. pumilus, all of these
they are all capable of autotrophic nitrification, as this will affect isolates show high rates of ammonia oxidation and CO2 fixation
our knowledge of carbon cycling as well as nitrogen budgets. The in laboratory experiments, using a highly efficient variant of the
experiments of Herndl et al. (2005) showed that at least some of hydroxypropionate/hydroxybutyrate cycle identified by Könneke
these archaea take up amino acids, suggesting that heterotrophic et al. (2014). Like many other chemolithoautotrophs, the AOA grow
or mixotrophic nutrition is occurring, and this is supported by later slowly despite their comparatively efficient CO2 fixation; most cul-
genomic analysis (Qin et al., 2014). Various studies have shown that tivated strains have laboratory doubling times in the range 19–34 h,
the AOA distribution varies with depth; they are in low and vari- although “Ca. N. brevis” doubles only every 4–4.6 days (Santoro
able abundance at the surface (except in polar seas), reach maximal and Casciotti, 2011).
levels in the mesopelagic, and decrease below that (review, Santoro
et al., 2018). The presence and abundance of different ecotypes of What is the global biogeochemical impact of AOA? The auto-
AOA is strongly affected by the flux of ammonia and dissolved trophic/mixotrophic metabolism of thaumarchaea clearly has great
organic compounds (Figure 5.4). impact on the carbon and nitrogen cycles in the mesopelagic and
deep ocean, particularly in areas with high export of organic mat-
Isolation and cultivation of new species. Despite the advances in ter, which provides the substrates for growth. These processes are
knowledge of metabolic capabilities obtained from metagenomic especially important in supplying the nitrite or nitrate used by
studies, culture of additional strains is needed to fully understand the denitrifying or anammox bacteria in the oxygen-minimum zones
physiological differences of diverse members of the AOA. Several (see Chapter 9). Denitrification leads to production of the green-
strains isolated from soil and wastewater treatment plants have also house gas N2O, and it is possible that this can also be formed as
been described, permitting comparison of genomes and metabolic a by-product of ammonia oxidation. Elucidating full understand-
characteristics. Within the genus Nitrosopumilus, there are consid- ing of planktonic and particle-associated archaeal metabolism and
erable differences in the genes shared by different species and there in situ measurements of rates of growth, carbon fixation, and N2O
are distinct ecotypes showing differences in growth rate, optimum production is an exciting challenge for marine microbiologists and
temperatures, tolerance of oxygen, and ammonium concentrations, biogeochemists. This is especially important in connection with
or other lifestyle characteristics. Several (including strain SCM1) the microbial carbon pump and the capacity of the deep oceans for
have now been fully characterized with phenotypic, genotypic, and long-term storage of carbon, and the effects of ocean hypoxia and
chemotaxonomic criteria and given formal species names (Table acidification due to rising CO2 levels.
5.2). Some appear to be fully autotrophic, and addition of a range

PHYLUM NANOARCHAEOTA
Nanoarchaeum is an obligate parasite
of another archaeon
Nanoarchaeum equitans was discovered as a member of hyperthermophilic microbial
communities near a submarine hot vent off Iceland. In cultures of Igniococcus, groups of
tiny cocci about 400 nm in diameter were seen attached to the surface of its larger cells
(Figure 5.5A). These were isolated, and 16S rRNA sequencing revealed unusual sequences
that could not be matched to known members of the Archaea. Nanoarchaeum was there-
fore assigned to the novel phylum Nanoarchaeota. Its taxonomic position was uncertain for
several years, and the status of this new phylum was questioned, but new phylogenomic
evidence confirms this as a valid phylum within the DPANN supergroup. Nanoarchaeum
appears to be an obligate parasite on the surface of Igniococcus, because all attempts to
grow N. equitans in pure culture have been unsuccessful, even using extracts of I. hospi-
talis. However, the growth rate of Igniococcus does not seem to be strongly affected by
the presence of Nanoarchaeum cells. Genome analysis shows that Nanoarchaeum has no
genes to support the same chemolithoautotrophic physiology as its host. Nanoarchaeum
162 Chapter 5

Figure 5.5 A. SEM showing larger

cells of Igniococcus with attached
cells of Nanoarchaeum equitans. Scale
bar = 300 nm. B. TEM of ultrathin
section showing close attachment of
Nanoarchaeum to the outer mem-
brane of Igniococcus. Credits: A.
Mircea Podar, Oak Ridge National
Laboratory. B. Reprinted from
Cerdeño-Tárraga (2009) with permis-
sion from Springer Nature.

obtains energy and metabolites, including membrane lipids, amino acids, nucleotides, and
cofactors from Igniococcus. As shown in Figure 5.5B, Nanoarchaeum attaches to this
outer membrane via fibrillar structures, and at these contact points the periplasmic space
of Igniococcus is greatly reduced, so that the inner and outer membranes are in contact
with each other. Transcriptional studies of cultured Igniococcus show that the expression
of membrane protein genes increases markedly following attachment of Nanoarchaeum,
suggesting that the parasite exploits these structures. Advanced imaging techniques reveal
direct contact between the cytoplasm parasite cells with the endomembrane system of the
host, allowing Nanoarchaeum to acquire nutrients.

Conclusions
Members of the domain Archaea exhibit a wide variety of morphology, physiology, and
metabolic types. In sediments, they play a major role in the carbon cycle in the production
and oxidation of methane. At hydrothermal vents, chemolithotrophs and organotrophs are
active at very high temperatures owing to the stability of their membranes, proteins, and
intracellular constituents. Although exploration of marine archaea has lagged behind that
of bacteria, there have been dramatic recent advances, revealing unexpected diversity,
with many new phyla described. The discovery of the Thaumarchaeota and methods used
to reveal their physiology, activities, and importance in nitrification and carbon fixation
provides a important reminder of the value of efforts to cultivate new organisms, com-
bined with a range of methodologies based on genomes, isotopes, and advanced imaging.
Such integrated approaches will yield further advances in understanding the properties
and functions of newly discovered archaea, revealing fresh insights into both the early
evolution of life on our planet, and the impacts of human activities some three billion
years later.

Da Cunha, V., Gaia, M., Gadelle, D. et al. (2017) Lokiarchaea are close
References and further reading relatives of Euryarchaeota, not bridging the gap between prokaryotes and
Archaeal phylogeny and evolution eukaryotes. PLoS Genet. 13: e1006810.

Adam, P.S., Borrel, G., Brochier-Armanet, C., and Gribaldo, S. (2017) The Eme, L., Spang, A., Lombard, J. et al. (2017) Archaea and the origin of
growing tree of Archaea: New perspectives on their diversity, evolution eukaryotes. Nat. Rev. Microbiol. 15: 711–723.
and ecology. ISME J. 11: 2407–2425. Guy, L. and Ettema, T.J.G. (2011) The archaeal ‘TACK’ superphylum and
the origin of eukaryotes. Trends Microbiol. 19: 580–587.
 Marine Archaea 163

Imachi, H., Nobu, M.K., Nakahara, N., et al. (2019) Isolation of an Wegener, G., Krukenberg, V., Riedel, D. et al. (2015) Intercellular wiring
archaeon at the prokaryote-eukaryote interface. bioRxiv, 726976. enables electron transfer between methanotrophic archaea and bacteria.
Kuypers, M.M.M., Blokker, P., Erbacher, J. et al. (2001) Massive expan- Nature 526: 587–590.
sion of marine Archaea during a Mid-Cretaceous oceanic anoxic event.
Igniococcus and Nanoarchaeum
Science 293: 92–95.
Lake, J.A., Henderson, E., Oakes, M., and Clark, M.W. (1984) Eocytes: A Cerdeño-Tárraga AM. (2009) Genome watch: What a scorcher! Nat. Rev.
new ribosome structure indicates a kingdom with a close relationship to Microbiol. 7: 408–409.
eukaryotes. Proc. Natl. Acad. Sci. USA 81: 3786–3790. Heimerl, T., Flechsler, J., Pickl, C. et al. (2017) A complex endomembrane
Nasir, A., Kim, K.M., Da Cunha, V., and Caetano-Anollés, G. (2016) system in the archaeon Ignicoccus hospitalis tapped by Nanoarchaeum
Arguments reinforcing the three-domain view of diversified cellular life. equitans. Front. Microbiol. 8: 1072.
Archaea 2016: 1851865. Nicks, T. and Rahn-Lee, L. (2017) Inside out: Archaeal ectosymbionts
Ren, M., Feng, X., Huang, Y. et al. (2019) Phylogenomics suggests oxygen suggest a second model of reduced-genome evolution. Front. Microbiol.
availability as a driving force in Thaumarchaeota evolution. ISME J. 13: 8: 384.
2150–2161. Waters, E., Hohn, M.J., Ahel, I. et al. (2003) The genome of Nanoarchaeum
Rivera, M.C. and Lake, J.A. (1992) Evidence that eukaryotes and eocyte equitans: Insights into early archaeal evolution and derived parasitism.
prokaryotes are immediate relatives. Science 257: 74–76. Proc. Natl. Acad. Sci. 100: 12984–12988.

Schloss, P.D., Girard, R.A., Martin,T. et al. (2016) Status of the archaeal Thaumarchaeota
and bacterial census: An update. MBio 7: e00201-16.
Bayer, B., Vojvoda, J., Reinthaler, T. et al. (2019) Nitrosopumilus adriaticus
Spang, A., Saw, J.H., Jørgensen, S.L. et al. (2015) Complex archaea that sp. nov. and Nitrosopumilus piranensis sp. nov., two ammonia-oxidizing
bridge the gap between prokaryotes and eukaryotes. Nature 521: 173–179. archaea from the Adriatic Sea and members of the class Nitrososphaeria.
Woese, C.R. and Fox, G.E. (1977) Phylogenetic structure of the prokaryotic Int. J. Syst. Evol. Microbiol. 69: 1892–1902.
domain: The primary kingdoms. Proc. Natl. Acad. Sci. 74: 5088–5090. Brochier-Armanet, C., Boussau, B., Gribaldo, S., and Forterre, P. (2008)
Woese, C.R., Kandler, O., and Wheelis, M.L. (1990) Towards a natural Mesophilic crenarchaeota: Proposal for a third archaeal phylum, the
system of organisms: Proposal for the domains Archaea, Bacteria, and Thaumarchaeota. Nat. Rev. Microbiol. 6: 245–252.
Eucarya. Proc. Natl. Acad. Sci. 87: 4576–4579. Brochier-Armanet, C., Gribaldo, S., and Forterre, P. (2011) Spotlight on
Zaremba-Niedzwiedzka, K., Caceres, E.F., Saw, J.H. et al. (2017) Asgard the Thaumarchaeota. ISME J. 6: 227–230.
archaea illuminate the origin of eukaryotic cellular complexity. Nature Daebeler, A., Herbold, C.W., Vierheilig, J. et al. (2018) Cultivation and
541: 353–358. genomic analysis of “Candidatus Nitrosocaldus islandicus,” an obligately
thermophilic, ammonia-oxidizing thaumarchaeon from a hot spring bio-
Euryarchaeota, methanogenesis film in Graendalur Valley, Iceland. Front. Microbiol. 9: 193.
and methanotrophy Herndl, G.J., Reinthaler, T., Teira, E. et al. (2005) Contribution of Archaea
Cui, M., Ma, A., Qi, H. et al. (2015) Anaerobic oxidation of methane: An to total prokaryotic production in the deep Atlantic Ocean. Appl. Environ.
‘active’ microbial process. Microbiol. Open 4: 1–11. Microbiol. 71: 2303–2309.
Evans, P.N., Boyd, J.A., Leu, A.O. et al. (2019) An evolving view of meth- Karner, M.B., DeLong, E.F., and Karl, D.M. (2001) Archaeal dominance
ane metabolism in the Archaea. Nat. Rev. Microbiol. 17: 219–232. in the mesopelagic zone of the Pacific Ocean. Nature 409: 507–510.
Haro-Moreno, J.M., Rodriguez-Valera, F., López-García, P. et al (2017) Könneke, M., Bernhard, A.E., de la Torre, J.R. et al. (2005) Isolation of an
New insights into marine group III Euryarchaeota, from dark to light. autotrophic ammonia-oxidizing marine archaeon. Nature 437: 543–546.
ISME J. 11: 1102–1117. Könneke, M., Schubert, D.M., Brown, P.C. et al. (2014) Ammonia-
Knittel, K., Losekann, T., Boetius, A. et al. Diversity and distribution oxidizing archaea use the most energy-efficient aerobic pathway for CO2
of methanotrophic Archaea at cold seeps. Appl. Environ. Microbiol. 71: fixation. Proc. Natl. Acad. Sci. USA 111: 8239–8244.
467–479. Martens-Habbena, W., Berube, P.M., Urakawa, H. et al. (2009) Ammonia
Martin-Cuadrado, A.-B., Garcia-Heredia, I., Moltó, A.G. et al. (2015) A oxidation kinetics determine niche separation of nitrifying Archaea and
new class of marine Euryarchaeota group II from the mediterranean deep Bacteria. Nature 461: 976–979.
chlorophyll maximum. ISME J. 9: 1619–1634. Mosier, A.C., Allen, E.E., Kim, M. et al. (2012) Genome sequence
McGlynn, S.E., Chadwick, G.L., Kempes, C.P., and Orphan, V.J. (2015) of “Candidatus Nitrosopumilus salaria” BD31, an ammonia-oxidiz-
Single cell activity reveals direct electron transfer in methanotrophic con- ing Archaeon from the San Francisco Bay estuary. J. Bacteriol. 194:
sortia. Nature 526: 531–535. 2121–2122.
Orsi, W.D., Smith, J.M., Wilcox, H.M. et al. (2015) Ecophysiology of Park, S.-J., Kim, J.-G., Jung, M.-Y. et al. (2012a) Draft genome sequence of
uncultivated marine euryarchaea is linked to particulate organic matter. an ammonia-oxidizing archaeon, “Candidatus Nitrosopumilus sediminis”
ISME J. 9: 1747–1763. AR2, from Svalbard in the Arctic Circle. J. Bacteriol. 194: 6948–6949.
Ruppel, C.D. and Kessler, J.D. (2017) The interaction of climate change Park, S.-J., Kim, J.-G., Jung, M.-Y. et al. (2012b) Draft genome sequence of
and methane hydrates. Rev. Geophys. 55: 126–168. an ammonia-oxidizing archaeon, “Candidatus Nitrosopumilus koreensis”
AR1, from marine sediment. J. Bacteriol. 194: 6940–6941.
Skennerton, C.T., Chourey, K., Iyer, R. et al. (2017) Methane-fueled syn-
trophy through extracellular electron transfer: Uncovering the genomic Peng, X., Fuchsman, C.A., Jayakumar, A. et al. (2016) Revisiting nitrifica-
traits conserved within diverse bacterial partners of anaerobic methano- tion in the Eastern Tropical South Pacific: A focus on controls. J. Geophys.
trophic archaea. MBio 8: e00530-17. Res. Ocean. 121: 1667–1684.
Sparrow, K.J., Kessler, J.D., Southon, J.R. et al. (2018) Limited contribu- Qin, W., Amin, S.A., Martens-Habbena, W. et al. (2014) Marine ammonia-
tion of ancient methane to surface waters of the U.S. Beaufort Sea shelf. oxidizing archaeal isolates display obligate mixotrophy and wide ecotypic
Sci. Adv. 4: eaao4842. variation. Proc. Natl. Acad. Sci. USA 111: 12504–12509.
164 Chapter 5

Qin, W., Heal, K.R., Ramdasi, R. et al. (2017) Nitrosopumilus mari- Wuchter, C., Abbas, B., Coolen, M.J.L. et al. (2006) Archaeal nitrification
timus gen. nov., sp. nov., Nitrosopumilus cobalaminigenes sp. nov., in the ocean. Proc. Natl. Acad. Sci. 103: 12317–12322.
Nitrosopumilus oxyclinae sp. nov., and Nitrosopumilus ureiphilus sp. nov., Wuchter, C., Schouten, S., Boschker, H.T.S., and Sinninghe Damste, J.S.
four marine ammonia-oxidizing archaea of the phylum Thaumarchaeota. (2003) Bicarbonate uptake by marine Crenarchaeota. FEMS Microbiol.
Int. J. Syst. Evol. Microbiol. 67: 5067–5079. Lett. 219: 203–207.
Santoro, A.E. and Casciotti, K.L. (2011) Enrichment and characterization
of ammonia-oxidizing archaea from the open ocean: Phylogeny, physiol- Other articles
ogy and stable isotope fractionation. ISME J. 5: 1796–1808.
DeLong, E.F. (1992) Archaea in coastal marine environments. Proc. Natl.
Santoro, A.E., Dupont, C.L., Richter, R.A. et al. (2015) Genomic and Acad. Sci. USA 89: 5685–5689.
proteomic characterization of “Candidatus Nitrosopelagicus brevis”: An
ammonia-oxidizing archaeon from the open ocean. Proc. Natl. Acad. Sci. Fuhrman, J.A., McCallum, K., and Davis, A.A. (1992) novel major archae-
USA 112: 1173–1178. bacterial group from marine plankton. Nature 356: 148–149.

Teira, E., Reinthaler, T., Pernthaler, A. et al. (2004) Combining catalyzed Huber, H. and Stetter, K.O. (1998) Hyperthermophiles and their possible
reporter deposition-fluorescence in situ hybridization and microautora- potential in biotechnology. J. Biotechnol. 64: 39–52.
diography to detect substrate utilization by bacteria and archaea in the Kashefi, K. and Lovley, D.R. (2003) Extending the upper temperature
deep ocean. Appl. Environ. Microbiol. 70: 4411–4414. limit for life. Science 301: 934.
Venter, J.C., Remington, K., Hoffman, J. et al. (2004) Environmental Santoro, A.E., Richter, R.A., and Dupont, C.L. (2019) Planktonic Marine
genome shotgun sequencing of the Sargasso Sea. Science 304: 66–74. Archaea. Ann. Rev. Mar. Sci. 11: 131–158.
Chapter 6
Marine Eukaryotic
Microbes

This chapter reviews the biology of protists and fungi, highlighting selected examples that
are particularly important in marine ecology. The protists are highly diverse and play a major
role in ocean food webs providing trophic links to zooplankton, both as primary producers
in the photic zone and as consumers in all parts of the water column and the benthos. Due
to their genomic complexity and associated technical difficulties, progress in application of
molecular techniques for study of eukaryotic microbes has lagged behind that of bacteria and
archaea, but new methods are revealing exciting new discoveries of hitherto unknown groups
and their importance. The study of fungi (mycology) in the marine environment is consider-
ably less developed than that of other microbes, but there is growing evidence of their impor-
tance in the carbon cycle as benthic and planktonic consumers of complex organic materials.

Key Concepts
• Microbial protists are highly diverse and range in size from <1 µm to >100 µm, with
some types forming macroscopic colonies.
• Classification of protists is complex and has been subject to frequent change, but mod-
ern schemes combine morphological, functional, and molecular information in which
multiple protist groups occur throughout the phylogenetic radiation of the eukaryotes.
• Many planktonic protists contain plastids and carry out photosynthesis, which contrib-
utes a large fraction of global carbon fixation; this primary production enters pelagic
food webs at several levels because of the wide range of sizes of protists.
• Many protists are phagotrophic, feeding by predation (grazing) on bacteria, archaea,
viruses, other protists, and even metazoa; they dominate the initial levels of food webs.
• Classical models of carbon and energy flow based on phototrophy or heterotrophy are
confounded because mixotrophy is extremely common among protists.
• Seasonal blooms of some protists can have major impacts on ocean–atmosphere inter-
actions and affect global climate processes; in other cases, blooms may have deleteri-
ous consequences for marine life and human health.
• Some protists (labyrinthulids and thraustochytrids) and marine fungi are saprotrophs,
important as degraders of organic matter in coastal environments; their abundance and
quantitative importance in carbon cycling has been underestimated.
• The protists and fungi include important parasites and parasites and pathogens of
marine life.
166 Chapter 6

WHAT’S IN A
MARINE PROTISTS
? NAME? PROTISTS,
PROTOCTISTS, Protists are a highly diverse collection of
AND PROTOZOA unicellular eukaryotic microbes
Even in the early nineteenth Today, the term “protist” is widely accepted as a term of convenience to describe a wide
century, biologists recognized that assortment of unicellular eukaryotic microbes. There is no clear definition—in a way, it eas-
there are simple forms of life that ier to say that they are eukaryotic organisms that are not animals, plants, or fungi. The Fungi
did not fit the classical defini- are not normally included in the protists because they form a separate monophyletic king-
tions of “plants” or “animals” and
dom, although the distinctions between this group and some other forms are very blurred. In
attempted to include them in a
some cases, protists may form colonies or filamentous aggregates, but these do not show tis-
classification of life using various
approaches. In 1969, Whittaker’s sue differentiation, so are not regarded as multicellular organisms. The designation “micro-
five-kingdom system for the bial eukaryotes” may also be used as a collective term; this would include members of the
classification of life grouped all Fungi. As members of the domain Eukarya, protists possess a defined nucleus bounded by a
unicellular eukaryotic organisms double nuclear membrane, which is continuous with a membranous system of channels and
into a single kingdom called the vesicles called the endoplasmic reticulum. This is the site of fatty acid synthesis and metabo-
Protista. As discussed in Chapter 1, lism and is also lined with ribosomes responsible for protein synthesis. The Golgi apparatus
the advent of molecular techniques (a series of flattened membrane vesicles) processes proteins for extracellular transport and is
transformed our understanding also responsible for the formation of lysosomes, membrane-bound vesicles containing diges-
of relationships between organ- tive enzymes that fuse with vacuoles in the cell, either for the digestion of food in phagocytic
isms and made it clear that the
vacuoles or for the recycling of damaged cell material.
Protista is not a valid phylogenetic
grouping of organisms; nor is the
Protoctista, which was used in Most protists demonstrate some form of sexual recombination as a regular component of
other classification systems. For their life cycle in which the fusion of the genomes of two cells and their nuclei is followed
many years, the Protozoa (defined by meiosis. Unlike multicellular organisms, which are usually diploid (i.e. they contain two
as a group of animal-like unicellular copies of each chromosome and reproduce by forming haploid gametes), protists vary greatly
organisms) and the Algae (defined in whether the diploid or haploid state is the predominant phase in the life cycle.
as simple plant-like organisms,
both unicellular microalgae, and The cytoskeleton is a system of hollow microtubules about 24 nm in diameter (composed
multicellular seaweeds) were recog- of the protein tubulin) and microfilaments about 7 nm in diameter (composed of two actin
nized as formal taxonomic groups
monofilaments wound around each other). Protists often display a variety of other fila-
and became the province of zoolo-
ments. Microtubules often extend under the surface and provide the basic shape of the cell,
gists and botanists respectively.
Many protists have nutritional or they may be grouped into bundles to support extensions of the cell. Microfilaments pro-
features typical of both plants and vide a strengthening framework for the cell. The cytoskeleton provides a mechanism for
animals, because they can carry movement of intracellular structures within the cell, for example, in the separation of chro-
out photosynthesis or engulf par- mosomes during nuclear division. Another function of the cytoskeleton, which is of par-
ticulate food (phagotrophy), while ticular importance in the marine amoebozoa, radiolarians, and foraminifera, is amoeboid
many are mixotrophic and carry movement due to changes in the cross-linking state of the protein actin. Rearrangement
out both processes. Therefore, the of the membrane also enables the process of phagocytosis, which brings food particles or
terms protists, algae, and protozoa prey organisms into the cell, enclosed within a food vacuole, which subsequently fuses
are not used to designate a formal with lysosomes.
taxonomic status, but they remain
useful to encompass groups of
organisms that share some biologi-
A defining feature of almost all eukaryotes is the presence of mitochondria—organelles
cal and biochemical characteristics. with an outer membrane and an extensively folded inner membrane, which evolved from
the primary endosymbiosis of an alphaproteobacterial ancestor with an ancestral eukaryotic
precursor cell (see Box 5.1). The nature of the infoldings (cristae) is distinctive of different
eukaryotic groups, and marine protists have either tubular cristae (as seen in alveolates and
stramenopiles) or flattened cristae (as seen in euglenids and kinetoplastids). A few types of
protists do not contain mitochondria but have alternative structures (mitosomes or hydro-
genosomes). Most of these are animal parasites, but free-living protists lacking mitochondria
may be important in marine anaerobic sediments, microbial mats, and particles.

Photosynthetic protists contain chloroplasts and are commonly referred to as algae. The
term microalgae is often used to indicate unicellular algae of microscopic dimensions to
distinguish them from larger multicellular types (macroalgae or seaweeds). Much evidence
indicates that the chloroplasts evolved via primary endosymbiosis, in which an ancestral
eukaryotic cell engulfed a cyanobacterium. Analysis of ribosomal RNA, the arrangement of
the internal membranes, and the nature of photosynthetic pigments are used in classification
of photosynthetic protists. Study of these features shows that chloroplasts have arisen on sev-
eral occasions during evolution via secondary and tertiary symbiosis events.
 Marine Eukaryotic Microbes 167

Protists show enormous diversity and

classification systems are regularly revised
As noted in Table 1.1, protistan microbes are found in the pico-, nano-, and microplankton
size classes, ranging in size from ~1 to 200 µm. Early studies of marine protists using light
microscopy revealed a wondrous diversity of forms, with many of the larger organisms in the
microplankton size class showing complex and beautiful forms, such as the specimens from
the 1872–1876 HMS Challenger expedition classified and drawn by Ernst Haeckel, which had
important influences on early twentieth-century art and architecture and continues to inspire
artists and designers (Figure 6.1).

The advent of scanning electron microscopy (SEM) allowed further exploration of these
distinctive features, producing striking images of surface architecture that can be used for
species identification (several examples occur later in this chapter). This enormous morpho-
logical diversity of the protists contrasts with what we have seen in discussion of the bacteria
and archaea, where there are a small number of basic cell shapes, but an enormous metabolic
diversity. That said, morphological features are of little use in distinguishing many species
of the smaller picoeukaryotes. Classification systems, based largely on morphological fea-
tures, were developed over many years by different groups of scientists who used two sepa-
rate International Codes of Nomenclature for naming protozoa (Zoological Code) and algae,
fungi, and plants (Botanical Code) leading to much confusion. Modern classification systems
no longer include these terms as formal categories, although they persist because they are
so embedded in general public understanding and are a convenient way of grouping certain
organisms that share general features (even if they are not particularly closely related).

There have been many attempts to develop phylogenetic trees that depict the relationships
between the many different eukaryotic lineages. Since the 1990s, the application of molecu-
lar methods has led to a fundamental shift in understanding of relationships between organ-
isms. Sequencing of 18S rRNA provided a basis for developing new phylogenetic trees, but,
as noted in Chapter 1, reliance on a single gene can produce many anomalous results, and
evidence of horizontal gene transfer and secondary symbiosis events confounds the inter-
pretation of relationships. By combining the sequences of various genes and amino acid
sequences of key proteins, together with morphological and biochemical evidence, consensus
phylogenetic trees grouping the eukaryotes into a number of supergroups were developed.
The advent of high-throughput sequencing (HTS) has led to many new insights into phy-
logenetic relationships, especially in the higher taxa, resulting in further major revisions of
classifications in 2005, 2012, and 2019. Figure 6.2 is a graphical representation showing the
phylogenetic relationships of the major groups of eukaryotes according to the 2019 formal

Figure 6.1 Artistic drawings of

(a) radiolarians and (b) foraminifera
published as prints by Ernst Haeckel
between 1899 and 1904. From the
collection Kunstformen der Natur
(available online at caliban.mpiz-
koeln.mpg.de).
168 Chapter 6

Figure 6.2 Schematic representa-

tion of the phylogenetic tree of
Eukarya. The branches (supergroups)
are labeled with common names of
major taxa; names in bold indicate
protists and fungi discussed in this
chapter. Adapted from Adl et al.
(2019), CC-BY-4.0.

classification, coordinated by the International Society of Protistologists. Inevitably, there is

still much disagreement among experts about the interpretation of phylogenetic relationships
and classification schemes. In particular, designation of the taxonomic ranks phylum, class,
and order is debatable, so the literature can be quite confusing. The tree indicates that evolu-
tionary radiation occurred very early in the history of eukaryotes. Marine microbes appear in
almost all the higher taxa envisaged in these classification systems. The huge diversity of pro-
tists makes a systematic coverage impossible in a book like this, so we will imagine a visit to
the “Protist Aquarium” like the approach we used in the preceding chapters on Bacteria and
Archaea, in this case grouping organisms under major functional or morphological themes,
rather than attempting to present them in a formal classification scheme.

The -omics approaches have some limitations

for understanding protist diversity
Like bacteria and archaea, most protists have not been cultivated and much of our knowledge
about their diversity in the marine environment has come from direct sequencing of DNA.
However, knowledge of the protists through the application of molecular methods has pro-
gressed much more slowly than for the Bacteria, for which the amount of sequence data is
hundreds of times higher than it is for protists. There are several reasons for this. Most micro-
bial eukaryotes tend to have smaller populations, and most have not been isolated and cul-
tivated. Only a very small number of complete protist genomes are available (Table 6.1); in
these, the genomes are many times bigger and more complex than bacteria and archaea (com-
pare with Table 3.1). Nuclear genomes contain multiple chromosomes, repeat sequences, and
large amounts of noncoding sequences which make interpretation difficult. Methods such
as clone libraries and tag sequencing are often biased in favor of larger cells such as marine
alveolates, including dinoflagellates and ciliates, which may have tens to hundreds of copies
of 18S rRNA genes. Methods of sample preparation often result in contamination with DNA
from fragmented cells of larger organisms. The niche specialization and ecological function
of protists is much harder to explain by the presence of particular genes—unlike bacteria and
archaea, where we can make reliable predictions based on the identification of genes con-
ferring metabolic functions or adaptation to particular environmental conditions like tem-
perature, salinity, or oxygen. Instead, eukaryotes have many complex structural adaptations,
behaviors, and dynamic interactions with their environment and other organisms. Of course,
we can use genomic information to predict that a protist is, say, photosynthetic, but we do not
have any genetic indicators of what makes an organism a grazing predator or an infectious
parasite, for example. Thus, the use of meta-omics approaches to predict ecological function
and interactions in situ is inherently more difficult for protists, although advances are now
being made (Box 6.1). To fully understand protist diversity, we will need a much larger body
of information obtained by combining traditional biological observations with cultivation
 Marine Eukaryotic Microbes 169

Table 6.1 Examples of marine protists for which whole genome sequences have been obtained

Organism Typea Genome Size Predicted

(Mb)b genesb

Ostreococcus tauri Prasinophyte (Mamiellophyceae) Photosynthetic 13.9 7932

Micromonas pusilla Prasinophyte (Mamiellophyceae) 21.9 10242

Photosynthetic

Bodo saltans Euglenid (Kinetoplastea) 39.9 18190

Bacterivorous

Monosiga brevicollis Choanoflagellate (Salpingoecidae) 41.7 9203

Bacterivorous

Schizochytrium sp. Thraustochytrid (Labyrinthulomycetes) 51.6 −

Osmotrophic

Fragilariopsis cylindrus Diatom (Bacillariophyceae) 76.2 18246

Photosynthetic

Emiliania huxleyi Haptophyte coccolithophore (Isochrysidales) 167.7 38544

Photosynthetic

Symbiodinium goreaui (type C1) Dinoflagellate (Suessiales) 1030 43403

Photosynthetic, coral symbiont.
a Common name of group, with class or order in parentheses. b Data from NCBI Genome Database.

and genome-sequencing of a wide range of species in order to establish a reference system

that we can map our newly identified sequences against.

Picoeukaryotes play a major role in ocean food webs

Despite these reservations, environmental genomic approaches have led to some paradigm-
shifting discoveries about the role of protists in marine processes. Most significant of these is
the discovery in 2001 of a hitherto unknown extra level of diversity amongst picoeukaryotes
(1 to 3 µm diameter). In these studies, DNA was extracted from filtered plankton cells and
amplified via PCR using general primers for 18S rRNA genes, followed by construction of
a clone library. Because eukaryotic plankton cells exist across such a broad range of sizes,
from micrometers to millimeters, the method of specimen collection is very important, so
the researchers used a combination of filters and prefilters to trap cells of different size frac-
tions and verified their findings by flow cytometry. Novel organisms, discovered in different
regions and at different depths, were affiliated to many diverse groups containing known,
cultured organisms in higher size ranges, with a range of lifestyles including photoautotrophs
(in the photic zone) and osmotrophic heterotrophs, grazing predators, and parasites through-
out the water column. We now know that these are prevalent in surface waters at up to 104
cells mL –1, decreasing to hundreds or tens of cells in the meso- and bathypelagic depths,
respectively. This has led to a major shift in understanding the ecological role of protists in
ocean food webs, both as primary producers through photosynthesis and as a trophic link
through grazing on other members of the picoplankton. Recent breakthroughs in develop-
ment of methods for analysis of protist communities by single-cell genomics are leading to
new insights into their diversity and role, discussed in Box 6.1.

Heterotrophic flagellated protists play a major

role in grazing of other microbes
Protists that possess one or more flagella for movement and feeding have traditionally been
grouped under the general term “flagellates,” although this characteristic has no phylogenetic
significance and is found as a component at some stage of the life cycle in about half of the
major protistan phyla. Eukaryotic flagella are composed of nine pairs of peripheral micro-
tubules and two pairs of central microtubules. One pair of the tubules in each pair contains
the protein dynein, which moves along an adjacent microtubule; rapid and repeated bending
170 Chapter 6

causes the flagellum to beat. The flagella are anchored in the cell via a basal body (kineto-
some), which varies greatly in its ultrastructure and has been used in previous classification
systems to distinguish different groups.

The introduction of epifluorescence microscopy and flow cytometry in the 1970s to 1980s led
to realization of the abundance and importance of heterotrophic and mixotrophic flagellates
in the nanoplankton (2–20 μm) size range as the major consumers of primary and secondary
production in marine systems. These unicellular organisms swim through the water using
flagella and ingest bacteria, archaea, and smaller protists via phagocytosis. Subsequently, as
discussed above, flagellates were also shown to be abundant in the picoplankton size range.
Grazing protists are also very abundant in benthic habitats such as mats and sediments. They
are also important in sea ice and as members of the surface mucus and intestinal microbiota
of invertebrates. Protistan grazing leads to the recycling of carbon, nitrogen, and phospho-
rus, since the smaller flagellates themselves are preyed upon by larger protists (e.g. ciliates
and dinoflagellates), providing a trophic link to metazoan zooplankton in the food web (the
microbial loop, discussed in Chapter 8). In addition, protists eliminate up to 30% of ingested
cell material that has not been digested—this reenters the DOM/POM pool in the ocean.
The flagellates are taxonomically very diverse, occurring in many distantly related different
phyla. Their ecological and biogeochemical functions are probably also very diverse, and it
is important to remember that grouping them as pico-, nano-, or macroplanktonic flagellates
(often abbreviated to just picoflagellates, nanoflagellates, or microflagellates) is only a con-
venience based on the use of the logarithmic scale for the plankton size classes and the pico-
and nano- prefixes do not have the same meaning as their usual scientific use (1 µm = 103
nm = 106 pm). Furthermore, the distinctions between these categories are blurred, and some
authors consider the microflagellates as organisms larger than 15 µm, rather than 20 µm
used in the Sieburth scale (Table 1.2). Examples of common grazing protists are shown in
Table 6.2 and Figure 6.3. Abundance and biomass vary greatly with seasonal and nutritional
conditions. In temperate regions, abundance increases sharply due to higher numbers of prey
associated with the spring bloom and decrease in early summer as the protists themselves
become subject to intensive grazing by metazoan zooplankton.

Heterotrophic flagellated protists have

different feeding mechanisms
Protists that feed on bacteria (bacterivores) by direct interception often show species-specific
preference for selection and uptake of particular sizes or types of prey and vary in their mode
of feeding. Different species may either be free-swimming or attached to particle such as algal
debris, fecal pellets, and marine snow. As a rule, free-swimming forms tend to consume larger
prey, while attached forms increase the probability of encounter with small prey by generation of
water currents. Examples of genera of heterotrophic nanoflagellates (HNF) are shown in Table
6.2. Some of the most active HNF are the bicosoecids, which are members of the Stramenopiles.
Cafeteria roenbergensis has been found in all ocean waters examined, occurring at densities up

Table 6.2 Examples of marine heterotrophic nano- and microflagellates

Taxonomic groups Examples of common genera Relative abundance

Pelagic Benthic

Euglenids Petulamonas, Peranema + ++

Kinetoplastids Bodo, Caecitellus + +++

Chrysomonads Spumella, Paraphysomonas +++ +

Bicosoecoids Cafeteria, Bicosoeca +++ +

Dinoflagellates Gymnodinium, Katodinium + +

Choanoflagellates Monosiga, Diaphonoeca +++ +

Cercomonads Cercomonas, Bodomorpha + +++

Data from Boenigk and Arndt (2002)

 Marine Eukaryotic Microbes 171

phagocytic vacuole Figure 6.3 Schematic diagram of

Cafeteria roenbergensis, illustrating
containing prey ingestion of prey bacteria propelled
bacteria being towards the mouthparts at the base
digested anterior flagellum of the flagellum by a water current
(red dotted line) generated by the
anterior flagellum. The posterior
flagellum trails during swimming
or may be attached to a surface, as
shown here. Adapted from image by
Dennis Barthel (derivative zapyon)
via Wikimedia Commons, CC-BY-3.0.

nucleus

posterior
flagellum
mitochondria
water current

to 105 mL−1 in coastal waters where there are high concentrations of bacterial prey, and it exerts
strong top-down control on bacterial populations. Its impact on populations of Archaea, which
would be important in the deep sea, is less clear; however, experimental studies have shown
that cultured C. roenbergensis can prey on both bacterial and thaumarchaeotal cells, although
feeding and growth rates vary according to the prey type. The cells, typically 3–10 µm, have
two flagella, one of which projects forwards to propel the cell in a spiral movement, while the
other flagellum trails behind. It is a suspension feeder, filtering suspended bacteria and other
particles from the water; the anterior flagellum beats about 40 times a second, creating a rapid
water current that concentrates suspended particles and carries bacteria towards a “mouthparts”
region at the base of the flagellum, where they are phagocytosed (Figure 6.3). This heterokont
pattern of a “tinsel” anterior flagellum and a smooth posterior, “whiplash” flagellum is a dis-
tinguishing feature of the motile stage in most stramenopiles. In nonmotile species, the poste-
rior flagellum attaches to the substrate. The cells have one nucleus and five mitochondria with
WHAT’S IN A
tubular cristae, as found in all stramenopiles. Unlike most eukaryotes, Cafeteria has a highly
compact mitochondrial genome, with a very small amount of noncoding DNA. A related spe-
cies, Pseudobodo tremulans, is found mainly in sediments.
i NAME? CAFETERIA
ROENBERGENSIS
Species names are often complex,
The group known as the euglenids provided one of the earliest challenges to traditional “zoo-” and students struggle with pro-
and “phyto-” based classification, since their nutrition may be phototrophic or heterotrophic, nouncing and retaining them, so it
or combine features of both (mixotrophy). Thus, past taxonomic systems have separated them always helps if the name has some
at phylum level on the basis that they are simple animals or plants respectively. They were memorable feature. Humor is rarely
classified as one of the main divisions of the supergroup known as the Excavata, although this used in the serious business of
systematics (and scientific writing
is not now recognized as a monophyletic group. In modern systems, the phylum Euglenozoa
generally), but the marine ecolo-
and class Euglenida encompasses several orders of phagotrophic and mixotrophic types, as
gist Tom Fenchel and taxonomist
well as photosynthetic types (Euglenophycae). Also within the excavates, the kinetoplastids David Patterson provide an inter-
possess a unique structure termed the kinetoplast near the base of the flagellum, easily vis- esting exception when they named
ible in the light microscope after applying DNA stains. This is a large, single mitochon- a new species of planktonic protist
drion containing a few interlocked large, circular chromosomes (maxicircles), which encode (Fenchel and Patterson, 1988).
mitochondrial enzymes, plus thousands of smaller minicircles of DNA. The minicircles do They wrote: “We found a new
not appear to encode any complete genes but are involved in post-transcriptional editing of species of ciliate during a marine
mRNA. The bodonids (e.g. Bodo spp.) are kinetoplastids with oval cells about 4–10 μm long field course in Rønberg [Denmark]
and are very common in coastal waters and sediments, often moving over surfaces using a and named it Cafeteria roenber-
trailing flagellum. A shorter flagellum propels water currents toward a cytopharynx lined gensis because of its voracious and
indiscriminate appetite, after many
with microtubules. The group of Euglenozoa protists known as the marine diplonemids has
dinner discussions in the local
recently been recognized as being one of the most diverse and abundant lineages of hetero- cafeteria.”
trophic protists in the oceans.
172 Chapter 6

Figure 6.4 A. TEM of the cho-

anoflagellate Kakoeca antarctica,
showing the basket-like lorica.
B. Schematic diagrams based
on 3D reconstruction of serial
ultrathin TEM, showing similar
structure of choanoflagellate
(left) and sponge (right) cells, but
with differences in organization
of membrane vesicles and other
cellular components. Credits: A.
Fiona Scott, Electron Microscopy
Unit, Australian Antarctic Division,
© Commonwealth of Australia. B.
Davis Laundon, Marine Biological
Association, Plymouth. (See
Laundon et al., 2019).

Larger (>20 µm) predatory protists, described as heterotrophic microflagellates (HMF),

include dinoflagellates, ciliates, and choanoflagellates, which are also very active grazers of
bacteria and other protists; they are discussed below. These are especially important in the
photic zone, where they are responsible for large-scale transfer of primary production from
the phytoplankton in the microbial loop. In the deep sea also, protistan grazing is estimated
to account for a daily turnover of up to 30% of the bacterial and archaeal cell stock.

Many protists are mixotrophic

The ability to combine the processes of feeding by phagotrophy with photosynthesis is
extremely common in marine protists, found in diverse taxa throughout the photic zone and
across the eukaryotic tree of life. Some organisms possess plastids permanently and supple-
ment their photosynthetic metabolism by ingesting bacteria or other microbes. Endosymbiosis,
involving extensive gene transfer and reorganization of cellular functions, occurred early in
the evolution of eukaryotes, leading to the presence of organelles in several groups. So, it is
not really surprising that many protists have retained the capacity for dual modes of nutri-
tion. However, many mixotrophs may not permanently contain photosynthetic ability, but can
“steal” it from phototrophs that they have ingested. This acquired phototrophy by transient
retention of functional algae or their plastids for photosynthesis is known as kleptoplastidy.
Usually, algal cells are broken up and digested in the food vacuole following phagocytosis. It
is not known how the plastids remain intact and functional while the rest of the prey cell is
digested. Many more persistent, facultative associations occur in other dinoflagellates, ciliates,
foraminifera, and radiolarians. Kleptoplastidy does not seem to involve extensive adaptations
to accommodate foreign cells or plastids—the hosts receive the benefit of additional carbon
with few “upkeep” costs. This might explain the development of features such as rapid motil-
ity, complex surface structures, or food storage systems in mixotrophs. Acquired phototrophy
is usually overlooked in ecological and biogeochemical models, and understanding the pro-
cess is important to explain how aquatic ecosystems respond to changing environments.

The choanoflagellates have a unique

morphology and feeding mechanism
About 100 species have been described from marine and estuarine waters throughout the
world. The choanoflagellates resemble the feeding cells found in sponges and are of special
 Marine Eukaryotic Microbes 173

interest to evolutionary biologists as they are phylogenetically the closest unicellular relatives
CHOANOCYTES
to the obligately multicellular animals (members of the Holozoa, as shown in Figure 6.2).
The choanoflagellates show strong similarities to the feeding cells (choanocytes) of sponges i PROVIDE INSIGHTS
INTO ANIMAL
(Figure 6.4). Study of the choanoflagellate cell organization is providing important insights
EVOLUTION
into the origins of multicellularity. The cells are ovoid or spherical, about 3–10 um diameter,
with a single flagellum that propels free-swimming cells though the water and draws a cur- Phylogenetic trees and comparison
rent of water through a funnel-shaped collar of 30–40 microvilli around the top part of the of genome sequences lead to the
cell, where they are trapped and ingested by the cell. Bacteria are trapped and taken into inference that the last common
food vacuoles in the cell. Some choanoflagellates produce a delicate basket-like shell (lorica) ancestor of multicellular animals
would have been choanoflagellate-
composed of silica around the cell (Figure 6.4).
like. The actin-rich collar cells sur-
rounding the whip-like flagellum
Dinoflagellates have several critical roles in marine systems are conserved across almost all ani-
mal phyla. In sponges, the closet
The most distinctive feature of the dinoflagellates is the presence of a transverse flagellum relative of choanoflagellates, they
that encircles the body in a groove, and a longitudinal flagellum that extends to the posterior are used for feeding, while in other
of the cell. This gives rise to a distinctive spinning motion during swimming, from which animals they are adapted for sen-
the name dinoflagellate is derived (from the Greek dinein, “to whirl”). Another distinctive sory and osmoregulatory functions.
feature of their cell structure is the presence of a layer of vesicles (alveoli) underlying the cell To investigate the origins of multi-
cellularity and cell differentiation,
membrane. In many species of dinoflagellates, these alveoli contain cellulose, which forms
Davis Laundon and colleagues at
a protective system of plates that fits together like a suit of armor. There are numerous forms the Marine Biological Association,
of such armored dinoflagellates, mainly in the genera Dinophysis (200 marine species) and Plymouth and University of
Protoperidinium (280 marine species), often with unusual morphology with spines, spikes, California, Berkeley used TEM
or horn-like projections. The geometry of the arrangement of plates is an important factor of ultrathin sections to gener-
in identification. Many other naked or unarmored dinoflagellates also occur, including those ate 3D reconstructions of cells of
that resemble species in the genera Gymnodinium, Gyrodinium, and Katodinium. Most dino- Salpingoeca rosetta (Laundon et al.,
flagellates are in the 20–200 μm microplankton size range; however, several mixotrophic 2019). This choanoflagellate has
genera have been identified that appear to be below 20 μm, while Noctiluca can reach a size a complex life cycle, transitioning
of 2 mm. between single and colonial collar
cell types. This revealed morpho-
logically and positionally distinct
One of the most striking features of the dinoflagellates is their possession of huge genomes,
populations of intracellular vesicles
which range is size from 1.2 to 272 Gb (for comparison, the human genome is 3.2 Gb). The and important differences between
huge diversity in genome size in different species cannot be explained in terms of differences single-celled and multicellular
in cell size or function. Genomes appear to contain large regions of repeated and interspersed choanoflagellates in structures
noncoding DNA sequences and there is evidence of extensive horizontal gene transfer. Most connected with cellular energet-
of the plastid genes have been transferred from the organelles to the nuclear genome. Unlike ics, membrane trafficking, and cell
other eukaryotes, the DNA of dinoflagellates occurs in a highly condensed form and most morphology (Figure 6.4). The dis-
genes are not subject to transcriptional regulation; instead a process of mRNA editing occurs. covery of cell-cell connections and
The technical difficulties of working with such large genomes means that genetic informa- cell differentiation in multicellular
tion about the dinoflagellates is limited. Only a few genomes have been sequenced, from spe- choanoflagellates reveals important
insights into the early evolution of
cies with relatively small genomes. including the coral symbiont Symbiodinium (Table 6.1).
the animals.
This has revealed the presence of gene families related to establishment and maintenance of
symbiosis (see Chapter 10).

At least 2000 dinoflagellate species are known, of which about 90% are marine, with dif-
ferent species adapted to life in many habitats per depth, temperature, and salinity. They
occur as benthic stages associated with sediments, corals, seaweeds, and other surfaces and
as free-floating members of the plankton. Representative images are shown in Figure 6.6.
Photosynthetic dinoflagellates contain chlorophylls a and c, plus carotenoids and xantho-
phylls, and make a significant contribution to carbon dioxide fixation and primary productiv-
ity in the oceans. This is especially true in coastal waters in higher latitudes, where strong
seasonality is affected by light, temperature, reduction in wind-driven turbulence, and input
of fresh water from melting ice promoting spring blooms of dinoflagellates and diatoms.

The xanthophyll pigments confer the golden-brown color typical of many photosynthetic dino-
flagellates and gives rise to the name zooxanthellae, used to describe the Symbiodiniaceae
dinoflagellates that form symbioses in invertebrates such as corals, anemones and clams (see
Chapter 10). Endosymbiotic interactions also occur within other protists, such as ciliates,
foraminiferans, and colonial radiolarians.
174 Chapter 6

BOX 6.1 RESEARCH FOCUS

Sailing towards a better understanding of heterotrophic ocean protists

As noted in the main text, our understanding of eukaryotic ocean Development of single-cell genomics. Unlike the advances made
microbes through the application of molecular methods in the with bacteria and archaea, an understanding of ecological roles of
oceans has been somewhat slower to emerge than knowledge of bac- eukaryotic microbes though analysis of genome sequencing has
terial and archaea. Among the many exciting discoveries emerging been impossible until the development of single-cell genomics
from analysis of samples from the Tara Oceans expeditions (p.55) (SCG). These methods are still in their infancy, but the problems of
are significant recent breakthroughs in our understanding of protist highly fragmented eukaryotic genomes are now being solved by co-
diversity, functions, and importance, following the development of assembly of single amplified genomes (SAGs) from several cells of
improved sampling methods and advanced molecular techniques. the same population and integration of SCG with metagenomic and
metatranscriptomic sequence data. Sieracki et al. (2019) analyzed
Diversity studies using rRNA gene sequencing. For example, de 900 SAGs from Tara surface seawater samples collected across the
Vargas et al. (2015) analyzed 18S rRNA gene sequences from 334 Indian Ocean and Mediterranean Sea. They used improved methods
photic zone samples identifying ~150,000 OTUs in different size of sample preservation and flow cytometry (FC) to sort cells based on
fractions, of which most had no close match in existing databases and chlorophyll fluorescence into those with plastids (likely phototrophs
about a third could not even be assigned to any of the known higher or mixotrophs) and those without plastids (likely heterotrophs). The
eukaryotic taxa. The greatest diversity occurred among heterotrophic taxonomic diversity of cells <5 µm in just hundreds of microliters
protists, particularly those from groups known to include predators, of water matched that found in earlier surveys filtering many liters
parasites, or symbiotic hosts—suggesting the importance of complex of samples. Sieracki et al. found 22 major taxa of aplastidic types,
ecosystems that are regulated mainly by interactions between organ- including Mamiellophyceae, Pelagophyceae, and Prymnesiophyceae,
isms rather than competition for resources or space. In another study, with varying relative abundance. Twelve major groups of aplastidic
Gimmler et al. (2016) analyzed >6 × 106 18S rRNA gene amplicons cells were found, dominated by marine stramenopiles (MAST),
assigned to the phylum Ciliophora from surface and mesopelagic Chrysophyceae. Bicosoecida and marine alveolates (MALV).
Tara samples. This study found a relatively low diversity of ciliates, Gawryluk et al. (2016) linked SCG with microscopy to describe the
with ~1300 OTUs, of which most were globally distributed. Ciliates diversity and abundance of marine diplonemids. SCGs generated
are highly abundant in the oceans and more is known about them from the Tara samples also led to new insights into the function of
than other groups because they have been studied microscopically MALVs, confirming their importance as parasites of a wide range
for many years, due to their larger size. Even so, more than half of the of hosts, including other protists, as well as diverse trophic strate-
OTUs were distantly related to previously known groups. gies (Strassert et al., 2018). Network analysis of metagenomic data

Figure 6.5 Analysis of functional

genes from marine heterotro-
phic SAG lineages shows that
they form a functional group
distinct from autotrophs and
other heterotrophs. Non-metric
multidimensional scaling (NMDS)
projection of a Bray–Curtis
distance matrix showing Pfam
motif occurrences in various
stramenopile genomes. Reprinted
from Seeleuthner et al. (2018),
CC-BY-4.0.
 Marine Eukaryotic Microbes 175

BOX 6.1 RESEARCH FOCUS

by Lima-Mendez et al. (2015) identified multiple interactions among distant (Figure 6.5). One interesting observation concerned genes
grazers, primary producers, symbionts, and parasites. encoding the flagellar protein components, a key gene that showed
signs of relatively recent loss from some MAST-3 members. Perhaps
Seeleuthner et al. (2018) assembled 900 SAGs from 40 single-cell these have adapted to a sessile or parasitic lifestyle. Some MAST-4
representatives of FC-sorted heterotrophic protists in the 0.8–5.0 lineages possessed highly expressed proteorhodopsin genes indicat-
um size range. Before genome assembly, steps were taken to remove ing that they may have photoheterotrophic capacity (see Box 3.1).
sequences from chloroplasts or mitochondria and more distantly Seeleuthner et al. found that all SAGs showed a wide range of car-
related organisms to distinguish members of three stramenopile bohydrate-active enzymes, but the gene content of chrysophytes and
clades—chrysophytes, MAST-3, and MAST-4—known to be abun- some MAST types suggested that they may not be bacterivorous. The
dant in the pico- and nanoplankton. From the 40 SAGs sequenced predicted enzyme profile of many of these protists suggested that
and annotated, Seeleuthner et al. used bioinformatics software to they also degrade chitin-containing organisms like fungi, diatoms,
predict the functional repertoires of these uncultured stramenopiles and crustaceans. Although the authors acknowledge some limita-
and compared them with genomes from previously cultured spe- tions to conclusions about geographical and seasonal distribution of
cies. This showed that genomes clustered primarily per the trophic these protists (because of sampling strategy of the Tara dataset), their
mode of the organisms, with the MAST lineages and chrysophytes finding clearly demonstrate that heterotrophic protists participate in
clade-H clustering together, even though they are phylogenetically various aspects of complex functional networks in ocean food webs.

Although traditionally grouped with the phytoplankton, heterotrophy or mixotrophy occurs in

about half of known dinoflagellates and such species possess a variety of feeding mechanisms,
including absorption of organic material, engulfment of other microbes by phagotrophy, or use
of a peduncle to suck out the contents of larger prey. These can prey on bacteria and larger phy-
toplankton, as well as zooplankton or even fish eggs and larvae; they therefore have major effects
on food web structure. Nonpigmented, heterotrophic dinoflagellates often comprise more than
half the biomass of the microzooplankton size class (20–200 μm) and are frequently the most
important grazers of blooms of diatoms and other microplankton. They also provide a major
food source for copepods. They may also possess very different forms in their life cycle. The
haploid stage, with cells reproducing by mitosis, appears to be the normal mode during active
growth in optimal environmental conditions. During cell division, the thecal plates may also be
divided, but in some species, they are lost completely before the naked cells divide and reform
the plates. During sexual reproduction, haploid gametes fuse to form a motile diploid planozy-
gote, which divides by meiosis to restore the vegetative stage. The planozygote, as well as haploid
cells, can form resting cysts which may have thickened cell walls enabling them to withstand
adverse conditions. Factors affecting the transition between sexual and asexual stages and the
survival and germination characteristics of cysts are very variable and mechanisms are poorly
understood. Endogenous clocks on an approximately annual cycle, nutrient composition and con-
centration, temperature, salinity, pollution, and hydrological conditions are all important factors
in development of dinoflagellate blooms, and processes differ for benthic and planktonic forms.
Understanding this process is especially significant for those dinoflagellates that are responsible
for the formation of deleterious blooms, which may have damaging effects on marine life and
humans; these are discussed in Chapters 12 and 13.

Dinoflagellates and other protists

undertake diel vertical migration
In regions of the open ocean where there is little upwelling, most photosynthetic dinoflagel-
lates and other motile protists such as raphidophytes are found in the epipelagic photic zone
during the day and migrate to deeper nutrient-rich waters at night. This vertical diel (diurnal)
migration in the water column can exceed 30 m in each direction; this is a remarkable two
million times the cell diameter. The onsets of both ascent and descent are regulated by an
endogenous biological clock, so that the cells “anticipate” sunrise and are in the optimum posi-
tion to start photosynthesizing near the surface as soon as light is available. As light declines
at dusk, they migrate to lower levels in the water column to take advantage of higher nutrient
levels for the dark phase. The diel movement of phytoplankton such as dinoflagellates and
raphidophytes is a tactic response to gravity, but during their daily migration, the cells encoun-
ter turbulence and gradients of temperature, nutrients, and the amount and spectral quality of
light. All of these factors can act as input signals for regulating the circadian rhythm.
176 Chapter 6

Figure 6.6 SEM images of dinofla-

gellates. A. Protoperidinium defec-
tum B. Protoperidinium incognitum.
C. Protoperidinium antarcticum.
D. Gonyaulax striata. Credit: Rick
van den Enden, Electron Microscopy
Unit, Australian Antarctic Division,
© Commonwealth of Australia.

TOPSY-TURVY
i PHYTOPLANKTON
CHANGE THEIR
SHAPE
Strong turbulence can cause dam-
age to flagella or cell structure, and
some organisms show adaptations
to cope with this. For example,
Ceratocorys alters its surface spines
Some dinoflagellates exhibit bioluminescence
to enhance sinking to deeper, About 2% of dinoflagellate species found in coastal waters are bioluminescent, with the
calmer layers, while Alexandrium best-known species being Noctiluca scintillans and Lingulodinium polyedrum (formerly
forms chains to alter swimming
Gonyaulax polyedra). They occur worldwide, but exceptionally high densities occur in
behavior. But how do organisms
certain tropical coastal waters and produce spectacular displays of sparkling light when
respond to turbulence over short
timescales? Sengupta et al. (2017) the surface of water is broken at night. (This is often erroneously called phosphorescence,
used an experimental millifluidic which is a different mechanism, caused by prolonged re-emission of radiation absorbed
chamber that was repeatedly by fluorescent molecules.) Bioluminescence in dinoflagellates usually consists of brief
flipped vertically to assess the flashes of blue–green light (wavelength about 475 nm), containing 108 photons and lasting
effects on upward swimming by about 0.01 s. Lingulodinium may also emit red light at 630–690 nm. The stimulus for light
Heterosigma akashiwo. Within min- emission is deformation of the membrane due to shear forces, such as agitation of water
utes of overturning, the microscale by fish, breaking waves, or the wake of a boat. In the laboratory, bioluminescence can be
turbulent eddies caused some stimulated by lowering the temperature or pH of the medium. The bioluminescent flash is
cells to begin to swim downward, preceded by an action potential, during which the inside of the membrane becomes nega-
while others continued to swim
tively charged. This leads to acidification of vesicles in the vacuolar membrane contain-
upwards. This is due to changes in
cell orientation relative to gravity.
ing the enzyme luciferase and the substrate luciferin. There are about 400 such vesicles
Reversal of swimming direction is (scintillons) in each cell and they occur as spherical evaginations of cytoplasm into the
caused by cells morphing from a cell vacuole, each containing luciferin complexed with a special binding protein. A tran-
pear-shaped cell to a more sym- sient pH change results from the opening of membrane proton channels in the scintil-
metric egg-shaped cell, which has lons; this activates the reaction by release of the luciferin from its complexed state so
greater stability and reorientates that it can be oxidized in an ATP-mediated reaction similar to that in bacteria (Figure
more quickly. This rapid behavioral 3.17B). Experimental studies of Lingulodinium cultured under various light conditions
response gives the phytoplankton have revealed that bioluminescence is regulated by a circadian rhythm and peaks in the
precise control over their migratory middle of the dark period. The circadian expression of bioluminescence involves the daily
behavior, increasing the likelihood
synthesis and destruction of the scintillons and component proteins; this is regulated at
that some members of the popula-
the translational level and may involve a clock-controlled repressor molecule that binds
tion will evade intermittent patches
of turbulence in the water column. to mRNA.
 Marine Eukaryotic Microbes 177

There are two main hypotheses for the ecological function of bioluminescence in dinofla-
gellates. One idea is that mechanical stimulation occurs when a predator such as a copepod
approaches a dinoflagellate cell—stimulating a brief, bright flash. This is thought to startle
the predator and lead to its disorientation. An alternative possibility, called “the burglar
alarm hypothesis,” is that light produced by luminescent prey attracts grazing predators,
which in turn sends a signal to larger predators, so that the grazers themselves become prey.
Grazing on bioluminescent dinoflagellates will decrease if the risk of consuming them results
in reduction of the net benefit that consumers receive. Thus, bioluminescence could reduce
grazing pressure by reduction of feeding efficiency and the species as a whole could benefit,
even though some mortality of individuals occurs. Various experimental approaches have
been used to determine which of these alternative hypotheses is correct, but there is no clear
answer and progress in understanding requires methods to measure interactions in situ under
natural conditions.

The ciliates are voracious grazers of

other protists and bacteria
This group of protists, within the supergroup Alveolata is distinguished by the posses-
sion of cilia in at least one stage in the life cycle—these have the same basic structure as
flagella, but the cilia often cover the cell surface or are arranged in groups; they beat syn-
chronously to provide movement or to create water currents to channel particulate food
into the cell. At least 8000 species of ciliates are known—many of which are marine—
and, although highly diverse, the group appears monophyletic in modern classification
systems. Marine ciliates are generally in the size range 15–80 μm, with some up to 200
μm. One group of ciliates, the tintinnids, produce a “house” called a lorica, which is
constructed from protein, polysaccharides, and accumulated particulate debris collected
from the water (Figure 6.7). Thus, tintinnid loricae are large enough to be collected in
fine-mesh plankton nets and were, therefore, one of the first groups of marine ciliates to be
studied. There has been intense interest in the ciliates in recent years because of growing
recognition of the essential role that most types play in the microbial loop of ocean food
webs by ingesting bacteria and other small protists and because they are large enough
to be preyed upon in turn by larger protists and mesozooplankton. Selective grazing on
particular prey types is an important factor in structuring the composition of microbial
communities in food webs. Consequently, there have been numerous studies of the abun-
dance and activity of marine ciliates, with a wide range of results. Typically, there are
about 1–150 ciliates per milliliter in seawater, with the highest numbers in coastal waters;
much higher concentrations occur in marine snow particles. Abundance varies greatly
with water depth, temperature, and nutrient concentration. Water stratification during the
different seasons is a major factor affecting ciliate numbers. Ciliates are also abundant in
benthic sediments and microbial mats.

Figure 6.7 Light micrographs

of ciliates. Scale bar = 20 µm. A.
Gastrocirrhus monilifer. B. Parafavella
parumdentata tintinnid. Credits: A.
Bill Bourland, used by permission
of David Patterson, http://micro-
scope.mbl.edu/. B. Diane Stoecker,
NOAA Fisheries Collection, via Flickr
CC-BY-2.0.
178 Chapter 6

The most common marine ciliates are spherical, oval, or conical cells with a ring of cilia sur-
rounding the cytostome (“mouth”), which they use to filter bacteria and small flagellates from
the surrounding seawater. Upon entering the cytostome, ingested food particles are engulfed
by phagosomes, which then fuse with lysosomes in the cytoplasm. The phagosomes become
acidified and enzymes contained in the lysosome lead to digestion. Nutrients pass into the
cytoplasm and undigested waste material is egested. Although most genera are phagotro-
phic, some are strict photosynthetic autotrophs and have lost phagotrophic capability. The
most important of these is Mesodinium rubrum (previously classified as Myrionecta rubra
in the Zoological Code), which can occur in large blooms in coastal waters and may make
a sizable contribution to primary production. The species is named for its association with
red tides, which discolor coastal waters, although it is non-toxic. This ciliate contains func-
tional organelles obtained from its prey, cryptophyte algae. Unlike the retention of plastids
(kleptoplastidy) mentioned above and observed in a wide range of ciliates, M. rubrum also
contains stolen mitochondria, cytoplasm, and a transcriptionally active nucleus, as well as
chloroplasts. There have been suggestions that M. rubrum can retain intact cryptophyte cells
as endosymbionts, although this is disputed. There are also many examples of mixotrophy in
other ciliates, arising from transient retention of ingested plastids.

A distinctive feature of the ciliates is the possession of two types of nuclei. The larger
macronucleus consists of multiple short pieces of DNA and is concerned largely with
transcription of mRNA and growth processes. The smaller micronucleus is diploid and
is responsible for an unusual type of sexual reproduction (conjugation), in which two cells
fuse for a short period and exchange haploid nuclei derived from the micronuclei by meio-
sis. The macronuclei disappear during this process. The result of conjugation is that each
partner ends up with one of its own haploid nuclei and one from its partner; these then
fuse, and the cells separate. In each cell, the diploid nucleus thus formed divides and dif-
ferentiates into micro- and macronuclei due to extensive rearrangement and amplification
of DNA sequences.

The haptophytes (prymnesiophytes) are some of the

most abundant components of ocean phytoplankton
The haptophytes, also known as prymnesiophytes, are very important members of the phy-
toplankton, some of which have profound influences on oceanic and atmospheric processes.
There are about 500 living species in 50 genera, with many additional species identified as
fossils in sedimentary rocks. Most species are marine and occur worldwide, with the greatest
diversity and abundance in tropical waters.

Haptophytes are photosynthetic and contain one or two plastids, together with the yellow–
brown accessory pigments diadinoxanthin and fucoxanthin. They often have a complex life
cycle, alternating between motile and nonmotile haploid and diploid stages. The cells contain
two slightly unequal (heterokont) flagella in one stage of their life cycle. They are also distin-
guished by the presence of a haptonema; this is a thin structure reminiscent of a flagellum,
but with a different structure and unknown function. The haptonema is composed of six to
seven microtubules in a ring or crescent, with a fold of endoplasmic reticulum extending out
within the flagellum.

The best-known members of the haptophytes are the coccolithophores, which have been
one of the most abundant forms of phytoplankton in the oceans for millions of years. Their
distinguishing feature is the presence of external scales or plates, formed within the Golgi
apparatus, that cover the cell surface in many species. The plates have distinctive shapes and
surface architecture in different species.

The best-known and most abundant example of coccolithophores is Emiliania huxleyi,

in which the typical cells are diploid, nonmotile, 4–5 μm in diameter, and covered with
about 30 calcified plates (coccoliths) (Figure 6.8B). The haploid stage occurs as motile
swarmer cells with noncalcified scales and another nonmotile naked cell type may also
occur as a mutant diploid stage. Each stage can reproduce vegetatively. The haploid stage
 Marine Eukaryotic Microbes 179

Figure 6.8 A. Satellite image of

Emiliania huxleyi bloom off south-
west England. B. SEM images of
cultured E. huxleyi cells at different
stages of development showing 6
(a), 8 (b–c), 10 (d), 14 (e), or 22 (f)
coccoliths. Red numbers indicate the
number of adjacent bordering coc-
coliths. Credits: A. Plymouth Marine
Laboratory Remote Sensing Group.
B. Reprinted from Xu et al. (2018),
CC-BY-4.0.

is very important in the resistance of E. huxleyi to viral infection, as discussed in Box 7.1.
Calcification takes place intracellularly in a specialized vesicle of the Golgi apparatus and
depends on a high flux of inorganic carbon (in the form of bicarbonate) and calcium. The
calcium-binding protein is a key factor in the calcification process and the gpa gene that
encodes this is often used as a genetic marker in studies of E. huxleyi community dynam-
ics. Calcite (one of the crystalline forms of calcium carbonate) is precipitated onto the
organic matrix of the vesicle and extruded to the surface to form a coccolith. The reasons
why coccolithophores calcify their surface plates is unclear. It is an energy-demanding
process and may have evolved to provide mechanical protection against grazing pressure,
but it may provide additional benefits such as protection from photodamage. Calcification
also enhances the efficiency of photosynthesis by lowering the internal pH of the cell,
increasing the concentrations of dissolved CO2 needed from maximum efficiency of the
CO2-fixing enzyme RuBisCO (p.83).

The coccoliths are arranged, usually in a single layer, to cover the entire surface area of the cell,
which changes during growth. Observation of cultured E. huxleyi cells shows that the number
180 Chapter 6

of coccoliths formed keeps pace with the increase in cell size during growth, as shown in
Figure 6.8B. To maintain full coverage of the surface, each coccolith interlocks with four to six
others; small coccoliths interlock with fewer and larger coccoliths, and vice versa. The optimal
configuration to enclose the volume of the cell is hexagonal tiling, following the same math-
ematical principles as the construction of bees’ honeycombs or formation of soap film bubbles.

Each year, massive blooms of E. huxleyi occur in the upper photic layers of temperate and
subtemperate waters. They are especially notable in the coastal waters of northern Europe
and Scandinavia following spring diatom blooms, where they are highly visible as milky-
white areas in satellite images (Figure 6.8A), because of the reflective properties of free
coccoliths released from the cells as they disintegrate after death. Under favorable condi-
tions, individual blooms may cover 105 km 2 and contain more than 1021 cells, account-
ing for almost 90% of all the phytoplankton in the upper water columns. The reflective
“signature” of the coccoliths is unique and allows precise measurement of the fate of the
bloom from satellite images. On a global scale, such dense blooms of E. huxleyi probably
transiently cover up to 2 × 106 km 2 in a typical year. E. huxleyi is present at relatively con-
stant, lower levels in many oligotrophic oceans, where they also contribute significantly to
primary productivity.

E. huxleyi has global distribution and has been intensively studied in recent years because
of its role in the global carbon cycle, being the largest global producer of calcium car-
bonate and, hence, a major sink for CO2. However, there are several complications to the
process and the exact contributions that E. huxleyi makes to carbon cycling and climate
change are not clear. The light-scattering properties of the coccoliths may have a small
albedo effect, reflecting solar radiation and helping to cool surface waters during a bloom.
Calcification leads to rapid removal of inorganic carbon as the plates settle through the
water column to form sediments. In fact, although E. huxleyi is the most dominant cocco-
lithophore, other species with larger, heavier coccoliths may be more important in the flux
of calcite. Various studies to estimate the global productivity of calcite in the modern ocean
have been conducted, leading to a consensus value of about one gigatonne (petagrams) of
calcite carbon a year. Only about half of this amount will be precipitated to the ocean floor
because in deep waters (over about 3000 m), changes in the water chemistry due to high
pressure cause the calcite to redissolve. Fossils of the coccolithophorid group first appeared
in the Jurassic period about 290 MYA and reached their greatest abundance and diver-
sity in the Late Cretaceous period, at the end of which a mass extinction of many genera
occurred. Accumulation of coccolithophorid plates on the seabed contributed to the forma-
tion of ocean sediments and rocks such as the Mesozoic limestones and chalks. The famous
White Cliffs of Dover in southern England are composed of a very fine-grained white chalk
derived mainly from coccoliths. Most modern coccolithophores are smaller and less heavily
calcified than ancient forms, but nevertheless remain highly diverse.

Dissolved carbon exists in three forms in seawater—dissolved CO2 gas, bicarbonate, and
carbonate, and the relative proportions of each depend on pH, temperature, and pressure.
When coccolithophore cells remove bicarbonate to form coccoliths, the pH drops and shifts
more of the carbon into the dissolved CO2 form. Calcification can be represented by the
formula:

Ca + 2HCO3 −  CaCO3  H 2O + CO2

Thus, coccolithophore blooms might increase global warming by causing release of CO2
back to the atmosphere, but this is complicated by the fact that the association of heavy
coccoliths with marine snow particles and the feces of grazing zooplankton could lead to
more rapid settling of associated organic material, thus accelerating the sequestration of fixed
carbon to the sediments. Understanding the fluxes involved is very important in view of ris-
ing atmospheric CO2 levels leading to ocean acidification, but despite many laboratory and
mesocosm experiments, there are no clear answers.

Two other prymnesiophytes, Chrysochromulina and Phaeocystis, have both been studied
extensively because of their ability to produce large nuisance blooms under certain conditions,
 Marine Eukaryotic Microbes 181

especially in northern Europe and Scandinavia. Blooms are linked to climatic conditions and
the input of pollutants from land runoff. Phaeocystis globosa aggregates into large colonies
surrounded by polysaccharide mucilage. This is responsible for the foam that commonly
affects seashores during summer, and excessive production of DMS (see p.248) during the
collapse of the blooms leads to sulfurous smells that beachgoers often wrongly confuse with
sewage pollution. DMS is also suspected of affecting the migration behavior of fish, which
seek to avoid it, while the mucilage has physical effects such as clogging of aquaculture nets
and desalination plant filters and pipes.

Diatoms are extremely diverse and abundant

primary producers in the oceans
The diatoms are one of most diverse groups of the protists, with in excess of 200 genera and
about 105 current species, with probably about the same number of fossil species. In mod-
ern classification schemes, diatoms are recognized as unicellular microbes belonging to the
stramenopiles in the class Bacillariophyceae. Diatoms range in size from 2–200 µm in length
and possess either radial or bilateral symmetry; these forms are termed centric or pennate,
respectively.

The defining feature of diatoms is enclosure of the cell within a hard but porous wall known
as a frustule, composed of two overlapping plates (valves or thecae) of hydrated silica (silicon
dioxide) that is polymerized intracellularly. The two valves overlap rather like the lid and
base of a petri dish. These opal-like structures are highly patterned, forming a variety of
shapes of great architectural beauty and are highly distinctive for identification (Figure 6.9).
The mechanism by which diatoms create these wondrous structures is of great interest. A
major goal is to link the genetic, biochemical, and physiological processes with mathemati-
cal models to explain the patterns of silica deposition, which has many potential applications
in the nanofabrication of silicon-based materials for electronics. Like dinoflagellates, diatom
genomes are complex and present many technical difficulties, but several examples have now
been sequenced. They contain evidence of their chimeric origin from their algal and hetero-
trophic ancestors, as well as extensive horizontal gene transfer from bacteria.

The nature of the frustule enclosing the diatom cell results in a unique reproductive cycle.
Reproduction is usually asexual, with each daughter cell constructing a new theca within the
old one, resulting in a progressive reduction in cell size with each division cycle. At a criti-
cal point—this is usually when the diatom is about one-third of the original size—sexual
reproduction occurs by formation of gametes via meiosis. The gametes fuse to form a zygote

Figure 6.9 Light micrographs of

marine diatoms. A. Chaetoceros
debilis. B. Melosira moniliformis.
C. Thalassiosira nordenskiioeldi. D.
Coscinodiscus centralis. Credits: D.
Cassis, T. Ivanochko, J. Shiller, B.
Moore-Maley, J. Kim, S. Huang, A.
Sheikh, G. Oka. Phytopedia: The
Phytoplankton Encyclopedia Project.
Published at www.eoas.ubc.ca/
research/phytoplankton/
182 Chapter 6

(auxospore), which increases in size and synthesizes new full-size thecae. Hence, sexual
BACTERIA HELP TO
i DISSOLVE DIATOM
SHELLS AND
reproduction accomplishes genetic variability as well as allowing the cell line to regain maxi-
mum size before the next round of vegetative cell division.
RECYCLE SILICA
Most diatoms are free-living in the plankton, but many attach to surfaces such as rocks,
When diatoms die, they aggregate marine plants, mollusks, crustaceans, and larger animals. The skin of some whales has been
and sink through the water col- shown to be covered with dense colonies of diatoms. Locomotion of some (mostly pennate)
umn. Some of the silica dissolves diatoms occurs in contact with surfaces, probably via the secretion of mucus through a slit
to form silicic acid, which is then (raphe) in their silica frustules but they are mostly nonmotile in the water column and their
available through upwelling to heavy cells sink rapidly, relying on turbulence in the surface layer to keep them in suspen-
support more diatom growth. The
sion. Some diatoms have pairs of thin spines projecting from the ends of the cells, which link
input of silicic acid from terrestrial
runoff is balanced by deposition in
with those of other cells to form long chains, thereby increasing buoyancy and forming large
sediments. Bacteria play a critical mats.
role in silica recycling by producing
extracellular enzymes that break The major pigments in this group are usually chlorophylls a and c and the carotenoid fuco-
down cell wall polymers. Bidle and xanthin. It is thought that the chloroplast in stramenopiles, which contains four membranes,
Azam (1999, 2001) showed that if arose by a secondary endosymbiosis event about 700–1200 MYA, involving a red alga ances-
bacteria are removed from seawa- tor and a heterotrophic eukaryote. Genome sequencing reveals that genes from red algae have
ter or inhibited by antibiotics, the been transferred to the nucleus of diatoms, which provides an explanation for the unusual
dissolution of silica from diatom metabolic properties of diatoms. For example, unlike green plants, diatoms have a complete
debris is very low. Diatom spe-
urea cycle and generate energy from the breakdown of lipids, and carbohydrates are stored as
cies vary in both the thickness of
a β1,3-linked glucose polysaccharide.
their silica shells and the extent of
protection by glycoprotein coats.
These factors determine the extent For many years, diatoms have been recognized as the dominant member of the marine phy-
to which the shells of different toplankton and therefore attributed with the major role in primary productivity and carbon
species remain intact. Bidle (2002) export. They are estimated to contribute ~40% of the total ocean primary productivity.
showed that temperature strongly They are especially important in seasonal productivity in well-mixed, nutrient-rich coastal
influences bacterial dissolution waters in higher latitudes, but we now realize that cyanobacteria and protists in the pico-
of silica. In polar regions, slow plankton size contribute a large fraction of photosynthetic productivity in open oceans.
bacterial activity means that more Environmental 18S rRNA gene datasets from the Tara Oceans surveys indicate that they
carbon and silica is sequestered in are the most abundant group of obligately phototrophic marine eukaryotes. In polar regions,
the sediments. Understanding this
diatoms play a critical role in the food chain, especially as their mixotrophic metabolism
process is important for model-
ing the role of the oceans in the
allows them to survive long periods of darkness before emerging as blooms in the spring.
removal of atmospheric CO2 and In some Antarctic samples, diatoms represent 25% of the sequences. Like dinoflagellates,
the effects of global warming on many open-ocean diatoms undertake a vertical migration from the nutrient-depleted photic
geochemical cycles. zone to deeper waters in order to pick up nitrate and other nutrients. The complex ecological
factors that determine community composition of the phytoplankton are discussed further
in Chapter 8.

Diatoms are largely responsible for the spring bloom along the continental shelf in temperate
waters and for seasonal blooms in regions of nutrient upwelling. As with other algae, a key
factor determining the size of blooms is the availability of nutrients, especially nitrogen and
phosphorus, but the concentration of silica is also a critical limiting factor for diatoms, and
most will not grow at silica concentrations below about 2 μM. Some species make frustules
with very high silica contents, and in some waters (e.g. the Antarctic circumpolar current),
the shells are very resilient to dissolution as they settle through the water. Blooms are gener-
ally followed by the exhaustion of nutrients and aggregation and sinking of diatoms. The
dynamics of diatom production and settlement are still poorly understood but are important
in understanding ocean biogeochemistry.

Diatoms began to accumulate on the seabed about 100 MYA and reached a peak of abun-
dance in the middle part of the Cenozoic period. Settlement of dead diatoms over millen-
nia led to thick deposits and formation of siliceous sedimentary rocks, which are mined as
diatomaceous earth. This has many industrial uses such as filtration compounds, abrasives,
insulating agents, pharmaceutical products, and insecticides. The small fraction of organic
matter that escaped remineralization accumulated in sediments, leading to the formation
of petroleum hydrocarbons. Today, diatoms are being exploited in biotechnology, including
applications in nanotechnology and the production of biofuels (see Chapter 13). Few diatoms
are toxic, but an important exception is the genus Pseudo-nitzschia, which produces domoic
 Marine Eukaryotic Microbes 183

acid. This is responsible for human illness associated with shellfish consumption and for
mortalities in marine mammals and seabirds (p.320).

Other stramenopiles may cause harmful blooms

Raphidophytes are photoautotrophic stramenopiles related to the diatoms. They have large
cells (50–100 μm) without cell walls and possess two flagella. Blooms of species such as
Heterosigma akashiwo, Chatonnella spp., and Fibrocapsa spp. have caused “red tides”
responsible for extensive fish mortalities in Japan, Scotland, and Canada. They are thought to
kill fish as a result of the production of a range of toxins, including reactive oxygen species,
hemolytic substances, and neurotoxins.

Other important stramenopile algae, with small cells about 2–3 μm in diameter, include
Aureococcus spp., Nannochloropsis spp., and Bolidomonas spp. Aureococcus anophagef-
ferens is notorious as a cause of regular “brown tides” in estuaries on the eastern US coast
since the 1980s. It is always present at low concentrations and blooms to high concentrations
under certain conditions, causing severe limitation of growth of seagrass (Zostera marina)
beds, because of shading caused by the dense blooms. This affects larval recruitment of fish,
scallops, and clams, with severe consequences for fisheries. Aureococcus is also suspected of
producing a toxin that may affect ciliary function in bivalve mollusks, although this has not
been isolated. Nannochloropsis has a very high content of polyunsaturated fatty acids, mak-
ing it an important food source for fish larvae. This feature is being exploited in development
of aquaculture feeds (see p.393).

Thraustochytrids and labyrinthulids are

active degraders of organic matter
These two groups of closely related stramenopiles are characterized by the production of a
slime net—a network of filaments produced as extensions of the cytoplasmic membrane that
serve as tracks for the cells to glide along. They were originally grouped with slime molds in
the Fungi, but their heterokont flagella, mitochondrial structure, and genetic features mark
them out as stramenopiles. They have been known for many years, but their ecology and
importance in marine systems are understudied.

Several studies have shown that thraustochytrids are widely distributed in the plankton in
coastal and open ocean, although they seem to occur at very low densities, perhaps only 1–100
mL−1. Along with bacteria, archaea, and fungi, these organisms are the only heterotrophic
members of the plankton which feed by osmotic absorption (osmotrophy) of dissolved organic
compounds rather than by phagocytosis of cells or particles. The slime net contains extracel-
lular enzymes—including amylase, cellulase, lipase, protease, phosphatase, pectinase, and
xylanase—that digest large polymeric organic compounds to smaller units for absorption.
The slime net can penetrate particles, such as marine snow or clumps of decaying cells from
algal blooms. These properties and the fact that their cells are many times bigger than bacte-
rial or archaeal cells mean that, despite their low numbers, they may have a comparable bio-
mass. This leads to a very significant and under-recognized role at the base of the food chain,
especially since they can digest highly refractory organic matter and are readily eaten by
ciliates and amoebae. This link to higher trophic levels is particularly important because the
thraustochytrids produce high levels of omega-3 polyunsaturated fatty acids (PUFA), which
are essential for the growth and reproduction of crustaceans and fish larvae. This feature is
being exploited in aquaculture biotechnology. Thraustochytrids also degrade plant detritus
such as algae and mangrove leaves, and a few species have been described as parasites of mol-
lusks. Genome sequences of a number of strains have been obtained, providing information
about the genetic basis of production of PUFA and hydrolytic enzymes. This reveals that some
strains of thraustochytrids produce low levels of the enzymes responsible for degradation of
refractory polymers and may depend on scavenging nutrients released by bacterial action.

Labyrinthulids are very similar in structure to the thraustochytrids but have not been found
in plankton. Instead, they are usually associated with the surfaces of marine macroalgae,
184 Chapter 6

plants, or animals as parasites or commensals. One species, Labyrinthula zosterae, causes

the “wasting disease” of the seagrass Zostera marina (p.324).

Photosynthetic prasinophytes are abundant

members of the picoplankton
Many algae in the microplankton groups discussed above have been described since the stud-
ies of pioneer marine biologists and oceanographers, as their relatively large size and distinc-
tive morphological features can be used for classification and identification. However, the
full importance of photosynthetic protists as components of the picophytoplankton (<3 µm)
has only been realized in the last few decades. Phototrophic types belonging to the green
algae (or Chloroplastida), stramenopiles, and haptophytes (prymnesiophytes) occur in the
photic zone of all the world’s oceans. Together with the cyanobacteria Prochlorococcus and
Synechococcus (see p.134), the picophytoplankton may constitute up to 80% of the chloro-
phyll-containing biomass in subtropical oceans. Studies in the Pacific Ocean have shown
that, although phototrophic picoeukaryotes comprise only a small proportion of the total
phytoplankton cells, they account for the majority of the net carbon production. They are
subject to heavy grazing pressure by other protists, probably because they lack a cell wall.
Thus, their role in marine food webs and the biological carbon pump (see Figure 8.1B) may
be much more important than previously realized because of the efficient capture of carbon
dioxide and its rapid transfer to food webs via grazing.

The most studied prasinophyte phototroph is Ostreococcus tauri in the family Mamelliaceae.
Originally discovered and given its genome name because it was found in waters used for
oyster (Ostrea) cultivation in France, O. tauri is now known to be ubiquitous in coastal waters
and in the open ocean. This is the smallest eukaryotic organism known: the tiny cell (about
0.8 μm diameter) has no cell wall or flagella and contains the nucleus, only one mitochon-
drion, and one chloroplast, tightly packed within the cell membrane (Figure 1.4B). Genome
analysis has revealed that it has a small and compact genome with 20 linear chromosomes
(Table 6.1). It is thought that one chromosome may be a sex chromosome, as it contains
genetic information for meiosis, although sexual reproduction has not been observed. Genes
on some chromosomes show little homology to other genes and appear to have been later-
ally transferred; they may encode surface proteins to protect the cell against predation. One
of the most important findings of genome analysis is the presence of genes for C4 photo-
synthesis. This route of CO2 fixation may be more efficient than the normal C3 route when
dense populations of cells are competing for available CO2. As with the cyanobacterium
Prochlorococcus, comparative genome analysis shows that different ecotypes of O. tauri are
adapted to different light and nutrient niches.

Another prasinophyte, Micromonas pusilla, is a major component of the picoplanktonic

community in several oceanic and coastal regions and shows summer blooms that may be
terminated by viral infection. Micromonas has a pear-shaped cell (~ 2 µm) and is actively
motile via a single flagellum, enabling it to swim toward light sources. Comparative genomic
analysis of strains presumed to be of the same species, but isolated from different geographic
regions, has shown remarkably large differences in genome sequences, with only about 90%
of their genes shared and the taxon undoubtedly encompasses several species. The green
algae were long thought to be a purely phototrophic group but Micromonas, like many other
protists, is also phagotrophic. They are probably the dominant bacterivore in the Arctic
Ocean, where studies have suggested that climate change is leading to higher temperatures
and input of fresh water is lessening the mixing processes that deliver nutrients from deep
to surface water. This seems to be favoring the picoplankton, as their smaller cells are more
competitive in acquiring scarce nutrients.

Amoebozoa are important grazers of

particle-associated bacteria
A wide range of phylogenetically diverse protists in various eukaryotic groups demonstrate
a crawling or amoeboid movement using pseudopodia. Most of the free-living amoebae
 Marine Eukaryotic Microbes 185

occur in the supergroup Amoebozoa in the eukaryotic phylogenetic tree (Figure 6.2).
Hundreds of species have been described, including specifically marine lineages. Planktonic
amoebae are sparsely distributed in the open-ocean water column, so their contribution to
energy flow via preying on bacteria, microalgae, and other protists within pelagic food
webs has largely been overlooked. However, they can reach high densities in more turbid
estuarine and coastal waters and are particularly associated with surfaces such as flocs
and marine snow particles, where the density can exceed that of the ciliated protists most
commonly associated with bacterivory. Their plastic cell shape may facilitate removal of
firmly attached bacteria by enabling them to penetrate crevices inaccessible to other graz-
ers. Some bacteria have evolved the ability to survive the killing mechanisms in the phago-
cytic vacuole and can multiply within host amoebae, forming stable symbiotic or parasitic
associations with their host.

Radiolarians and foraminifera have highly

diverse morphologies with mineral shells
These two groups are classified as members of the supergroup Rhizaria (Figure 6.2). They
are characterized by an amoeboid body form, using pseudopodia for locomotion and feeding.
The cells are usually less than 1 mm in size, but some species are the largest protists known,
with diameters up to several cm.

Radiolarians have existed since the beginning of the Paleozoic era about 600 MYA and
produce a great diversity of beautiful, intricate shapes (Figures 6.1, 6.10A) used in iden-
tification of more than 4000 species. Cells are typically 0.1–0.2 mm. Two main groups
known as the Polycystina and Acantharea are recognized. They are characterized by stiff,
needle-like pseudopodia arranged in radial symmetry (from which they derive the name
radiolarian) and internal skeletons made of silica. Larger species are often associated with
surfaces and may contain algal symbionts that provide some nutrients to the cell, while
the smaller types occur throughout the water column and in deep-sea sediments. Densities
vary greatly, ranging from 10 4 cells cm-3 in some parts of the subtropical Pacific Ocean to
less than 10 cells cm-3 in the Sargasso Sea. The silica skeletons of the polycystine radio-
larians deposit as microfossils and are second only to diatoms as a source of silica in
sediments. The cell body consists of a central mass of cytoplasm surrounded by a capsu-
lar wall. This contains pores through which the cytoplasm extrudes into an extracapsular

Figure 6.10 A. The radiolarian

Astrolithium cruciatum. B. A radiolar-
ian (unidentified) ingesting a tintin-
nid Proplectella. C. A globigerinid
foraminiferan. D. A large (~20 cm)
xenophyophore sampled from the
seabed of the Atlantic Ocean, Mid-
Atlantic Ridge. Credits: A–C. John R.
Dolan, NOAA Fisheries Collection,
via Flickr CC-BY-2.0. D. NOAA Photo
Library; IFE, URI-IAO, UW, Lost City
Science Expedition, 2005.
186 Chapter 6

cytoplasm and forms the stiffened pseudopodia. The cytoplasm moves by streaming and

i
GIANT BLOBS captures other protists and small zooplankton, which are then surrounded and digested in
FROM THE DEEP! food vacuoles. Hydrated, polymerized silicon dioxide (opal) is deposited within a frame-
A 2011 video exploration of the work of the cytoplasm. Reproduction occurs asexually by binary fission or sexually via
Marianas Trench using free-falling the production of haploid gametes. The Acantharea produce skeletons of strontium sulfate
drop cameras led to identifica- (celestite) rather than silica.
tion of giant xenophyophores at a
depth of 10.6 km. The discovery The Foraminifera is a phylum or class of the Rhizaria whose members form a multi-cham-
attracted much interest in the bered shell within a cytoplasmic envelope produced by pseudopodia; the shapes of these are
popular press, who reported the
highly distinctive for species identification (Figure 6.10C). The chambers are connected by
discovery with headlines implying
it was a new life form—reminis-
openings—from which forms derive their name from the Latin for “hole-bearers”—and have
cent of 1950s horror movies. In sealed pores that face the external environment, through which cytoplasmic spines stretch
fact, xenophyophores were first for long distances to form a net for the capture of prey, which can include bacteria, phyto-
described in 1883 and are the plankton, and small metazoan animals. Cells are usually <1 mm in size, but larger species,
dominant megafaunal organism including multinuclear forms known as xenophyophores up to 20 cm across, occur in the
on much of the deep ocean floor. deep sea (Figure 6.10D). These have been found to be abundant on the sea floor of all ocean
New research has been stimulated basins, especially in submarine trenches on seamounts beneath highly productive waters,
by the need for baseline studies where they can reach densities of 20 organisms m−2. They appear to feed by scooping up
during exploration of the eastern sediment particles using secreted mucus threads. These build up an agglutination of waste
equatorial Pacific, which includes
material around the cells that traps particles, from which the group gets its name (Greek:
a large area being investigated by
“bearer of foreign bodies”). The larger types are often found in association with animals such
deep-sea mining companies for
exploitation of metal nodules on as isopods, worms, and echinoderms and are thought to provide habitat structure promoting
the seafloor. They are extremely “hotspots” of deep-sea diversity.
fragile and difficult to study, but
Gooday et al. (2018) examined the About 5 × 10 4 species of foraminifera have been documented, of which about four-fifths are
internal structure of six species of fossils. Many forams contain other protists as endosymbionts, including green algae, red
xenophyophores obtained from a algae, diatoms, and dinoflagellates. Foraminifera make up only a very small component
depth of ~4 km using non-destruc- of the plankton, but their discarded skeletons are the major source of calcareous depos-
tive micro-CT 3D imaging. This its, accumulating as the “globigerian ooze” over vast areas of the ocean floor. Massive
showed that the great majority of deposits accumulated during the tertiary period, about 230 MYA, and these have become
the organisms’ structure is com-
uplifted over time and exposed as limestone beds in Europe, Asia, and Africa. The main
posed of masses of waste material
and less than 5% of the “body” is
application of the study of the many diverse forms of foraminifera is in biostratigraphy,
living cytoplasm, organized in a a branch of geology which assigns relative ages of rock strata by recording fossil assem-
network of organic tubes. blages. This technique is widely used in oil exploration and drilling projects and for recon-
structing the history of oceanic conditions during geological eras (paleoceanography and
paleoclimatology).

MARINE FUNGI
The Fungi form a distinct monophyletic group
on a branch within the Nucletmycea
The Fungi are a large and very diverse group of organisms ranging from microscopic forms
(chytrids, molds, and yeasts) to large hyphal networks. Many terrestrial fungi are familiar,
as mushrooms and toadstools formed as reproductive structures, but only the microscopic
forms occur in marine environments. In the phylogenetic tree of eukaryotes (Figure 6.2),
the fungi form a distinct monophyletic group on a branch within the Nucletmycea, part of
the Opisthokonta supergroup. They are a sister group to animal cells, which are believed to
have diverged ~1.5 BYA. The classification of fungi has traditionally been based on mor-
phological features, especially the structure and properties of their systems for reproducing
asexually and sexually by the production of spores (sporulation). Most marine fungi belong
to three of the well-known major fungal phyla, namely Ascomycota, Basidiomycota, and
Chytridiomycota. The use of genetic analysis based on rRNA gene sequences in both the
small and large subunits, as well as the intergenic transcribed spacer (ITS) has led to a
revised classification of the Fungi into 18 phyla, although there is still considerable uncer-
tainty about the placement of certain taxa within these divisions and novel marine taxa have
been identified.
 Marine Eukaryotic Microbes 187

Fungi are heterotrophic osmotrophs, feeding by the secretion of extracellular enzymes

and absorbing the products of digestion of macromolecules. The majority are saprotophs,
which decompose complex organic materials in the environment. Others are important
parasites of algae and other protists, plants, and animals. A distinctive characteristic is
the possession of a thick, tough cell wall composed of chitin, a polymer of N-acetyl glu-
cosamine. Many fungi are filamentous, forming a multicellular network of hyphae that
extends by growth at the tip. The hyphae may have cross-walls (septa) separating the indi-
vidual cells, but in some species the filament is coenocytic, containing numerous nuclei
formed by division without formation of septa. On aerial surfaces, the hyphae ramify
over and into the substratum forming a network of branched filaments (mycelium) and
produce branches bearing asexual spores (conidia). They may also form sexual spores in
macroscopic fruiting bodies. Some of these distinctive reproductive structures are seen in
marine fungi when they colonize surfaces such as submerged wood or vegetative litter, but
very little is known about the organization of hyphae and reproductive structures in fungi
in the water column. Alternatively, many species are unicellular (yeasts) and reproduce
by budding cell division.

The Chytridiomycota (commonly referred to as chytrids) are characterized by the production

of motile reproductive spores (zoospores) produced within a sac-like structure (their name
is derived from the Greek for “little pot”). These zoospores swim rapidly using a posterior
whiplash flagellum, and are rapidly dispersed in water and moist habitats, explaining their
abundance in aquatic systems. Chytrids are mostly unicellular or may form colonies with
short hyphae. The Basidiomycota are mostly filamentous forms, although some yeasts are
common in marine habitats. They are named for their distinctive sexual reproductive struc-
ture—the basidium, meaning “little pedestal”—with external spores formed by meiosis. The
largest phylum of the Fungi is Ascomycota, again named for their distinctive cup like sac, the
ascus (meaning “sac” or “wineskin”) that holds the sexual spores. In all groups, some species
reproduce asexually.

Fungi are increasingly recognized to be major

components of the marine microbiome
There are approximately 1.2 × 105 known species of fungi, most of which have been
considered as being primarily terrestrial organisms. Until recently, their importance in
marine systems had not been fully recognized and they have not received the depth
of study afforded to other groups of marine microbes. Nevertheless, a small group of
researchers dedicated to study of the fungi has cataloged the diversity of marine fungi
over 150 years using painstaking culture methods and morphological examination,
especially from coastal habitats such as salt marshes and tropical mangroves. Sea foam
has also been a particularly rich source of inocula for marine mycologists, as fungal
spores and hyphal fragments become aggregated and trapped in air bubbles in the poly-
meric matrix of foam formed by wind and wave action. Figure 6.11 shows an example
of some of the diverse fungal colonies obtained from a single sample. The advent of
molecular identification techniques has improved our knowledge of marine fungi greatly.
However, the methods employed so successfully in global surveys to assess diversity and
distribution of bacteria and archaea have had some limitations when applied to fungi.
Nevertheless, the importance of marine fungi is increasingly recognized for their role in
nutrient cycling, as parasites of marine organisms, and as a source of natural products.
Over 1100 marine species have now been described, which probably represents <1% of
the true diversity. When concerted efforts using modern techniques are made to discover
fungi in marine habitats, it becomes obvious that they occur almost everywhere we look.
Examples of common genera are shown in Table 6.3. Many of these predominant gen-
era were already familiar from soil, plant, and freshwater mycology studies. However,
the use of HTS in marine samples has revealed many novel sequences and previously
unknown taxa, especially those belonging to the Rozellomycota and Chytridiomycota,
which are early divergent branches of the Fungi. Habitats range from the surface of
coastal and ocean waters to the deep sea and subsurface sediments, from polar to tropical
188 Chapter 6

Figure 6.11 Morphological diversity

of a collection of fungi isolated from
a marine sponge, Ircinia variabilis,
growing on agar plates. Reprinted
from Amend et al. (2019), CC-BY-4.0.

Table 6.3 Predominant fungi in marine habitats determined by cultivation and cultivation-independent methods.
(Data from Grossart et al., 2019)

Habitat Dominant fungal group Common genera (phylum, class in parentheses)

Deep sea Filamentous fungi and yeasts Aspergillus, Cladosporium, Penicillium (Ascomycota, Eurotiomycetes)
Mycosphaerella, Alternaria. Aureobasidium (Ascomycota,
Dothidiomycetes)
Cadophora, Candida, Pichia (Ascomycota, Saccharomycetes)
Cryptococcus (Basidiomycota, Tremellomycetes)
Rhodotorula, Rhodosporidium (Basidiomycota, Ustilaginomycetes)
Malassezia (Basidiomycota, Malasseziomycetes)
Ganoderma (Basidiomycota, Agaricomycetes),

Sub-sea floor Yeasts and filamentous fungi Exophiala (Ascomycota, Eurotiomycetes)

Candida (Ascomycota, Saccharomycetes)
Cryptococcus, Trichosporon (Basidiomycota, Tremellomycetes)
Rhodotorula, Rhodosporidium (Basidiomycota, Ustilaginomycetes)
Malassezia (Basidiomycota, Malasseziomycetes)

Hydrothermal vents Yeasts and chytridiomycetes Aureobasidium (Ascomycota, Dothidiomycetes)

Malassezia (Basidiomycota, Malasseziomycetes)
Rhodotorula (Basidiomycota, Ustilaginmycetes)
Exophiala (Ascomycota, Eurotiomycetes)

Coastal and ocean water Ascomycota, Chytridiomycota, Candida, Debaryomyces (Ascomycota, Saccharomycetes)
Basidiomycota and Cryptococcus, Trichosporon (Basidiomycota, Tremellomycetes)
Rozellomycota Rhodotorula, Rhodosporidium (Basidiomycota, Ustilaginomycetes)
Malassezia (Basidiomycota, Malasseziomycetes)
Phaeosphaeria (Ascomycota, Dothidiomycetes)
Aspergillus, Cladosporium (Ascomycota, Eurotiomycetes)
Chytridium, Rhizophydium, Lobulomyces, Spizellomyces
(Chytridiomycota, Chytridiomycetes)
Clade LKM11 (Rozellomycota)

Polar systems Chytridiomycota, Rozellomycota, Rhizophydium, Lobulomyces, Podochytrium, Rhizoclosmatium,

Basidiomycota and Ascomycota Cladochytrium, Cyclopsomyces, Mesochytrium, Polychytrium
(Chytridiomycota, Chytridiomycetes)
Clade LKM11 (Rozellomycota)
Glaciozyma, Sporobolomyces (Basidiomycota, Microbotryomycetes)
Mrakia (Basidiomycota, Tremellomycetes)
Cadophora, Kluyveromyces, Candida (Ascomycota,
Saccharomycetes)
Cladosporium, Penicillium (Ascomycota, Eurotiomycetes)
Rhodotorula (Basidiomycota, Ustilaginomycetes)
Epicoccum, Aureobasidium (Ascomycota, Dothidiomycetes)
Basidiobolus (Zoopagomycota, Basidiobolomycetes)
 Marine Eukaryotic Microbes 189

regions, on surfaces and particles of organic material, and in association with a wide
FROM DANDRUFF TO
range of other organisms.
i DEEP-SEA VENTS
A major challenge has been defining fungal species that are truly marine. Many fungi Fungi in the genus Malassezia have
associated with freshwater and terrestrial habitats like soil and plants can grow in sea- been known for many years to be
water, for example, when decomposing detritus finds its way into the sea from coastal associated with human skin condi-
runoff or rivers. Terrestrial fungi also produce massive numbers of aerial spores, which tions such as dandruff and eczema.
can be blown into the sea. Indeed, many of the fungal species (especially ascomycetes and These fungi lack ability to synthe-
basidiomycetes) identified in marine samples are often well characterized from soil and size fatty acids and use extracel-
lular lipases to obtain these. It
plant sources. Such “accidental” introductions would not identify an isolate as “marine,”
was thought that Malassezia spp.
although physiological and gene expression studies show that many of these fungi are had evolved into a narrow niche
equally at home on land and in the sea and might be considered as amphibious or aero- associated with the skin of mam-
aquatic. In laboratory experiments, most fungi can grow at NaCl concentrations found malian hosts but sequencing of
in seawater since they have efficient systems for controlling cell turgor and function in environmental DNA samples from
hypertonic environment. Phylogenetic analysis shows that many marine fungi group with diverse environments has revealed
lineages from terrestrial habitats, suggesting that there have been repeated transitions that they are extremely wide-
from terrestrial to marine habitats. To be recognized as truly marine, a fungus must rely spread and hyper-diverse. As well
on the marine habitat for part or all of its life cycle. We need to demonstrate that it is as occurring in plants and soils,
regularly associated with marine habitats and is able to grow and reproduce (sporulate) sequences related to Malassezia
have been detected in deep-sea
in the sea. Further evidence of this comes from demonstration of in situ transcription
sediments, hydrothermal vents,
and metabolic activity that can be directly associated with the fungus and from genomic corals, and other invertebrates
evidence of adaptations to the marine environment (or to hosts, for symbionts and para- (Amend, 2014). An obvious ques-
sites). There is now sufficient evidence to conclude that fungi are a significant component tion facing researchers is whether
of the marine microbiome, but despite growing evidence of the diversity of fungal spe- these marine sequences might
cies in marine samples, reliable estimates of their abundance, contribution to metabolic have arisen from contamination
processes, and ecosystem functions remain limited. However, improved methodology and of samples by aerial skin flakes
coordinated efforts by the research community are leading to rapid advances, some of from the experimenters, but this
which are discussed in Box 6.2. was ruled out by the detection of
distinctive sequences of Malassezia
Besides their activities in natural processes, fungi have important roles in anthropo- RNA (which degrades very rapidly)
and by using stringent protocols to
genic pollution. In particular, their ability to degrade complex hydrocarbons means that
exclude contamination (Orsi et al.,
they occur at natural oil seeps and show increased activity after accidental oil spills 2013). Amend (2014) concluded
(Box 13.2). They are also involved in the colonization of plastic debris in the oceans; that members of the Malassezia
this can result in slow degradation or alteration of the physical properties of the plastic, lineage are among the most wide-
which can have important influences on its long-term fate and effects on marine pro- spread fungi on the planet.
cesses (Box 13.3).

Conclusions
This chapter has shown that eukaryotic microbes are extremely diverse and have critical roles
in ocean processes, as well as providing important insights to the evolutionary history of our
oceans. Phototrophic protists are major contributors to primary productivity, while phago-
trophic predators of other microbes and saprotrophic fungi provide conduits for the flow of
nutrients into marine food webs. One of the most significant discoveries of recent years is
the large extent to which many protists mix these nutritional strategies, with profound influ-
ences on ecology. Recent advances in methods for the application of molecular techniques
to the study of protists and fungi is leading to greatly improved knowledge of their diversity
and functional roles in ocean processes. As well as their role in food webs and biogeochemi-
cal cycles, the activities of protists and fungi have many significant implications for human
welfare, both negative (e.g. diseases of marine life or harmful blooms) and positive (for bio-
technological exploitation).
190 Chapter 6

BOX 6.2 RESEARCH FOCUS

Exciting times for marine mycology

Chytrid parasites promote trophic links in aquatic food webs. in freshwater systems. Model experimental systems to investigate
Numerous studies have shown that marine fungi are associated with interactions between marine fungal parasites, phytoplankton hosts,
a variety of different organisms, including protists (Gutiérrez et al., and zooplankton predators are now being developed.
2016); macroalgae (Wainwright et al., 2017); and corals, sponges,
and other invertebrates (Paz et al., 2010; Yarden, 2014; Ainsworth Nevertheless, the results of several in situ surveys suggest that para-
et al., 2017). Their role as parasites and pathogens is an area of sitism by marine chytrids does have similar food web effects. A
active exploration. Although there are many examples in macro- major insight was obtained by Gutiérrez et al. (2016), who inves-
ecology of the impact of infectious agents such as viruses on com- tigated the occurrence of infection of diatoms in the Humboldt
munity structure and diversity due to mass mortalities, parasites current system off central Chile. This is one of the most produc-
are frequently missing or under-represented in food web models. tive marine ecosystems due to upwelling of nutrients, which sus-
Lafferty et al. (2008) provides examples of various ways in which tains high primary production, mainly by blooms of large diatoms
parasites should be fitted into food webs and how they affect their such as Skeletonema and Thalassiosira. Infection by chytrid
topology and dynamics. parasites was observed using SEM and EFM, revealing attached
spherical structures (sporangia) formed from recently settled zoo-
The association of chytrids with phytoplankton is an area of spe- spores, and branched rhizoid structures penetrating the diatom
cial interest, because chytrids are dominant obligate or facultative cell (Figure 6.12A). Measurements of abundance of these diatoms
parasites in aquatic systems. They infect a variety of algae, other showed that they increased during the upwelling of cold, low-
protists, and cyanobacteria (Frenken et al., 2017). Kagami et al. oxygen waters during the austral spring, peaking in November at
(2014) investigated the role of chytrids in aquatic food webs, par- ~5 × 104 cells per liter. Abundance of chytrids followed the same
ticularly in the transfer of material from poorly grazed phytoplank- seasonal pattern, ranging from 2–391 per liter for attached, and
ton to zooplankton—a component of the food web that they termed from 111–395 per liter for detached sporangia. The ratio of attached
the “mycoloop.” When chytrid zoospores infect their host, they to detached sporangia and the number of diatoms both showed a
penetrate the cell and feed, before forming numerous motile zoo- sharp decline at the beginning of summer. This coincided with an
spores. Agha et al. (2016) conducted experiments with freshwater increase in the abundance of Chaetoceros, which became the domi-
cyanobacteria and the zooplankter Daphnia. In four out of five fit- nant diatom at the start of summer. The decline of diatom blooms in
ness parameters, Daphnia performed better when feeding on chy- upwelling systems is usually attributed to depletion of nutrients and
trid-parasitized filamentous cyanobacteria than in the absence of grazing, but Gutiérrez et al. concluded that specificity of infection
infection. The zoospores released from the infected phytoplankton by chytrids partly accounts for the decline of particular diatoms.
provide a highly nutritional food source for the zooplankton. Also, They calculated the biomass of detached zoospores at 3.7–10.1 µg
chytrid infection breaks up the cyanobacterial filaments, which C L−1 and suggest that these provide a significant conduit of energy
makes them more edible for Daphnia and the death and decay of within the pelagic trophic food web in this marine upwelling sys-
the cyanobacteria increases the biomass of heterotrophic bacteria, tem, transferring organic carbon to zooplankton, which are inhib-
which provide an additional food source. Such additional trophic ited from grazing on Skeletonema and Thalassiosira due to their
links between primary and secondary production processes may be production of toxic aldehydes.
especially important when certain phytoplankton species bloom.
In another recent study, Frenken et al. (2018) showed that rotifers Diatoms are also the most prominent and diverse component of algal
declined in cultures of a large filamentous blooming cyanobacte- blooms in polar sea ice systems (Gradinger et al., 2010). Hassett
rium because they could not ingest them, but they grew rapidly if a and Gradinger (2016) provided the first evidence linking diatom
chytrid parasite that infected the cyanobacteria was present. Again, populations with parasitism by Chytridiomycota, which increased
this was due to the zooplankton feeding on the lipid-rich chytrid under higher light penetration due to reduced snow cover on the
zoospores. To date, these experimental studies have been conducted ice. The spatially constrained labyrinthine brine channels in sea ice

Figure 6.12 Fungi detected in

the coastal upwelling ecosystem
off Chile. A. SEM of chytrid spo-
rangia attached to the diatom
Skeletonema sp., showing pen-
etration of filament by rhizoids.
B. Epifluorescence microscopy
of septate hyphae detected by
Calcofluor staining. Credit: courtesy
of Marcelo Gutiérrez, Universidad
de Concepción.
 Marine Eukaryotic Microbes 191

BOX 6.2 RESEARCH FOCUS

(p.22) are thought to be a major reservoir of fungal diversity, and using amplification of DNA with eukaryotic primers and block-
to investigate this, Hassett et al. (2017) enumerated chytrids in sea ing techniques that reveal the types of prey ingested. Maloy et al.
ice across the Western Arctic using microscopic counts. They also (2013) showed that fungi constituted 16% of the diet of bivalve lar-
identified fungal diversity using HTS targeting the 18S rRNA gene, vae and Hu et al. (2015) showed the presence of several genera of
supplemented by deep sequencing of the 28S rRNA gene. Analysis Ascomycota and Basidiomycota in coral reef copepods (although
of over 7 million unique sequences revealed a broad range of diverse chytrids were not found in these studies).
Ascomycota, Basidiomycota, and Chytridiomycota. These Arctic
chytrids were mainly assigned to the new order Lobulomycetales. Annual seasonal increases in abundance of chytrids linked to
In areas with light penetration due to low snow cover, chytrids were diatom blooms were also observed in a HTS- and qPCR-based
observed to be actively parasitizing large pennate diatoms, with the study over six years of seasonal variation in coastal plank-
species Pleurosigma elongatum showing infection rates of ~25%, tonic fungi at the Western English Channel Station L4, off the
suggesting a major role in trophic transfer. Further studies by this coast of Plymouth, conducted by Taylor and Cunliffe (2016).
group used novel methods of assessing the biomass of chytrid fungi Assemblages of chytrids, and the more dominant Ascomycota
in sea ice and seawater, by integrating information from CARD- and Basidiomycota, were dynamic and linked to fluctuations in
FISH analysis using Chytridiomycota-specific probes with the con- particulate organic carbon and nitrogen, temperature, and salin-
centrations of ergosterol, a major component of fungal membranes ity (linked to input from rivers). Although the method used to
(Hassett et al., 2019). This confirmed that the biomass of chytrids in determine abundance (Q-PCR of the 18S rRNA gene) could not
sea ice and seawater is similar to the biomass estimates of marine be used to produce carbon biomass estimates, the rapid changes in
taxa considered necessary to support marine food webs and eco- abundance again suggest that fungi are responsible for significant
system functions. carbon turnover.

The role of fungi in marine carbon cycling. Estimates of biomass Most fungi are osmotrophic saprotrophs feeding by absorbing DOC
composition in other marine habitats also confirms that the impor- generated by the action of the extracellular enzymes that they pro-
tance of fungi has been overlooked. Gutiérrez et al. (2010) used duce. Gutiérrez et al. (2011) had previously shown that abundance
0.22 µm membrane filtration to capture particles from seawater and of mycoplankton in the Humboldt current system was positively
stained them with a chitin stain (Calcofluor) before examining with correlated with processing of high molecular-weight polymers. In
epifluorescence microscopy (Figure 6.12B). This showed numerous a follow-up to their study of the Western English Channel fungi,
hyphae and fungal aggregates, accounting for 0.03–0.12 ug C L−1 in Cunliffe et al. (2017) sought to characterize mycoplankton that
the top 15 m of the water column in the Humboldt system, decreas- actively utilize algal-derived polysaccharides. They combined sta-
ing with depth. These authors also showed seasonal variability in ble isotope probing (SIP) with analysis of 18S rRNA genes in L4
fungal carbon associated with changes in phytoplankton biomass— seawater incubated with 12C and 13C-labeled polysaccharide micro-
fungal biomass was comparable to that of bacteria during upwelling gels produced from cultures of the diatom Phaeodactylum tricornu-
(Gutiérrez et al., 2011). Mycelial growth increases the surface to tum. Comparison of the gene libraries from the 12C (control) and 13C
volume ratio and gives fungi the possibility to mobilize intracellu- incubations showed that the latter were enriched in fungal sequences
lar content and substrates in heterogeneous environments. Hyphae relative to other plankton groups, indicating that they had assimi-
also present a microhabitat for other microorganisms. Bochdansky lated carbon from the polysaccharide microgels. Dominant OTUs
et al. (2017) used gentle gravity filtration to collect slow-sinking belonged to the genera Malassezia and Cladosporium. Cultures of
marine snow from bathypelagic depths of ~1000–3000 m and used Cladosporium F2 were isolated on agar plates containing a glu-
CARD-FISH with various eukaryotic primers to assess abundance can identical to the algal polysaccharide laminarin and shown to
of different groups. They found that approximately one fungal cell possess strong β-glucosidase activity. Analysis of the 6-year data-
occurred for every 1000 bacterial or archaeal cells (revealed by set of fungal diversity from the L4 site showed that abundance of
DAPI epifluorescence). By extrapolation, this puts fungal biomass specific Cladosporium-related OTUs correlated with blooms of
on a par with that of bacteria or archaea and suggests that fungi either Leptocylindricus or Chaetoceros diatoms, both of which are
have a previously overlooked significant role in degradation of known TEP producers. Cunliffe et al. (2017) also analyzed the Tara
organic matter in the deep sea, especially because of their propen- Oceans dataset and found that Cladosporium OTUs accounted for
sity to degrade complex organic compounds. Bochdansky et al. also 26% of the mycoplankton communities, almost always coinciding
suggest that penetration of the marine snow by hyphal filaments with high phytoplankton biomass.
may help to stabilize these particles, facilitating their transport to
deep waters. In a HTS study of the vertical distribution of eukary- Conclusions. These examples of research indicate that fungi are
otic microbes from the surface to the hadal zone of the Marianas an abundant and highly active component of many marine habi-
Trench, Xu et al. (2018) found that fungal sequences were most tats. In a review of a “state of the art” discussion of marine mycol-
abundant at 1759 m. ogy, Amend et al. (2019) concluded: “We are in exciting times for
marine fungal functional biology and ecology, and studies of these
Further evidence that fungi form a major component of marine communities will very likely force us to rethink global biogeo-
food webs comes from molecular dietary analysis of zooplankton chemical cycles.”
192 Chapter 6

References and further reading González, J.M. & Suttle, C.A. (1993) Grazing by marine nanoflagellates
on viruses and virus-sized particles: Ingestion and digestion. Mar. Ecol.
Protist—diversity, and systematics Prog. Ser. 94: 1–10.
Adl, S.M., Bass, D., Lane, C.E. et al. (2019) Revisions to the classifica- Jezbera, J., Hornak, K., & Simek, K. (2005) Food selection by bacteriv-
tion, nomenclature, and diversity of eukaryotes. J. Eukaryot. Microbiol. orous protists: Insight from the analysis of the food vacuole content by
66: 4–119. means of fluorescence in situ hybridization. FEMS Microbiol. Ecol. 52:
351–363.
Ainsworth, T.D., Fordyce, A.J., & Camp, E.F. (2017) The other microeu-
karyotes of the coral reef microbiome. Trends Microbiol. 25: 980–991. Johnson, M.D., Lasek-Nesselquist, E., Moeller, H.V. et al. (2017)
Mesodinium rubrum: The symbiosis that wasn’t. Proc. Natl. Acad. Sci.
Caron, D.A., Countway, P.D., Jones, A.C. et al. (2012) Marine protistan
USA 114: E1040–E1042.
diversity. Annu. Rev. Mar. Sci. 4: 467–493.
Laundon, D., Larson, B.T., McDonald, K. et al. (2019) The architecture of
de Vargas, C., Audic, S., Henry, N. et al. (2015) Ocean plankton. Eukaryotic
cell differentiation in choanoflagellates and sponge choanocytes. PLOS
plankton diversity in the sunlit ocean. Science 348: 1261605.
Biol. 17: e3000226.
Gast, R.J., Sanders, R.W. & Caron, D.A. (2009) Ecological strategies
Medina, L.E., Taylor, C.D., Pachiadaki, M.G. et al. (2017) A review of protist
of protists and their symbiotic relationships with prokaryotic microbes.
grazing below the photic zone emphasizing studies of oxygen-depleted water
Trends Microbiol. 17: 563–569.
columns and recent applications of in situ approaches. Front. Mar. Sci. 4: 105.
Keeling, P.J. & Campo, J. Del (2017) Marine protists are not just big bac-
Pernice, M.C., Forn, I., Gomes, A. et al. (2015) Global abundance of
teria. Curr. Biol. 27: R541–R549.
planktonic heterotrophic protists in the deep ocean. ISME J. 9: 782–792.
Lima-Mendez, G., Faust, K., Henry, N. et al. (2015) Determinants of com-
Seeleuthner, Y., Mondy, S., Lombard, V. et al. (2018) Single-cell genomics
munity structure in the global plankton interactome. Science 348: 1262073.
of multiple uncultured stramenopiles reveals underestimated functional
Millette, N.C., Grosse, J., Johnson, W.M. et al. (2018) Hidden in plain diversity across oceans. Nat. Commun. 9: 310.
sight: The importance of cryptic interactions in marine plankton. Limnol.
Sengupta, A., Carrara, F. & Stocker, R. (2017) Phytoplankton can actively
Oceanogr. Lett. 3: 341–356.
diversify their migration strategy in response to turbulent cues. Nature
Moreira, D. & López-García, P. (2002) The molecular ecology of micro- 543: 555–558.
bial eukaryotes unveils a hidden world. Trends Microbiol. 10: 31–38.
Sieracki, M.E., Poulton, N.J., Jaillon, O. et al. (2019) Single cell genomics
Moreira, D. & López-García, P. (2003) Are hydrothermal vents oases for yields a wide diversity of small planktonic protists across major ocean
parasitic protists? Trends Parasitol. 19: 556–558. ecosystems. Sci. Rep. 9: 6025.
Nissimov, J.I. & Bidle, K.D. (2017) Stress, death, and the biological glue of
sinking matter. J. Phycol. 53: 241–244. Mixotrophy
Rocke, E., Pachiadaki, M.G., Cobban, A. et al. (2015) Protist community graz- Mcmanus, G.B., Schoener, D., & Haberlandt, K. (2012) Chloroplast sym-
ing on prokaryotic prey in deep ocean water masses. PLoS One 10: e0124505. biosis in a marine ciliate: Ecophysiology and the risks and rewards of host-
ing foreign organelles. Front Microbiol 3: 321.
Sherr, B.F., Sherr, E.B., Caron, D.A. et al. (2007) Oceanic protists.
Oceanography 20: 130–134. Qiu, D., Huang, L., & Lin, S. (2016) Cryptophyte farming by symbiotic
ciliate host detected in situ. Proc. Natl. Acad. Sci. 113: 12208–12213.
Stoecker, D.K., Johnson, M.D., de Vargas, C., & Not, F. (2009) Acquired
phototrophy in aquatic protists. Aquat. Microb. Ecol. 57: 279–310. Stoecker, D.K., Hansen, P.J., Caron, D.A., & Mitra, A. (2017) Mixotrophy
in the Marine Plankton. Annu. Rev. Mar. Sci. 9: 311–335.
Strassert, J.F., Karnkowska, A., Hehenberger, E. et al. (2018) Single cell
genomics of uncultured marine alveolates shows paraphyly of basal dino- Stoecker, D.K., Johnson, M.D., de Vargas, C., & Not, F. (2009) Acquired
flagellates. ISME J. 12: 304. phototrophy in aquatic protists. Aquat. Microb. Ecol. 57: 279–310.
Worden, A.Z. & Not, F. (2008) Ecology and diversity of picoeukaryotes. Ward, B.A. (2019) Mixotroph ecology: More than the sum of its parts.
In: Microbial Ecology of the Oceans, D.L. Kirchman (ed.). Wiley-Liss Proc. Natl. Acad. Sci. USA 116: 5846–5848.
Inc., New York, pp. 159–205.
Coccolithophores
Heterotrophic flagellates and grazing Balch, W.M. (2018) The ecology, biogeochemistry, and optical properties
Boenigk, J. & Arndt, H. (2002) Bacterivory by heterotrophic flagellates: of coccolithophores. Annu. Rev. Mar. Sci. 10: 71–98.
Community structure and feeding strategies. Ant. van Leuevenhoek 81: Monteiro, F.M., Bach, L.T., Brownlee, C. et al. (2016) Why marine phyto-
465–480. plankton calcify. Sci. Adv. 2: e1501822.
De Corte, D., Paredes, G., Yokokawa, T. et al. (2018) Differential response Taylor, A.R., Brownlee, C., & Wheeler, G. (2017) Coccolithophore cell
of Cafeteria roenbergensis to different bacterial and archaeal prey charac- biology: Chalking up progress. Annu. Rev. Mar. Sci. 9: 283–310.
teristics. Microb. Ecol. 78: 1–5.
Xu, K., Hutchins, D., & Gao, K. (2018) Coccolith arrangement follows Eulerian
del Campo, J., Not, F., Forn, I. et al. (2013) Taming the smallest predators mathematics in the coccolithophore Emiliania huxleyi. Peer J. 6: e4608.
of the oceans. ISME J. 7: 351–358.
Fenchel, T. & Patterson, D.J. (1988) Cafeteria roenbergensis nov. gen., Diatoms
nov. sp., a heterotrophic microflagellate from marine plankton. Mar.
Microb. Food Webs 3: 9–19. Bidle, K.D. (2002) Regulation of oceanic silicon and carbon preservation
by temperature control on bacteria. Science 298: 1980–1984.
Gawryluk, R.M., del Campo, J., Okamoto, N. et al. (2016) Morphological
identification and single-cell genomics of marine diplonemids. Curr. Biol. Bidle, K.D. & Azam, F. (1999) Accelerated dissolution of diatom silica by
26: 3053–3059. marine bacterial assemblages. Nature 397: 508–512.

Gimmler, A., Korn, R., de Vargas, C. et al. (2016) The Tara Oceans voy- Bidle, K.D. & Azam, F. (2001) Bacterial control of silicon regeneration
age reveals global diversity and distribution patterns of marine planktonic from diatom detritus: Significance of bacterial ectohydrolases and species
ciliates. Sci. Rep. 6: 33555. identity. Limnol. Oceanogr. 46: 1606–1623.
 Marine Eukaryotic Microbes 193

Gradinger, R., Bluhm, B.A., Hopcroft, R.R. et al. (2010) Marine life in the arc- Frenken, T., Alacid, E., Berger, S.A. et al. (2017) Integrating chytrid fun-
tic. In: Life in the World’s Oceans. Wiley-Blackwell, Oxford, UK, pp. 183–202. gal parasites into plankton ecology: Research gaps and needs. Environ.
Kröger, N. & Poulsen, N. (2008) Diatoms—From cell wall biogenesis to Microbiol. 19: 3802–3822.
nanotechnology. Annu. Rev. Genet. 42: 83–107. Frenken, T., Wierenga, J., van Donk, E. et al. (2018) Fungal parasites of
Leblanc, K., Queguiner, B., Diaz, F. et al. (2018) Nanoplanktonic dia- a toxic inedible cyanobacterium provide food to zooplankton. Limnol.
toms are globally overlooked but play a role in spring blooms and carbon Oceanogr, 63: 2384–2393.
export. Nat. Commun. 9: 953. Golubic, S., Radtke, G. & Campion-Alsumard, T.L. (2005) Endolithic
fungi in marine ecosystems. Trends Microbiol. 13: 229–235.
Dinoflagellates Grossart, H.-P., Van den Wyngaert, S., Kagami, M. et al. (2019) Fungi in
Akbar, M., Ahmad, A., Usup, G., & Bunawan, H. (2018) Current knowl- aquatic ecosystems. Nat. Rev. Microbiol. 17: 339–354.
edge and recent advances in marine dinoflagellate transcriptomic research. Gutiérrez, M.H., Jara, A.M. & Pantoja, S. (2016) Fungal parasites infect
J. Mar. Sci. Eng. 6: 13. marine diatoms in the upwelling ecosystem of the Humboldt current sys-
Figueroa, R.I., Estrada, M., & Garcés, E. (2018) Life histories of microal- tem off central Chile. Environ. Microbiol. 18: 1646–1653.
gal species causing harmful blooms: Haploids, diploids and the relevance Gutiérrez, M.H., Pantoja, S., Quiñones, R.A. & González, R.R. (2010)
of benthic stages. Harmful Algae 73: 44–57. First record of filamentous fungi in the coastal upwelling ecosystem off
Gómez, F. (2012) A quantitative review of the lifestyle, habitat and trophic central Chile. Gayana 74: 66–73.
diversity of dinoflagellates (Dinoflagellata, Alveolata). System. Biodiv. 10: Gutiérrez, M.H., Pantoja, S., Tejos, E. & Quiñones, R.A. (2011) The role of
267–275. fungi in processing marine organic matter in the upwelling ecosystem off
Grossart, H.P., Levold, F., Allgaier, M. et al. (2005) Marine diatom species Chile. Mar. Biol. 158: 205–219.
harbour distinct bacterial communities. Environ. Microbiol. 7: 860–873. Hassett, B.T., Borrego, E.J., Vonnahme, T.R. et al. (2019) Arctic marine
Haddock, S.H.D., Moline, M.A. & Case, J.F. (2010) Bioluminescence in fungi: Biomass, functional genes, and putative ecological roles. ISME J.
the sea. Annu. Rev. Mar. Sci. 2: 443–493. 13: 1484.

Hastings, J.W. (2007) The Gonyaulax clock at 50: Translational control of Hassett, B.T., Ducluzeau, A.-L.L., Collins, R.E. & Gradinger, R. (2017)
circadian expression. Cold Spring Harbor Symp. Quant. Biol. 72: 141–144. Spatial distribution of aquatic marine fungi across the western Arctic and
sub-arctic. Environ. Microbiol. 19: 475–484.
Liu, H., Stephens, T.G., González-Pech, R.A. et al. (2018) Symbiodinium
genomes reveal adaptive evolution of functions related to coral-dinoflagel- Hassett, B.T. & Gradinger, R. (2016) Chytrids dominate arctic marine fun-
late symbiosis. Commun. Biol. 1: 95. gal communities. Environ. Microbiol. 18: 2001–2009.

Valiadi, M. & Iglesias-Rodriguez, D. (2013) Understanding biolumines- Hu, S., Guo, Z., Li, T. et al. (2015) Molecular analysis of in situ diets of
cence in dinoflagellates—How far have we come? Microorganism 1: 3–25. coral reef copepods: Evidence of terrestrial plant detritus as a food source
in Sanya Bay, China. J. Plankton Res. 37: 363–371.
Wisecaver, J.H. & Hackett, J.D. (2011) Dinoflagellate genome evolution.
Annu. Rev. Microbiol. 65: 369–387. Kagami, M., Miki, T. & Takimoto, G. (2014) Mycoloop: Chytrids in
aquatic food webs. Front. Microbiol. 5: 166.
Foraminifera and radiolarians Lafferty, K.D., Allesina, S., Arim, M. et al. (2008) Parasites in food webs:
The ultimate missing links. Ecol. Lett. 11: 533–46.
Anderson, O.R. (2001) Protozoa, radiolarians. In: Encyclopedia of Ocean
Sciences, J. Steele, S. Thorpe & K. Turekian (eds.). Academic Press, pp. Maloy, A.P., Culloty, S.C. & Slater, J.W. (2013) Dietary analysis of
2315–2319. small planktonic consumers: A case study with marine bivalve larvae. J.
Plankton Res. 35: 866–876.
Gooday, A.J., Sykes, D., Góral, T. et al. (2018) Micro-CT 3D imaging reveals
the internal structure of three abyssal xenophyophore species (Protista, Nilsson, R.H., Anslan, S., Bahram, M. et al. (2019) Mycobiome diver-
Foraminifera) from the eastern equatorial Pacific Ocean. Sci. Rep. 8: 12103. sity: High-throughput sequencing and identification of fungi. Nat. Rev.
Microbiol. 17: 95–109.
Lampitt, R.S., Salter, I. & Johns, D. (2009) Radiolaria: Major exporters of
organic carbon to the deep ocean. Glob. Biogeochem. Cycles 23: GB1010. Orsi, W., Biddle, J.F. & Edgcomb, V. (2013) Deep sequencing of subsea-
doi: 10.1029/2008GB003221 floor eukaryotic rRNA reveals active fungi across marine subsurface prov-
inces. PLoS One 8: e56335.
Murray, J. (2008) Ecology and Applications of Benthic Foraminifera.
Cambridge University Press. Paz, Z., Komon-Zelazowska, M., Druzhinina, I.S. et al. (2010)
Diversity and potential antifungal properties of fungi associated with a
Fungi Mediterranean sponge. Fungal Divers. 42: 17–26.
Taylor, J.D. & Cunliffe, M. (2016) Multi-year assessment of coastal plank-
Agha, R., Saebelfeld, M., Manthey, C. et al. (2016) Chytrid parasitism
tonic fungi reveals environmental drivers of diversity and abundance.
facilitates trophic transfer between bloom-forming cyanobacteria and zoo-
ISME J. 10: 2118.
plankton (Daphnia). Sci. Rep. 6: 35039.
Sherwood, A.R., Smith, C.M. &Amend, A.S. (2017) Fungi associated with
Amend, A. (2014) From dandruff to deep-sea vents: Malassezia-like Fungi
mesophotic macroalgae from the ‘Au‘au Channel, west Maui are differen-
are ecologically hyper-diverse. PLoS Pathog. 10: e1004277.
tiated by host and overlap terrestrial communities. Peer J. 5: e3532.
Amend, A., Burgaud, G., Cunliffe, M. et al. (2019) Fungi in the marine envi-
Wainwright, B.J., Zahn, G.L., Spalding, H.L., et al. (2017) Fungi associated
ronment: Open questions and unsolved problems. MBio 10: e101189-18.
with mesophotic macroalgae from the ‘Au ‘au Channel, west Maui are dif-
Bochdansky, A.B., Clouse, M.A. & Herndl, G.J. (2017) Eukaryotic ferentiated by host and overlap terrestrial communities. Peer J. 5: e3532.
microbes, principally fungi and labyrinthulomycetes, dominate biomass
Xu, Z., Wang, M., Wu, W. et al. (2018) Vertical distribution of microbial
on bathypelagic marine snow. ISME J. 11: 362–373.
eukaryotes from surface to the hadal zone of the Mariana Trench. Front.
Cunliffe, M., Hollingsworth, A., Bain, C. et al. (2017) Algal polysaccha- Microbiol. 9: 2023.
ride utilisation by saprotrophic planktonic marine fungi. Fungal Ecol. 30:
Yarden, O. (2014) Fungal association with sessile marine invertebrates.
135–138.
Front. Microbiol. 5: 228.
Chapter 7
Marine Viruses

The true abundance and ecological importance of viruses in marine ecosystems and global
processes has only been elucidated in the last few decades. Recent discoveries have led to
the emergence of virus ecology as one of the most exciting and fastest developing branches
of marine science. The world’s oceans are estimated to contain more than 1030 viruses—
they are the smallest and most abundant biological entities in marine ecosystems. All forms
of cellular life—from bacteria to whales—are susceptible to viral infection. Through their
interactions with all types of marine organisms, viruses play a critical role in the structuring
of marine communities, in ocean processes, and in biogeochemical cycles. In this chapter,
the focus is primarily on the properties and activities of marine viruses with respect to their
interactions with other members of the plankton, especially bacteria, microalgae, and other
protists. Viruses that infect other marine organisms, such as invertebrates, fish, marine ver-
tebrates, or macroalgae are considered in Chapter 11. Human pathogenic viruses that may be
introduced into the marine environment via pollution are discussed in Chapter 12.

Key Concepts
• Viruses contain either DNA or RNA and are obligate intracellular parasites that take
over the biosynthetic machinery of host cells.
• There are more than 1030 viruses in the oceans; marine viruses are extremely diverse
in host range, size, structure, and genomic composition.
• Viruses that infect bacteria (phages) and other planktonic microbes are responsible
for the turnover of microbial communities, with consequent major effects on nutrient
cycles and biogeochemical processes.
• Viruses are responsible for structuring the diversity of microbial communities and
manipulating life histories through co-evolution with their hosts and widespread
genetic exchange.
• Analysis of marine viral metagenomes (viromes) shows that groups of viral genotypes
are globally distributed, although their relative abundance is affected by local selection
factors.
• Studies of marine viromes are contributing to a reappraisal of the nature and evolution
of viruses and cells.
196 Chapter 7

Viruses are highly diverse non-cellular microbes

Viruses are non-cellular biological entities. They cannot be described as microorganisms,
but they are included in the more encompassing term microbes. Virus particles, termed “viri-
ons,” are composed of nucleic acid surrounded by a protein coat (capsid). A fundamental
difference in the makeup of viruses compared with cells is that viruses contain only one type
of nucleic acid, either DNA or RNA, whereas cells contain both. Also, they do not contain
ribosomes and therefore cannot synthesize proteins. Therefore, viruses are obligate intracel-
lular parasites, that rely on the biochemical machinery of the cell they infect to complete the
flow of genetic information, synthesize proteins, and replicate their virions. Since all forms
of cellular life are susceptible to virus infection, we can reasonably speculate that every type
of marine organism—from bacteria to whales—is a host to at least one type of virus. Many
organisms are known to be hosts to several different viruses. Table 7.1 shows a list of repre-
sentative marine virus families and their hosts; as can be seen, virions show great variation
in size and exist in various morphological forms.

Various schemes have been developed to classify the different types of viruses. In the early
days of virology in the mid-twentieth century, viruses were classified largely on the basis of
their hosts (plants, animals, or humans). As knowledge of these accumulated in the 1960s and
1970s, the structure and replication of viruses became the main criteria for classification. The

?
WHAT’S IN Baltimore classification scheme devised in the 1970s divided viruses into seven groups, based
A NAME? on the nature of their genome and their replication strategy. Since the 1990s, the International
As in other microbes with extensive Committee on Taxonomy of Viruses (ICTV) has developed rules for classifying and naming
genetic diversity, the concept of viruses using taxonomic divisions and Latin species names, but these differ from the bino-
a “species” of virus is problem- mial format used for cellular organisms. The organization reflects the phylogenetic ancestry,
atic. The ICTV provides a formal structure, replication, hosts, and transmission vectors of different viruses. Thus, in the latest
definition of a viral species as (2018) ICTV taxonomy, there are six classes (suffix -viricetes), 14 orders (-virales), 143 fami-
a monophyletic group whose lies (-viridae), 846 genera (-virus), and 4958 species of viruses. Some orders and families are
properties can be distinguished divided into multiple sub-groups.
from those of other species by
multiple criteria. Examples of
Viral genomes can be circular, linear, or segmented; with either a DNA or RNA genome,
formal names for cultured viral
species are Cafeteria roenbergensis both of which can be either single- or double-stranded. A key factor in the replication strat-
virus (genus Cafeteriavirus, family egy of single-stranded viruses—used in the Baltimore classification scheme—is whether the
Mimiviridae) and Emiliania huxleyi nucleic acid is positive or negative sense. Positive-sense RNA can serve as viral mRNA
virus 86 (genus Coccolithovirus, fam- directly, whereas negative-sense RNA must be converted to a complementary mRNA by
ily Phycodnaviridae). Surprisingly, RNA-dependent RNA polymerase. As shown in Table 7.1, the size of virions in different
the nomenclature chosen by viral groups varies greatly; the smallest are only ~20–30 nm in diameter. The largest known
virologists does not follow the viruses belong to a clade known as the nucleocytoplasmic large DNA viruses (NCLDVs), and
binomial Genus species system the family Mimiviridae contains the largest of them all—some virions of Tupanvirus are up
used in the rest of biology (see to 2.3 µm in length. We will return to these so-called “giant viruses” later. The size of viral
p.118), Virus strains may have
genomes is very variable, ranging from about 3.2 kb in the Hepapadnaviridae to >1200 kb
informal names or acronyms
reflecting their host. For example,
in the Mimiviridae.
the species of Cafeteriavirus and
Coccolithovirus named above are The virion contains identical protein subunits (capsomeres), which self-assemble to form the
usually referred to as CroV and capsid, a shell-like structure surrounding the nucleic acid. The arrangement of the capso-
EhV-86 respectively; RDJLΦ1 is meres and the overall morphology of the virion when visualized in the electron microscope is
an acronym for Roseobacter virus a key factor in the identification and classification of viruses. Helical viruses are rod-shaped
RDJL1 (genus Xiamenvirus, fam- or filamentous, with a single type of capsomere arranged around a central cavity. Negative
ily Siphoviridae); and HaRNAV is charges on the nucleic acid bind it to positive charges on the protein helix. Many viruses have
short for Heterosigma akashiwo an icosahedral structure—a symmetry that is the optimum way of forming a closed shell-
RNA virus (Genus Marnavirus,
like structure for identical subunits. Enveloped viruses are surrounded by an outer mem-
family Marnaviridae). The great
brane containing lipids and carbohydrates derived from the host and proteins encoded by
majority of marine viruses can-
not be cultured (because in most the viral genome; the envelope is often involved in the infection of host cells by the virus.
cases their hosts also cannot be Many viruses have complex symmetry combining these different features. The best-known
cultured). This fact, and the vast examples of this type are the tailed phages that have an icosahedral head containing the viral
amounts of sequence data emerg- genome, which is bound to a helical tail with a base plate with protein fibers, which serves
ing from metagenomic studies, as a molecular syringe for delivery of the viral genome into the cell. An enormous variety of
presents a major challenge to the genomic structures can be seen among different viruses. Although we now know that there
meaningful classification of viruses are millions of different virus types, only about 5000 have been described in any detail.
(Simmonds, 2015).
 Marine Viruses 197

Table 7.1 Examples of viruses infecting marine organisms

Virus family Morphology of virion Size of Marine host(s)

virion (nm)a
Double-stranded DNA viruses
Baculoviridae Enveloped rods, some with tails 200−450 × 100−400 Crustaceans
Corticoviridae, Tectiviridae, Icosahedral with spikes 60−75 Bacteria
Herpesviridae Pleomorphic, icosahedral, enveloped 150−200 Mollusks, fish, corals mammals,
turtles
Iridoviridae Round, icosahedral 190−200 Mollusks, fish
Lipothrixviridae Thick rod with lipid coat 40 × 400 Archaea
Mimiviridae Icosahedral with microtubule-like 450–650 b Protists, corals (?), sponges (?)
projections
Myoviridae Polygonal head (icosahedral) with 50−110 (head) Bacteria
contractile tail (helical)
Nimaviridae Enveloped, ovoid with tail-like 120 × 275 Crustaceans
appendage
Papovaviridae Round, icosahedral 40−50 Mollusks
Phycodnaviridae Icosahedral 130−200 Algae
Podoviridae, Siphoviridae Icosahedral with noncontractile tail 60 (head) Bacteria
Single-stranded DNA viruses
Microviridae Icosahedral with spikes 25−27 Bacteria
Parvoviridae Round, icosahedral 20 Crustaceans
Double-stranded RNA viruses
Birnaviridae Round, icosahedral 60 Molluscs, fish
Cystoviridae Icosahedral with lipid coat 60−75 Bacteria
Reoviridae Icosahedral, some with spikes 50−80 Crustaceans, molluscs, fish,
protists
Totiviridae Round, icosahedral 30−45 Protists
Single-stranded RNA viruses (positive sense)
Caliciviridae Round, icosahedral 35−40 Fish, mammals
Coronaviridae Rod-shaped with projections 200 × 42 Crustaceans, fish, seabirds
Dicistroviridae Round, icosahedral 30 Crustaceans
Leviviridae Round, icosahedral 26 Bacteria
Marnaviridae Round, icosahedral 25 Algae
Nodaviridae Round, icosahedral 30 Fish
Picornaviridae Round, icosahedral 27−30 Algae, crustaceans,
thraustochytrids, protists (?),
mammals
Togaviridae Round, with outer fringe 66 Fish
Single-stranded RNA viruses (negative sense)
Bunyaviridae Round, enveloped 80−120 Crustaceans
Orthomyxoviridae Round, with spikes 80−120 Fish, mammals, seabirds
Paramyxoviridae Various, mainly enveloped, 60−300 × 1000 Mammals
filamentous
Rhabdoviridae Bullet-shaped with projections 45−100 × 100−430 Fish
aFor rod-shaped virions, dimensions are shown as diameter × length. b The virion of Tupanvirus has a large cylindrical tail (450 × 550 nm)
attached to the capsid (Figure 7.8). The average length of the complete virion is 1.2 µm, but some can reach 2.3 µm in length.
198 Chapter 7

Early studies of free marine viruses in the plankton were made during the late 1970s
“PLANET VIRUS”—
i 75 MILLION BLUE
WHALES, 10 MILLION
and 1980s by direct observation of filtered seawater samples, using transmission electron
microscopy (TEM). Samples can be concentrated onto grids by ultracentrifugation and
negatively stained with uranyl acetate or another similar heavy-metal stain. Strictly, we
LIGHT YEARS
should refer to structures observed in this way as “virus-like particles” (VLPs), because
The consensus estimates for viral TEM cannot indicate whether the particles contain genetic information and thus whether
density in different waters allow they are capable of infecting a host cell. In addition, bacteria can produce large numbers
us to calculate the estimated of outer membrane vesicles which may have a similar size to virions. Although technically
total number of viral particles in straightforward, observing and enumerating VLPs in seawater samples requires great care
the ocean. The total volume of
and control of variables such as the density of suspended particles. Despite many techni-
the oceans is ~1.3 × 1021 liters,
cal difficulties with sample preparation in early studies, this method led to the gradual
and assuming that the average
abundance of viruses is ~3 × 109 realization in the late 1980s that viruses are highly abundant in the marine environment.
per liter, then the total number of The TEM method is also very valuable because it shows the morphology of viruses; the
viruses in the oceans is ~4 × 1030 majority of free VLPs observed in seawater are phages (see below) having capsids with
(Suttle, 2005). This is about 10−15 pentagonal or hexagonal icosahedral three-dimensional symmetry. Both tailed and non-
times the total number of bacterial tailed forms occur, and appendages such as capsid antennae or tail fibers can sometimes
and archaeal cells, making viruses be seen (Figure 7.1). VLPs vary in size, with most phage particles being 30–100 nm in
the most abundant biological enti- diameter, thus forming part of the fraction referred to as dissolved organic matter (DOM,
ties in the ocean, comprising about see Figure 1.6). The giant viruses have particle sizes measuring hundreds of nanometers
94% of all particles containing and are larger than many bacteria—the largest known (Tupanvirus) measures 1.25 µm.
nucleic acids. Although their small
The method of sample preparation has a great influence on the observed morphology, and
size means that they only account
for 5% of ocean biomass, the total
it is generally easier to obtain details of virus morphology from isolates in laboratory cul-
carbon reservoir contained within ture—in the few cases in which this is now possible—rather than from free VLPs obtained
marine viruses is 0.2 Gt (Pg)—the directly from seawater samples.
equivalent of ~75 million blue
whales (Suttle, 2007). If the viruses Another approach developed for enumerating VLPs in sea water is an adaptation of epi-
were stacked end to end, they fluorescence microscopy, in which the sample is treated with a fluorochrome that binds to
would span >10 million light years. nucleic acids (Table 2.1). Brightly fluorescent stains such as Yo-Pro-1 and SYBR Green
Astonishingly, viruses attached to 1® have been used successfully in many studies, with VLPs appearing as very small bright
organic particles can also be swept dots that can be distinguished from the larger stained cells of bacteria and other micro-
into the atmosphere by aero- scopic plankton (Figure 7.2). This is a relatively rapid and inexpensive method, although
sols—deposited at 8 x 106 m2 in
caution is needed to avoid overestimating viral abundance, because free nucleic acids
the upper atmospheric boundary
layer—and are transported long
bound to colloidal particles or within vesicles may also appear as fluorescent dots; con-
distances before falling back to the versely, large VLPs might be confused with bacteria, resulting in an underestimation.
Earth’s surface in rain and dust. This method may also result in underestimations of viral density because RNA viruses
Thus, genetically similar viruses are and single-stranded DNA viruses do not stain well with current methods. Since viruses
dispersed to all parts of the planet
(Reche et al., 2018).

Figure 7.1 Transmission electron

micrographs of phages. A. Mature
phages inside a cultured bacterium
before lysis. B. Free viruses after lysis.
C. Cyanophage 9303-2 (Myoviridae).
D. Cyanophage 9313-4 (Siphoviridae).
Credits: A, B. Mikal Heldal and
Gunnar Bratbak, University of Bergen.
C, D. Bin Ni, and Matthew Sullivan,
Chisholm Lab, MIT.
 Marine Viruses 199

Figure 7.2 Assays for lytic phages.

A. Cultures of Ostreococcus tauri host
strain OTH95 inoculated with dilu-
tions of seawater samples contain-
ing O. tauri virus. After incubation,
lysis  or no lysis ☒ is visible (◇
shows sham-inoculated controls). B.
Schematic of plaque assay. The bac-
terial host is grown in broth culture
and mixed with a sample contain-
ing the phage. This is resuspended
in molten top agar, which contains
a low concentration of agar so that
phage can diffuse through the gel to
infect adjacent bacteria when poured
onto a plate of agar medium contain-
ing appropriate nutrients for growth
of the host. After incubation, clear
areas of lysis appear as plaques in the
lawn of bacterial growth. By plat-
ing different dilutions of the phage
suspension, the density of infective
phage can be calculated as plaque-
forming units (PFU) per mL, analo-
gous to the viable colony count of
are below the limits of resolution of the light microscopy, fluorescence microscopy pro- bacteria. C. Example of plaque assay
vides no information about the morphology of viruses—indeed, the only reason that they of O. tauri virus. Similar techniques
can be seen with this method is because of the halo of bright light emitted by the fluoro- can be used for the propagation
chromes. Many experimental and field studies have used flow cytometry (Figure 2.4) as a of viruses infecting archaea, algae,
high-throughput method to enumerate viruses alongside the community structure of their and other protists, providing suit-
potential hosts in the plankton. It is possible to discriminate various host and viral popu- able media for growth are available.
lations based on their fluorescence and scatter signals after staining with fluorochromes, Credit: A, C: Derelle et al. (2008),
and this can be applied to analysis of seasonal and spatial variation in viral dynamics in CC-BY-2.0.
mesocosm and open-water studies. Recent improvements include the use of virus-specific
probes for detection of infected cells and the use of solid-phase single-molecule multiplex
PCR (polony method).

Based on numerous studies of seawater from various geographic sites and different depths
using TEM and epifluorescence microscopy since the 1990s, a consensus value for the den-
sity of viruses in seawater of about 107 per milliliter has become established. However, there
is considerable seasonal and geographic distribution, and it is hard to generalize to all depths
and locations. In general, the viral abundance increases with the productivity of the system.
In the open ocean, virus density declines rapidly below a depth of 250 m to a relatively
constant value of about 106 per milliliter. Counts in highly productive coastal waters are
usually higher than in the open ocean and are not so dependent on depth, with typical densi-
ties of 108 per milliliter. Viruses are also found in high densities in marine sediments and
sea ice. The distribution of viruses in the water column generally mirrors the productivity
and density of host populations of bacterioplankton, as these are the most abundant hosts.
As a rough approximation, it is generally assumed that there are about ten times as many
viruses as bacteria in most seawater samples. In the photic zone of very oligotrophic waters,
cyanophages infecting photosynthetically active cyanobacteria such as Prochlorococcus and
Synechococcus form the dominant group. The distribution of viruses infecting eukaryotic
algae also generally mirrors the abundance of the population of algal hosts, but this is a small
proportion of the total. However, recent meta-analyses of previous virus abundance studies
have questioned the validity of the assumed ~10:1 linear virus to host ratio and indicate that
it is better represented as a power-law function. The estimates of VLPs to bacteria ratios are
obtained from bulk counts of relatively large samples of water, which conceals large spatial
variations in host and viral diversity. When host cell densities are high, the VLPs to bacteria
ratios are often considerably less than ten. As noted previously, seawater is a highly hetero-
geneous environment, and microbes show microscale aggregation around nutrient sources.
200 Chapter 7

Indeed, in studies of abundance at small spatial scales, changes in the density of VLPs are
THE BIG PROBLEM
i OF SMALL VESICLES
AND GTAS
very dynamic, and large fluctuations can occur over short timescales as a result of synchro-
nized lysis of host cells and different rates of degradation of released virus particles depend-
ing on environmental conditions.
Recently, the dominant phyto-
plankton Prochlorococcus and
diverse heterotrophic bacteria in Phages are viruses that infect bacterial and archaeal cells
the picoplankton were shown to
release large numbers of mem-
When viruses that infect bacteria were discovered in 1915, they were called bacterio-
brane vesicles (MVs) produced by phages—a term meaning “bacteria-eaters.” Subsequently, viruses that infect archaeal cells
“blebbing” of the outer membrane have also been discovered, but no specific term to denote them has been coined. Therefore,
of the cell envelope (Biller et al., the abbreviated term “phage” is usually used to indicate viruses infecting members of either
2014). These can reach densities of domain. Prefixes indicating a particular host range are also used: terms such as “vibrio-
105 –106 MVs mL-1, which is similar phages” (infecting Vibrio spp.), “cyanophages” (infecting cyanobacteria), or “roseophages”
to the lower range of estimates (infecting members of the Roseobacter clade) are used. The best-known isolated phages
for VLP concentrations. They are dsDNA viruses comprising a head (nucleocapsid) connected to a tail structure used for
have similar dimensions to many adsorption to the cell and injection of the nucleic acid into the host cell. These phages may
phages, which can possibly explain
be classified into a number of major groups, namely the Myoviridae (with a contractile tail,
the high numbers of non-tailed
Figure 7.1C); the Podoviridae (with a very short tail); and the Siphoviridae (with a long flex-
VLPs in TEM reported by Brum et
al. (2013). MVs may also contain ible tail, Figure 7.1D). However, a wide range of other morphologies occur in marine phages
DNA, meaning that they could be and global sampling of ocean waters and quantification by electron microscopy has con-
mistaken as VLPs in fluorescence cluded that non-tailed phages comprise 51–92% of VLPs and that the abundance of phages
counts (Soler et al., 2015). Further with RNA genomes has probably been significantly underestimated. Only a few hundred
problems arise from the fact that types of archaeal viruses have been isolated in culture, mostly associated with hyperther-
MVs are built from host cell pro- mophiles or halophiles. This small number of archaeal isolates contains a much higher level
teins and lipids but may package of morphological diversity and genome content than bacterial viruses, and it is speculated
viral DNA, whereas another type of that viruses infecting ancestral cells diverged in early evolution due to differences in the cell
particle known as a gene transfer
envelope of bacteria and archaea.
agent (GTA) is an incomplete virion
that may contain fragments of
host DNA. Both types of particles
Although the first marine phages were described in the 1950s, the significance of early
can therefore be involved in gene findings was not appreciated, and it was not until the late 1980s that serious attention was
transfer between viruses and cells, paid to this field. The relatively slow development of this area of research may be linked
or vice versa, as well as leading to to the long-held fallacy that microbial populations in the oceans are insignificant and
potential overestimates of virus that, by association, viruses must also be unimportant. As discussed in Chapter 4, early
abundance. marine microbiologists seriously underestimated the abundance and diversity of bacteria
because of reliance on inappropriate culture methods. The classical method of enumerat-
ing infective phages and viruses infecting protists is based on the formation of plaques
of lysis in lawns of susceptible hosts grown on agar plates (Figure 7.3) or by dilution in
liquid media and enumeration by the most probable number (MPN) method. In liquid
culture, reduction in the turbidity of cultures can be measured by spectrophotometry
as an indicator of cell lysis. These methods can be used with phages that infect easily
cultivated heterotrophic and phototrophic bacteria but are not yet possible for the many
bacteria and archaea which cannot yet be cultured. (Plaque assay and MPN methods can
also be used to enumerate viruses of protists). It was not until reliable methods for direct
observation and molecular analysis were developed in the study of both marine bacteria
and their phages that it was realized that, far from being insignificant players in marine

Figure 7.3 Epifluorescence micro-

graph of a filtered (0.02-µm Anodisc)
water sample stained with SYBR
green for enumeration of virus-like
particles (VLPs). The smallest bright
dots are VLPs, the next largest
are bacterial or archaeal cells. The
two larger cells shown are protists.
Credit: Jed Fuhrman, University of
Southern California.
 Marine Viruses 201

systems, phages play a critical and central role in ocean ecology and food web dynamics,
although it must be remembered that these methods do not directly indicate the number
of infective viruses.

The life cycle of phages shows a number of distinct stages

The first step in the life cycle of phages is adsorption to the host cell surface, which may be
reversible, followed by irreversible binding to a specific receptor on the host cell surface. For
those marine phages that have so far been propagated in cultivated bacteria, most show speci-
ficity for particular bacterial species and sometimes for particular strains. These findings are
similar to those seen in other well-known viruses and are usually thought to be due to the
molecular specificity of virus receptors on the host cell surface, the presence of restriction
enzymes, or compatibility of the replication processes. However, there are indications that
these highly specific interactions may be something of an artifact introduced by the assay
system in vitro, and it is thought that some marine phages may have a relatively broad host
range in the natural environment. For example, some cyanophages have been shown to infect
both Synechococcus and Prochlorococcus and some vibriophages infect several species of
Vibrio. If correct, this has significant implications for the possibility of genetic exchange
between different organisms and for the role of phages in determining bacterial commu-
nity structure. Very little is known about the nature of the receptors for phage adsorption
to marine bacteria. It is possible that broad host range phages target conserved amino acid
sequences of proteins on the cell surface of different bacterial types and that the tail fibers
can recognize more than one type of receptor. Recent work suggests that some phages exploit
membrane proteins of host cells as a mechanism for entry (see p.237).
CYANOPHAGES
In most cases, enzymes in the tail or capsid of the phage attack the bacterial cell wall, forming i CARRY GENES FOR
PHOTOSYNTHESIS
a small pore through which its nucleic acid enters. The phage genetic material then remains
in the cytoplasm or is integrated into the host cell genome. In the lytic cycle (Figure 7.5a), this Sequencing of phages infecting
is followed by expression of phage proteins, phage genome replication, and formation of the Synechococcus and Prochlorococcus,
capsids and other parts of the virion. When assembly is complete, most phages cause lysis of showed that some contain pho-
the host cell by producing enzymes that damage the cytoplasmic membrane and hydrolyze tosynthesis genes of cyanobacte-
the peptidoglycan in the cell wall. rial origin, including psbA (Mann
et al., 2003; Lindell et al., 2004).
In the host, the reaction center
The “burst size” of phage infection is a term used to describe the average number of prog-
proteins are rapidly turned over
eny virus particles released from the host cell at the time of lysis. Knowing this value is due to light-induced damage and
important for modeling the dynamics of viral infection of host populations. The larger the replacement by new proteins—this
burst size, the smaller the number of host cells that need to be lysed to support viral pro- repair process is shut down dur-
duction. Burst size can be measured directly by observing visibly infected cells using TEM ing phage infection. It is “in the
(Figure 7.1B), or it can be calculated using models of virus to host ratios and theoretical rates interest” of the phage for photo-
of contact and infection required to maintain virus production. There is a wide variation in synthesis rates to remain high in
results, depending on the location of the study and the methods used, but values in the range order to provide energy for virion
of 10–50 progeny viruses per cell are typical for in situ studies in natural communities, with production. During infection,
higher values occurring for copiotrophic bacterial hosts in nutrient-rich waters. For those these phage genes are expressed
along with essential capsid genes
phage-host systems that can be propagated in laboratory culture, the burst sizes are usually
(Lindell et al., 2005). Presence
much larger (average about 180) because in vitro grown bacteria are bigger and support
of the genes may enhance the
greater virus densities. In Synechococcus, a large burst size of about 100–300 occurs in reproductive fitness and host range
culture. of the phage (Sullivan et al., 2006).
Analysis shows that PSII genes
have been transferred via HGT into
Lysogeny occurs when the phage genome phages on multiple occasions in
is integrated into the host genome evolution and that considerable
genetic rearrangements have taken
Lytic phages take over the host cell, replicate their nucleic acid, and cause lysis of the host place between host and phage
cell after assembly of the virus particles. However, another outcome is seen when phages, genes. Phages therefore serve as
known as temperate viruses, infect the cell (Figure 7.5b). The phage genome replicates along a reservoir of genetic diversity for
with the host DNA, but it is not expressed. Often, the silent viral genome is stably integrated their hosts. A wide range of such
auxiliary metabolic genes (AMGs)
into the bacterial genome; this latent state is known as a prophage. In other cases, the phage
involved in numerous functions
genome remains in the cytoplasm in a circular or linear form. Bacteria infected with these
have now been identified in marine
phages are known as lysogenic, because under certain conditions the bacteria enter the lytic phages (Breitbart et al., 2018).
cycle and release infective virus particles.
202 Chapter 7

BOX 7.1 RESEARCH FOCUS

Models for the co-evolution of bacteria and their phages

An “arms race” of co-evolution—the “Red Queen” hypothesis. and less virulent phage in the community, due to alternating viral
According to ecological theory, competition between species in the selection for resistant mutants and competitive selection for faster-
same ecological niche should lead to extinction and reduced biodi- growing hosts.
versity. But the opposite occurs, leading Hutchinson (1961) to pose
the “paradox of the plankton,” questioning how so many plankton The “kill the winner” (KtW) hypothesis. Thingstad (2000)
species can co-exist while competing for the same nutrients. We developed a highly influential concept based on a mathematical
know now that this exceptional biodiversity can be partly explained model of traditional host-parasite dynamics incorporating an ide-
by the microscale structuring of the marine environment (p.120), alized food web of bacterial and protist hosts, plus predatory graz-
but the impact of phages as predators also has a major impact on ing protists. The “winner” was conceived as the most active host
promoting the exceptional diversity of microbial ecosystems. Since population—not necessarily the most abundant. In this model, the
phages replicate more rapidly than their hosts, and lytic infection outcome is that virulent phages reduce populations of their sus-
results in death of the host, why don’t phages kill all the hosts and ceptible hosts to a low steady-state level, independent of the host’s
cause their own extinction in the process? A common explanation growth rate and allowing multiple species for each nutrient type
is that there is a constant co-evolution of the host and virus—an (Maslov and Sneppen, 2017). Laboratory experiments involving
“arms race” via mutation and counter-mutation—resulting in con- enrichment or depletion of viruses have provided some support for
tinual selection pressure for the host to evolve resistance and for the the hypothesis, with strong effects on particular taxa of bacteria
virus to evolve to overcome that resistance. This is referred to as (e.g. Hewson and Fuhrman, 2006), but effects on overall bacte-
the “Red Queen Hypothesis,” so named because in Lewis Carroll’s rial community composition are less clear-cut, possibly because
Alice in Wonderland sequel, Through the Looking Glass, the Red of the combined effects of phage-induced lysis and protistan graz-
Queen says, “It takes all the running you can do, to keep in the same ing. Maslov and Sneppen propose a more dynamic interpretation
place.” Despite wide recognition of this model, co-evolutionary of KtW, in which phage infections result in abrupt and severe col-
experiments indicate that this arms race does not continue indefi- lapses of bacterial populations when they become large. In their
nitely, due to genetic and metabolic constraints that lower the fit- model, the total population of all species fluctuates around the
ness of hosts. Resistance in the host and virulence in the phage both carrying capacity of the environment and the overall diversity
carry a fitness cost; for example, changes in surface phage receptors remains high. When viewed over a long period of time, this leads,
might affect host transport mechanisms associated with resistance counter-intuitively, to higher diversity of microbial communities
and vice versa (Avrani et al., 2012). Therefore, this arms race may that are exposed to frequent and severe collapses induced by phage
give way to fluctuating selection. In this model, fast-growing hosts infection. Maslov and Sneppen (2017) conclude that in natural eco-
will dominate the population, but the consequent increased contact systems, this density-dependent selection will be supplemented by
rates with phages result in increased mortality and the host popula- other drivers of diversity and modulated by the effects of spatial
tion will decline. Eventually, resistant mutant hosts will increase, heterogeneity of nutrients.
despite their lower growth rate. Abundance of the dominant phage
will also decrease, allowing the faster-growing susceptible hosts to Game of thrones—what determines the succession of hosts
increase in number again. Avrani et al. explain that, over time, this and phages? In the KtW model, new phages emerge to infect
results in oscillation of resistant and susceptible hosts and virulent newly dominant hosts. Breitbart et al. (2018) propose an analogy

Figure 7.4 Proposed models for explaining the succession of marine phages. Reprinted from Breitbart et al. (2018) with
permission from Springer Nature.
 Marine Viruses 203

BOX 7.1 RESEARCH FOCUS

to conceptualize the hierarchy of bacterial and phage interactions. stay in power, and any changes in the throne will favor bacteria
They imagine a “royal family” of dominant bacteria and phages— belonging to royal lineages that have escaped their abundant phage
this will vary between different locations or “kingdoms”—and a attackers.
population of all the other bacteria and phages that are the “com-
moners.” On the “death” of the royal family—the decline of the Switching to a temperate lifestyle—the “Piggyback-the-
bacterial monarch along with its associated phages—two possibili- winner” hypothesis. The importance of the lytic and temperate
ties exist for succession. In one scenario (Figure 7.4a), all bacterial cycles of phage biology must be factored into these attempts to
hosts and their respective phage hosts have an equal opportunity to understand phage and host selection and evolution. Knowles et al.
become royalty and to fill the empty niche. Each of the commoners (2016) conducted a meta-analysis of reports of virus to bacteria
is relatively rare, so it is unlikely that the same host and phage sig- ratios across diverse habitats, showing a consistent trend for lower
nature sequences would be abundant at different times. An alter- densities of VLPs at high host densities. Experimental analysis
native scenario (Figure 7.4b) assumes that members of the royal showed that viral densities were more consistent with temperate
family are dominant because they are optimized to that specific than lytic cycles at increasing host abundances. Knowles et al.
niche (“blue blood?”). In this case, the bacterial successor to the included metagenomic analysis focusing on 24 coral reef viromes,
monarch is much more likely to be a descendant of the royal fam- showing a correlation between the abundance of host bacteria
ily (“next in line to the throne?”) than a commoner that has scaled and abundance of genes that are hallmarks of temperate phages.
the rank-abundance plot. Therefore, the next phages are likely to Based on this finding of “more microbes, fewer viruses,” they pro-
be variants of previous royal family phages that have overcome posed a “Piggyback-the-Winner” (PtW) model, in which phages
the host’s newly developed resistance. Breitbart et al. note that this integrate into their hosts’ genomes as prophages when microbial
evolutionary arms race is supported by analysis of microdiversity abundances and growth rates are high. Switching to the temper-
in marine bacteria and their phages in culture-based experiments ate life cycle reduces the control of phage predation on bacterial
and in time-series datasets from natural marine systems. While abundance and confers protection on the host cell from infection
further time-series studies are needed to test their hypothesis, by a closely related phage (superinfection exclusion). This is in
Breitbart et al. conclude that initial evidence supports their royal contrast with the conclusions of many previous studies on lysis
family model, whereby hosts that are well-suited to the niche will and lysogeny.

The mechanisms that determine whether the phage enters the lytic or lysogenic cycle
has been well studied in the lambda phage of Escherichia coli and a few other examples,
where there appears to be a quorum-sensing like process (p.102) mediated by the phage
and a peptide signal molecule. Similarly, there have been many studies of the process by
which expression of the prophage genes controlling the lytic pathway switch are repressed
until induced by the agents or treatments noted below. However, the molecular events of
lysogeny in marine bacteria are largely unexplored and complete systems—temperate
phage, lysogenic host, and uninfected host—have only been described for a few marine
bacteria. Factors such as host growth rates, nutrient depletion, and phage density are
thought to be major factors in the “decision” of some phages whether to enter the lytic or
lysogenic cycle.

Little is known about the mechanism in marine phages by which the expression of the
prophage genes controlling the lytic pathway switch is repressed. It is thought that about
1 in 105 lysogenic cells will revert to the lytic cycle without an obvious induction trigger,
so that there is a constant low-level production of temperate phages. Experimental condi-
tions such as exposure to ultraviolet light, temperature shifts, peroxides, antibiotics, or
pollutants have all been shown to induce a large proportion of prophage-containing cells
to enter the lytic cycle. These external stressors generally trigger the cell’s SOS response,
which is initiated by bacteria to prevent DNA damage. In laboratory experiments, the
antibiotic mitomycin C is the most commonly used agent for inducing the lytic cycle.
One method of assessing the proportion of the host population in the lytic and lysogenic
states is to measure the abundance of genes that are characteristic of a temperate phage,
such as the integrase and excisionase enzymes used at the start and end of the prophage
integration process. These can be measured in metagenomic analyses which compare
DNA libraries from ambient-free virus particles with libraries prepared after prophage
induction.

Integration and excision of the phage genome has very important evolutionary conse-
quences because it provides a natural mechanism, known as specialized transduction, by
204 Chapter 7

which specific host genes can be transferred from one cell to another when part of the host
DNA becomes incorporated into the mature virus particle. The presence of a prophage also
confers resistance to infection of the host bacterium by viruses of the same type (phage
immunity) because repressor proteins, which prevent replication of the prophage genome,
also prevent replication of incoming genomes of the same (or closely similar) virus.
Prophage genes can also affect the phenotypic characteristics of the host cell in other ways,
including effects on the general reproductive fitness of the host or alteration of the host phe-
notype through expression of different genes. An important example of phenotypic modi-
fication or lysogenic conversion is the role of phages in the acquisition of virulence factors
enabling colonization and toxicity in Vibrio cholerae, a bacterium found in coastal and
estuarine waters and responsible for the human disease cholera (p.333). Phage-mediated
virulence of bacteria may be very widespread in other Vibrio spp. and other pathogenic
bacteria. Lysogenic immunity and phage conversion both carry a strong selective value
for the host cell and could provide phages with a strategy to survive periods of low host
density or low metabolic activity when the probability of encountering a new host and
initiating a lytic cycle factor is low. Since many marine environments contain low levels of
slow-growing bacterioplankton and because free viruses are inactivated quite quickly (see
below), virulent lytic phages could become rapidly depleted. For all these reasons, it would
be reasonable to consider lysogeny to be a very common state of virus infections in nature.
Indeed, some studies with cultivable bacteria have shown that lysogeny is more common
in bacteria (up to 50%) in samples from offshore environments than nutrient-rich coastal
environments, and in studies of deep-water communities the frequency of lysogenic cells
appears to be negatively correlated with bacterial production. However, some other studies
have produced conflicting results and we have no knowledge at all about the importance
of lysogeny in bacteria that have not yet been cultured. Another form of hidden infection
may occur, in which the viral nucleic acids remain in the host cell for an extended period,
but the lysis of the host cell is delayed. Such pseudolysogeny (Figure 7.5c) may be due to
nutrient deprivation and low metabolism of the host cell and so might also be a common
trend in oligotrophic marine waters. It is hoped that the application of new molecular biol-
ogy techniques will reveal the true importance of lysogeny and pseudolysogeny in marine
virus-host interactions.

Figure 7.5 Possible outcomes of

infection of a bacterial cell by a DNA
phage. a) In the lytic cycle, viral genes
are expressed, DNA is replicated, and
host machinery is used to make the
components of the virus particles.
These self-assemble into mature
virions which are released. b) The
DNA of temperate phages may be
incorporated and replicated with the
host genome as a prophage. The host
cell is said to be lysogenic because it
may be triggered to enter the lytic
cycle under certain conditions. c) In
pseudolysogeny, the phage genome
remains unintegrated for an extended
period, usually due to nutrient deple-
tion. Reprinted from Feiner et al.
(2015) with permission from Springer
Nature.
 Marine Viruses 205

Loss of viral infectivity arises from damage

to the nucleic acid or capsid
Usually, a virus will lose its infectivity before showing obvious signs of degradation.
However, since most marine viruses are studied by microscopic or flow cytometric enumera-
tion of VLPs, the term “virus decay” reflects the observation of a decline in numbers of VLPs
over time in the absence of new viral production. Many of the studies of virus inactivation
in water were originally carried out in connection with the health hazards associated with
sewage-associated viruses (such as enteroviruses or coliphages; in waters for swimming or
cultivation of shellfish (p.366). Subsequently, the results of these studies have been applied to
the population dynamics of autochthonous marine viruses. A wide range of physical, chemi-
cal, and biological factors can influence virus infectivity and decay. Different studies have
produced various estimates of decay rates, but a value of about 1% per hour is typical in
natural seawater kept in the dark. Visible light and ultraviolet (UV) irradiation are by far the
most important factors influencing virus survival, and in full-strength sunlight, the decay
rate may increase to 3–10% per hour, and in some circumstances can be as high as 80%. UV
light has its greatest effect in the upper part of the water column but is probably still effective
down to about 200 m in clear ocean water. Even in very turbid coastal waters, viral inactiva-
tion by light can be observed down to several meters. Such high rates of inactivation would
lead to the conclusion that there are no or very few infective viruses in the top layer of water.
However, extensive repair of UV-induced damage can occur by mechanisms encoded either
by hosts or their viruses. Another important factor in decay is the presence of particulate mat-
ter and enzymes such as proteases and nucleases produced by bacteria and other members of
the plankton. These involve complex interactions, because adsorption of viruses to marine
snow particles or TEPs can also afford some protection from decay. Virus inactivation does
not proceed at a constant rate and it appears that there is variation in the resistance of viral
particles to damaging effects, presumably because of minor imperfections in the capsid.
Thus, over time, inactivation will lead to a slowly decaying low level of infective particles,
but we do not know the rate of this for ecologically important viruses.

Measurement of virus production rates is important

for quantifying virus-induced mortality
Many studies in viral ecology have attempted to measure the effect of viruses on microbial
mortality and the rate of production of new virions. This enables estimates of the proportion
of primary and secondary production that is “turned over” by viral lysis. It is possible to use
filtration and high-speed centrifugation to obtain a pellet of planktonic microbes, which can
be embedded in resin and sectioned for TEM. By examining such samples from various loca-
tions, it has been found that about 1−4% of microbial cells contain mature, fully assembled
VLPs. Since VLPs can only be seen within infected host cells in the final stages of the lytic
cycle of infection—this stage usually represents about a quarter of the life cycle—it is pos-
sible to estimate the total proportion of plankton that are infected at any one time. Another
method used in early studies of virus production was to measure the incorporation of radio-
actively labeled precursors such as 3H-thymidine, 14C-leucine, or 32P-phosphate into virus
particles, which can be separated from cells and cell debris by filtration. The virus dilution
(reduction) method has been widely used; in this method, free virus particles are removed
from a water sample by filtration and the number of new particles occurring in the water
over time is measured by epifluorescence microscopy or flow cytometry. After adjustments
for the relative abundance of host cells in the sample, the production rate and the original
percentage of infected cells can be calculated. Each of these methods has advantages and
disadvantages and no single method gives precise estimates of virus-mediated mortality.
(Obviously, these methods cannot detect prophage-infected cells). Nevertheless, despite vari-
ability depending on the method, location, and time of sampling, the unequivocal conclusion
from various studies is that viral mortality has a highly significant impact on mortality of
microbes that is at least as significant as grazing by protists and zooplankton. A consensus
value is that about up to 40% of marine bacteria are killed each day by viral action. Between
20–30% of bacteria in the oceans are probably infected by phages at any one time, and an
estimated 1023 viral infections occur every second. In the case of algae, laboratory experi-
ments and mesocosm studies have indicated that viral infection can account for nearly 100%
206 Chapter 7

of mortality of bloom-forming microalgae such as E. huxleyi. Using modified dilution meth-

ods to distinguish the effects of mortality due to viruses from mortality attributed to protistan
grazing, viral mortality of Micromonas pusilla has been estimated at 9–25% standing stock
per day and in this case, it is likely that there is a more stable co-existence of viruses with
their microalgal hosts. As with bacteria, precise values for the effects of viral mortality on
natural populations of microalgae are hard to obtain, but there is no doubt that it has a very
significant impact on primary production, climate processes, and global biogeochemistry.

Viral mortality “lubricates” the biological pump

Viral infection of heterotrophic and autotrophic bacteria, archaea, fungi, microalgae, and
other protists seems to be the most important factor influencing nutrient cycles in the oceans.
This is important because it leads to the release of massive amounts of organic material
and essential elements from these microbes into the dissolved organic pool from where it
is metabolized by heterotrophs. Because the contents of cells lysed by viruses are rich in
nitrogen and phosphorus, this “viral shunt” speeds up the recycling of nutrients, enhances
the rate of microbial respiration, and reduces the amount of organic material available to
higher trophic levels through protistan grazing in the microbial loop (see Figures 8.3 and
8.4). However, viral lysis is now recognized as having more complex effects, because it leads
to the release of high molecular weight polymeric substances from cells. These contribute
to the aggregation of algal flocs and marine snow (Figure 1.9), thus accelerating the export
of carbon-containing compounds to the ocean floor via the biological pump. This has been
termed the “viral shuttle” and seems to be particularly associated with viral lysis of certain
microbial groups in oligotrophic ocean regions. The importance of these processes is consid-
ered further in Chapter 8.

Nucleocytoplasmic large DNA viruses (NCLDVs) are

THE DEEP IMPACT OF
i THE VIRAL SHUNT important pathogens of microalgae and other protists
One of the most studied family of viruses is the Phycodnaviridae, a family of large dsDNA
Biogeochemical cycles in deep-sea
ecosystems are determined by the viruses infecting a wide host range of freshwater and marine algae. Although genetically
activities of bacteria and archaea diverse, they share some morphological similarities, with a dsDNA-protein core usually sur-
that largely depend on organic rounded by a lipid membrane and a capsid (120–220 nm diameter) with icosahedral sym-
matter exported from the ocean metry. Genomes of several species have been obtained and range in size from 100–560 kb.
surface. Microbial life is abundant There are currently six genera recognized in ICTV taxonomy. The best known of these is
in deep-sea sediments and consti- Coccolithovirus, with a single species that infects the important bloom-forming prymne-
tutes a major fraction of the total siophyte coccolithophorid Emiliania huxleyi. It causes cell lysis and contributes to the rapid
microbial carbon on Earth, most
collapse of blooms (Figure 6.9), with major ecological and biogeochemical consequences,
of which appears to be unused by
higher trophic levels. This paradox
as discussed in Box 7.2. Raphidovirus species infect a number of significant bloom-form-
is explained by Danovaro et al. ing raphidophytes, including Chrysochromulina, Aureococcus, Heterosigma, Heterocapsa,
(2008) who studied viral activity and Phaeocystis. Apart from Chrysochromulina, all of these are the cause of harmful algal
at 232 sites at different locations blooms (HABs). Virus-induced collapse of large blooms of Phaeocystis resembles that of the
and depths, showing that over E. huxleyi–Coccolithovirus interaction and has similar important roles in carbon and sulfur
80% of the heterotrophic produc- cycling via DMSP release (Box 9.2). Collapse of blooms can also cause release of gelatinous
tion is shunted by viral lysis into polysaccharides leading to extensive formation of foam that has a nuisance (but not toxic)
labile dissolved organic material effect on beaches and in coastal waters. Of even greater significance is the large-scale shift
(DOM), producing between 0.37 in bacterial populations following bloom collapse, which can lead to anoxia. Members of the
and 0.63 Gt (Pg) of carbon a year.
genus Prasinovirus infect picophytoplankton, including Ostreococcus (the smallest known
Viruses therefore play a major role
in deep-sea processes by injecting
eukaryote, (Figure 1.3) and Micromonas, which is a prominent member of the picophyto-
huge amounts of organic material plankton in oceanic and coastal regions. Viruses specific to Micromonas pusilla cause lysis
into the ecosystem—stimulating of up to 25% of the host’s daily population but do not cause the “bloom and bust” dynamics
microbial metabolism and acceler- seen in the E. huxleyi–Coccolithovirus interaction. There are nine species of Phaeovirus,
ating biogeochemical processes. In which infect brown macroalgae (seaweeds), such as Ectocarpus and Feldmannia. In addition,
a recent study of sediments from there are more than 20 Chlorovirus species that infect the freshwater protists Acanthocystis
the Black Sea, Cai et al. (2019) and Paramecium and the cnidarian animal Hydra viridis.
showed high viral production 37 m
below the seafloor in sediments Evolutionary analysis of the genomes of the phycodnaviruses shows that they belong to a
~6000 years old, with densities up monophyletic group of large DNA viruses infecting eukaryotes—the nucleocytoplasmic large
to 1.8 × 1010 viruses cm−3.
DNA viruses (NCLDVs). Other viruses within this group include poxviruses, iridoviruses,
 Marine Viruses 207

and mimiviruses, including a range of plant and animal pathogens. These viruses replicate
inside the nucleus and/or cytoplasm and their large genomes encode hundreds of genes of
unknown function. Unlike many viruses, NCLDVs can carry genetic information for entire
biosynthetic pathways. Genome analysis of the NCLDVs shows that they have a very ancient
origin and co-evolution with their hosts—divergence of the major families of NCLDVs
appears to have occurred prior to the radiation of the major eukaryotic lineages 2–3 billion
years ago. Figure 7.7 shows examples of NCDLV infections of plankton.

The mechanism of infection of host cells differs amongst the NCLDVs, with the mechanisms
of entry and exit showing large variations. Among the phycodnaviruses, Chlorovirus pos-
sesses a lipid-containing membrane (underlying the capsid), which fuses with the host cell
membrane after enzymic digestion of the algal cell wall, followed by injection of the viral
DNA and associated proteins. Phaeovirus also fuses with the host cell, in this case infecting
only the spores or gametes of its seaweed host, which lack a cell wall. Coccolithovirus infects
E. huxleyi cells using a mechanism that resembles that seen in enveloped animal viruses,
involving entry of the entire nucleocapsid into the cytoplasm following either fusion of the
envelope with the cell membrane or an endocytotic process.
Figure 7.7 Electron microscopy
Remarkable new insights into the origin and evolution of these NCDLV viruses and of NCLDV infections of plankton
their importance in marine ecosystems have followed the description of the giant virus cells. A. SEM of E. huxleyi cells (~5
Mimivirus and the discovery that related viruses are very abundant in marine ecosystems mm diameter) and shed coccoliths,
as pathogens of heterotrophic eukaryotes, as well as the photosynthetic plankton currently showing a virion of Coccolithovirus
described. EhV-86 (~180 nm diameter). B.
Cryotomography image of an
isolated virion; three lipid bilayer-
like membranes surrounding the
capsid (average width = 4.5 nm. C.
Multiple virions of Prasinovirus OtV
attaching to a cell of Ostreococcus
tauri. D. Intact OtV virion (contain-
ing DNA) attached to cell membrane
of O. tauri, 30 min after infection.
E. Mature virion of the Klosneuvirus
BsV. The particle contains at least
six layers, with a DNA-containing
core surrounded by a core wall and
inner membrane, with a putative
membrane under a double-capsid
layer. The top vertex of the virion
contains a possible stargate struc-
ture like that in Mimivirus ApMV. F.
BsV virion assembly and maturation
in cell of Bodo saltans: lipid vesicles
migrate through the virion factory
where capsid proteins attach for the
proteinaceous shell. Vesicles burst
and accumulate at the virus factory
periphery where the capsid assem-
bly completes (black arrow). Once
the capsid is assembled, the virion is
filled with the genome and detaches
from the virus factory. Internal
structures develop inside the virion
in the cell’s periphery where mature
virions accumulate until the host cell
bursts. Credits: A. reprinted from
Michaelson et al. (2010) with permis-
sion of Elsevier; B. reprinted from
Schatz et al. (2014), CC-BY-3.0; C.,
D. reprinted from Derelle et al. 2008,
CC-BY-2.0; E., F. reprinted from Deeg
et al. (2018), CC-BY-4.0.
208 Chapter 7

BOX 7.2 RESEARCH FOCUS

“Bloom and bust”—the life cycle of Emiliania huxleyi and Coccolithovirus (EhV)

An arms race between host and virus. Massive natural blooms high host diversity that helps to buffer the effects of virus infec-
of the coccolithophorid prymnesiophyte alga Emiliania huxleyi tion and environmental change. A further twist to the tale of this
develop annually in many temperate and sub-temperate coastal and cat is supplied by Mordecai et al. (2017), who invoked the analogy
oceanic waters, before collapsing suddenly with the release of mil- of Schrödinger’s Cat (from the famous “simultaneously dead and
lions of coccoliths their calcium carbonate shells (cocoliths) due to alive” thought experiment in quantum theory). Although the hap-
lysis following infection by coccolithoviruses (Figure 6.9). Because loid E. huxleyi cells are thought to escape infection, they contain
of the importance of this process in global ocean biogeochemical viral lipids and viral RNA, but not viral DNA—hence the virus
cycles, this virus-host system has been explored extensively in the cannot progress to capsid formation and lysis in this stage, which
natural setting, in laboratory experiments, and in large-scale meso- they designate a haplococcolithovirocell. To complicate things fur-
cosms (Figure 2.14). Several different strains of E. huxleyi viruses ther, a recent study by Johns et al. (2019) also indicates that the cal-
(EhVs) have been characterized from different locations (Schroeder cification state of diploid E. huxleyi cells indeed plays a protective
et al., 2002; Wilson et al., 2002, Nissimov et al., 2017) and genome role from infection by serving as a physical barrier from virus par-
sequences are available for 13 strains. Extensive studies of commu- ticles. Nevertheless, where infectious viruses go between blooms—
nity dynamics and genetic diversity structure in the open ocean and when host densities are extremely low and distance between cells
mesocosm studies were carried out in different years using PCR- and EhV particles are extremely vast—remains an enigma!
DGGE analysis of genes encoding the calcium-binding protein
(GPA) of E. huxleyi, and the major capsid protein (MCP) and DNA EhVs manipulate host lipid metabolism. One of the most remark-
polymerase genes of EhVs (Martinez et al., 2007, 2012; Schroeder able features of the EhV genome is the presence of a cluster of
et al., 2003). These studies revealed a periodic annual succession of seven genes encoding an almost complete sphingolipid biosyn-
identical E. huxleyi genotypes, which in turn determined the suc- thesis pathway (SBP) (Wilson et al., 2005; Monier et al., 2009).
cession of viral genotypes. By monitoring the MCP and GPA genes Virally encoded glycosphingolipids (vGSLs) are essential for EhV
every 4 h during a bloom, Sorensen et al. (2009) showed that the infection (Vardi et al., 2009; Rosenwasser et al., 2014; Ziv et al.,
viral community was highly dynamic, with fluctuations in detect- 2016) and transcriptional studies show expression of viral SPB
able levels of different viral genotypes occurring at very short genes during infection in laboratory (Allen et al., 2006) and natu-
intervals. Thus, there is a constant “Red Queen”-style arms race ral mesocosm studies (Pagarete et al., 2009). Electron microscopy
in which there is continual selection pressure for the host to evolve studies have revealed that EhV infects cells using a mechanism that
resistance and for the virus to evolve to overcome that resistance, resembles that of enveloped animal viruses, involving entry of the
like that for bacteria–phage interactions discussed in Box 7.1. entire nucleocapsid into the cytoplasm, following either fusion of
the viral envelope with the cell membrane or an endocytotic pro-
Escape from infection. Until the study by Frada et al. (2008), most cess (Mackinder et al., 2009). During viral replication, capsids
studies on E. huxleyi had focused on the diploid cells, which are are transported through and assembled in the host cytoplasm, and
coated in calcified coccoliths. Noncalcified motile haploid cells leave the cell via a budding process. These processes require exten-
also occur but these were overlooked until flow cytometric analysis sive manipulation of lipid metabolism during host–virus interac-
showed the appearance of smaller cells in populations of E. hux- tions (Pagarete et al., 2009, 2011; Malitsky et al., 2016; Ziv et al.,
leyi undergoing viral lysis (Jacquet et al., 2002). Frada et al. (2008) 2016) and indeed, vGSLs comprise >80% of the lipid content of
showed that three strains of cultured diploid E. huxleyi were all sen- the EhV envelope (Fulton et al., 2014)). In addition, purified vGSL
sitive to five EhV strains, whereas haploid stages were not infected. from an EhV virion suppresses host growth in a dose-dependent
Electron microscopy and PCR amplification of the gene encoding manner (Vardi et al., 2009) and EhVs use vGSL enriched mem-
MCP revealed that viruses did not adsorb to haploid cells, perhaps brane lipid domains called “lipid rafts” for entry and exit from the
because a different cell surface structure prevents the enveloped host cell (Rose et al., 2014). In view of the importance of vGSLs in
virus from fusing with the cytoplasmic membrane. Additional the infection process, and previous observations (Nissimov et al.,
experiments showed that the motile, haploid cells only appeared in 2016) that closely-related EhVs can outcompete one another during
the late stages of infected cultures, suggesting that viral infection infection, Nissimov et al. (2019) characterized the dynamics, diver-
triggers meiosis (Frada et al., 2017). The authors called this strategy sity and catalytic production of vGSLs in a range of EhV strains.
of escape from infection the “Cheshire Cat” phenomenon—another Interestingly, the laboratory infections suggest that fast-infecting
allegorical reference to Alice in Wonderland, in which the fictional virulent EhV strains with higher rates of vGSL production and uti-
cat disappears and appears without its head. The authors discuss lization of intracellular substrates for sphingolipid production have
the implications of releasing host evolution from pathogen pressure, a competitive advantage at high host densities in laboratory experi-
acting to counter the co-evolutionary arms race. Perhaps this is why ments. However, slow-infecting EhVs appear to co-exist and some-
E. huxleyi populations can survive bloom and bust dynamics—the times dominate their faster counterparts in natural North Atlantic
escape strategy creates a reservoir of resistant haploid cells that populations. Nissimov et al. (2019) conclude that although the
may persist in the environment for some time, eventually mating biochemical diversity of glycosphingolipid biosynthesis is a major
to produce new diploid E. huxleyi genotypes. Thus, selection for driver of the competitive ecology of EhV strains, additional factors
sexual reproduction as an antiviral mechanism helps to maintain in the natural environment likely influence the relative abundance
 Marine Viruses 209

BOX 7.2 RESEARCH FOCUS

of fast- and slow-infecting EhV genotypes, which can influence occurs by surveillance of the stress molecules nitric oxide and reac-
affect their co-existence. Coupled with mathematical ecosystem tive oxygen species and cell signaling pathways. The importance
modeling, it was concluded that some of these factors likely include of PCD in infection of E. huxleyi by Coccolithovirus is illustrated
differences in removal rates of fast-infecting viruses compared to in Figure 7.6. The virus hijacks the process of PCD in the host cell
slow-infecting ones into the deep ocean, and access to a larger array by triggering the upregulation of host metacaspase enzymes, which
of susceptible hosts available to slower-infecting EhV genotypes. are highly specific proteases that play a central role as “execution-
ers” in PCD (Bidle, 2016). In particular, the process of autophagy is
Assisted suicide? The sudden crash—from densities up to ~105 controlled by the vGSLs and is essential for virus assembly and exit
cells mL−1—that is so characteristic of E. huxleyi blooms is usu- from the cells. Schatz et al. (2014) used a variety of microscopic
ally attributed to viral infection. Mature EhV virions are released and biochemical methods to show that EhV initiate upregulation
by budding, so why do infected cells undergo such rapid lysis? In of autophagy-related genes, which are essential for propagation
fact, Mimivirus unicellular phytoplankton undergo senescence and of the virus and generation of a high burst size. Specific inhibi-
cell death as a natural process of their life cycle due to cell aging tors of autophagy did not affect EhV DNA replication but led to a
and adverse conditions such as nutrient depletion, high light, oxida- large drop in the assembly and production of extracellular virions.
tive stress, or excessive salinity. A “cellular suicide” phenomenon Nissimov and Bidle (2017) highlight the great significance of PCD
known as programmed cell death (PCD) occurs—this is a funda- in ocean biogeochemical cycles as it relates to stress, because it
mental developmental process with ancient origins in all forms can determine whether the products of photosynthesis by phyto-
of life (Bidle, 2016). Inhibitors of these enzymes were shown to plankton are transferred predominantly to the microbial loop for
prevent production of EhV virions (Bidle et al., 2007). The major regeneration in surface waters, or exported to deep waters via the
processes are apoptosis (morphological changes including cell biological pump (see Chapter 8). Further complexity is added by
shrinkage, DNA fragmentation, and cell membrane blebbing) and bacteria associated with the E. huxleyi phycosphere, whose patho-
autophagy (in which phagosomes interact with lysosomes to engulf genic effects are mediated by DMSP, as discussed in Box 9.2.
and digest cellular constituents). A complex system of regulation
“Cross-kingdom thievery and metabolic thuggery”.
It has been known for some time that SBP genes were
transferred from the host to EhV by horizontal gene
transfer (HGT) (Wilson et al., 2005; Monier et al., 2009),
but the comparative analysis of the 13 EhV genomes by
Nissimov et al. (2017) revealed evidence of even more
extensive HGT. They showed that 84% of the EhV cod-
ing sequences (CDSs) have close similarities to those in
the Eukarya. In addition, results of Nissimov et al. (2017)
indicate that many EhV genes may have had their ori-
gin in a wide range of other protists, some phylogeneti-
cally very distant from the Isochrysidiales (the order to
which E. huxleyi belongs). Perhaps EhVs (or their ances-
tors) have (or had, in their evolutionary past) alternative
hosts. Furthermore, 16% of EhV genes were found to be
very similar to those in the Bacteria. Indeed, many bac-
teria exist in close association with E. huxleyi cells and
their phycosphere, and colonization of the exopolymers
exuded by the algal cell during virus infection may facili-
tate evolutionary relationships such as HGT between the
algal cell, its virus, and the co-habiting, co-occurring
bacterial community. Nissimov et al. (2017) conclude by
highlighting the central role of viruses in genetic trans-
fer across the three domains of life, but acknowledge the
challenge of unwinding these relationships to provide
Figure 7.6 Importance of PCD in interactions between an E. huxleyi further insight into the ecology and evolution of viruses
calcified diploid cell and Coccolithovirus. Credit: reprinted from Bidle and their hosts.
(2016) with permission from Elsevier.
210 Chapter 7

Other giant viruses are abundant

pathogens of heterotrophic protists
The designation “giant virus” was first applied following the discovery of a giant, dsDNA
virus with a 650 nm icosahedral capsid and a genome of 1.2 Mb, which was isolated from
Acanthamoeba polyphaga (La Scola et al., 2003). It was originally thought to be a bacte-
rium due to its size and was therefore named Mimivirus (derived from mimicking microbe).
Several related giant viruses were isolated soon afterward, including Mamavirus from
another Acanthameoba species and Megavirus from marine sources. These viruses have
sufficient information to allow them to carry out most, but not all, parts of their life cycle.
They possess some genes for energy production but are dependent on the host ribosomes for
translation of mRNA into proteins (Suzan-Monti et al., 2006). The life cycle thus resembles
that of Rickettsia but is clearly viral in nature as it does not replicate through cell division,
but by assembly of preformed units. Most DNA viruses of eukaryotes insert their DNA into
the nucleus of host cells, where it is replicated. However, after infection of Acanthamoeba,
Mimivirus enters an eclipse period, during which host cells are instructed to build a large
organelle-like “virion factory” in which metabolic processes leading to the synthesis of new
virus particles are coordinated. The function of these intracytoplasmic virion factories has
been compared with that of a cell nucleus, because they both contain the apparatus for DNA
replication and transcription but lack the ability for energy production and mRNA translation.
This discovery has led to revival of the hypothesis that evolution of the eukaryotic nucleus
may have occurred by infection of an ancestral bacterial cell by a virus. Other evidence sup-
ports the concept that DNA viruses are the origin of replication proteins in eukaryotic cells.

After the discovery of Mimivirus, reexamination of the Sargasso Sea and GOS metagenomes
(see p.55) revealed large numbers of sequences homologous to NCLDV genes and found
that members of the family Mimiviridae are the second most abundant group after phages,
suggesting that they commonly infect plankton. Further analysis showed multiple sequences
of specific genes that clustered with four of the six NCLDV families. Because these metage-
nomes were derived from filters designed to capture particles between 0.1 and 0.8 µm, it was
presumed that the sequences must have come either from free giant viruses of similar size to
Mimivirus or from infected picoeukaryotes.

A very large (300 nm capsid diameter) lytic virus infects the abundant protist roenbergensis.
This is a motile phagotrophic flagellate with a major role as a grazer of bacteria, viruses, and
other picoplankton (p.171). The genome of the C. roenbergensis virus (CroV) was obtained
by pyrosequencing and de novo assembly in 2010. The 0.73 Mb genome contains 544 puta-
tive protein-coding sequences (CDS), most of which are of unknown function. The diverse
array of genes includes a large number associated with DNA replication, translation (includ-
ing tRNAs), and transcriptional control mechanisms, indicating that replication and propa-
gation of CroV is relatively autonomous in comparison with many other viruses. Like other
mimiviruses, elements of the genome are thought to have evolved by HGT from eukaryotic
hosts and bacteria.

BsV is a giant virus infecting the heterotrophic kinetoplastid flagellate Bodo saltans virus.
This has a similar role to C. roenbergensis, and the two genera were initially confused.
Based on the recruitment of sequences from metagenomes, BsV appears to be one of the
most abundant members—and the first to be isolated and cultured—of a subfamily of the
Mimiviridae (proposed name Klosneuvirinae), which are the most abundant group of giant
viruses in ocean metagenomes. It has an icosahedral particle size ~300 nm and has a genome
size of 1.39 Mb with 1227 CDS and complex replication machinery (Figure 7.7). Almost all
components of translation have been lost, including tRNAs, but it does carry tRNA repair
genes making it likely that it depends on the host’s tRNAs during infection. The genome
reveals an array of toxins and DNA cutting enzymes, which likely prevent replication of
other competitor viruses. It has also acquired a membrane fusion system from its host via
HGT, which is thought to enable release of the viral genome into the cytoplasm. Recently,
strains of a new genus called Tupanvirus (a sister clade to the amoebal mimiviruses) have
been isolated from soda lake and deep ocean sediments. Besides the exceptional size of their
virions (Figure 7.8), they are remarkable because the genome encodes an almost complete
apparatus for the translation of all 20 standard amino acids in proteins. Astonishingly, there
 Marine Viruses 211

Figure 7.8 Electron microscopy of

Tupanvirus from soda lake. A. SEM
of virion. B. TEM highlights the
inner elements of virion. C. Stargate
vertex transversally cut. D. Capsid
transversally cut. E. Tail transversally
cut. F., G. Late stages of infection of
Acanthamoeba castellanii showing
mature viral factories (VF). Arrows
highlight tail formation associated
with the VF. Reprinted from Abrahão
et al. (2018), CC-BY-4.0.

CAN A VIRUS
? HAVE A VIRUS?
The Sputnik virus, a dsDNA virus
with a small genome (~18 kb),
was described as a parasite of the
Acanthamoeba polyphaga mimivirus
are ORFs for up to 70 tRNA, 20 ORFs related to aminoacylation and transport, 11 factors for (Mamavirus) (La Scola et al., 2008).
all translation steps, and factors related to tRNA/mRNA maturation and ribosome protein It was therefore called a virophage
(“virus eater|”). Since then, more
modification. Only the genes encoding the ribosome itself are missing. Tupanvirus appears
than a dozen other virophages
to have a broad host range and infects amoebas and other protists in experimental laboratory
have been identified, including
systems, although the natural hosts of Tupanvirus and its abundance and distribution in the two marine examples: Mavirus
ocean environment are currently unknown. in association with the Cafeteria
mimivirus CroV (Fischer and Suttle,
2011) and PgVV in association with
RNA viruses also infect protists the Phaeocystis phycodnavirus PgV
(Santini et al., 2013). About half of
It is generally assumed that DNA viruses are the most dominant viruses in the ocean, but
the ORFs (16 in CroV, 20 in PgVV)
this assumption may be influenced by methodological constraints because RNA viruses with
are “ORFans,” with no known func-
small genomes are more difficult to observe and enumerate using electron microscopy or tion. When first described, it was
flow cytometry, and metagenomic methods for their detection have not been fully developed. wrongly assumed that Sputnik was
Nevertheless, RNA viruses appear to be very diverse, and numerous ecologically impor- replicating inside the Mamavirus.
tant examples occur in the major virus groups, as shown in Table 7.1. PCR-based screening Based on a comparative pro-
of seawater samples using conserved sequences of the RNA-dependent RNA polymerase teomics approach, Sobhy (2018)
gene has demonstrated that distinct groups of picorna-like viruses are abundant, widespread, concludes that virophages could
and persistent. Metagenomic analysis based on reverse-transcribed shotgun sequencing has enter cells independently and exist
shown that most sequences are unrelated to known RNA viruses. Some RNA viruses are as latent viruses within host cells
well-known because of their economic importance as pathogens in marine fish and shellfish and only initiate replication when
giant viruses infect the host cell.
aquaculture (Chapter 11). However, until recently, there has been little information about the
The virophage shares resources
role of RNA viruses in marine ecology as pathogens of planktonic microbes. Only a small with the giant virus, but because
number of RNA phages infecting bacteria are known, but several RNA viruses infecting both viruses depend on the host
protists have now been described. A ssRNA virus called HaRNAV in the family Reoviridae cell transcription machinery, this
infects the raphidophyte Heterosigma akashiwo. HaRNAV has been investigated extensively cannot be construed as “a virus
in view of the importance of this virus in controlling blooms of this organism, which is infecting a virus.” Nevertheless,
responsible for large fish kills. Electron microscopy of infected cells shows that they may they are parasites—co-infection
contain as many as 2 × 104 viral particles. Genome sequencing led to the definition of a new results in reduced fitness of the
virus family called the Marnaviridae. These viruses seem to be widespread and may play an giant virus, often accompanied by
important role in population dynamics of other marine protists. A dsRNA virus has also been an increased survival rate of the
infected host organism.
isolated from the prasinophyte Micromonas pusilla.
212 Chapter 7

Diatoms are of major importance in ocean food chains, being one of the main components
of the phytoplankton responsible for primary productivity, especially in coastal waters and
at high latitudes. One might imagine that the silica frustule of diatoms would protect them
from viral infection, but some viruses have been identified as lytic agents of bloom-forming
diatoms. Small viruses presumably enter via pores in the frustule. Novel ssRNA viruses have
been isolated from Rhizosolenia setigera in temperate coastal waters. Another ssRNA virus
has also been described in Chaetoceros spp., where infection leads to a burst size of more
than 104 virions per cell. As infection of diatoms is likely to have a major impact on primary
production and ocean processes, further study of the role of viruses in population dynamics
is necessary. For example, we do not know the extent to which viral infection is involved in
the collapse of diatom blooms, which is a major contributor to the formation of marine snow
and silicon cycling. Nevertheless, there is evidence to suggest that, as in E. huxleyi–EhV
infections, diatom viruses also contribute to the aggregation of material following infection
and potentially influence the vertical flux of carbon into deep waters.

Dinoflagellates form another major group of the phytoplankton of special importance in

coastal waters, with many forming harmful blooms. Two distinct types of ssRNA virus
(HcRNAVs) (as well as the dsDNA Phycodnaviridae virus HcV mentioned above) have now
been described from Heteroplasma circularisquama. There appears to be a clear correlation
between density of HcRNAV and host population dynamics. Again, the burst size of infected
cells seems to be very large. VLPs have been observed at high densities in heterotrophic and
mixotrophic protists in Arctic sea-ice brines. Densely packed small icosahedral VLPs have
been observed in myxosporean parasites of fish, and viruses have been described as putative
pathogens of sarcodines, rhizopods, and radiolarians.

A novel icosahedral ssRNA virus (possibly a picornavirus) that affects the thraustochytrid
protist Schizochytrium sp. has also been described. Since thraustochytrids have special
importance in marine ecology because of their high content of omega-3 polyunsaturated
fatty acids, viral infection of this group could be very significant in ocean food webs and
biotechnological applications.

Viral mortality plays a major role in structuring

diversity of microbial communities
As well as influencing microbial diversity by HGT and genetic rearrangements between host
and virus, viruses influence host diversity at ecological scales. The discovery in the 1990s
that phages are so numerous and responsible for high levels of bacterial mortality led to the
development of models to explain how this might be influencing community structure. These
approaches subsequently extended also to interactions between viruses and algae. The prob-
ability of an encounter between a host cell and a virus is more likely at high host densities
because increased contact rates between a virus and its host are determined by their relative
abundances. Also, some viruses may be very specific, and small changes in host cell surface
receptors or replication processes could make them resistant to infection by a particular virus
genotype (but note the earlier discussion suggesting that some phages probably have a broad
host range). The density dependence and host range factors mean that viruses should prefer-
entially infect the most common hosts: abundant hosts are more susceptible and rare hosts are
less susceptible. Therefore, viruses may control excessive proliferation of hosts that have an
advantage in nutrient acquisition and growth, encouraging a high diversity in the host popula-
tion so that less competitive, but virus-resistant, microbes survive. This has been developed
into a mathematical model termed the “kill the winner” hypothesis, discussed in Box 7.1.
 Marine Viruses 213

BOX 7.3 RESEARCH FOCUS

New ideas about the nature and evolution of viruses

Perhaps the most wondrous discovery in virology in the past decade (2013) emphasizes the importance of the virocell as “cradles of new
has been that of the giant viruses. With each new giant virus dis- genetic information” and the idea that cells are giant “pickpockets
covered—many of them in the marine environment—new sur- of viral genes” during replication or recombination—a phenomenon
prises about their structure, replication, and interaction with their of immense importance that is overlooked by many evolutionary
hosts has emerged. During the same period, the widespread use of biologists. Furthermore, as Forterre emphasizes, we can only truly
high-throughput sequencing and metagenomics has confirmed the understand the ecology and evolution of viruses by studying viro-
enormous diversity of viruses, and deciphering their genomes has cells to elucidate the entry, intracellular, and exit stages of the viral
given rise to new ideas about the nature and evolution of viruses life cycle—complementing the vast bodies of metagenomic infor-
and how this has influenced the evolution of all other forms of life. mation. This adds further weight to the importance of overcoming
the “unculturability” of most marine microbes discussed in Box 2.1.
Are viruses alive? This question provokes great differences of Rosenwasser et al. (2016) also highlight the realization that “virocell
opinion whenever it is discussed—ask a group of virologists and metabolism is a unique metabolic state, determined by modulation
they will probably split into three roughly equal groups answering of host-derived metabolic genes and the introduction of virus-
“alive,” “not alive,” or “something in between.” There is no clear- encoded auxiliary metabolic genes (vAMGs)” (see Box 7.2) and that
cut answer, simply because there is no completely satisfactory defi- study of virocells will “expand the current virion-centric approaches
nition of what we mean by “life.” In the early days of virology at the to quantify the impact of marine viruses on microbial food webs.”
end of the nineteenth and early twentieth century, viruses were con-
sidered to be infectious agents—originally thought to be liquid, but The origin of viruses revisited. One view for the origin of viruses
subsequently shown to be small particles—that behaved like bacte- is that they developed from degeneration of cells that were intra-
ria, and they were perceived as simple living entities. However, in cellular parasites of other cells; another is that they have evolved
the 1930s, the view of viruses changed to one in which they were within cells via the escape of nucleic acids that have become par-
regarded as purely chemical structures composed of proteins and tially independent of their cellular origin as “selfish” genetic enti-
nucleic acids. By focusing our attention on the virus particles (viri- ties. Koonin et al. (2006) reviewed the evidence for an alternative
ons), the dogma grew that they cannot be “alive” in the usual sense scenario—that viruses developed before the evolution of cellular
of the word. However, when a virus infects a cell, it is clear that it organisms. Interestingly, this idea had been first put forward in the
is part of a living system, even though it is not self-sustained and 1920s by Felix d’Herelle following his discovery of phages, and
depends on the host cell to enable it to replicate. by J.B.S. Haldane in his 1928 treatise The Origin of Life. Koonin
et al. (2006) showed how analysis of genes with key roles in viral
We can regard life as an emergent, complex state—a collection of replication and capsid formation can be used to identify viral hall-
non-living molecules which need to reach a critical level of com- mark genes that are shared by apparently unrelated viruses, but
plexity and interactions before it achieves the status of “living.” never found in cellular organisms (except as proviruses integrated
Villareal (2004) gives the analogy of human consciousness— into the genome). They use this to construct a concept of an ancient
for which a whole fully functional brain is needed, not just the “virus world” with extensive gene mixing that predated the emer-
nerve cells—as another example of an emergent complex system. gence of cellular organisms. Claverie (2006) proposed a model in
Villareal argues that this is more than just a philosophical ques- which the initial step was infection of an RNA-based cell to form
tion, because the tendency to categorize viruses as non-living has a primitive nucleus, which becomes transformed into a DNA-based
had the unintended, serious consequence of leading a generation cell because of the selective advantages of DNA biochemistry. He
of biologists to ignore their key importance in evolution—a major proposed that new pre-eukaryotic viruses were created by rapid
oversight from which we are only just escaping. reassortment of genes from the viral and cellular pools before the
evolution of a stable eukaryotic cell with a fully developed DNA
Viruses, virions, and virocells. An important concept to grasp is nuclear genome. Forterre (2006) suggested that the three domains
that a virus can only reveal its biological (living?) nature when it is of life originated from different RNA-based cell precursors contain-
replicating inside a host cell. A viral particle (virion) might be inac- ing DNA viral genomes. The discovery that mimiviruses contain
tive and incapable of infecting a cell—this is one of the problems of the DNA encoding the molecules needed to translate mRNA into
attempting to enumerate “viruses” by fluorescence microscopy or proteins prompted the idea that giant viruses descended from an
flow cytometry. Forterre (2012) introduced the concept of the “viro- ancient free-living cell that no longer exists—a “fourth domain of
cell” to describe a cell (a bacterium, archaeon, or eukaryote) that life” (Colson et al., 2012). These ideas have provoked much debate
has been transformed by viral infection so that its major function and alternative explanations have been proposed (e.g. Moreira and
becomes the propagation of viral genes, rather than cell division. López-García, 2009). Recently, Schulz et al. (2017) discovered a
(In some cases, normal cellular processes can continue alongside new group of giant viruses (Klosneuviruses) whose genomes indi-
viral replication). With this concept, the virocell is a cellular organ- cate an expanded translation system, including aminoacyl tRNA
isms, so it can clearly be regarded as the “living form” of the virus, synthetases for all 20 amino acids. Schulz et al. considered that the
whereas the virions are akin to the seeds or spores of multicellu- evolutionary history of these translation proteins is incompatible
lar organisms—packages of genetic information that will only with an origin in an ancient cellular ancestor or fourth domain of
be expressed under the right environmental conditions. Forterre life and suggest that they have been acquired relatively recently in
214 Chapter 7

BOX 7.3 RESEARCH FOCUS

a piecemeal fashion, but disagreement about interpretation of the their Acanthamoeba hosts, mimiviruses enter an eclipse period,
results has not settled the debate (Reardon, 2017). The first member during which host cells are instructed to build a large organelle-like
of the klosneuviruses to be cultivated was the Bodo saltans virus “virion factory” in which metabolic processes leading to the syn-
(BsV, Deeg et al., 2018) and genome analysis provides further evi- thesis of new virus particles are coordinated (La Scola et al., 2003).
dence that the translational machinery of the klosneuviruses has not The function of these intracytoplasmic virion factories has been
been derived from an ancient common ancestor. The size of the BsV compared with that of a cell nucleus because they both contain the
genome appears to have grown rapidly by the duplication of genes apparatus for DNA replication and transcription but lack the abil-
at the end of the genome—a feature noted in other mimiviruses and ity for energy production and mRNA translation. Bell (2001) had
dubbed the “genomic accordion” (Boyer et al., 2011). However, BsV previously proposed that the eukaryotic nucleus may have evolved
appears to have lost most of the genes for its translational machin- from an infection of an ancestral bacterial cell by a virus, and his
ery, including all tRNAs. This loss might be explained by the fact controversial viral eukaryogenesis hypothesis was revamped in the
that BsV infects the kinetoplastids, which have unusual RNA modi- light of the mimivirus discovery. Bell (2009) suggests that “the
fications. Deeg et al. suggest that strong evolutionary pressure is first eukaryotic cell was a multimember consortium consisting of
placed on the giant viruses by evolutionary virus-host arms races, a viral ancestor of the nucleus, an archaeal ancestor of the eukary-
with competition between related viruses for shared hosts. By con- otic cytoplasm, and a bacterial ancestor of the mitochondria.”
trast, Abrahão et al. (2018) isolated the Tupanviruses, which have Mimiviruses and other NCDLV possess several other features that
an even more complete translation machinery than the klosneuvirus are common to the eukaryotic nucleus (Bell, 2013; Villarreal and
genomes studied by Schulz et al. (2017). The giant viruses continue DeFilippis, 2000) and the recent discovery of the formation of a
to provide surprises with each new discovery. nucleus-like compartment for viral replication during infection of
bacterial cells (Pseudomonas) infected with the very large phage
Did the eukaryotic cell nucleus evolve from a virus? Most DNA 201φ2-1 (Chaikeeratisak et al., 2017) lends further support to the
viruses of eukaryotes insert their DNA into the nucleus of host cells, viral eukaryogenesis hypothesis.
where it is replicated. However, it was shown that after infection of

Marine viruses show enormous genetic diversity

As with other microbial groups, the application of culture-independent methods has led to
major advances in our understanding of the level of diversity of marine viruses. One of the
problems in the study of viral diversity is that there is no universal marker like the ribosomal
RNA genes used for other microbes. However, some success has been made by using repre-
sentative signature genes that are sufficiently conserved to be used as markers of particular
viral groups. For example, extensive virus diversity studies focused on the variation in the
sequences of structural proteins, such as the g20 capsid protein gene in cyanophages or the
major capsid protein (MCP) in phycodnaviruses. Primers for parts of the viral DNA or RNA
polymerase genes are also used in PCR amplification reactions for phycodnaviruses or picor-
naviruses, respectively. Such approaches can be used for identifying and “fingerprinting”
uncultured viruses in environmental samples, for example by PCR-DGGE or TGGE analysis
(p.??). Construction of clone libraries and RFLP analysis and quantitative PCR (p.??) has also
given some useful results for community analysis. These studies show that there is a very
high level of diversity of single genes, even within these restricted viral groups. Attempts to
overcome limitations of assessing diversity based on single genes (as previously discussed
for other microbes in Chapters 4–6) have been made using microarrays (59) or pulsed field
gel electrophoresis (48).

Major advances in the study of viral diversity were also made possible using high-through-
put metagenomic sequencing. The genome sequences of a large number of cultured marine
phages have been obtained, providing insight into metabolic functions and replication strate-
gies. In addition, direct sequencing of environmental samples is helping us to understand
the extent of viral diversity and biogeography in their natural setting. New developments in
single-cell genomics and metagenome assembled genomes (p.56) of viruses is leading to vast
increases in the amount of sequence data, which in turn catalyzes further exploration of viral
diversity. In most cases however, the hosts of newly identified viruses remain unknown. As
noted earlier, there is increasing evidence that many marine viruses have a relatively wide
host range and co-infection by different viruses may be common.
 Marine Viruses 215

Viromes are creators of genetic HOW MANY

diversity and exchange
As noted previously, besides affecting microbial population dynamics, viruses influ-
? TYPES OF MARINE
VIRUSES EXIST?
ence diversity in the marine environment because of genetic exchange. Analysis of viral A coordinated investigation involv-
metagenomes (viromes) has unexpectedly revealed numerous genes involved in metabolic ing metagenomic assemblies of
pathways, indicating that they are a reservoir of genetic information, which is important 3.95 terabytes of sequencing
in the evolution and adaptation of their hosts to different ecological niches. Lysogeny of data resulted in the Global Ocean
bacteria by temperate phages may lead to the introduction and expression of new genes Viromes 2.0 dataset, with sampling
into a host and excision of a prophage can lead to transduction of genes to new hosts. In from >80 sites, from the surface to
4000 m, including new sequences
addition, a process known as generalized transduction can occur, in which the enzymes
collected as part of the Malaspina
responsible for packaging viral DNA into the capsid may mistakenly incorporate host and Tara Oceans studies. Nearly
DNA. These virions are defective and cannot induce a lytic infection, but DNA can be 2 x 105 populations of DNA viruses
passed from one host to another and may recombine with the DNA of the recipient host. were found, with the most diverse
We have known for some time that viruses are vectors for HGT, but we now know that communities in tropical and tem-
this has powerful effects on the evolution of both the microbial hosts and the viruses. perate surface waters, as well as
Genetic information is moved by viruses from organism to organism and throughout the the Arctic Ocean. Because of the
biosphere. As discussed in Box 7.2, genes encoding entire metabolic pathways may move difficulties of defining viral spe-
between viruses and their hosts. cies, Gregory et al. (2019) used a
method of genotypically clustering
sequenced viruses sharing >95%
Unlike defective phages, which are “genuine” virions that are empty or have mistak-
of their DNA. More than 90% of
enly packaged host DNA, another class of virus-like entities called gene transfer agents
the viral “species” designated in
(GTAs) has recently been discovered in Rhodobacter and some other members of the this way couldn’t be identified
Alphaproteobacteria. Some authors also refer to these as generalized transducing agents. as belonging to any currently
GTAs are different to the defective phages described above because they seem to function named virus family. Communities
only to transfer random fragments of DNA between cells. They are smaller than phages appeared to be mapped into five
(about 30–40 nm head diameter), and they carry less DNA (about 4 kb) than would be ecological zones based on tem-
required to encode the protein components of the virion. There are no negative effects perature and depth, although viral
associated with gene transfer to the recipient. It appears that some species of bacteria may macro- and microdiversity did
produce these GTA particles as a dedicated mechanism of gene transfer and it has been not follow the latitudinal diversity
suggested that they are cellular structures akin to flagella or pili. The gene cluster encoding gradient. Like most studies, this
investigation did not evaluate the
the GTA is widespread in the Alphaproteobacteria and may have evolved from a defective
diversity of RNA viruses. Culley
prophage—through loss of replication, regulatory, and lysis genes—long before the diver- (2018) reviews the new sampling,
gence of the major phylogenetic groups of bacteria. methodological, and bioinformat-
ics advances that will be needed to
reveal this hidden diversity.

Conclusions
This chapter has illustrated the rapid pace at which the field of marine virology is mov-
ing. Viruses have emerged to take “center stage” in marine ecology and biogeochemical
processes, through their effects on plankton composition and production. On a more funda-
mental level, recent metagenomic studies and the genomic analysis of isolated viruses have
revealed that they provide an unprecedented reservoir of genetic diversity and play a major
role in the evolution of life. The discovery of extensive gene transfer and the metabolic effects
of viral infection provides important insights into evolution and adaptation to environmental
change. Viruses feature again in the next three chapters, which include further discussion of
their involvement in nutrient cycles, in specific biogeochemical cycles, and as disease agents
of marine organisms other than microbes.

Brum, J.R., Schenck, R.O., & Sullivan, M.B. (2013) Global morphological
References and further reading analysis of marine viruses shows minimal regional variation and domi-
Diversity, ecology, and importance nance of non-tailed viruses. ISME J. 7: 1738–1751.

Avrani, S., Schwartz, D.A., & Lindell, D. (2012) Virus-host swinging Cai, L., Jørgensen, B.B., Suttle, C.A. et al. (2019) Active and diverse
party in the oceans: Incorporating biological complexity into paradigms viruses persist in the deep sub-seafloor sediments over thousands of years.
of antagonistic coexistence. Mob. Genet. Elem. 2: 88–95. ISME J. 13: 1857–1864.

Baran, N., Goldin, S., Maidanik, I., & Lindell, D. (2018) Quantification Chaikeeratisak, V., Nguyen, K., Egan, M.E., et al. (2017) The phage
of diverse virus populations in the environment using the polony method. nucleus and tubulin spindle are conserved among large Pseudomonas
Nat. Microbiol. 3: 62. phages. Cell Rep. 20: 1563–1571.
216 Chapter 7

Culley, A. (2018) New insight into the RNA aquatic virosphere via Breitbart, M. & Rohwer, F. (2005) Here a virus, there a virus, everywhere
viromics. Virus Res. 244: 84–89. the same virus? Trends Microbiol. 13: 278–284.
Danovaro, R., Dell’Anno, A., Corinaldesi, C. et al. (2008) Major viral Krishnamurthy, S.R., Janowski, A.B., Zhao, G. et al. (2016) Hyperexpansion
impact on the functioning of benthic deep-sea ecosystems. Nature 454: of RNA bacteriophage diversity. PLoS Biol. 14: e1002409.
1084–1087. Lindell, D., Jaffe, J.D., Johnson, Z.I. et al. (2005) Photosynthesis genes in
Derelle, E., Ferraz, C., Escande, M.L., et al. (2008) Life-cycle and genome marine viruses yield proteins during host infection. Nature 438: 86–89.
of OtV5, a large DNA virus of the pelagic marine unicellular green alga Lindell, D., Sullivan, M.B., Johnson, Z.I. et al. (2004) Transfer of photo-
Ostreococcus tauri. PLoS One 3: e2250. synthesis genes to and from Prochlorococcus viruses. Proc. Natl. Acad.
Feiner, R., Argov, T., Rabinovich, L., et al. (2015) A new perspective on Sci. USA 101: 11013–11018.
lysogeny: Prophages as active regulatory switches of bacteria. Nat. Rev. Mann, N.H., Cook, A., Millard, A. et al. (2003) Marine ecosystems:
Microbiol. 13: 641. Bacterial photosynthesis genes in a virus. Nature 424: 741.
Gregory, A.C., Zayed, A.A., Conceição-Neto, N. et al. (2019) Marine Ofir, G. &Sorek, R. (2018) Contemporary phage biology: From classic
DNA viral macro- and microdiversity from pole to pole. Cell 177: models to new insights. Cell 172: 1260–1270.
1109–1123.e14.
Paul, J.H. (2008) Prophages in marine bacteria: Dangerous molecular time
Hewson, I. & Fuhrman, J.A. (2006) Viral impacts upon marine bacterio- bombs or the key to survival in the seas? ISME J. 2: 579–589.
plankton assemblage structure. J. Mar. Biol. Assoc. UK 86: 577–589.
Sullivan, M.B., Lindell, D., Lee, J.A. et al. (2006) Prevalence and evolu-
Johns, C.T., Grubb, A.R., Nissimov, J.I., et al. (2019) The mutual interplay tion of core photosystem II genes in marine cyanobacterial viruses and
between calcification and coccolithovirus infection. Environ. Microbiol. their hosts. PLOS Biol. 4: 1344–1357.
21: 1896–1915.
Knowles, B., Silveira, C.B., Bailey, B.A., et al. (2016) Lytic to temperate Coccolithovirus infection of E. huxleyi
switching of viral communities. Nature 531: 466.
Allen, M.J., Forster, T., Schroeder, D.C. et al. (2006) Locus-specific gene
Martínez, J.M., Schroeder, D.C., & Wilson, W.H. (2012) Dynamics and expression pattern suggests a unique propagation strategy for a giant algal
genotypic composition of Emiliania huxleyi and their co-occurring viruses virus. J. Virol. 80: 7699–7705.
during a coccolithophore bloom in the North Sea. FEMS Microbiol. Ecol.
81: 315–323. Bidle, K.D. (2016) Programmed cell death in unicellular phytoplankton.
Curr. Biol. 26: R594–R607.
Maslov, S. & Sneppen, K. (2017) Population cycles and species diversity
in dynamic Kill-the-Winner model of microbial ecosystems. Sci. Rep. 7: Bidle, K.D., Haramaty, L., Barcelos e Ramos, J. & Falkowski, P. (2007)
39642. Viral activation and recruitment of metacaspases in the unicellular cocco-
lithophore, Emiliania huxleyi. Proc. Natl. Acad. Sci. USA 104: 6049–6054.
Michaelson, L.V., Dunn, T.M., & Napier, J.A. (2010) Viral trans-dominant
manipulation of algal sphingolipids. Trends Plant Sci. 15: 651–655. Frada, M., Probert, I., Allen, M.J. et al. (2008) The “Cheshire Cat” escape
strategy of the coccolithophore Emiliania huxleyi in response to viral
Munn, C.B. (2006) Viruses as pathogens of marine organisms—From infection. Proc. Natl Acad. Sci. USA 105: 15944–15949.
bacteria to whales. J. Mar. Biol. Assoc. UK 86: 1–15.
Frada, M.J., Rosenwasser, S., Ben-Dor, S. et al. (2017) Morphological
Nissimov, J.I. & Bidle, K.D. (2017) Stress, death, and the biological glue of switch to a resistant subpopulation in response to viral infection in the
sinking matter. J. Phycol. 53: 241–244. bloom-forming coccolithophore Emiliania huxleyi. PLoS Pathogens 13:
Reche, I., D’Orta, G., Mladenov, N. et al. (2018) Deposition rates of viruses e1006775.
and bacteria above the atmospheric boundary layer. ISME J. 12: 1154. Fulton, J.M., Fredricks, H.F., Bidle, K.D. et al. (2014) Novel molecular
Rohwer, F. & Thurber, R. V (2009) Viruses manipulate the marine envi- determinants of viral susceptibility and resistance in the lipidome of
ronment. Nature 459: 207–212. Emiliania huxleyi. Environ. Microbiol. 16: 1137–1149.
Simmonds, P. (2015) Methods for virus classification and the challenge of Jacquet, S., Heldal, M., Iglesias-Rodriguez, D. et al. (2002) Flow cytomet-
incorporating metagenomic sequence data. J. Gen. Virol. 96: 1193–1206. ric analysis of an Emiliania huxleyi bloom terminated by viral infection.
Aquat. Microb. Ecol. 27: 111–124.
Soler, N., Krupovic, M., Marguet, E. & (2015) Membrane vesicles in natu-
ral environments: A major challenge in viral ecology. ISME J. 9: 793–796. Laber, C.P., Hunter, J.E., Carvalho, F. et al. (2018) Coccolithovirus facili-
tation of carbon export in the North Atlantic. Nat. Microbiol. 3: 537.
Suttle, C.A. (2005) Viruses in the sea. Nature 437: 359–361.
Mackinder, L.C.M., Worthy, C.A., Biggi, G. et al. (2009) A unicellular
Suttle, C.A. (2007) Marine viruses—Major players in the global ecosys-
algal virus, Emiliania huxleyi virus 86, exploits an animal-like infection
tem. Nat. Rev. Microbiol. 5: 801–812.
strategy. J. Gen. Virol. 90: 2306–2316.
Thingstad, T.F. (2000) Elements of a theory for the mechanisms con-
Malitsky, S., Ziv, C., Rosenwasser, S. et al. (2016) Viral infection of the
trolling abundance, diversity, and biogeochemical role of lytic bacterial
marine alga Emiliania huxleyi triggers lipidome remodeling and induces
viruses in aquatic systems. Limnol. Oceanog. 45: 1320–1328.
the production of highly saturated triacylglycerol. New Phytol. 210: 88–96.
Winter, C., Bouvier, T., Weinbauer, M.G. & Thingstad, T.F. (2010) Trade-
Martinez, J.M., Schroeder, D.C., Larsen, A. et al. (2007) Molecular
offs between competition and defense specialists among unicellular plank-
dynamics of Emiliania huxleyi and cooccurring viruses during two sepa-
tonic organisms: The “killing the winner” hypothesis revisited. Microbiol.
rate mesocosm studies. Appl. Environ. Microbiol. 73: 554–562.
Mol. Biol. Rev. 74: 42–57.
Monier, A., Pagarete, A., de Vargas, C. et al. (2009) Horizontal gene trans-
Phages fer of an entire metabolic pathway between a eukaryotic alga and its DNA
virus. Genome Res. 19: 1441–1449.
Biller, S.J., Schubotz, F., Roggensack, S.E. et al. (2014) Bacterial vesicles
in marine ecosystems. Science 343: 183–186. Mordecai, G.J., Verret, F., Highfield, A. et al. (2017) Schrödinger’s
Cheshire Cat: Are haploid Emiliania huxleyi cells resistant to viral infec-
Breitbart, M., Bonnain, C., Malki, K., & Sawaya, N.A. (2018) Phage pup- tion or not? Viruses 9: 51.
pet masters of the marine microbial realm. Nat. Microbiol. 3: 754–766.
 Marine Viruses 217

Nissimov, J.I., Napier, J.A., Allen, M.J., & Kimmance, S.A. (2016) Deeg, C.M., Chow, C.-E.T., & Suttle, C.A. (2018) The kinetoplastid-infect-
Intragenus competition between coccolithoviruses: An insight on how a ing Bodo saltans virus (BsV), a window into the most abundant giant
select few can come to dominate many. Environ. Microbiol. 18: 133–145. viruses in the sea. Elife 7: e33014.
Nissimov, J.I., Pagarete, A., Ma, F. et al. (2017) Coccolithoviruses: A Fischer, M.G., Allen, M.J., Wilson, W.H., & Suttle, C.A. (2010) Giant
review of cross-kingdom genomic thievery and metabolic thuggery. virus with a remarkable complement of genes infects marine zooplankton.
Viruses 9: 52. Proc. Natl. Acad. Sci. U S A 107: 19508–19513.
Nissimov, J.I., Talmy, D., Haramaty, L. et al. (2019) Biochemical diversity Fischer, M.G. & Suttle, C.A. (2011) A virophage at the origin of large
of glycosphingolipid biosynthesis as a driver of Coccolithovirus competi- DNA transposons. Science 332: 231–4.
tive ecology. Environ. Microbiol. 21: 2182–2197. Forterre, P. (2010) Giant viruses: Conflicts in revisiting the virus concept.
Nissimov, J.I., Vandzura, R., Johns, C.T. et al. (2018) Dynamics of trans- Intervirology 53: 362–378.
parent exopolymer particle production and aggregation during viral infec- Koonin, E.V. & Yutin, N. (2010) Origin and evolution of eukaryotic large
tion of the coccolithophore, Emiliania huxleyi. Environ. Microbiol. 20(8): nucleo-cytoplasmic DNA viruses. Intervirology 53: 284–292.
2880–2897.
La Scola, B., Desnues, C., Pagnier, I. et al. (2008) The virophage as a
Pagarete, A., Allen, M.J., Wilson, W.H. et al. (2009) Host-virus shift of the unique parasite of the giant mimivirus. Nature 455: 100–4.
sphingolipid pathway along an Emiliania huxleyi bloom: Survival of the
fattest. Environ. Microbiol. 11: 2840–2848. La Scola, B., Robert, C., Jungang, L. et al. (2003) A giant virus in amoe-
bae. Science 299: 2033.
Pagarete, A., Le Corguillé, G., Tiwari, B. et al. (2011) Unveiling the tran-
scriptional features associated with coccolithovirus infection of natural Santini, S., Jeudy, S., Bartoli, J. et al. (2013) Genome of Phaeocystis glo-
Emiliania huxleyi blooms. FEMS Microbiol. Ecol. 78: 555–564. bosa virus PgV–16T highlights the common ancestry of the largest known
DNA viruses infecting eukaryotes. Proc. Natl. Acad. Sci. USA 110:
Rose, S.L., Fulton, J.M., Brown, C.M. et al. (2014) Isolation and character- 10800–10805.
ization of lipid rafts in E miliania huxleyi : A role for membrane microdo-
mains in host-virus interactions. Environ. Microbiol. 16: 1150–1166. Sobhy, H. (2018) Virophages and their interactions with giant viruses and
host cells. Proteomes 6: 23.
Rosenwasser, S., Mausz, M.A., Schatz, D. et al. (2014) Rewiring host lipid
metabolism by large viruses determines the fate of Emiliania huxleyi, a Suzan-Monti, M., La Scola, B. & Raoult, D. (2006) Genomic and evolu-
bloom-forming alga in the ocean. Plant Cell 26: 2689–2707. tionary aspects of Mimivirus. Virus Res. 117: 145–155.
Schatz, D., Shemi, A., Rosenwasser, S. et al. (2014) Hijacking of an
autophagy-like process is critical for the life cycle of a DNA virus infect-
Viruses and evolution
ing oceanic algal blooms. New Phytol. 204: 854–863. Bell, P.J.L. (2001) Viral eukaryogenesis: Was the ancestor of the nucleus a
Schroeder, D.C., Oke, J., Hall, M. et al. (2003) Virus succession observed complex DNA virus? J. Molec. Evol. 53: 251–256.
during an Emiliania huxleyi bloom. Appl. Environ. Microbiol. 69: Bell, P. (2013) Meiosis: Its origin according to the viral eukaryogenesis
2484–2490. theory. In Meiosis. IntechOpen. doi: 10.5772/56876.
Schroeder, D.C., Oke, J., Malin, G. & Wilson, W.H. (2002) Coccolithovirus Broecker, F. & Moelling, K. (2019) What viruses tell us about evolution
(Phycodnaviridae): Characterisation of a new large dsDNA algal virus and immunity: Beyond Darwin? Ann. N. Y. Acad. Sci. 1447: 53–68.
that infects Emiliania huxleyi. Arch. Virol. 147: 1685–1698.
Chaikeeratisak, V., Nguyen, K., Khanna, K. et al. (2017) Assembly of a
Sorensen, G., Baker, A.C., Hall, M.J. et al. (2009) Novel virus dynamics in nucleus-like structure during viral replication in bacteria. Science 355:
an Emiliania huxleyi bloom. J. Plankton Res. 31: 787–791. 194–197.
Vardi, A., Van Mooy, B.A.S., Fredricks, H.F. et al. (2009) Viral glyco- Claverie, J.M. (2006) Viruses take center stage in cellular evolution.
sphingolipids induce lytic infection and cell death in marine phytoplank- Genome Biol. 7, 110.
ton. Science 326: 861–865.
Colson, P., De Lamballerie, X., Fournous, G., & Raoult, D. (2012)
Wilson, W.H., Allen, M.J., Schroeder, D.C. et al. (2005) Complete genome Reclassification of giant viruses composing a fourth domain of life in the
sequence and lytic phase transcription profile of a Coccolithovirus. new order Megavirales. Intervirology 55: 321–332.
Science 309: 1090–1092.
Forterre, P. (2006) Three RNA cells for ribosomal lineages and three
Wilson, W.H., Tarran, G.A., Schroeder, D. et al. (2002) Isolation of viruses DNA viruses to replicate their genomes: A hypothesis for the origin of
responsible for the demise of an Emiliania huxleyi bloom in the English cellular domain. Proc. Nat. Acad. Sci. USA 103: 3669–3674.
Channel. J. Mar. Biol. Assoc. UK 82: 369–377.
Forterre P. (2012) Virocell concept. In: The Encyclopedia of Life Science.
Yamada, Y., Tomaru, Y., Fukuda, H., & Nagata, T. (2018) Aggregate for- John Wiley & Sons, Ltd.
mation during the viral lysis of a marine diatom. Front. Mar. Sci. 5: 167.
Forterre, P. (2013) The virocell concept and environmental microbiology.
Ziv, C., Malitsky, S., Othman, A. et al. (2016) Viral serine palmitoyl- ISME J. 7: 233–236.
transferase induces metabolic switch in sphingolipid biosynthesis and is
Hutchinson, G.E. (1961) The paradox of the plankton. Am. Nat. 95:
required for infection of a marine alga. Proc. Natl. Acad. Sci. USA 113:
137–145.
E1907–E1916.
Koonin, E.V., Senkevich, T.G., & Dolja, V.V. (2006) The ancient Virus
Giant viruses and virophages World and evolution of cells. Biol. Dir. 1: 29.

Abrahão, J., Silva, L., Silva, L.S. et al. (2018) Tailed giant Tupanvirus pos- Moelling, K. & Broecker, F. (2019) Viruses and evolution—Viruses first?
sesses the most complete translational apparatus of the known virosphere. A personal perspective. Front. Microbiol. 10: 523.
Nat. Commun. 9: 749. Moreira, D. & López-García, P. (2009) Ten reasons to exclude viruses
Boyer, M., Azza, S., Barrassi, L. et al. (2011) Mimivirus shows dramatic from the tree of life. Nat. Rev. Microbiol. 7: 306–311.
genome reduction after intraamoebal culture. Proc. Natl. Acad. Sci. USA Reardon, S. (2017) Giant virus discovery sparks debate over tree of life.
108: 10296–10301. Nature News (online) doi: 10.1038/nature.2017.21798.
218 Chapter 7

Rosenwasser, S., Ziv, C., Van Creveld, S.G., & Vardi, A. (2016) Virocell Villarreal, L.P. (2004) Are viruses alive? Sci. Am. 291: 100–105.
metabolism: Metabolic innovations during host–virus interactions in the Villarreal, L.P. & DeFilippis, V.R. (2000) A hypothesis for DNA viruses
ocean. Trends Microbiol. 24: 821–832. as the origin of eukaryotic replication proteins. J. Virol. 74: 7079–7084.
Schulz, F., Yutin, N., Ivanova, N.N. et al. (2017) Giant viruses with an Villarreal, L.P. & Witzany, G. (2010) Viruses are essential agents within
expanded complement of translation system components. Science 356: the roots and stem of the tree of life. J. Theor. Biol. 262: 698–710.
82–85.
Wessner, D.R. (2010) Discovery of the giant mimivirus. Nat. Ed. 3: 61.
Chapter 8
Microbes in Ocean
Processes—Carbon
Cycling

The aim of this chapter is to bring together areas explored in the preceding chapters on eco-
physiology and diversity of marine microbes, emphasizing their importance in carbon cycling
in the oceans and their global biogeochemical significance. Chapters 3–6 have discussed the
activities of various individual types of marine microbes in major processes including pho-
totrophy, chemolithotrophy, nitrogen fixation, heterotrophic breakdown of organic material,
and oxidation–reduction transformations of major elements. These processes occur in seawa-
ter and sediments, as well as in specialized habitats such as hydrothermal vents, cold seeps,
and epi- or endobiotic associations. In this chapter, and in Chapter 9, the focus is primarily on
the role of planktonic microbes. This exciting area of work, in which the activities of micro-
biologists and chemical and physical oceanographers come together, has led to spectacular
paradigm shifts in our view of the importance of ocean microbes.

Key Concepts
• Microbes play a central role in carbon cycling in the oceans, both as primary producers
and as consumers.
• Fixed carbon from primary production is released as dissolved organic material
(DOM) which is metabolized by heterotrophic bacteria, archaea, and fungi.
• Aggregates of particulate organic matter (POM) are exported to the deep ocean and
sediments via the biological pump.
• Degradation of labile components of DOM results in a large residue of refractory DOM
that has accumulated over millennia in the deep oceans via the microbial carbon pump.
• Grazing of heterotrophs by protists and filter feeding by tunicates leads to the transfer
of fixed carbon to higher trophic levels via the microbial loop.
• Viral lysis of bacteria, archaea, and eukaryotic organisms catalyzes nutrient regenera-
tion in the upper ocean by diverting the flow of carbon from the food chain into a semi-
closed cycle of bacterial uptake and release of nutrients, as well as promoting carbon
flux due to release of aggregative polymers.
• Marine phytoplankton are responsible for about half of the global CO2 fixation, with
cyanobacteria and picoeukaryotes having a dominant role in oligotrophic gyres and with
blooms of eukaryotic phytoplankton being most important in mixed, turbulent waters.
• There is increasing recognition of the importance of chemolithoautotrophic bacteria
and archaea in “dark” CO2 fixation.
220 Chapter 8

BOX 8.1 RESEARCH FOCUS

The role of microbes in the ocean carbon cycle

Milestones in discovery. This Box highlights a small selection of Pomeroy of the University of Georgia, entitled “The ocean’s food
milestone research papers that have led to major paradigm shifts web: a changing paradigm” (Pomeroy, 1974; >1423 citations) is
in microbial ecology and oceanography. Space does not permit a widely credited as being one of the most influential advances in
full historical treatment of the development of modern ideas about our thinking about the role of microbes in the movement of energy
carbon cycling and ocean food webs, but I will outline a few of the and nutrients in marine systems. The main arguments in this paper
most important highlights in the last few decades. These built on were: (1) that the main primary producers in the oceans are small
the pioneering work by early marine microbiologists including the phototrophs <60 μm in size rather than the larger phytoplankton
German scientists Ernst Haeckel, Bernard Fischer, and Wilhelm previously recognized from studies with traditional plankton net
Benecke, who first described marine microbes in the late nineteenth sampling; (2) that microbes are responsible for most of the meta-
and early twentieth century. A succession of scientists in the mid- bolic activity in seawater; and (3) that dissolved and particulate
twentieth century—including Haldane Gee, Claude Zobell, Selman organic matter (DOM, POM) form an important source of nutrients
Waksman, Holger Jannasch, E. J. Ferguson Wood, Cornelius B. consumed by heterotrophic microbes in the marine food web. In
van Niel, and John Sieburth—established research programs in the 1980s, new analytical techniques to measure in situ bacterial
the USA at Scripps Institute for Oceanography, Hopkins Marine activity were developed. Jed Fuhrman and Farooq Azam (Scripps)
Station, and Woods Hole Oceanographic Institution (WHOI), pioneered the measurement of the incorporation of radiolabeled
which laid the foundations for the modern subject. In this abbrevi- amino acids in field experiments, showing that a large proportion
ated account of developments, I name a few individuals whose work of marine bacterial communities are metabolically active, consum-
is especially associated with these conceptual advances, but it is ing up to half of the photosynthetically fixed carbon (Fuhrman and
important to recognize that progress in this field has only been pos- Azam, 1982; >1575 citations). These new methods were also used
sible by the collaborative, interdisciplinary research of numerous by Barry and Evelyn Sherr of the University of Georgia to reveal
microbiologists, oceanographers, biogeochemists, mathematical the importance of the consumption of a large fraction of bacterial
modelers, and other scientists that underpins our knowledge. production by small phagotrophic protists (Sherr et al., 1989; >229
citations), providing evidence of a direct link between bacterial
New methods reveal the importance of microbes in carbon production and higher components of the food web.
cycling. Until the mid-1970s, marine bacteria were regarded as
of little importance other than as decomposers of detritus. The The “microbial loop” concept. Evidence about the role of the
classic view of trophic interactions in the oceans was of a simple heterotrophic bacterioplankton and protistan grazing steadily
food chain, in which primary production by photosynthesis is due accumulated until a series of seminal papers were published in the
mainly to algae like diatoms and dinoflagellates that are large early 1980s by Farooq Azam (Scripps), together with other micro-
enough to be trapped in traditional plankton nets. A simple food bial oceanographers and ecologists from Denmark, South Africa,
chain was envisaged, in which these algae are consumed by cope- Norway, and Germany. The concept of the “microbial loop” was
pods (a diverse group of small crustaceans), which in turn are con- developed to explain the flow and cycling of dissolved organic mat-
sumed by larger zooplankton, eventually reaching fish at the end of ter (DOM) in the oceans (Azam et al., 1983). The paper has had one
the food chain. In this scenario, phytoplankton was assumed to be of the greatest impacts of all papers in the aquatic sciences (>5432
consumed as rapidly as it is produced, with all the primary produc- citations) and led to a true paradigm shift—a sudden and dramatic
tion going through herbivorous zooplankton. Bacteria did not fea- change in our thinking about the role of ocean microbes. Twenty-
ture at all in this food chain. Estimates of bacterial abundance were five years after publication, Tom Fenchel (University of Aarhus,
orders of magnitude lower than we now know to be the case, fitting Denmark) one of the original authors on the paper, reflected on
with the prevalent view at that time that most of the oceans are why the paper was so influential (Fenchel, 2008: 222 citations). He
biological “deserts” with low-nutrient fluxes, low biomass of phy- points out that the general idea had been expressed more or less
toplankton, and low productivity. In 1977, John Hobbie of WHOI explicitly in several earlier papers and attributes the impact of the
published one of the most important methods papers in marine paper to the fact that the simple descriptive name “microbial loop”
microbiology (Hobbie et al., 1977; >5085 citations) describing the was used for the first time—suddenly clarifying and connecting the
use of controlled pore-size filters and epifluorescence microscopy various studies and catalyzing further research.
for the visualization of marine bacteria. This also led to the dis-
covery of Synechococcus (Waterbury et al., 1979; >933 citations), The original microbial loop concept has been continuously devel-
a previously overlooked cyanobacterium now recognized as one of oped as a result of new discoveries made possible by technical
the major contributors to primary productivity in a large part of advances, involving numerous teams of microbiologists and ocean-
the oceans. We quickly realized that the oceans are, in fact, teem- ographers. For example, the development of flow cytometers that
ing with microbes—typically 106 –107 mL-1—in the picoplankton could be used on research vessels led to the discovery of photo-
size class (<2–3 μm). Until the early 1980s, the perceived small synthetic Prochlorococcus by Penny Chisholm of MIT (Chisholm
populations of bacteria were thought to be largely dormant and this et al., 1988; <808 citations). We now know that this is one of the
perspective only changed as a result of the application of new tech- major primary producers in the oceans, and possibly the most abun-
niques, as discussed in Chapter 2. A paper in 1974 by Lawrence dant photosynthetic organism on Earth, yet it escaped detection
 Microbes in Ocean Processes—Carbon Cycling 221

BOX 8.1 RESEARCH FOCUS

until 1988. Since the late 1980s, the large biomass and diversity concerns about the effects on ocean processes of increasing atmo-
of marine microbes has been revealed through the application of spheric CO2 levels. Studies sought to answer questions about the
molecular techniques described in earlier chapters, and the pace of ability of the ocean to absorb increasing CO2 and the effects of
discovery has been dramatic. The previously unimagined impor- associated ocean acidification and global warming. This led to the
tance of archaea, picoeukaryotes, and mixotrophs as components realization that the integrated activity of the microbial loop and
of the plankton has emerged in the last few years, whilst metage- viral shunt has the effect of altering the nutrient composition of
nomic techniques have led to the discovery of new organisms, new DOM as it cycles through these pathways. Heterotrophic bacteria,
energy-generating mechanisms, and the unraveling of metabolic archaea, and fungi break down labile (i.e. readily transformed)
processes in organisms that have not yet been cultured. organic compounds, leading to a residue of increasing amounts
of “inedible” material which is unavailable as a source of energy
The “viral shunt” concept. A highly significant addition to the for further microbial respiration. This material is known as recal-
microbial loop concept occurred in the 1990s, with recognition of citrant DOM (RDOM) and includes complex polymers such as
the role of viruses in controlling community structure in the plank- those from cell walls, or compounds that have been modified by
ton and nutrient cycling. As discussed in the preceding chapter, the photochemical action. It accumulates in the deep ocean where it is
importance of marine viruses was unknown until the late 1970s, sequestered for hundreds or thousands of years. Various multi-dis-
when pioneering studies by Jed Fuhrman’s group (University of ciplinary teams published results from analysis of the origins, dis-
Southern California) first recognized their role in controlling bac- tribution, and analysis of RDOM collected during research cruises
terial populations in the sea (Proctor and Fuhrman, 1990; >986 and this led to the realization that there is a huge, ancient reser-
citations). Growing knowledge of the role of viruses in nutrient voir of organic carbon compounds in the deep ocean. Again, many
cycles led Steven Wilhelm and Curtis Suttle (University of British scientists were involved in this breakthrough, but Nianzhi Jiao of
Columbia) to assign the moniker “viral shunt” to the function of Xiamen University, China is credited with making the conceptual
viral lysis in transferring organic carbon from microbes back into advance explaining how microbes “pump” bioavailable carbon into
the DOM pool (Wilhelm and Suttle, 1999; >894 citations). this reservoir of relatively inert compounds. A seminal paper (Jiao
et al., 2010); >675 citations) published with other leading microbial
The “microbial carbon pump” concept. The most recent major oceanographers and ecologists first described the concept of the
paradigm shift in the ocean carbon cycle grew from extensive microbial carbon pump.
research carried out in the 1990s—largely driven by growing

Physical factors and biotic processes determine

the fate of carbon in the oceans
The unique chemical properties of carbon mean that it is present in many different inorganic
and organic forms, playing a fundamental role in abiotic and biotic ocean processes. The
major reservoir of carbon occurs as carbonate minerals in sedimentary rocks in the Earth’s
crust and in organic compounds such as coal, oil, and natural gas. The turnover of this car-
bon by natural geological processes is very slow, occurring over timescales of millennia.
However, in the last few hundred years, human activities have accelerated the flux of carbon
to the biosphere because of the use of fossil fuels and changes in land use. The atmosphere
currently contains ~858 Gt (Pg) of carbon and the concentration has risen rapidly since pre-
industrial times to ~405 ppm (parts per million), its highest level in the past 800,000 years.
The oceans are estimated to contain ~662 Gt (Pg) of dissolved carbon, about three-quarters
of that in the atmosphere. Understanding the role of marine microbes in the fate of carbon in
the oceans is a critical component of assessing the impact and possible mitigation of climate
change caused by anthropogenic CO2 emissions.

Gaseous CO2 is highly soluble in water and forms carbonic acid (H2CO3-). At the current
natural pH of seawater (7.8–8.2), this mostly dissociates rapidly to form bicarbonate (HCO3-,
p.12). Absorption of CO2 at the interface between ocean and atmosphere and its transfer to
deeper layers is driven to a large extent by physical processes—circulation of water occurs as
a result of tilted density gradients caused by changing temperature and salinity at the surface,
wind forcing and mixing, and the Coriolis effect due to the Earth’s rotation. In the North
Atlantic and Southern Oceans, cold, saline, and dense water with a high CO2 concentration
sinks vertically to depths of 2000–4000 m and then distributes through the ocean basins. The
mass of water sinking to the deep ocean is replaced by upwelling to the surface elsewhere.
As the water warms, CO2 escapes to the atmosphere again. This process for CO2 circulation
in the oceans is called the “solubility pump.”
222 Chapter 8

Figure 8.1 Biological processes in

the sequestration of atmospheric
CO2 and transfer of organic material
to the deep ocean. In the biological
pump, CO2 is fixed by phytoplankton
and sinks to the deep sea as par-
ticulate organic material (POM). In
the microbial carbon pump (MCP),
dissolved organic matter (DOM) is
taken up by bacteria and archaea,
which enter food webs via protist
grazing (microbial loop). Viral lysis of
microbes returns POM to the DOM
pool (viral shunt). Transformation of
DOM by microbial activity results in
the formation of recalcitrant (refrac-
tory) RDOM, which persists for 100s
or 1000s of years in the deep ocean.
Reprinted from Zhang et al. (2018)
by permission of Oxford University
Press; image courtesy of Glynn
Gorick.

In addition to purely physical factors, the “biological carbon pump” is responsible for massive
redistribution of organic carbon in the oceans (Figure 8.1). Fixation of CO2 by phytoplankton
in the upper ocean leads to primary production by incorporation of carbon into cellular mate-
rial. Phytoplankton cells may be consumed by zooplankton or fish, so some of that carbon
enters the food chain directly. Significant amounts of fixed carbon are released as DOC from
phytoplankton cells by exudation and when they die and break up, either by natural senescence
or due to viral lysis. Polymeric substances (TEPs and EPS, see p.12) and cell debris formed in
the photic zone form POM, which aggregate (together with calcium carbonate and silica shells
of protists) to form larger particles of marine snow (also see Figures 1.7–1.8). These settle into
deeper waters by gravity, vertical migrations of zooplankton, and subduction of water masses.
On its passage through the water column, a large fraction of the fixed organic carbon is
respired by microbes and will be returned to the atmosphere over a 10–1000+ year timescale,
varying with depth, as shown in Figure 8.1. About 25% of the primary production reaches the
seafloor. There, most of the carbon compounds are broken down by microbial fermentation
and respiration, fueling sulfate-reducing bacteria and methanogenic archaea. About 0.3% of
primary production will be buried in deep sediments and over many millennia, this residue is
transformed by increasing heat and pressure to form vast reservoirs of hydrocarbons.

The microbial loop, viral shunt, and microbial carbon pump components of the carbon
cycling system illustrated in Figure 8.1 provide both a sink for fixed carbon, as well as a link
to higher trophic levels. About half of the DOM resulting from phytoplankton production
passes through the microbial loop, acting as a source of energy, carbon, and other nutrients
for heterotrophic organisms. This results in greater retention of dissolved nutrients in the
upper ocean but, as mentioned above, microbial use of the carbon compounds alters the
elemental composition and results in the RDOM, which is transported to the deep ocean.
These processes are discussed in more detail below.

Marine phytoplankton are responsible for about

half of the global primary production
Primary production can be defined in terms of the total amount of CO2 fixed into cellular
material during daylight (gross primary production) or as the fraction that remains after
 Microbes in Ocean Processes—Carbon Cycling 223

losses due to respiration by the phytoplankton, which occurs both day and night (net primary
production). Another concept, net community production, has been introduced to encompass
the effects of respiration by heterotrophic organisms. Oceanographers also use various other
ways of measuring productivity (the rate of production) in terms of new, export, and regener-
ated (recycled) production. Both the biomass of phytoplankton and the rate of gross and net
production are affected by a range of factors, which vary greatly over both long-term and
short-term timescales, owing to the interplay of hydrographic conditions (currents, upwell-
ing, and turbulence) with light conditions (latitude, season, cloud cover) and nutrient avail-
ability (especially nitrates, iron, and phosphorus). Temperature increases both photosynthetic
and respiratory rates; but this effect is complicated in the oceans because, although most
upwelling waters have a low temperature, the stimulation of photosynthetic rates by their
increased nutrient levels overcomes the inhibitory effects of cooling.

Quantifying photosynthetic productivity has been a central question in biological ocean-

ography for more than a century. Various methods are available, and these may produce
different results depending on the parameter measured and the timescale employed. The
most common laboratory experiments measure incorporation of radioactively labeled
bicarbonate (H14CO3 –) into organic material, comparing the effects of incubation in light
and dark conditions. Oxygen concentration can be measured by high-precision titration,
and changes indicate O2 evolution by oxygenic photosynthesis and O2 consumption by
respiration. Alternatively, specialized equipment can be used to follow 18O2 evolution from
H218O, indicating photosynthesis without interference from respiratory processes. Clearly,
the discovery of widespread anoxygenic phototrophic bacteria (p.77) means that earlier
assumptions based on the use of O2 evolution need to be reconsidered, as these bacteria are
thought to be responsible for about 5% of primary production. All methods have certain
drawbacks, and it is necessary to make allowances for isotope effects and differences in
the photosynthetic quotient (ratio of HCO3 – assimilated to O2 generated), which varies
according to species and nutrient availability. It is important to recognize that growth of
phytoplankton is not a simple process concerned only with these processes; growth also
depends on assimilation of other nutrients and many metabolic transformations to pro-
duce macromolecules in the cell. Measuring the physiological rates of photosynthesis in
a natural setting is difficult. Most experiments are conducted in laboratory cultures with
model representative organisms, or by enclosing seawater containing natural communities
in bottles, which are then held at various depths in the field. Photosynthetic rates are mea-
sured over time under different environmental conditions, including light. Batch culture
and bottle experiments obviously have inherent limitations because nutrients are quickly
depleted in the absence of diffusion and artifacts can be introduced during deployment and
retrieval of samples, by contamination, or by alteration of the spectral qualities of light
entering the containers. Mesocosm experiments in large natural enclosures containing sev-
eral cubic meters of seawater, or semi-continuous and chemostat culturing in the laboratory
can overcome some of these problems (see p.61).

Since the distribution of primary production is highly dynamic and influenced by many fac-
tors, reliable estimates of productivity throughout the oceans have only been possible since
the advent in the 1970s of remote sensing, using the spectral scanners on satellites, which
measure ocean color. The NASA SeaWiFS satellite-borne sensor, which operated from
1977–2010, and its successors operated by international space agencies, have produced a vast
database of global ocean color and other measurements that can be used to track regional and
temporal variability in the phytoplankton communities, allowing estimates of primary pro-
duction, as shown in the example data in Table 8.1 and Figure 8.2. In clear waters, with little
suspended material, the main factor influencing the ocean color is light absorption by the
phytoplankton pigments, especially chlorophyll. Therefore, water appears bluer when there
are low densities of phytoplankton, and greener when there are high densities. By measuring
the different absorbance properties of chlorophyll a and other photosynthetic pigments, it
is possible to map surface phytoplankton biomass daily (cloud cover permitting). The rela-
tionships between chlorophyll concentration, biomass of phytoplankton, and gross and net
productivity are complex. Furthermore, the depth of the mixed layer, which can vary fivefold,
has to be accounted for when estimating pools and fluxes. Mathematical models and different
algorithms based on in situ and experimental measurements of the various parameters are
then used to compute these values.
224 Chapter 8

The total net annual primary production of the world’s oceans is ~45–50 Gt (Pg) of carbon
per year, which is similar to the total productivity of all terrestrial ecosystems. However, on
an area basis, the annual global marine productivity is about 50 g C m-2, which is only one-
third that of terrestrial productivity. This discrepancy is due to the lower utilization of solar
radiation by ocean phytoplankton than terrestrial plants, largely because of differences in the
availability of nutrients and the effect of suspended particles—including the phytoplankton
cells themselves—in absorbing light.

There are wide geographical and seasonal

variations in primary production
As shown in Table 8.1 and Figure 8.2, the most productive regions are upwelling regions
associated with currents flowing along the west coasts of continents; namely, the Canary
Current (off Northwest Africa), the Benguela Current (off southern Africa), the California
Current (off California and Oregon), the Humboldt Current (off Peru and Chile), and the
Somali Current (off Western India). Because of this, all these currents support major fisher-
ies. Around the equator, winds and the Coriolis effect generate a divergence current resulting
in upwelling of denser, nutrient-rich water. There is also large-scale upwelling in the Southern
Ocean resulting from replacement of the northward flow of water driven by strong winds and
the Coriolis effect. The least productive areas are the subtropical gyres of the ocean basins.

Around coastlines, photosynthetic activity is generally high owing to the input of nutrients
from rivers, sediments, and wind-blown dust. Seaweeds, seagrasses, and salt-marsh plants in
the coastal zone contribute about 5% of total global primary production. Tropical coral reefs
are locally very productive because of the dense concentrations of photosynthetic dinoflagel-
lates in many corals, but because they occupy such a small area, they probably contribute less
than 1% to global primary production.

Although productivity on a global basis is fairly constant across the year, it is highly seasonal
in some oceanic regions. High latitude temperate regions, especially the North Atlantic, char-
acteristically show a spring bloom. Deep convective mixing occurs as the surface layer cools

Table 8.1 Net primary production of the oceans, estimated from SeaWiFS data
using a vertically generalized production model. There is a slight discrepancy in
the total production estimates between the geographical and seasonal
estimates. (ICMS, 2000).

Gt (Pg) C fixed % of total

Annual production in different oceans

Pacific 19.7 42.8

Atlantic 14.5 31.5

Indian 8.0 17.3

Antarctic 2.9 6.3

Arctic 0.4 0.9

Mediterranean 0.6 1.2

Total global annual production 46.1 100.0

Seasonal estimates of global ocean production

March–May 10.9 23.0

June–August 13.0 28.2

September–November 12.3 26.7

December–February 11.3 22.1

Total global annual production 47.5 100.0

 Microbes in Ocean Processes—Carbon Cycling 225

Figure 8.2 Average global con-

centration of surface chlorophyll
a measured by the SeaWiFS spec-
tral scanner from 1997–2003.
Chlorophyll concentrations (logarith-
mic scale) are higher near coastlines
due to terrestrial runoff and nutrient
upwelling, and higher in the north-
ern than the southern hemisphere.
Very low levels occur in the ocean
gyres. Dotted ellipses indicate
approximate locations of high-nutri-
ent low-chlorophyll (HNLC) regions
(see p.236). Credit: SeaWiFS Project,
NASA Earth Observatory.

during the cold, dark, windy winter—bringing nutrients to the surface layers and promoting
a rapid increase in photosynthesis as light levels increase during spring. Increased stratifica-
tion and nutrient depletion lead to reduction in productivity during summer, even though
light levels are at their highest. During autumn, vertical mixing occurs again and a small
secondary peak of production results. In tropical seas, seasonal effects are much less pro-
nounced, except where there is a seasonal upwelling of nutrients, such as that occurring in
the Arabian Sea.

Remote sensing can now also differentiate size classes among the phytoplankton, based
on the models that allow estimation of photosynthetic rates relevant to cell sizes combined
with in situ measurements of abundance using flow cytometry. This reveals differences in
geographic distribution and the relative contribution of different groups to the total produc-
tion. Microphytoplankton (larger diatoms and dinoflagellates) are abundant mainly at mid to
high latitudes and are patchily distributed. Nanophytoplankton (smaller diatoms and flagel-
lates) are moderately abundant globally, but concentrations are higher at equatorial and mid
to high latitudes and relatively low in subtropical gyres. Picophytoplankton (cyanobacteria
and picoeukaryotes) mainly dominate oligotrophic gyres. Detailed analysis of monthly and
annual variations in distributions of different size classes and correlation with other oceano-
graphic parameters reveals how phytoplankton biomass and production vary over different
periods—such information is a critical component of models to predict the impact of climate
change on ocean processes.

Carbon is rarely a limiting factor in productivity, since HCO3− is abundant in seawater and
most phytoplankton have mechanisms to concentrate it to improve photosynthetic efficiency
(p.83), but nitrogen, phosphorus, silicon, trace metals, and some trace organic compounds
such as B vitamins can all be limiting nutrients, thus affecting production. The availability of
iron has special significance, as discussed in Chapter 9.

The penetration of water by light of different wavelengths limits photosynthesis to the upper
100–200 m of clear ocean waters, and considerably less in the presence of suspended mate-
rial. Different phototrophic organisms are adapted to use light of different wavelengths due
to different types of chlorophyll and ancillary pigments, and this affects their distribution in
the water column. Indeed, the traditional definition of the limit of the euphotic zone as the
depth at which light intensity is reduced to 1% of its surface level is no longer valid, since
some ecotypes of Prochlorococcus are photosynthetic at light levels less than 0.1% of surface
irradiance. Very high irradiances can inhibit photosynthesis, especially due to damage to
the photosystem by ultraviolet light, and organisms possess photoprotective mechanisms to
overcome these effects (p.107). The water depth at which the light intensity is just enough to
balance the incorporation of CO2 by photosynthesis and its loss due to respiration is known
as the compensation depth. This varies according to species composition of the phytoplank-
ton, geographic region, season, light penetration during the day, and nutrient availability.
226 Chapter 8

Clouds and atmospheric dust have a marked effect on light penetration, and the density of
phytoplankton itself is a major factor due to self-shading. Usually, maximum primary pro-
duction does not occur at the very surface of the ocean, but several meters deeper where light
and nutrient levels are optimal.

Dark ocean carbon fixation makes a major

contribution to primary production
Although photosynthesis by the phytoplankton in the surface layer of the oceans is the most
obvious source of marine primary production, experiments measuring incorporation of
14CHO - and 3H leucine showed that there is substantial fixation of dissolved inorganic carbon
3
(DIC) in the mesopelagic ocean, perhaps reaching a level half that of phytoplankton export
production. As discussed above, much of the DOC transferred to the deep ocean is refrac-
tory and is therefore unable to support heterotrophic microbial growth. Furthermore, flux
measurements of the transfer of particulate organic carbon (POC) show that it is not enough
to meet the energy and biosynthesis demands of the standing stock of deep ocean microbes.
Some of this shortfall is made up by the ability of heterotrophs to incorporate DIC by anaple-
rotic reactions—these are “top-up” mechanisms used to maintain the supply of intermediates
in the TCA cycle. But this can only supply a small fraction of the carbon requirements of
heterotrophs. Therefore, the recent discovery of chemoautotrophy as a source of new fixed
carbon provides an explanation for the enigma that most of these microbes are metabolically
active and clearly finding plenty of nutrients to live on.

The discovery that about a fifth of all the picoplankton in the deep ocean are chemolithoauto-
trophic archaea (phylum Thaumarchaeota, Class Nitrosphaera), which fix CO2 using energy
from ammonia oxidation (AOA, see Box 5.2) provided a possible explanation. However, cal-
culations of the amount of ammonia reaching the mesopelagic zone—the only source is the
breakdown of sinking POM—show that it is much lower than that needed to fuel the levels of
autotrophy observed. The amount of carbon fixed by AOA is only about 10% of the estimate
for dark ocean autotrophy.

Recently, it was discovered that the organisms responsible for this missing 90% of new
organic carbon production are nitrite-oxidizing bacteria (NOB). Screening of single ampli-
fied genomes (SAGs) from the North Atlantic found that the major group of NOB in the
deep ocean belongs to the phylum Nitrospinae. These bacteria contain genes encoding nitrite
oxido-reductase and metagenomic and metatranscriptomic evidence supports nitrite oxida-
tion as the only energy source for these bacteria. Experiments measuring H14CO3- incorpora-
tion indicate that they could fix ~1 Gt (Pg) C per year globally—about a tenth as much as the
export production from the surface waters.

Classic food chain and microbial loop

processes occur in the epipelagic
As noted earlier, the classic view of marine trophic interactions was a simple pyramidal
food chain, roughly summarized as microalgae (mainly diatoms) → copepods → fish and
whales. Development of the microbial loop concept has placed much greater emphasis on
the importance of DOM and the activity of bacteria, archaea, grazing protists, fungi, and
viruses, culminating in our modern view of the ocean food web represented schemati-
cally in Figure 8.3. This shows that there are trophic interactions at multiple levels, with
microbial processes occupying center stage. However, elements of the classic model are
still valid, and their relative importance depends on hydrographic and environmental con-
ditions. In weakly stratified water masses, with a high degree of mixing and turbulence—
such as polar and coastal temperate seas during the spring bloom—the food web is often
based strongly on the dominance of organisms in the microplankton size class or greater
(i.e. diatoms, dinoflagellates, colonial prymnesiophytes, and crustacean grazers). By con-
trast, the large areas of the ocean in the tropics and subtropics that have highly stratified
waters with constant low-nutrient levels have food webs that are strongly dependent on
microbial loop processes dominated by the activities of nanoplankton (small protists), pico-
plankton (bacteria and very small protists), and viruses. As a generalization, the relative
 Microbes in Ocean Processes—Carbon Cycling 227

Figure 8.3 Schematic representa-

Algae
aee nd
Cyanobacteria andd tion of ocean food webs, showing
picoeukaryotess the central importance of dissolved
organic material (DOM) and the
microbial loop. The left side of the
Zooplankton Heterotrophic figure represents the “classic” food
DOM bacteria, archaea, fungi chain, whilst the right side rep-
resents microbial loop processes.
Solid arrows show the main routes
Fissh
Fish Nanoflagellates
for trophic transfer of fixed carbon.
Dashed blue lines indicate the release
Ciliates
of cellular material as DOM by viral
Mucus net feeders lysis of plankton, release of exudates,
or cell breakage during feeding by
zooplankton. Dashed orange lines
POM indicate the formation of particulate
Classic food chain Microbial loop
organic material (POM) from aggre-
gation of fecal pellets, polymers and
influence of the pico- or nanoplankton size classes of microbes is greatest in oligotrophic cellular debris. Mucus net feeders
waters, because conditions of low-nutrient flux exert positive selection for small size (see such as salps and larvaceans can
p.71). The relative importance of activity of larger particle-feeding protists, zooplankton, feed directly on very small microbes,
and fish becomes progressively greater as nutrient levels in the water column increase. A including virus particles. For clar-
useful analogy may be made between the behavior of food webs in nature and the behavior ity, not all trophic connections and
groups of organisms are shown.
of laboratory cultures. Food webs dominated by the microbial loop tend to behave like
Images of organisms and viruses are
chemostat (steady state) cultures. Despite rapid multiplication rates of some of the com-
illustrative only, and not to scale.
ponent organisms, species abundance and the concomitant geochemical processes in the
oligotrophic ocean gyres seem remarkably stable if viewed over long timescales. By con-
trast, food chains dominated by microalgae and copepods typical of spring bloom events
have dynamics that resemble batch laboratory cultures. Biomass, photosynthetic activity,
and respiration increase rapidly and then decline owing to nutrient depletion, cell death,
and the pressure of copepod grazing, leading to a greater export of photosynthetically fixed
carbon to deeper waters through the sedimentation of phytoplankton aggregates, zooplank-
ton fecal pellets, and cell debris. Certain hydrographic, nutritional, and climatic conditions
can lead to the very sudden development of massive, dense blooms, such as dinoflagellates
or diatoms and prymnesiophytes, which then disappear equally suddenly, often as a result
of viral lysis (see Chapter 7).

The microbial loop results in retention

of dissolved nutrients
About 50% of the daily net production from photosynthesis enters the ocean system as DOM,
which supports the growth of heterotrophic microbes via the microbial loop, resulting in
greater retention of dissolved nutrients in the upper layers of the ocean. Some DOM is formed
from direct extracellular release of carbohydrates, amino acids, lipids, and organic acids
from phytoplankton cells as they grow. The amounts released by this route are very variable
but are generally highest in the most photosynthetically active regions at high light intensi-
ties. Extracellular release might result from overproduction of photosynthate when CO2 fixa-
tion provides more organic material than can be incorporated into growing cells because of
nutrient limitation. There is probably also a constant leakage of low molecular weight (MW)
compounds across cell membranes owing to the steep concentration gradient caused by very
low solute concentrations in seawater. Under some circumstances, phytoplankton cells can
lose 5–10% of cell mass a day. The greatest contribution to the DOM pool comes from the
release of dimethylsulfoniopropionate (DMSP) from microalgae—it has been suggested that
no other single compound contributes as much to the DOM pool in surface waters—the for-
mation and fate of this compound is discussed in the next chapter. A large amount of DOM
and POM is released by protistan grazing on plankton. Partially digested particles of prey,
unabsorbed small molecules, and digestive enzymes are released as colloidal material dur-
ing the egestion process, when food vacuoles fuse with the external membrane. DOM and
POM are also released during grazing as a result of “sloppy feeding,” when phytoplankton
cells are broken apart by the mouthparts of zooplankton such as copepods. However, by far
228 Chapter 8

the most important source of DOM is the lysis of phytoplankton and heterotrophic microbes
by viruses. For example, up to 80% of the total photosynthetic production can be released as
DOM during the collapse of phytoplankton blooms.

The biological pump transports fixed carbon

to the deep ocean and sediments
The fecal pellets of zooplankton and fish, together with the fragments and polymeric sub-
stances from organisms that have died due to senescence or viral lysis contribute to the
formation of aggregates, and these settle to deeper waters as marine snow. Thus, some of
the carbon fixed in the upper layers is removed to deeper waters where the turnover time
increases greatly. As it sinks, some of the carbon in POM is remineralized to CO2 by the
respiratory activities of microbes, zooplankton, or fish following ingestion of particles. As
shown in Figure 1.7, viral lysis and the activity of extracellular enzymes leads to a substan-
tial release of DOM from particles as they sink. The vertical movement of organic matter is
accentuated by the daily migration of plankton over hundreds of meters in the water column.

Salps, larvaceans, and other mucus net feeding animals also provide a mechanism for export
of POC produced by small phytoplankton. These are small, pelagic barrel-shaped tunicates
that pump large volumes of seawater through their body for movement by jet propulsion and
trap food particles in a sticky mucus net that is rolled up and passed to the digestive tract.
Salps can trap and consume submicron particles, including cyanobacteria and heterotrophic
bacteria, at very high rates. The role of salps in the food chain has been overlooked until
recently, but we now know that they are a major food source for many species of fish, marine
mammals, turtles, and seabirds—stable isotope analysis shows that they are the dominant
source of nutrition for some species. Thus, they provide a direct trophic link between micro-
bial processes and higher organisms. In addition, salps package undigested material into
large, dense fecal pellets rich in carbon that provide a route for rapid sequestration of POC to
deep waters, amplified by the vertical migration of salps to a depth of 600–800 m at night.
The remains of dead salps also provide a large fraction of the carbon reaching the sea floor.

A large amount of fixed carbon—possibly as much as a quarter of net primary production—

is also removed to the ocean floor as CaCO3 from the walls of calcifying plankton such as
coccolithophores and foraminifera. In waters deeper than 3000 m, some of this CaCO3 dis-
solves to HCO3−, but in shallower waters the remains of these calcifying organisms lead to
the formation of calcareous sediments. Together, all the processes described constitute the
biological pump depicted in Figure 8.1, which results in the transport of fixed carbon from
the atmosphere to deep ocean waters.

Carbon export of primary production may change

due to ocean warming and acidification
Many studies conducted in the past few years have documented changes in the distribution
of plankton, which appear to be occurring due to the warming of the surface ocean due to
increasing levels of atmospheric CO2. This leads to changes in stratification, salinity, and pH
of the water column, which is expected to have marked effects on the production and export
of fixed carbon.

Alterations in seawater chemistry as a result of ocean acidification (OA, p.12) will have a large
effect on the process of calcification by organisms such as coccolithophores (p.180). Some
studies based on projections of future OA scenarios indicate that such calcifying organisms
could begin to experience difficulties in maintaining their CaCO3 skeletons as early as 2050,
especially in polar oceans where the state of CaCO3 saturation is lower. However, research
has also indicated that there is great variation in the responses of different species (and
strains within species) to changes in ocean pH and CO2 levels. Because coccolithophores
fix CO2 during photosynthesis, but also release CO2 when they make the plates of calcite
(coccoliths) that surround the cell, the balance between these processes is very important in
the global carbon cycle. Increased CO2 levels may enhance photosynthetic carbon fixation of
some types of phytoplankton and it is possible that calcifying plankton might benefit at the
 Microbes in Ocean Processes—Carbon Cycling 229

expense of some other groups. Some mesocosm studies of the response of Emiliania huxleyi
blooms to high CO2 levels simulating end of the century conditions have shown a reduction
in calcification rates when they were applied, but other evidence based on analysis of geo-
logical sediment cores suggests that coccolithophores have gradually increased their overall
calcification rates in response to past increases in CO2 levels. Some experimental studies have
shown that some strains of E. huxleyi increased both photosynthesis and calcification rates
at high CO2 levels. Enhancement of photosynthesis in this and other phytoplankton might
result in stimulation of the microbial loop through release of excess carbon as DOC and the
microbial carbon pump might be effective in sequestering this to deep waters as RDOM.
Alternatively, greater formation of sticky, polysaccharide-containing, transparent exopoly-
mer particles (TEPs) might contribute to the formation of marine snow that settles more
quickly, transporting POC and CaCO3 to the seafloor via the biological pump. Thus, there are
two mechanisms by which increased productivity might provide a partial negative feedback
mechanism whereby some of the increased CO2 levels dissolved in seawater is removed.
Further evidence for a possible negative feedback mechanism comes from observations of the
cyanobacterium Trichodesmium, which increases its rate of nitrogen fixation very markedly
at high CO2 levels. This could enhance the productivity of oligotrophic oceans that are cur-
rently limited by nitrogen and increase the flux of carbon in the biological pump.

An alternative scenario is that warming, freshening (due to melting of polar ice), and changes
in stratification of surface waters will lead to a shift towards the dominance of small pho-
totrophs, whose consumption by grazing heterotrophic protists results in retention of fixed
carbon in the surface layers rather than transfer to higher trophic levels. This would result ARE SALPS A
in a positive feedback mechanism in which microbial respiration remineralizes much of the
DOC to CO2. Numerous studies have shown an increasing abundance of picophytoplankton
? FAST TRACK FOR
CARBON EXPORT?
such as cyanobacteria and prasinophytes replacing microplankton such as diatoms and coc-
The discovery that salps can feed
colithophores. This is generally predicted to lead to a reduction in carbon export via the bio-
on very small picoplankton, which
logical pump due to lower sinking rates of picophytoplankton, which are smaller and lighter includes primary producers, as
(posssibly with positive buoyancy) than the negatively buoyant silica-containing diatoms or well as heterotrophic bacteria
calcite-containing coccolithophores. However, we know now that aggregation of small phy- and picoeukaryotes (Sutherland
toplankton due to production of TEPs means that particles can be large enough to be con- et al., 2010), means that large
sumed by zooplankton and removed to deeper waters by vertical migration and the formation populations of salps might divert a
of fecal pellets containing undigested carbon compounds. substantial fraction of the carbon
and energy from primary produc-
tion away from heterotrophic
Ingestion of bacteria by protists plays remineralization by transferring
it rapidly to deep waters, provid-
a key role in the microbial loop ing a mechanism for sequestering
The consumption of bacteria by heterotrophic and mixotrophic protists (Chapter 6) helps to CO2. Warming of the Southern
explain the regulation of bacterial biomass at near-constant levels. Bacterivory constitutes a Ocean (SO) is predicted to cause
mechanism by which nutrients contained in very small bacterial cells are made available to an increase of salps over krill. Salps
can form large swarms and may
larger planktonic organisms. In addition to its importance as a “top-down” control on bacte-
therefore be important in the verti-
rial production involved in heterotrophic recycling, grazing of cyanobacteria by protists has a
cal export of POM. Iversen et al.
direct influence on primary production. It should be noted that, for many species of grazing (2017) concluded that a large pro-
protists, bacteria are not the sole diet. Feeding experiments and analysis of the food vacu- portion of salp fecal pellets would
ole contents of protists show that many can feed on larger photosynthetic and heterotrophic probably break up before reaching
protists as well as inanimate organic particles (by phagotrophy) or dissolved compounds the deep ocean—“floaters” rather
(by osmotrophy). As discussed in Chapter 6, the most active and abundant bacterivorous than “sinkers”—but their increased
protists are flagellates in the pico- and nanoplankton size classes, with most being less than activity here might make produc-
5 μm. Flagellated protists, including dinoflagellates, cryptomonads, euglenids, and ciliates tion from picophytoplankton more
in the microplankton (20–200 μm) class are also active bacterivores. These organisms feed accessible to other zooplankton,
which can feed on undigested
by the generation of water currents with cilia and flagella and they can process hundreds
material in the pellets. In a linked
of thousands of body volumes per hour. As noted in Chapter 6, many of these protists are
study, Cabanes et al. (2017) con-
mixotrophic. Larger protists such as radiolarians and foraminifera are also bacterivorous, but cluded that increased dominance
their importance is less well quantified. (Chapter 6 contains micrographs of representative of salps in the SO could lead to
examples of these organisms.) greater export of POC, but this
would be associated with greater
Some metazoan zooplankton can also consume bacteria directly. The larvae and juvenile removal of iron that would reduce
stages of copepods have been shown to consume labeled bacteria (1–5 μm) in experimental its availability as a required nutrient
studies, but they are possibly less efficient grazers of the very small bacteria that dominate for photosynthesis.
230 Chapter 8

pelagic systems. As noted above, tunicates such as larvaceans trap bacteria in fine mesh
structures constructed of gelatinous mucus, and the discarded structures and fecal pellets
make a major contribution to the formation of marine snow. Indeed, since many pelagic bac-
teria and picoeukaryotes associate with particles of marine snow rather than exist as freely
suspended forms, they may be consumed by a wide range of larger metazoan zooplankton
and small fish when they ingest these particles, although they be unable to feed directly on
individual cells.

An additional key component of the microbial loop is the release of DOM through protistan
grazing. Large amounts of DOM—up to 25% of ingested prey carbon—are egested by pro-
tists, and some of this is readily metabolized and enters the DOM pool for further recycling
by heterotrophs, whilst the remainder enters the sink of long-term refractory material. The
relatively high respiration and excretion rates of protistan grazers mean that the cycling of
“lost” photosynthetic products through the microbial loop is rather inefficient.

Detailed measurements of grazing rates are technically difficult. In one type of experimental
study, the uptake of fluorescently labeled prey or 3H– or 14C−labeled prey bacteria and their
accumulation in food vacuoles is followed using microscopy. Another approach is the dilu-
tion method, in which a dilution series of seawater is prepared, so that the natural growth
rate of the prey bacteria is unaltered, whilst the predator to prey ratio is reduced. The rate of
change in bacterial density is then monitored during incubation at different dilution levels.
Separation of bacteria and their grazers in different size classes may also be achieved by
filtration. Using such techniques, many studies have attempted to evaluate the impact of pro-
tistan grazing on the density, dynamics, and structure of bacterial populations in the oceans.
In low productivity waters, grazing seems to be the dominant factor in balancing bacterial
production. However, in higher productivity waters, viral lysis will be favored because of
the higher bacterial population densities, as discussed below. Other factors such as removal
by benthic filter-feeding animals may also be important in controlling bacterial numbers in
coastal regions and estuaries.

Up to a certain limit, which depends on their own size, protistan grazers show a preference
for larger prey as food. Therefore, one consequence of grazing pressure is to encourage the
domination of microbial assemblages by small cells. As discussed on p.71, most bacteria in
oligotrophic pelagic environments are less than 0.6 μm, with many less than 0.3 μm, and
many of the smallest bacteria may be in a state of metabolic maintenance rather than active
cell division. Thus, the interesting concept arises that protistan grazers are preferentially
removing the larger, actively growing and dividing bacteria, leaving a large stock of smaller
bacteria that are growing slowly, if at all. However, selective bacterivory of larger, active
bacteria seems to stimulate the growth of other bacteria by the release of regenerated nutri-
ents such as ammonium and phosphate. Use of high-resolution video techniques shows that,
as well as chance encounters, flagellates may actively select their prey. Species-specific dif-
ferences in the processing of food particles explain the coexistence of various bacterivorous
nanoflagellates in the 3–5 μm size range and indicate the existence of specific predation
pressure on different bacteria. Motility, surface characteristics, and toxicity can all affect
the outcome of bacterium-protist interactions at the various stages of capture, ingestion, and
digestion.

Bacteria are also susceptible to predation by other bacteria. The only well-studied example
is Bdellovibrio (p.132), but it is likely that many other types remain to be discovered; these
could have considerable ecological importance in nutrient dynamics in marine systems.

The viral shunt catalyzes nutrient

regeneration in the upper ocean
When the microbial loop was first conceived, we had little idea of the abundance and activ-
ity of marine viruses. Now, we know that viral lysis of primary producers and heterotrophic
organisms act as a short circuit that disrupts the flow of nutrients into higher trophic levels—the
so-called viral shunt. As illustrated in Figure 8.1 and 8.3, viral lysis leads to the release of cell
contents, much of which enters the DOM pool and is readily recycled by heterotrophs. However,
 Microbes in Ocean Processes—Carbon Cycling 231

cell fragments and high MW polymers are more recalcitrant to breakdown. For example, phage
lysis of bacteria leads to release of cell envelope fragments containing embedded bacterial
outer membrane proteins, which are relatively resistant to proteolytic degradation. Some com-
ponents, such as cell wall polymers of algal cells lysed by viruses are also very refractory. The
quantity and dynamics of these less labile components resulting from viral lysis are not yet
fully known. Mathematical models can be constructed to compare the effects of different levels
of viral mortality on nutrient budgets. At a level of 50% bacterial mortality from viruses, the
overall level of bacterial production and respiration rate is increased by about a third, compared
with that occurring with zero mortality due to viruses. In these models, a high level of protistan
grazing leads to carbon input to the higher trophic levels of the food chain (animals), whereas a
high level of viral lysis diverts the flow of carbon from the food chain into a semi-closed cycle
of bacterial uptake and release of organic matter. Because cell fragments, viruses, and dissolved
substances do not sink—unless they aggregate into larger particles—one effect of viral lysis is
to maintain carbon and inorganic nutrients such as nitrogen, phosphorus, and iron in the upper
levels of the ocean. Viral lysis also contributes to the microscale heterogeneity of seawater
through the release of polymeric substances and the dissolution of material from marine snow.
As noted in Box 7.2, viral infection of E. huxleyi blooms also directly enhances the efficiency
of the biological pump by stimulating TEP production, TEP aggregation, TEP sinking, and
zooplankton grazing. There is growing recognition that this production of sticky viral lysates
and promotion of aggregation plays a greater role in carbon flux than previously realized—a
“shuttle” rather than a “shunt” effect. As discussed in Chapter 7, both lytic and lysogenic cycles
of infection occur when phages infect marine bacteria. Although the lysogenic state is com-
mon and has important consequences for genetic transfer, induction in natural marine systems
seems to be relatively low, so the majority of viral infection seems to occur via encounter of
bacteria with active virions that initiate the lytic cycle. This process is highly dynamic and is
affected by the virus to host ratio, the contact rates between virus and host cells, the rate of virus
adsorption, the rate of viral replication, the degree of host resistance, the physiological state of
infected host cells, the availability of nutrients for the production of viable virus particles, the
burst size of virus progeny, and the rates of viral decay. As discussed in Chapter 7, estimation
of the impact of viral lysis on bacterial and microalgal populations produces very variable
results, depending on the techniques used, but there is no doubt that viruses are responsible for a
significant turnover of ocean microbes, perhaps 20–40% per day overall. As noted above, viral
lysis is relatively more important in eutrophic coastal waters, whereas protistan grazing is more
important in oligotrophic ocean waters.

Microbial processes alter the composition of DOM

Labile DOM from primary production by autotrophs is rapidly metabolized by heterotro-
phic microbes using extracellular or cell-associated ectoenzymes (p.91). Different organisms
vary greatly in their preference for particular substrates depending on their possession of
membrane transport systems; for example, among the dominant Alphaproteobacteria, SARll
favors uptake of amino acids, whilst the Roseobacter clade favors carbohydrates. Degradation
of semi-labile material such as complex polymers occurs more slowly and often depends on
activity of specific organisms such as Bacteroidetes (p.92), fungi, or thraustochytrids (p.183)
or the syntrophic activity of consortia of microbes. Different amounts of specific types of
organic compounds can have a major influence on community structure of bacterioplankton.

Organic material is thus incorporated into microbial cells until it is returned to the DOM
pool by viral infection and lysis, as discussed above. This secondary material will be less
easily metabolized—an effect which will be amplified with each cycle of absorption,
incorporation, and release by viral lysis, ultimately resulting in RDOM which is recal-
citrant to further degradation—RDOM is usually defined as having a lifetime of >100
years. Different researchers have used various methods to calculate the annual rate of
formation of this refractory carbon, resulting in a consensus value around 0.2 Gt (Pg) per
year, with a total pool of ~662 Gt (Pg). Studies of the composition of RDOM shows that
at least 25% is of bacterial origin, with much of it occurring as refractory forms of neutral
sugars, amino sugars, and amino acids. The d-isomers of amino acids are particularly
prevalent since they are major components of bacterial cell walls and require conversion
by racemase enzymes before they can be metabolized. Certain types of microbial lipids
232 Chapter 8

Figure 8.4 Schematic representa-

tion of the viral shunt. The viral shunt
moves material from phototrophs
(green arrows) and heterotrophs (red
arrows) and into particulate organic Viral
matter (POM) and dissolved organic lysis Heterotrophic
matter (DOM). During this process DOM microorganisms
there is a stoichiometric effect, such (66:16:1) (69:16:1)
that the C:N:P ratio (numbers of
atoms in parenthesis) of the POM
and DOM pools are altered. Material
that is exported to deeper waters
by the viral shunt is more carbon
rich than the material from primary POM
production, increasing the efficiency (804:40:1) Phytoplankton
of the biological pump. Redrawn (106:16:1)
from Suttle (2007) by permission of Carbon
Springer Nature. Nitrogen
EXPORT TO DEEP WATERS Phosphorus

and derived alkane compounds are probably the most resistant and oldest forms of DOM
and are very resistant to degradation. They persist in sediments for millions of years—for
example, the use of crenarchaeol as a biomarker was used to establish the importance of
archaeal autotrophy by Thaumarchaeota in the deep ocean (Box 5.2). Degradation products
of carotenoid pigments (carboxyl-rich alicyclic molecules) are also a significant component
of ancient DOM, with a probable lifespan of >1500 years. Abiotic processes, especially
photochemical reactions at the ocean surface, also participate in conversion of DOM to a
refractory form.

The combined activity of heterotrophic microbes and the recycling of nutrients via cycles of
viral lysis leads to transformation in the elemental composition of DOM, because the residue
after repeated cycles is enriched in carbon content relative to the ratios of nitrogen and phos-
phorus (Figure 8.4).

Eutrophication of coastal waters

stimulates microbial activity
Nutrient enrichment of estuaries and coastal waters is a growing problem as a result of
anthropogenic sources of pollutants such as runoff from heavily fertilized land, sewage, and
animal wastes from agriculture and aquaculture. The impact of eutrophication depends on
the source, nature, and level of nutrient inputs, as well as hydrographic factors (especially
tidal flushing and mixing) and other physical factors (especially light and temperature).
Increased nutrient loading—particularly from nitrate and phosphate fertilizers—stimulates
phytoplankton growth beyond the point at which it is controlled by zooplankton grazing and/
or viral infection. For example, massive cyanobacterial blooms (e.g. Nodularia, Microcystis,
and Oscillatoria) occur regularly in the Baltic Sea, and eutrophication is probably a major
factor in the increased occurrence of harmful dinoflagellate blooms (p.319) as well as exces-
sive growth of macroalgae. Active microbial loop processes convert excess primary produc-
tion, but these too may be overwhelmed, and large amounts of decaying detritus and particles
of organic material will sink toward the seafloor. Bacterial decomposition leads to a heavy
demand for oxygen, resulting in mass mortality of benthic animals and fish. The number
of such “dead zones” in shallow coastal waters has approximately doubled since the 1960s.
Ironically, schemes to reduce the use of fossil fuels by production of bioethanol from crops
such as corn may increase the spread of coastal dead zones unless runoff from increased
fertilizer use can be controlled. Bacterial decomposition results in oxygen depletion, and the
overlying water column may become hypoxic, with oxygen levels too low to support fish and
many invertebrate animals. Even slightly reduced oxygen levels may disrupt food chains or
cause stress-induced disease in economically important fish and shellfish. As well as these
local effects, the expansion of dead zones may have serious consequences for climate change.
Heterotrophic denitrification leads to the production of the greenhouse gas nitrous oxide
 Microbes in Ocean Processes—Carbon Cycling 233

(p. 244); elevated atmospheric concentrations of this gas could further exacerbate the effects
of global warming and contribute to ozone loss, causing an increase in exposure to harmful
ultraviolet radiation. This is discussed further in Chapter 9.

Conclusions
This chapter has shown how research has revealed the central role and inter-connectedness
of microorganisms and viruses through their activities as components of carbon cycling
systems—the biological pump, microbial loop, viral shunt/shuttle, and microbial carbon
pump. Whilst ‘omics studies have led to major advances in understanding of processes in
the surface ocean, knowledge of deep ocean microbiology (especially the functional biol-
ogy of planktonic archaea) is still sparse and extensive, long-term microbiome time-series
studies at different locations and depths are needed. Increased use of autonomous sampling
and monitoring devices and development of large-scale mesocosms for experimental studies
will be very valuable. This is an area of intensive research activity and coordinated method-
ological approaches are leading to further insights into possible future trends. Better under-
standing will come from improved knowledge of the rates of primary production in different
size classes and taxonomic groups of phytoplankton, and from better quantification of the
activities of heterotrophic microbes in the remineralization of DOM or its transformation to
RDOM. We also need better knowledge of the decay rates of marine viruses, and how they
will be affected by climate change and other altered conditions. Increased efforts are now
needed to gain further understanding of the integration of these systems, and how they will
respond to changes in the physical, chemical, and hydrographic properties of the oceans as
a result of pollution by CO2, microplastics, other pollutants, and terrestrial nutrient input.

Herndl, G.J., Reinthaler, T., Teira, E. et al. (2005) Contribution of Archaea

References and further reading to total prokaryotic production in the ocean. Appl. Environ. Microbiol.
Azam, F., Fenchel, T., & Field, J.G. (1983) The ecological role of water- 71: 2303–2309.
column microbes in the sea. Mar. Ecol.–Prog. Ser. 10: 257–262. Hirata, T., Aiken, J., Hardman-Mountford, N. et al. (2008) An absorption
Cabanes, D.J.E., Norman, L., Santos-Echeandía, J. et al. (2017) First eval- model to determine phytoplankton size classes from satellite ocean colour.
uation of the role of salp fecal pellets on iron biogeochemistry. Front. Mar. Remote Sensing Environ. 112: 3153–3159.
Sci. 3: 289. Hobbie, J.E., Daley, R.J., & Jasper, S. (1977) Use of nuclepore filters for
Carlson, C.A., del Giorgio, P.A., & Herndl, G.J. (2007) Microbes and counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol.
the dissipation of energy and respiration: From cells to ecosystems. 33: 1225.
Oceanography 20: 89–100. ICMS (2000) ICMS Ocean Productivity Study. Rutgers, The State
Cermeño, P., Chouciño, P., Fernández-Castro, B. et al. (2016) Marine pri- University of New Jersey. Institute of Marine and Coastal sciences. http://
mary productivity is driven by a selection effect. Front. Mar. Sci. 3: 173. marine.rutgers.edu/opp/
Chisholm, S.W., Olson, R.J., Zettler, E.R. et al. (1988) A novel free-liv- Iversen, M.H., Pakhomov, E.A., Hunt, B.P. et al. (2017) Sinkers or float-
ing prochlorophyte abundant in the oceanic euphotic zone. Nature 334: ers? Contribution from salp pellets to the export flux during a large bloom
340–343. event in the Southern Ocean. Deep Sea Res. Part II Top Stud. Oceanogr.
138: 116–125.
Diaz, R.J. & Rosenberg, R. (2008) Spreading dead zones and conse-
quences for marine ecosystems. Science 321: 926–929. Jiao, N., Herndl, G.J., Hansell, D.A. et al. (2010) Microbial production
of recalcitrant dissolved organic matter: Long-term carbon storage in the
Dittmar, T. & Arnosti, C. An inseparable liaison: Marine microbes and
global ocean. Nat. Rev. Microbiol. 8: 593–599.
nonliving organic matter. In: Microbial Ecology of the Oceans, 3rd edition
(ed. Gasol, J.M. & Kirchman, D.L.), Wiley-Blackwell, pp. 189–288. Jürgens, K. & Massana, R. (2008) Protistan grazing on marine bacterio-
plankton. In: Microbial Ecology of the Oceans, 2nd edition (ed. Kirchman,
Dutkiewicz, S., Hickman, A.E., Jahn, O. et al. (2019) Ocean colour signa-
D.L.). Wiley, pp. 383–342.
ture of climate change. Nat. Commun. 10: 578.
Kolber, Z. (2007) Energy cycle in the ocean: Powering the microbial
Fenchel, T. The microbial loop—25 years later. J. Exp. Mar Bio. Ecol.
world. Oceanography 20: 79–88.
366: 99–103.
Laber, C.P., Hunter, J.E., Carvalho, F. et al. (2018) Coccolithovirus facili-
Fuhrman, J.A. & Azam, F. (1982) Thymidine incorporation as a measure
tation of carbon export in the North Atlantic. Nat. Microbiol. 537–547.
of heterotrophic bacterioplankton production in marine surface waters:
Evaluation and field results. Mar. Biol. 66: 109–120. Legendre, L., Rivkin, R.B., Weinbauer, M.G. et al. (2015) The microbial
carbon pump concept: Potential biogeochemical significance in the glob-
Hansell, D.A. & Carlson, C.A. (eds.) (2014) Biogeochemistry of Marine
ally changing ocean. Prog. Oceanogr. 134: 432–450.
Dissolved Organic Matter. Academic Press.
Madin, L.P., Kremer, P., Wiebe, P.H. et al. (2006) Periodic swarms of the
Henschke, N., Everett, J.D., Richardson, A.J., & Suthers, I.M. (2016)
salp Salpa aspera in the slope water off the NE United States: Biovolume,
Rethinking the role of salps in the ocean. Trends Ecol. Evol. 31:
vertical migration, grazing, and vertical flux. Deep-Sea Res. I: Oceanogr.
720–733.
Res. Pap. 53: 804–819.
234 Chapter 8

Nagata, T. (2008) Organic matter-bacteria interactions in seawater. In: Sherr, B.F., Sherr, E.B., & Pedrós-Alió, C. (1989) Simultaneous measure-
Microbial Ecology of the Oceans, 2nd edition (ed. Kirchman, D.L.). ment of bacterioplankton production and protozoan bacterivory in estua-
Wiley, pp. 207–242. rine water. Mar. Ecol. Prog. Ser.54: 209–219.
Pachiadaki, M.G., Sintes, E., Bergauer, K. et al. (2017) Major role of Sherr, E. & Sherr, B. (2008) Understanding roles of microbes in marine
nitrite-oxidizing bacteria in dark ocean carbon fixation. Science 358: pelagic food webs: A brief history. In: Microbial Ecology of the Oceans,
1046–1051. 2nd edition (ed. Kirchman, D.L.). Wiley, pp. 27–44.
Pernthaler, J. (2005) Predation on prokaryotes in the water column and its Sutherland, K.R., Madin, L.P., & Stocker, R. (2010) Filtration of submi-
ecological implications. Nat. Rev. Microbiol. 3: 537–546. crometer particles by pelagic tunicates. Proc. Natl Acad. Sci. USA 107:
Pernthaler, J. & Amann, R. (2005) Role and fate of heterotrophic microbes 15129–15134.
in pelagic habitats: Focus on populations. Microbiol. Mol. Biol. Rev. 69: Suttle, C.A. (2007) Marine viruses—Major players in the global ecosys-
440–461. tem. Nat. Rev. Microbiol. 5: 801–812.
Pomeroy, L.R. (1974) The ocean’s food web: A changing paradigm. Talmy, D., Beckett, S.J., Taniguchi, D.A. et al. (2019) An empirical model
Bioscience 24: 409–504. of carbon flow through marine viruses and microzooplankton grazers.
Pomeroy, L.R., Williams, P.J.L., Azam, F., & Hobbie, J.E. (2007) The Environ. Microbiol. 21: 2171–2181.
microbial loop. Oceanography 20: 28–33. Talmy, D., Beckett, S.J., Zhang, A.B. et al. (2019) Contrasting controls on
Proctor, L.M. & Fuhrman, J.A. (1990) Viral mortality of marine-bacteria microzooplankton grazing and viral infection of microbial prey. Front.
and cyanobacteria. Nature 343: 60–62. Mar. Sci. 6: 182.

Rahlff, J. (2019) The virioneuston: A review on viral–bacterial associa- Waterbury, J.B., Watson, S.W., Guillard, R.R.L., & Brand, L.E. (1979)
tions at air–water interfaces. Viruses 11: 191. Widespread occurrence of a unicellular, marine, planktonic, cyanobacte-
rium. Nature 277: 293–294.
Reinthaler, T., van Aken, H.M., & Herndl, G.J. (2010) Major contribution
of autotrophy to microbial carbon cycling in the deep North Atlantic’s Wilhelm, S.W. & Suttle, C.A. (1999) Viruses and nutrient cycles in the sea.
interior. Deep Sea Res. Part II Top. Stud. Oceanogr. 57: 1572–1580. Bioscience 49: 781–788.

Richardson, T.L. (2019) Mechanisms and pathways of small-phytoplank- Yamada, Y., Tomaru, Y., Fukuda, H., & Nagata, T. (2018) Aggregate for-
ton export from the surface ocean. Ann. Rev. Mar. Sci.11: 57–74. mation during the viral lysis of a marine diatom. Front.Mar. Sci. 5: 167.

Robinson, C. (2008) Heterotrophic bacterial respiration. In: Microbial Zhang, C., Dang, H., Azam, F. et al. (2018) Evolving paradigms in biologi-
Ecology of the Oceans, 2nd edition (ed. Kirchman, D.L.). Wiley, pp. cal carbon cycling in the ocean. Natl. Sci. Rev. 5: 481–499.
299–327. Zubkov, M. & Hartmann, M. Ecological significance of microbial tro-
Schiffer, J.M., Mael, L.E., Prather, K.A. et al. (2018) Sea spray aerosol: phic mixing in the oligotrophic ocean: The Atlantic Ocean case stud-
Where marine biology meets atmospheric chemistry. ACS Central Sci. 4: ies. In: Microbial Ecology of the Oceans, 3rd edition (ed. Gasol, J.M. &
1617–1623. Kirchman, D.L.), Wiley-Blackwell, pp. 99–122.
Chapter 9
Microbes in Ocean
Processes—Nitrogen,
Sulfur, Iron,
Phosphorus, and
Silicon Cycling

This chapter integrates closely with the discussion of carbon and energy cycling in the preced-
ing chapter and builds on the knowledge of ecophysiological processes and transformations
by the different microbial groups introduced in Chapters 3–6. In the first section, the role
of nutrients as limiting factors in ocean productivity is considered, with particular emphasis
on the importance of iron. Then, the significance of nitrogen cycling in ocean processes is
discussed, with emphasis on recent discoveries of new nitrogen-fixing organisms, anaerobic
ammonia oxidation, and the role of archaea in nitrification, which have led to major revi-
sion of conventional ideas. Sulfur cycling is considered in the third section, mainly from the
perspective of the transformations of organic sulfur compounds, the implications for climate
control, and the structuring of microbial communities. This is followed by a brief overview
of recent developments in our understanding of the uptake of inorganic and organic phos-
phorus. The final section considers the role of silica in the life of diatoms. In all cases, major
advances in our understanding have occurred in the past few years due to the integration of
biogeochemical studies with genomic and metagenomic approaches.

Key Concepts
• Phytoplankton growth may be limited by the availability of nitrogen, phosphorus, and
silicon; co-limitation and effects on different community members often occurs.
• The availability and distribution of iron is a critical factor in ocean productivity.
• Oceanic nitrogen fixation is a greater contributor to the nitrogen cycle than previously
thought and knowledge of the diversity, distribution, and importance of diazotrophic
cyanobacteria is increasing.
• Fixed nitrogen is returned to the inorganic pool by ammonification and nitrification,
which fuel growth of heterotrophs and phytoplankton.
• Nitrate reduction, denitrification, and anammox reactions return nitrogen to its ele-
mental form and other gases; the balance of these processes carried out by networks
of nitrogen-transforming microbes has important effects in oxygen minimum zones.
236 Chapter 9

• A key component of the marine sulfur cycle is the production of dimethylsulfonio-

propionate (DMSP) by algae which is consumed by numerous heterotrophic bacteria;
products from its breakdown have many effects on community structure and ocean
ecology and contribute to climatic effects.
• Marine microbes have diverse adaptations to phosphorus limitation; these are impor-
tant factors in microbial evolution and community dynamics.
• Silicon is an essential limiting nutrient for the growth of diatoms and its availability
affects competition between diatoms and other phytoplankton, with major influences
on operation of the biological pump for carbon export from the photic zone.

NUTRIENT LIMITATION
Key elements act as limiting nutrients for phytoplankton
Carbon, nitrogen, and phosphorus are required by all phytoplankton and silicon is required
by diatoms and some other organisms as major constituents of cellular material. Therefore,
their availability affects the production of organic material that enters the food webs dis-
cussed in Chapter 8. Carbon availability is not normally a limiting nutrient, because of the
high concentration of HCO3- in seawater, but most surface waters have low levels of inor-
ganic N, P, and Si. All organisms also require iron, as it is a key component of many enzyme
systems and electron transfer proteins. Although cells incorporate only miniscule amounts
of Fe, its low concentration in seawater is often the most important factor determining pro-
ductivity. Liebig’s Law of the Minimum is a principle derived from crop science, holding
that growth is not dependent on the total amount of resources available, but by the single
resource in shortest supply. Study of nutrient cycling in the oceans reveals more complex
interactions. Limitation of phytoplankton may occur by multiple nutrients, as well as light.
Such co-limitation can operate at the level of individual cells, at the population level, or at the
community level. Diverse species of plankton may be limited by different nutrients and may
be affected differently at various stages of their life cycle. Furthermore, the use of these and
other nutrients is closely linked in mixed natural communities, so it is not easy to untangle
these complex interactions.

Productivity of surface waters shows

WHAT IS THE
? REDFIELD RATIO?
marked geographical variations
Early ideas that available nitrogen is the most important limiting nutrient in the oceans arose
During a series of cruises from
WHOI, Alfred Redfield analyzed largely because most of the initial studies of biological oceanography were carried out in
thousands of marine samples temperate North Atlantic waters. During the spring bloom, dominated by diatoms, inves-
from different ocean regions. In a tigators observed that the increase in phytoplankton biomass and concentration of nitrate
keynote paper published in 1934, were inversely proportional. They showed that phytoplankton growth in seawater increased
he reported that the average with experimental addition of ammonium, but not of phosphate. In addition, some phosphate
elemental ratios of C:N:P in POM remained even when the nitrate levels reached zero. At the onset of the spring bloom in
and dissolved nutrients are nearly temperate waters, phytoplankton biomass—as indicated by chlorophyll concentrations—is
constant at 106:16:1. These find- uniform throughout the mixed layer, but as stratification develops, the maximum chloro-
ings have been hugely influential
phyll levels occur in deeper waters where nitrate levels are higher. New production is that
in our understanding of biogeo-
portion of primary production dependent on new sources of nitrogen to the photic zone,
chemical cycles. Many subse-
quent studies have revealed that, including nitrate from the deep layers, nitrogen fixed by diazotrophs, and terrestrial runoff.
although the average C:N:P ratio Regenerated production is that portion of primary production supported by nitrogen recycled
is generally stable, this can conceal by heterotrophic decomposition in the photic zone.
significant differences in temporal
and geographical deviations and This view of limiting nutrients has been challenged by detailed study of open-ocean eco-
differences between phytoplank- systems. In particular, remote sensing has confirmed the existence of ocean expanses which
ton species. Some organisms are have much lower phytoplankton biomass (indicated by chlorophyll concentrations <0.5 µg
adapted to sustained growth under L-1 chlorophyll) than would be expected considering the relatively high (>2 µM) concen-
low-resource conditions, while trations of nitrate and phosphate. These high-nutrient, low-chlorophyll (HNLC) expanses
others respond to high-nutrient
occupy about 20% of the surface area of the oceans in the eastern equatorial and subarctic
concentrations with rapid growth
regions of the Pacific and the whole of the Southern Ocean Figure 8.2. Nitrate and phosphate
(blooming).
are never significantly depleted, and the low-chlorophyll anomaly is largely explained by the
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 237

low levels of available iron. The trace metals zinc and cobalt may also be co-limiting. These
COULD MARINE
regions tend to be dominated by pico- and nanophytoplankton size classes; larger phyto-
plankton are less prevalent, especially diatoms and the cyanobacterium Trichodesmium. In ? VIRUSES BE
RESPONSIBLE FOR
some areas, such as the eastern Mediterranean, phosphorus appears to be the key limiting
IRON CYCLING?
nutrient, while silicon availability can be limiting for diatom growth in coastal waters and
may lead to harmful algal blooms (p.254). Many phages recognize and
infect their bacterial hosts via
the cell-surface TonB receptors
Ocean microbes require iron for siderophores. As iron is of
such importance as an essential
Bioavailable iron occurs in extremely low concentrations in seawater due to its low solubility at micronutrient, receptors are highly
~pH 8. Nitrogen fixation and photosynthesis are especially dependent on iron because the key conserved, favoring their use by
components in these processes—the nitrogenase enzyme complex and the photosystem-cyto- phages as an entry route. The tail
chrome complex respectively—rely on iron-containing proteins. Over 99.9% of “dissolved” fibers of phages of enteric bacteria
iron is tightly bound to organic compounds; it occurs mainly as colloidal Fe(OH)3, which are known to contain small num-
is very insoluble, precipitates rapidly and adsorbs strongly to organic particles (Figure 9.1). bers of iron ions, which are respon-
Microbes need special mechanisms to acquire this iron and many bacteria achieve this by sible for recognition of the TonB
production of siderophores (p.92), which bring iron into the cell via binding of the organic binding sites. It seems probable
that marine phage tails also contain
complexes to surface receptors and specialized transport mechanisms to bring the iron into
iron, leading Bonnain et al. (2016)
the cell. Some bacteria are also able to acquire the iron bound to siderophores produced by to propose a mechanism dubbed
other species. Photochemical reactions with α-hydroxy acid-containing siderophores (such the “Ferrojan Horse Hypothesis”—
as aquachelins) may lead to an increased bioavailability of the siderophore-complexed iron. derived from the “Trojan Horse”
In the nitrogen-fixing cyanobacterium Trichodesmium, iron deprivation leads to formation metaphor, whereby the target bac-
of the puff-like trichome colonies from individual filaments (Figure 4.12) and these serve to terium “invites” the phage attacker
collect and concentrate iron-containing dust particles. Although some novel mechanisms for with a “gift” of iron. Bonnain et al.
iron acquisition have been identified in some members of the Gammaproteobacteria, their calculate that the ~107 phage per
general significance for marine bacteria is not clear, and almost nothing is known about mL could contain as much as 70%
iron uptake in archaea. Eukaryotic phytoplankton do not appear to synthesize siderophores of the colloidal iron (0.02–0.2 µm)
in the surface ocean and that
but can take up iron via a cell-surface ferrireductase enzyme that liberates iron bound to
cellular iron stocks are recycled
organic compounds such as porphyrins, which are released from cells through zooplankton
into the tail fibers of new phage,
grazing and viral lysis. Phagotrophic eukaryotes such as flagellates can acquire iron from which therefore act as organically
ingested bacteria. Depending on the chemical nature of available iron complexes, microbes complexed “dissolved” iron. If
could therefore be competing for iron and the outcome will affect the separation of ecologi- supported by further research, this
cal niches and community composition. This will have consequences for the fate of carbon concept will have major impacts on
in iron-limited waters. Siderophore-bound iron may be the major source of the element in ocean biogeochemistry.
regions dominated by cyanobacterial photosynthesis and the microbial loop (such as the
oligotrophic tropical and subtropical oceans), while porphyrin-complexed iron may be the
major source in coastal regions dominated by diatoms and zooplankton grazing. Another
important factor may be the production of storage proteins to sequester scarce elements like
iron within cells. We know that some bacteria produce ferritin-like compounds to store iron
when it is more abundant than is needed for immediate use, but these compounds have not
yet been detected in aquatic bacteria. Microalgae are known to synthesize phytochelatins,
which can store a range of trace metals, but these do not seem to sequester iron. An inter-
esting observation is that domoic acid—a toxin produced by the diatom Pseudo-nitzschia
(p.320)—binds iron and is produced in greater amounts when cultures are grown with high
iron concentrations.

Terrestrial runoff, dust, and volcanic ash

are major sources of iron input
Low concentrations of iron have been proposed to explain the paradox of the low productivity
of HNLC regions. As noted above, these oceanic regions show a much lower rate of photo-
synthesis than would be expected from the availability of nitrate. The major source of iron is
terrestrial, and it follows that most coastal regions receive regular input from rivers, runoff
from weathering of rocks, and wind-borne transport of dust. The extent of this input depends
on the geology of the land, and some upwelling coastal regions can be iron-deficient. By con-
trast, the HNLC regions occur at great distances from the continents and will only receive
iron inputs via wind-blown dust or from remobilization of iron in sediments and upwelling of
deep waters. Episodic input of iron also occurs following volcanic eruptions. Iron concentra-
tions in surface water (<200 m) decrease with distance from land. Some of the experimental
238 Chapter 9

Figure 9.1 Iron cycling in the

oceans. Arrows show the transfer
of iron and the acquisition mecha-
nisms employed by marine microbes.
POM = particulate organic matter.

evidence in support of the iron-limitation theory is discussed in Box 9.1, together with contro-
versial proposals to use iron fertilization of the oceans in geoengineering, as a way of remov-
ing excess CO2 from the atmosphere as a countermeasure for global warming.

Hydrothermal vents and glacial melting

also supply iron to the oceans
Recently, iron input sources other than wind-borne dust have been discovered. There is an
annual flux of ~109 mol Fe of dissolved iron from hydrothermal vent plumes. This becomes
associated with particulate organic complexes that can be transported to provide a major
source of iron in the deep ocean. Bacteria in the region of these vents show high expression
of genes associated with siderophore synthesis and uptake, leading to the hypothesis of a
“microbial iron pump” that transfers inorganic iron to the organic phase, which may be dis-
persed throughout the oceans (Figure 9.2).

Iron also enters the ocean via meltwater from sediment released from the base of glaciers and
icebergs. Ice sheets at the poles contain large amounts of iron accumulated from dust in snow
over many millennia. The current melting of the Greenland ice sheet is estimated to release
~0.4–2.5 Mt (Tg) of bioavailable particulate iron into the North Atlantic each year, while the
Antarctic ice sheet is releasing ~0.1 Mt (Tg). It is conceivable that this additional iron could
boost phytoplankton growth, leading to sequestration of atmospheric CO2—thus acting as
a potential feedback mechanism to limit further warming. However, its impact would be
affected by the input of large amounts of fresh melt water, which is likely to alter plankton
communities and ocean circulation with unpredictable effects.

Figure 9.2 The proposed microbial

iron pump in deep-sea hydrother-
mal plumes. Uptake of Fe (primarily
Fe(III)) is conducted by dominant
chemosynthetic, methanotrophic
and heterotrophic populations.
Subsequent dispersal of Fe may
occur as Fe-siderophore complexes
or via whole cells or POC or DOC
produced through cell lysis. Bacterial
siderophores may also be dispersed
away from the plume and contribute
to the pool of strong iron-binding
ligands in the deep ocean. Reprinted
from Li et al. (2014), with permission
of Springer Nature.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 239

BOX 9.1 RESEARCH FOCUS

Can artificial fertilization of the oceans with iron

help to mitigate global warming?

The iron hypothesis. In 1990, John Martin of the Moss Landing that the iron enrichment experiments provided much useful infor-
Marine Laboratory, California proposed that climatic changes in gla- mation about ocean processes, but the design and timescale of the
cial and interglacial periods of Earth’s history were caused by chang- experiments were simply not appropriate to determine if aOIF could
ing patterns of deposition of iron-rich dust, resulting in modification of be a successful strategy for carbon sequestration. They proposed that
CO2 exchange between the atmosphere and oceans. He developed the the next generation of experiments should be conducted on a much
ultraclean analytical techniques necessary to study trace metals and larger scale, over thousands of square kilometers, and with observa-
used bottle experiments to show that addition of small amounts of iron tions using a wide range of techniques made over periods of months
led to rapid growth of phytoplankton in water from HNLC regions of rather than days or weeks. In addition, they proposed that experiments
the ocean. Martin is credited with a famous quip made in a lecture: should take place in semi-closed areas of water formed by eddies—
“Give me half a tanker of iron and I will give you an ice age” (Weier, stable circulating water masses, vertically coherent down to the bot-
2010), which provoked intense scientific controversy and public inter- tom with limited exchange with the surrounding ocean—and new
est. Martin planned experiments to enrich areas of the ocean with iron, approaches to the introduction of iron compounds should be devel-
but he died shortly before his hypothesis could be tested. His work was oped in order to provide better evidence for the effect of iron fertiliza-
carried on, and the first experiment (IronEx I) in artificial ocean iron tion on carbon sequestration. The most convincing evidence that aOIF
fertilization (aOIF) was conducted in 1993 in the equatorial Pacific off might be effective comes from the EIFEX trial, conducted in 2004
the Galapagos Islands, by addition of ferrous sulfate dispersed in the in a 150 km2 patch in the core of a mesoscale eddy of the Antarctic
wake of a ship’s propeller as it sailed over the area. In this patch, the Circumpolar Current (ACC). By using an extended period of observa-
surface of the sea turned from blue to green—the idea that addition of tion and detailed analysis of the depletion of dissolved and particulate
iron would stimulate phytoplankton growth in the ocean had been vin- carbon in the trophic zone, Smetacek et al. (2012) provided powerful
dicated. Oceanographers developed plans for more experiments. evidence that over half the biomass stimulated by iron addition in the
mixed layer sank to a depth >1000 m, due mainly to sinking aggre-
Iron enrichment experiments—what have they shown? Thirteen gates formed by the diatom Chaetoceros dichaeta.
mesoscale aOIF experiments have been conducted between 1990–
2012—four in the northwest Pacific, two in the equatorial Pacific, and Possible risks of OIF. Since its inception, there has been strong oppo-
seven in the Southern Ocean (SO), and a detailed analysis of these is sition to OIF. Environmental pressure groups argue that the deliberate
provided in the review by Yoon et al. (2018). Any significant gains alteration of marine ecology—stimulating blooms of relatively large
in CO2 sequestration would have to come from the SO because it phytoplankton that are usually not abundant—could shift the food
has very low iron levels, and the largest reservoirs of unused mac- web in unpredictable ways and distort biogeochemical and ecological
ronutrients in the upper and deep mixed layers. Most experiments relationships throughout the system (Strong et al., 2009). Some scien-
employed similar methodology, in which hundreds or thousands of tists argue that aOIF could stimulate the growth of toxic algae such as
kg of iron sulfate plus an inert tracer (sulfur hexafluoride) were added Pseudo-nitzschia, producing the toxin domoic acid (DA, p.320) that
over patches of ocean a few kilometers in diameter. Over periods of could potentially lead to levels of DA sufficient to cause mortalities in
a few days to weeks, scientists on board research vessels measured seabirds and mammals or necessitate the closure of fisheries to pro-
biomass and diversity of phytoplankton, together with a wide range of tect public health (Trick et al., 2010). There have been concerns that
biochemical, geochemical, hydrographic, and physical parameters. In production of other climate-relevant gases (N2O, DMS, CH4, and vola-
most experiments, phytoplankton blooms dominated by diatoms have tile organic compounds) could be enhanced by aOIF, or rapid bacte-
been stimulated with up to 15-fold increases in the chlorophyll content rial decomposition of stimulated blooms could lead to anoxic “dead
of surface waters, providing strong evidence that iron is a growth-lim- zones” (reviewed by Yoon et al., 2018). Iron addition might also disturb
iting factor. However, estimates of the amount of carbon drawn down the nitrogen-fixing activity of Trichodesmium, with potential major
into deeper water as blooms sink from the surface have been variable effects on ocean ecology (Walworth et al., 2018). Many scientists argue
and interpreted as much lower than predicted; although the SOIREE, strongly that these risks are not worth taking, because there is not
SEEDS I, and EisenEx experiments were terminated for logistical rea- enough evidence that OIF would be effective in making a significant
sons after 2–3 weeks duration—too soon to detect sinking. EIFEX impact on atmospheric CO2 levels. The Royal Society (2009) consid-
was the only experiment in which vertical flux could be followed in a ered the feasibility of aOIF as part of a wider review of planetary-
vertically coherent water column from surface to bottom. Recycling scale geoengineering approaches to control increased CO2 levels and
of fixed carbon by microbial loop processes may return CO2 to the global warming. They concluded that to have even a modest impact,
atmosphere after a short period, defeating the goal of long-term CO2 we would need to remove several Gt (Pg) of CO2 per year, maintained
sequestration. Watson et al. (2008) reviewed the limitations of these over decades and more probably centuries. A report by the National
first-generation experiments and concluded that the timescale and spa- Research Council, USA (2015) also argued strongly against OIF, while
tial extent of the experiments conducted were severely hampered by supporting investigation of some other approaches to geoengineering.
logistical difficulties and expense associated with carrying out this
type of work —large scientific teams are needed for shipboard experi- Commercial interests and the moratorium on aOIF experiments.
ments, many miles from land. Smetacek and Naqvi (2008) noted that Early models based on the potential of fertilization of low produc-
the total ship time dedicated to the first 10 years of aOIF experiments tivity regions were promoted enthusiastically by some scientists and
was only 11 months, compared with more than 2500 months to other entrepreneurs. They suggested that the technique might accelerate the
oceanographic research cruises in this period. These authors argued removal of billions of tons of atmospheric CO2—helping to control
240 Chapter 9

BOX 9.1 RESEARCH FOCUS

the effects of global warming, as well as having potential benefits geoengineering and the potential threat of runaway commercial-
for enhancement of fisheries—while providing direct economic ben- ization.” He makes an impassioned case for further experiments to
efits. Following United Nations initiatives for reduction of CO2 emis- explore the ecology of sinking diatoms in the ACC.
sions, the concept of a global market in the trading of “carbon offset
credits” emerged. Subsequent treaties on climate change imposed by The future of aOIF and other ocean solutions to address climate
governments have placed strict caps and taxes on emissions, encour- change. Interest in aOIF continues, and new proposals for enhanced
aging entrepreneurs to establish companies with the aim of profiting experiments emerge from time to time, but opposition remains strong.
from ocean fertilization while mitigating the consistent rise in atmo- Yoon et al. (2018) of the Korea Polar Research Institute conducted
spheric CO2 levels. Such iron fertilization takes place in international the most comprehensive review to date of past aOIF experiments and
waters and is subject to treaties and laws administered by the London published detailed design guidelines for a planned long-term experi-
Convention through the International Maritime Organization. In 2007, ment in the SO (Figure 9.3), modeled on the successful EIFEX exper-
they ruled that commercial iron “dumping” and carbon trading based iment. The open peer review comments published with this paper
on OIF ocean iron fertilization should be prohibited unless research provide valuable insight into current thinking. However, the funding
provides the scientific foundation to evaluate risks and benefits. This for this experiment was withdrawn. The Oceaneos Marine Research
moratorium was subsequently supported by nearly 200 countries at Foundation (Canada) is planning a large-scale release of iron off the
the UN Convention on Biological Diversity in 2008. coast of Chile, emphasizing its potential value in ocean restoration
(“seeding”) on a local scale to stimulate depleted fisheries, rather
Scientific opinion about aOIF remains divided. Leading ocean- than global geoengineering designed to remove CO2; however, some
ographers and marine ecologists have presented conflicting views scientists suspect that the foundation is seeking to profit from an
of aOIF. Strong et al. (2009) argued that it is “time to move on” unproven and potentially harmful activity (Tollefson, 2017). Emerson
because further iron enrichment experimentation “will not resolve (2019) has recently proposed research to investigate the feasibility of
any remaining debate about the risks of iron fertilization for geo- using aircraft to distribute a fine dust containing biogenic hydrous
engineering … and is both unnecessary and potentially counter- iron oxides (produced by cultivating iron-oxidizing bacteria in shal-
productive, because it diverts scientific resources and encourages low ponds), timed to coincide with seasonal phytoplankton blooms in
… inappropriate commercial interest.” By contrast, Smetacek and meso-scale eddies in HNLC regions. Gattuso et al. (2018) assess 13
Naqvi (2008) and Buesseler et al. (2008) argue that continued multi- global- and local-scale ocean-based measures, including aOIF, that
agency research of large-scale experiments is necessary to study the could significantly reduce the magnitude and rate of ocean warm-
relationship between phytoplankton growth and the effects of graz- ing, ocean acidification, and sea-level rise, as well as their impacts
ing and breakdown of biomass, before we can determine whether on marine ecosystems and ecosystem services. A high-level review
significant amounts of CO2 are removed from the atmosphere and by a UN expert advisory body on environmental protection proposes
whether large-scale iron fertilization could be useful. They argue that urgent development of a streamlined, coordinated framework for
future experiments must be designed to produce proper evidence of assessing and regulating future aOIF and other marine geoengineer-
effective long-term CO2 sequestrations so that society can weigh up ing initiatives, emphasizing the importance of a distinction between
the risks aOIF against the likely benefits. Smetacek (2018) deplores research to assess feasibility and the possible commercial benefits
the “vilification of OIF … due to the negative public perception of from their deployment (Boyd and Vivian, 2019a,b).

Figure 9.3 Schematic diagram of the Korean Iron Fertilization Experiment in the Southern Ocean (KIFES) representing the
experiment target site (eddy structure) and survey methods (underwater sampling systems, multiple sediment traps, sub-
bottom profilers, sediment coring systems, and satellite observations). Reprinted from Yoon et al. (2018), CC-BY-3.0.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 241

Whales and seabirds play a major role in

supply of iron to phytoplankton
In the last decade, there has been growing recognition of the important role that whales play
in nutrient cycling in the oceans. By their diving and surfacing activities, the large bod-
ies of whales induce physical vertical mixing of the water column, which is important in
areas of highly stratified ocean, while bottom-feeding results in suspension of sediments and
recycling of nutrients. Sperm whales feed mostly on cephalopods at great depths and only
defecate or urinate when they surface to breathe. The excreta are rich in nitrogen, phosphate,
and iron from their prey and this “whale pump”—operating in the opposite direction to the
normal biological pump—serves as a natural source of fertilization by transporting nutrients
from deep to surface waters. Because much of the body mass of these whales is blubber,
adult animals only incorporate a very small proportion of the iron from their prey and the
feces therefore contain high concentrations of iron. This has particular importance in HNLC
areas of the Southern Ocean, where it will stimulate new phytoplankton production. A sperm
whale defecates ~50 Mt tonnes of iron as dispersible liquid plumes into the photic zone each
year, so the current population of ~12,000 sperm whales could facilitate the sequestration of
0.5 Mt (Tg) of C yr-1. Although this is only ~0.05% of atmospheric CO2, deaths of whales
(“whale fall” – also see p.269) along migration routes also provides significant transfer of
carbon to the ocean floor. The current whale population is probably <10% of the historical
level and as low as 1% for the giant blue whales. This is due to the impact of whaling—espe-
cially the massive commercial operations in the nineteenth and twentieth centuries which
resulted in the killing of millions of whales of many species. Therefore, depletion of whale
stocks has undoubtedly had a major impact on ocean productivity and carbon sequestration,
providing a strong argument for conservation through prohibition of whaling, leading to res-
toration of this process and other ecosystem functions. Other marine mammals such as seals
and seabirds such as penguins, petrels, and albatrosses also contribute to iron and phosphate
cycling and this can be especially important in sustaining productivity in regions where their
population density is high (Box 9.2).

THE NITROGEN CYCLE

There have been major shifts in our
understanding of the marine nitrogen cycle
Nitrogen is a key element that constitutes about 12% by dry weight of all living cells because
of its role in the structure of proteins, nucleic acids, and several other cell materials. In the
environment, nitrogen exists in various oxidation states from -3 in ammonium (NH4+) to
+5 in nitrate (NO3−); the role of microbes in these oxidation-reduction transformations was
introduced in Chapter 3. While the principal aspects of the nitrogen cycle have been under-
stood for more than 120 years, major changes in our understanding of how it functions in the
marine environment have resulted from recent discoveries of previously unknown micro-
bial processes. As shown in Figure 9.4, six nitrogen-transforming processes can be recog-
nized, namely: nitrogen fixation, assimilation, ammonification, nitrification, denitrification,
and anerobic ammonia oxidation (anammox). The key metabolic reactions involved in these
processes were discussed in Chapter 3. One of the biggest challenges facing oceanogra-
phers is determining a budget for the input and output of nitrogen via nitrogen fixation and
nitrate reduction processes, respectively. As can be seen from the estimated fluxes shown in
Figure 9.4, marine processes have a dominant role in the global nitrogen cycle.

Diazotrophs incorporate atmospheric

nitrogen into organic material
Nitrogen gas (N2, N≡N) forms 78% of the modern atmosphere and must be fixed in a reduced
state into organic material before it can used in cellular processes. Only a small number
of microbes (diazotrophs) are capable of carrying out this transformation (see Figure 3.9).
For many years, the rate of N2 fixation in the surface waters of the oceans was thought to
be negligible compared with other sources of N2 such as upwelling of nitrate from deep
waters; it was also thought to be only a small fraction of the N2 fixed in soils on land. It
242 Chapter 9

has always been a mystery that there seem to be so few ocean organisms that can exploit
the abundant supplies of gaseous N2 under the evolutionary pressure of severe shortages of
dissolved inorganic nitrogen compounds, although it must be recognized that diazotrophy
is energetically expensive and depends on a large use of ATP. However, biogeochemical
evidence indicates that diazotrophy in the large portion of the biosphere occupied by the
tropical and subtropical oceans is much more important than previously realized and is
currently estimated at ~180-200 Mt (Tg) y-1, exceeding the flux due to terrestrial microbes
and industrial production of ammonia for fertilizers (Figure 9.4). In regions where primary
production is N-limited, diazotrophy supports up to half of new production. In the 1960s,
classical microbiological and physiological techniques using isotope tracer techniques led
to the suggestion that the large filamentous cyanobacterium Trichodesmium—highly abun-
dant as colonial blooms in tropical seas (see Figure 4.12)—could carry out N2 fixation.
For some time, this was discounted because it was not clear how Trichodesmium could fix
nitrogen without forming heterocysts, structures known to protect the nitrogenase enzyme
from O2 (p.136). However, use of a sensitive acetylene reduction assay technique that could
be used in the field proved that Trichodesmium makes a significant contribution to the
global N2 fixation by temporal and spatial separation of N2 fixation and photosynthetic
O2 production. However, Trichodesmium diazotrophy did not seem to be high enough to
match estimates of global N2 fixation. A major breakthrough came from the application
of molecular biological approaches to detect the presence of nitrogenase (nif ) genes in
water samples and the discovery of new groups of very small, abundant unicellular diazo-
trophic cyanobacteria occurring as free-living forms, or in association with heterocys-
tous diatoms (Figure 4.11), tintinnids, and radiolarians. A discovery of special interest is
the symbiosis between members of the uncultivated cyanobacterial clade UCYN-A (“Ca.
Atelocyanobacterium thalassa”). This forms a symbiotic association with a single-celled
prymnesiophyte algal host Braarudosphaera bigelowii and related species (Figure 4.13).
The cyanobacterium, which has an extremely reduced genome (Table 3.1) is dependent on
its algal host to provide fixed carbon compounds. In return, the host receives fixed nitro-
gen. UCYN-A appears to use host-supplied carbohydrates and fix nitrogen in the daytime,
with close coupling to the energy provided by host photosynthesis. Diverse lineages of the
organism have now been identified, including a separate type UCYN-A2, whose genome
has the same gene deletions as UCYN-A, but has some marked differences. It appears that
there are distinct pairs of hosts and UCYN-A strains, with differences in the size and num-
ber of UCYN-A cells associated with the host. UCYN-A, together with Trichodesmium
and the diatom-diazotroph associations, makes a major contribution to ocean N2 fixation.
Diazotrophic Alpha- and Gammaproteobacteria, as well as Planctomycetes, have also been
identified in ocean metagenomes, but little is currently known about their contribution to
nitrogen budgets. Nitrogen fixation is also thought to be carried out by particle-associated
anaerobic bacteria and archaea, including those inhabiting the gut of copepods; as these
are highly abundant members of the zooplankton, this could be a significant source of fixed
nitrogen into ocean food webs. As discussed in Chapter 10, diazotrophic bacteria are also
very important as symbionts of bivalves in sediments and seeps.

Figure 9.4 Overview of nitrogen

inventory and transformation, show-
ing marine, terrestrial, and anthropo-
genic fluxes. These processes do not
form one balanced nitrogen cycle,
as often depicted. Reprinted from
Kuypers et al. (2018), with permis-
sion of Springer Nature.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 243

The input of new organic nitrogen is balanced by losses due to denitrification (see below) and
understanding of how these processes are coupled has proved elusive. Many complex factors
determine the geographical and seasonal distribution of the various types of diazotrophic
bacteria, including temperature and the availability of either phosphorus or iron as limiting
nutrients. Recent research has shown that most marine N2 fixation takes place in the subtropi-
cal ocean gyres, which have low levels of nutrients; whereas water column denitrification
dominates in the eastern tropical Pacific and Arabian Sea, as shown in Figure 9.5. However,
UCYN-A is not limited to subtropical waters and has recently been shown to be present and
active in N2-fixation in the Western Arctic and Bering Seas. Because of the rapid changes in
Arctic ecosystems due to global warming, it is possible that selection may be occurring for
UCYN-A and other unicellular diazotrophs normally associated with warmer waters.

Fixed nitrogen is returned to the inorganic

pool by ammonification and nitrification
When organisms die and decompose, the amino groups of proteins, nucleotides, and other
organic N-containing compounds are returned to the inorganic pool via remineralization.
The excretory products of animals also contain urea or uric acid. The first stage in this
process is ammonification, carried out by many types of heterotrophic bacteria and fungi,
resulting in the production of NH3 or NH4+. Because it is highly reduced, ammonium can
be easily assimilated by many microbes and used to synthesize amino acids, which are then
transformed to other compounds via transamination reactions. This is a less energy-demand-
ing process than the uptake of nitrate, which requires large amounts of NADPH and ATP
because of its high oxidation state.

Nitrification is usually a two-stage process, consisting of (1) oxidation of ammonia to nitrite

and (2) oxidation of nitrite to nitrate, carried out by different organisms and enzymes (p.81).
Recently, members of the bacterial genus Nitrospira have been shown to be capable of com-
plete ammonia oxidation (“comammox”) to nitrate. Ammonia-oxidizing archaea (AOA) and
bacteria (AOB) are highly diverse, abundant, and active in areas critical for global nitrogen
cycling, including the base of the photic zone, mesopelagic and sub-oxic waters, and coastal
sediments. As a group, the marine AOA (Thaumarchaeota, such as Nitrosopumilus) have
a high degree of metabolic versatility, and in addition to their role in the nitrogen cycle,
their contribution to carbon fixation via autotrophy means that measurement of their activi-
ties in different geographical regions and at different depths is a high priority for research.
AOA dominate environments with low levels of ammonia, because of their high affinity for
this substrate (see Box 5.1), whereas NOB appear more prevalent in environments with high
ammonia concentrations. Nitrite-oxidizing bacteria are diverse and widely distributed.

Figure 9.5 Links between marine

nitrogen fixation and denitrification.
Nitrogen fixation occurs mainly in
the subtropical gyres (yellow areas)
downstream of the tropical upwell-
ing regions in the upper 100 m while
denitrification occurs primarily at
depths of 100–1000 m in the Indian
Ocean and the eastern tropical Pacific
Ocean (red areas). The high levels of
nitrogen fixation in the gyres contrib-
ute to the formation of biomass that
has high N:P ratios, which sinks to
the ocean’s interior (blue arrows) and
compensates for the loss of nitrogen
caused by denitrification. Reprinted
from Gruber (2019) with permission
of Springer Nature.
244 Chapter 9

Assimilation of ammonium and nitrate fuels

growth of phytoplankton and other microbes
Nitrate is a major nitrogen source for protists, fungi, bacteria, and archaea, many of which
possess assimilatory nitrate reductases in their cytoplasm and ATP-dependent transporters
in their membranes. In the photic zone, phytoplankton assimilate ammonia as their major
source of nitrogen for cell growth. They are thus competing for available ammonia with
heterotrophic microbes and with AOA and AOA, with the consequence that ammonia is
consumed almost as rapidly as it is formed. However, most phytoplankton, including diatoms
and Synechococcus, respond to the high-nitrate concentrations found at high latitudes and in
coastal and open-ocean upwelling, with rapid growth leading to the formation of seasonal
or transient blooms. They can also utilize nitrate, if it is available, using assimilatory nitrate
reductase enzymes. When nitrate is plentiful due to upwelling, community shifts favor the
preferential use of nitrate. This network of nitrogen-transforming microbes is illustrated in
Figure 9.6a.

Nitrate reduction, denitrification, and anammox reactions

return nitrogen to its elemental form and other gases
Denitrification is an anaerobic respiratory pathway resulting in the removal of combined
nitrogen from the ocean, returning N2 to the atmosphere. Nitrous oxide and nitrogen dioxide
can form under certain conditions. The first stage in this process is dissimilatory reduction of
nitrate to nitrite. Although the denitrification process only occurs under low-oxygen condi-
tions, potential denitrifiers belonging to a very wide diversity of heterotrophic microbes can
be identified in almost all environments in both oxic and anoxic waters and sediments, and
many of these organisms can also grow aerobically using oxygen as an alternative electron

Figure 9.6 Potential nitrogen-

transforming microbial networks in
marine ecosystems. (a) The open-
ocean gyres are vast nutrient-limited
regions in which nitrogen flux is
nearly balanced due to extensive
recycling. In the sunlit surface
waters, cyanobacteria mainly assimi-
late ammonium and/or DON for
growth. Viral lysis and zooplankton
grazing release DON, which is subse-
quently mineralized back to ammo-
nium by heterotrophic bacteria,
protists, and animals. Nitrogen-fixing
bacteria provide additional ammo-
nium. In deeper waters, ammonium
is oxidized to nitrate. Some of this
nitrate diffuses up into the surface
waters and is assimilated by phyto-
plankton. (b) In OMZs, upwelling
stimulates primary productivity in
the surface waters. Aerobic miner-
alization of sinking POM depletes
oxygen in the underlying waters.
Nitrifying communities adapted
to low-oxygen conditions oxidize
ammonia to nitrite and nitrate. The
OMZs are major regions of nitrogen
loss owing to the activity of anam-
mox and denitrification, which
involves complex communities of
bacteria and archaea. Reprinted from
Kuypers et al. (2018), with permis-
sion of Springer Nature.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 245

acceptor. Some abundant organisms such as Beggiatoa reduce nitrate fully to ammonium,
DIFFERENCES IN
while many (including SAR11) only reduce it to nitrite.
i NITRATE USAGE BY
PROCHLOROCOCCUS
Since the 1960s, measurements of the levels of ammonium in anoxic basins have indicated
that there are much lower levels of ammonium than might be expected to accumulate dur- Surprisingly, most cultivated strains
ing the intense remineralization (either as aerobic or N-dependent respiration) supposed to of Prochlorococcus—the dominant
occur there, suggesting that there is some unknown oxidation mechanism for its removal member of the phytoplankton in
by anaerobic microbial activity. The discovery of anammox (anaerobic ammonia oxidation) subtropical and tropical oligo-
explains this deficit and has necessitated a major reevaluation of the ocean nitrogen cycle. trophic oceans—were found to
be unable to grow using nitrate
Only a very narrow phylogenetic group of bacteria in the class Planctomycetes are known
because they lack genes for assimi-
to carry out this process, which depends on a specialized cellular structure with a mem-
latory nitrate reductase, leading to
brane that protects the cytoplasm from hydrazine, a highly reactive intermediate formed dur- the assumption that they depend
ing the oxidation of ammonia to nitrogen using nitrite as the electron acceptor (Figure 3.9). entirely on regenerated nitro-
Anammox bacteria grow extremely slowly—perhaps doubling only every 14 days or so—and gen through growth using NH4+.
many oceanographers believed that they would be unlikely to thrive in natural environments. However, analysis of ocean metage-
However, several studies have indicated that, in sediments and in some anoxic basins, anam- nomes revealed micro-diverse
mox is responsible for at least as much nitrogen loss as occurs by denitrification. It is likely lineages that possess nitrite and
that nitrate reduction and anammox are coupled and some of the ammonium needed for the nitrate assimilation genes (Martiny
anammox reaction can be provided via the process known as dissimilatory nitrate reduction et al., 2009). The genes are more
(DNRA). Another source is remineralization of organic matter linked to sulfate reduction. prevalent in some regions than
others, and Martiny et al. specu-
late that this is related to nitrogen
availability. There may be selec-
Diverse microbial metabolic processes occur tion of lineages with the ability to
in oxygen minimum zones (OMZs) acquire nitrate in regions with low
N concentrations, whereas in high
Up to half of the global loss of nitrogen from the oceans due to the processes of denitrifica-
N regions, this may be outweighed
tion and anammox occurs in the nitrite-rich oxygen minimum zones (OMZs). In most parts by the advantage of possessing
of the ocean, a suboxic layer (usually defined as <20 µM dissolved oxygen) typically occurs a smaller genome. Berube et al.
at intermediate depths (usually 1000–1500 m). This usually corresponds to the pycnocline, (2015) later derived axenic culture
where there is change in density and accumulation of high levels of particulate organic mat- of Prochlorococcus strains that can
ter, resulting in a zone of intense heterotrophic bacterial activity which depletes oxygen. grow using nitrate as sole source of
In some areas of the ocean, most notably the eastern part of the Pacific Ocean and in the N (~17% slower than with NH4+).
Arabian Sea, high rates of phytoplankton productivity results in intense microbial respira- Analysis of 41 Prochlorococcus
tion creates an extensive oxygen-deficient layer of water at a much shallower depth (typically genomes showed that genes for
50–200 m). Here the oxygen concentrations at their core are often about 50 times lower than nitrate assimilation have been
obtained multiple times and exist
those at the “classical” oxygen minimum. Defining the extent of the OMZs depends on the
in distinct lineages.
definition adopted; if defined by an oxygen concentration below 5 µM, they occupy about
0.05% of the volume of the world’s oceans, whereas if defined at the suboxic concentration
of below 20 µM, they occupy about 1% of the volume. OMZs are almost devoid of multicel-
lular organisms, except for a few with special adaptations, and are dominated by anaerobic
microbes. The major permanent open-ocean OMZs, in which extremely low-oxygen lev-
els and extensive loss of nitrogen occur, are the eastern tropical North Pacific (ETNP), the
eastern tropical South Pacific (ETSP) near the equator off Chile and Peru, and the northern
Indian Ocean, as shown in Figure 9.7. Recently, another permanent OMZ has been identified

Figure 9.7 Location of the major

oxygen minimum zones (OMZs).
The scale bar shows the oxygen
concentration at 200 m depth. 1,
Southwest African continental mar-
gin; 2, Arabian Sea; 3, Bay of Bengal;
4, eastern tropical North Pacific;
5, eastern tropical South Pacific; 6,
Black Sea. Credit: NOAA, National
Oceanographic Data Center, World
Ocean Atlas.
246 Chapter 9

in deeper waters of the eastern subtropical North Pacific (25°N–52°N) off the west coast of
MIGRATING
i ZOOPLANKTON
EXCRETE THE
North America. In addition, OMZs form during the winter in higher latitudes, such as the
western Bering Sea and the Gulf of Alaska. Oxygen concentration is not uniform throughout
the OMZ. A core region occupies about 10% of the volume of the whole OMZ where oxygen
SUBSTRATE FOR
levels can be so low that they are undetectable using the most sensitive instruments (<3 nM).
ANAMMOX
Some authors refer to this as an anoxic marine zone (AMZ), others designate it an oxygen
Analyses of NH4+ production by depleted zone (ODZ). There is usually a narrow oxycline (about 10–20 m thick) at the top and
DNRA and remineralization of bottom of the OMZ.
organic matter in oxygen mini-
mum zones (OMZs) have revealed Detailed studies of OMZs reveal that they may not be uniformly anoxic, because physical
an enigma—these processes do
process such as eddies can introduce low concentrations of oxygen, creating niches that allow
not seem to generate enough
ammonium to account for the level
aerobic microbial processes to continue in apparently anoxic conditions. There can be suf-
of anammox activity observed. ficient light below the oxycline to support growth of oxygenic Prochlorococcus, leading to a
Bianchi et al. (2014) show that this secondary chlorophyll maximum, as shown in Figure 9.8. The oxygen produced is quickly
can be explained by the diurnal scavenged by aerobic microbes. Methanotrophs, feeding on methane produced in continental
vertical migration (DVM) of zoo- shelf sediments that diffuses into the OMZ, are also active. These are usually regarded as
plankton and micronekton. These strict aerobes, although some clades have been recognized that use nitrate or nitrite as termi-
authors used acoustic measure- nal electron acceptors. Others can obtain oxygen from the dismutation of nitric oxide.
ments to show that the animals
spend the daytime in the oxygen- The anoxic or almost anoxic conditions result in reduction of nitrate, leading to high con-
deficient waters of the OMZ, where
centrations of nitrite. This is further reduced to the gases nitrous oxide (N2O) via the pro-
they excrete products of NH4+ and
cess of denitrification or to N2 via DNRA/anammox. Analysis of metagenomes and SAGs
urea (which is quickly converted
to NH4+ by bacterial metabolism).
has revealed that OMZs may contain high proportions (up to 40%) of bacteria belonging to
Although the total amount of pri- subclades of SAR11. These are normally considered to be aerobic heterotrophs (see p.41),
mary production transported from but lineages able to utilize nitrate as a terminal oxidant instead of oxygen have evolved in
the photic zone is low, it is con- the OMZs.
centrated within a limited depth
range of anoxic water. Bianchi et al. Many studies have investigated the relative importance of the processes of denitrification and
devised models to show that this anammox in OMZs, because N2O is a greenhouse gas with a global warming potential nearly
input accounts for the “missing” 300 times that of CO2 over 100 years. Since the two processes are regulated by different
NH4+ that supports the levels of
anammox observed. Subsequent
modeling of the global effects of
DVM suggest that it also drives
a significant and efficient flux of
carbon to the mesopelagic zone, of
similar magnitude to the transport
of DOC (Aumont et al., 2018).

Figure 9.8 Chemical and bio-

logical gradients in the core of the
ETNP OMZ, showing representative
profiles of oxygen, nitrate, nitrite,
methane, and chlorophyll a concen-
trations (left) and microbial richness
and abundance (right). Reprinted
from Bertagnolli and Stewart (2018),
with permission of Springer Nature.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 247

factors and are likely to be affected differently by climate change, this could lead to major
EXPANSION OF
shifts in ocean biogeochemistry and ecology. Results of different studies have produced vari-
able results, with some concluding that anammox is the dominant process in the ETSP and i COASTAL OMZS
Arabian Sea OMZs, accounting for up to 80% of removal of fixed nitrogen. Other studies Rising sea temperatures mean that
suggest denitrification can be dominant. As well as differences due to variations in methods, seawaters hold less oxygen. In
the process of denitrification is likely to have a much patchier temporal and spatial distribu- coastal areas, this is compounded
tion, because it is fueled by high input of organic matter. Surprisingly, nitrite oxidation can by massive inputs of nitrates, phos-
also occur under nearly anoxic conditions at the core of the OMZ and novel species of the phates, and organic nutrients from
sewage and fertilizers in terrestrial
phylum Nitrospinae have recently been implicated in this process.
runoff. This drives the microbial
processes leading to oxygen
All of the known nitrogen transformation processes work alongside each other in the OMZs, depletion and denitrification, with
forming a complex network, as shown in Figure 9.6b. Studies of the diversity of microbial consequent reductions in biodiver-
adaptations to the niche conditions found in OMZs and their influences on biogeochemical sity. The volume of anoxic water
cycles are increasingly important to predict the possible effects of ocean warming and ocean bodies in OMZs is estimated to
acidification. have quadrupled since the 1950s,
and oceanographic models predict
further declines in the twenty-
Microbial processes in sediments are a first century, even with ambitious
major contributor to nitrogen cycling declines in fossil fuel use (Breitburg
et al., 2018). The coastlines of
Nitrogen-containing POM that reaches the seafloor may be permanently buried, while a many industrialized and develop-
range of anaerobic heterotrophs are responsible for production of NH4+, which is available for ing countries now have numerous
regeneration by other organisms, or may be coupled to nitrate production in the oxic layers of hypoxic zones (O2 <2 mg L-1) and
sediments. POM exported to sediments in deep waters usually has a much higher C:N ratio some areas, including the Gulf of
Mexico, estuaries on the US East
than plankton in the upper ocean because of the extensive recycling of labile nitrogen-rich
Coast, and the Baltic Sea, have
compounds in the microbial loop and viral processes (see Figure 8.4). Nitrogenous com-
large areas where O2 concentra-
pounds reaching the seafloor are predominantly proteins, chitin, and other polymers with a tions are so low that they are desig-
relatively high molecular weight that have escaped degradation by bacterial enzymes during nated “dead zones.” Upwelling of
their descent to the seafloor. Organisms in the sediment may also absorb nitrate, and this anoxic waters due to changes in
process is especially important in coastal waters receiving high levels of nitrate from agri- atmospheric forcing can result in
cultural runoff. As noted on p.128, in some regions nitrate can serve as an electron acceptor mass mortalities of sea life close to
for bacteria such as Thioploca and Thiomargarita. These form dense colonies underneath shore, such as the emergence of
anoxic waters with high levels of nitrate from natural upwelling, and form filaments that a dead zone off the Oregon coast
stretch up to collect nitrate which is then reduced to NH4+ using sulfide as electron donor. in 2006, where there had been
Sediments on the seafloor are usually anoxic below the top few millimeters of the surface and no previous records of O2 defi-
cits (Chan et al., 2008). Oxygen
are therefore a major site for the anammox and denitrification processes previously described
depletion is leading to the inability
for the loss of fixed nitrogen. Bioturbation by worms, crustaceans, echinoderms, and mol- of an increasing fraction of the
luscs plays a major role in altering the structure and chemical composition of the sediment oceans to support diverse animal
through burrowing and irrigation. Such organisms are often described as “ecosystem engi- assemblages.
neers.” In particular, burrows introduce fresh oxygenated water into the sediments, which
greatly stimulates heterotrophic decay of organic matter, with concomitant effects on nitro-
gen cycling. Microbial community composition varies greatly, depending on the sediment
type and the species of bioturbating animals. Nitrogen cycling also occurs on the surface of
sediments in coastal and shelf seas in relatively shallow waters, where the penetration of suf-
ficient light allows establishment of communities of microalgae and cyanobacteria. Although
nitrate is the main source of nitrogen for these communities, some nitrogen fixation may
take place within microbial mats. Benthic nitrogen fixation by symbiotic or closely coupled
diatoms, cyanobacteria, and other diazotrophs helps to explain the high productivity of some
benthic systems, such as seagrass beds and coral reefs, as discussed in Chapter 10.

THE SULFUR CYCLE

The oceans and sediments contain large
quantities of sulfur compounds
Sulfur is an essential element for all organisms, constituting about 1% of cellular mass,
mainly occurring in the amino acids methionine and cysteine and in various coenzymes
and metalloproteins. The oceans and sediments contain a high concentration of inorganic
sulfur compounds in the various oxidation states. The most important oxidation states are
–2 (sulfhydryl R-SH and sulfide HS−), 0 (sulfur S0), and +6 (sulfate SO42−). The oceans are
248 Chapter 9

the largest reservoir of sulfur (in the form of sulfate) in the biosphere. As noted on p.87, it is
DOES THE
? BREAKDOWN
OF DMSP HAVE
necessary to distinguish between dissimilative and assimilative sulfate reduction. The dis-
similative process, for energy generation, leads to the production of hydrogen sulfide and is
restricted to the sulfate-reducing bacteria (including many members of the Desulfobacterota
A DEFENSIVE
(Deltaproteobacteria, see p.129), which are a key part of the community in anaerobic sedi-
FUNCTION?
ments. By contrast, many protists, bacteria, and archaea can assimilate sulfide produced dur-
Wolfe et al. (1997) concluded ing ATP-driven reduction of sulfate and incorporate it into organic sulfur compounds. The
that grazing of Emiliania huxleyi metabolic processes involved in microbial transformations of these states, and examples of
by protists results in mixing of the key taxa involved in sulfate reduction and sulfide oxidation in sediments, mats, vents, and
enzyme DMSP lyase and its sub- seeps were discussed in detail in Chapters 3 and 4, respectively. Sulfur oxidation–reduction
strate DMSP, producing DMS and
reactions are also of major importance in symbiotic associations of bacteria with marine
acrylate. When different protistan
grazers were presented with a mix-
invertebrates (Chapter 10).
ture of E. huxleyi strains possessing
high and low levels of DMSP lyase,
they preferentially selected strains
Metabolism of organic sulfur compounds is
with low enzyme activity. The especially important in surface waters
highly concentrated acrylate that
Organisms need sulfur for the synthesis of sulfur-containing amino acids and other essen-
is produced in this reaction was
thought to act as a deterrent in tial compounds. Despite the abundance of sulfate in seawater, most marine microbes obtain
view of its toxic effects. However, sulfur from recycling of organic sulfur compounds produced by phytoplankton, especially
Strom et al. (2003) concluded dimethylsulfoniopropionate (DMSP, (CH3)2S+CH2CH2COO −), which is produced from the
that dissolved DMSP itself inhibits amino acid methionine by many algae, especially dinoflagellates and prymnesiophytes. Some
grazing and suggested that this of the bloom-forming prymnesiophytes, such as Phaeocystis globosa and Emiliania huxleyi
may have exerted strong selec- (see p.181) and some dinoflagellates, are particularly potent producers of the compound, with
tive pressure in the evolution of intracellular concentrations of up to 300 mM. DMSP appears to have many functions. It acts
high levels of DMSP production as a compatible solute that provides protection against osmotic stress and has cryoprotectant,
in bloom-forming algae such as E. antioxidant, and defensive functions. It may have evolved as a metabolic relief mechanism
huxleyi. Strains of E. huxleyi with
when algae are undergoing unbalanced growth, in order to eliminate excess reducing power
high DMSP lyase activity also
appear resistant to Coccolithovirus and reduced sulfur. Synthesis of DMSP and another compatible solute, glycine betaine, may
infection (Schroeder et al., 2002). be regulated by nitrogen levels. Some DMSP is exuded naturally from healthy algae, but
Evans et al. (2006) showed that this probably represents only 1–10% of the cellular contents—most DMSP enters the water
both DMS and acrylic acid affected column as result of disruption of the cells by zooplankton grazing or by viral lysis. Although
viral infectivity and concluded widespread, not all phytoplankton produce DMSP, and its intracellular concentrations vary
that the DMSP system functions considerably in different groups, even within genera. The enzymes for DMSP synthesis are
as an antiviral defense, protecting found in the mitochondria and chloroplasts. Until recently, DMSP production was thought
remaining members of the popula- to be a unique property of phytoplankton, but the key enzyme DsyB is now thought to have
tion from infection. The mecha- originated in bacteria, as discussed in Box 9.2.
nism is unknown.
The annual global input of sulfur by DMSP production and release from phytoplankton has
been estimated at over 1000 Gt (Pg). DMSP is the main vehicle for the cycling of organic
sulfur compounds in the oceans. The high carbon content of DMSP means that it also plays
a very important role in the flow of carbon in the marine food web—it is probably the most
important single organic compound transferring fixed carbon to the microbial loop processes
(Figure 8.3).

Many algae also incorporate large amounts of sulfur into sulfated polysaccharides such as
mucus and cell-wall components. These compounds have important functions in defense
against grazing and the sequestration of metals. Because they are quite recalcitrant to micro-
bial attack, these compounds form a significant component of marine snow aggregates.

DMSP production leads to release of the

climate-active gas dimethyl sulfide (DMS)
The fate of DMSP is a very important factor in marine ecology and ocean–atmosphere inter-
actions. When DMS was first discovered, it was assumed that phytoplankton produced the
gas directly, but we now know that it is generated by the action of a group of DMSP lyase
enzymes, which are widespread in many marine microbes and break down DMSP into DMS
and acrylate (Figure 9.9). DMSP-producing algae often contain DMSP lyase themselves, but
this is in a separate cellular compartment, so it is thought that the enzyme only encounters
the substrate when cells are broken apart by grazing or viral lysis.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 249

Figure 9.9 Microbial transforma-

tions and the fates of dimethyl-
sulfoniopropionate (DMSP) in the
ocean, resulting in production of
dimethyl sulfide (DMS) and prod-
ucts that are assimilated by different
bacterial groups for cell growth. A.
Representation of the hypotheti-
cal mechanism by which release of
DMS gas to the atmosphere results
in the formation of cloud condensa-
tion nuclei, producing cloud albedo
that may constitute a feedback
process that regulates phytoplankton
growth. Figures show the estimated
percentage of DMSP transformed
at each stage. B. Principal trans-
formations of DMS by microbial
activity. Arrows are labeled with
the principal enzymes known for
the cleavage and demethylation
pathways. DddD, a class III acyl CoA
transferase enzyme, which is thought
to perform the initial step in DMS
release, as an alternative to DMSP
lyase; MeSH, methanethiol; MMPA,
3-methiolpropionate; MPA, 3-mer-
captopropionate; X-CH3, unidenti-
fied intermediate with terminal
methyl group. Most of the DMSP is
broken down by microbial activity;
80–90% is demethylated and enters
the microbial loop, while 10–20% is
metabolized by the DddD cotrans-
ferase enzyme or degraded by DMSP
lyases to DMS and acrylate, which
are also metabolized by bacteria.

The flux to the atmosphere is ~21 Gt (Pg) y-1 (similar to S emissions from burning of fossil
fuels) where it is oxidized to sulfates, which act as nuclei for condensation of water vapor
causing cloud formation. Therefore, DMS has an albedo function affecting temperature and
light availability to surface waters. DMS production was found to be related to phytoplankton
growth and is therefore controlled by light, temperature, and nutrient availability. This led to
the “CLAW hypothesis,” in which algal DMS production is envisaged as providing a feedback
mechanism that regulates climate. This idea became one of the “planks” for the influential
Gaia Hypothesis developed by James Lovelock and Lynn Margulis, which proposes that bio-
logical and physical components of Earth interact to maintain the climatic and biogeochemi-
cal conditions of the planet as relatively stable (in comparison with the history of other planets
in our solar system). As shown in Figure 9.9A, one version of the model envisages a decline in
phytoplankton growth as cloud increases owing to a rise in the level of DMS, causing a reduc-
tion in phytoplankton production until cloud formation diminishes again.
250 Chapter 9

BOX 9.2 RESEARCH FOCUS

Transformations of DMSP are major drivers of diverse ocean processes

The smell of sea. Remember those summer holidays? Getting Alphaproteobacteria (Howard et al., 2006; Reisch et al., 2008) and
close to the beach, we can detect a very distinctive tangy odor— subsequently found in other clades, with indications of transmis-
the evocative “smell of the sea.” This is largely caused by the gas sion by HGT. Demethylation is now recognized as the major DMSP
DMS, produced from DMSP from marine algae. As well as evok- catabolism pathway, converting >80% of dissolved DMSP. Again,
ing seaside memories, DMSP and DMS have far-reaching effects in DmdA is most prevalent in members of the Alphaproteobacteria,
biogeochemical processes and marine ecology. and analysis of metagenomes shows that more than half of the
bacteria in marine surface waters contain genes for DMSP meth-
DMSP lyases. The nature of the enzymes that degrade DMSP varies ylation; many of these also have photoheterotrophic potential via
greatly among different microbes (Figure 9.9B). Seven gene fami- proteorhodopsins (Moran et al., 2012; see Box 3.1). In a transcrip-
lies encoding DMSP lyases have been identified; these all produce tomic study, Vila-Costa Vila-Costa et al. (2010) showed that enrich-
DMS and acrylate, but they seem to be completely unrelated as they ment of surface water with DMSP led to an increased expression
have almost no sequence similarity and differ in their requirement of genes supporting heterotrophic activity. Lidbury et al. (2016)
for metal cofactors. The DddP and DddQ enzymes are the most showed that R. pomeroyi and other marine roseobacters associated
widely distributed and occur mainly in members of the marine with algal blooms can also use the enzyme trimethylamine mono-
Roseobacter clade (alphaproteobacterial order Rhodobacterales), oxygenase (Tmm), in the presence of methylated amines, to oxidize
but they also occur in other groups due to HGT. The respective DMS (produced by DMSP lyases) to dimethyl sulfoxide (DMSO).
genes are highly represented in marine metagenomes. In cultures
of Ruegeria pomeroyi and Roseovarius nubinhibens, expression Interactions between DMSP-producing algae and phycosphere-
of dddP and dddQ was induced by the presence of the substrate associated bacteria. Barak-Gavish et al. (2018) observed that
DMSP (Bullock et al., 2017). Algae also produce DMSP lyase; the inoculation of E. huxleyi cultures with a consortium of bacteria iso-
aspartate racemase protein Alma1 was first isolated from E. hux- lated from grazing copepods collected during a bloom in the North
leyi and subsequently found in a wide range of phytoplankton, and Atlantic led to rapid death of the algae. Application of antibiotics
some bacteria (Alcolombri et al., 2015). The DMSP lyase enzymes prevented cell death, implicating an algicidal effect by bacteria. A
appear to have different catalytic mechanisms for carrying out the strain of Sulfitobacter was isolated and shown to cause death and
same reaction, indicating separate evolutionary paths to this activ- lysis of 90% of the algal cells in co-culture, whereas uninfected
ity, suggesting that there were multiple ancestral enzymes and that E. huxleyi only showed ~40% mortality after 15 days. Electron
there are strong selective pressures to maintain this function in dif- microscopy revealed membrane blebbing in the early stages of
ferent groups of organisms (Bullock et al., 2017). cell death. Coculturing with a strain of Marinobacter resulted in
similar proliferation of bacteria, but no increase in algal cell death.
Alternative routes for DMSP catabolism. A significant fraction Barak-Gavish et al. showed that the Sulfitobacter was closely asso-
of DMSP catabolism is diverted away from DMS production and ciated with the dynamics of naturally occurring bloom. The experi-
towards degradation and incorporation by chemoheterotrophic bac- menters identified the production of volatile methanethiol (MeSH),
teria. DMSP is estimated to supply up to 15% of the total bacte- DMS, and dimethyl disulfide in infected cultures, whereas non-
rial carbon demand and nearly all of the bacterial demand in the infected control cultures produced only DMS (from the activity of
surface waters of the ocean—probably no other single compound DMSP lyase). Further analysis of the interplay between DMSP pro-
contributes as much carbon to the DOM pool (Yoch, 2002). The duction and the growth and pathogenicity of Sulfitobacter suggests
gene dmdA encoding the enzyme responsible for DMSP demethyl- that the assimilation of reduced sulfur, via MeSH, into methionine
ation was first recognized in Roseobacter and SAR11 clades in the and cysteine leads to synthesis of bacterial algicidal compounds;

Figure 9.10 Model of the possible routes by

which algal DMSP promotes bacterial virulence
in the E. huxleyi phycosphere. During interaction,
Sulfitobacter D7 consumes E. huxleyi–derived DMSP
and transforms it into MeSH, which facilitates
bacterial growth. DMSP and its metabolic products
can promote production of QS molecules which
may enable chemoattraction to algal cells and
production of bacterial algicides. Reprinted from
Barak-Gavish et al. (2018), CC-BY-4.0.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 251

BOX 9.2 RESEARCH FOCUS

this has also been shown in the phycosphere-associated roseobacter ensure effective propagation, as occurs in viruses infecting cya-
Phaeobacter, which also kills E. huxleyi during senescence (Wang nobacteria and E. huxleyi. Raina et al. (2010) suggest that sulfur
et al., 2016). Thus, the balance between competing DMSP catabolic cycling plays a major role in structuring the microbial communi-
pathways, driven by microbial interactions, may regulate oceanic ties associated with the coral holobiont and that DMS production
sulfur cycling and DMS flux to the atmosphere (Figure 9.10). provides a local feedback mechanism by increasing cloud cover and
albedo effect over coral reefs (as in Figure 9.9A). In 2013, this group
Production of DMSP in bacteria and algae. Curson et al. (2017) made the surprising discovery that coral juveniles produce DMSP
demonstrated the production of DMSP in a range of cultivated in the absence of algal symbionts, a finding which “overturn[s] the
marine bacteria and identified the dysB gene encoding the key paradigm that photosynthetic organisms are the sole biological
methyltransferase enzyme responsible for its biosynthesis. Curson source of DMSP, and highlight[s] the double jeopardy represented
et al. (2018) identified numerous functional dsyB homologs (DSYB) by worldwide declining coral cover, as the potential to alleviate
in the genomes or transcriptomes of all marine prymnesiophytes, thermal stress through coral-produced DMSP declines correspond-
most dinoflagellates, and some diatoms. Phylogenetic analysis of ingly” (Raina et al., 2013).
the DsyB/DSYB proteins led Curson et al. to suggest that DMSP
production originated in bacteria and was transferred to eukary- We smell the seaside … seabirds smell lunch. Experimental
otes, either through endosymbiosis at the time of mitochondrial studies by Gabrielle Nevitt and coworkers at the University of
origin from the alphaproteobacterial ancestor, or more recently by California–Davis have shown that foraging procellariform sea-
HGT. Using NanoSIMS tracking of 34SO2, they showed that DMSP birds such as albatrosses, shearwaters, and petrels fly in response
is produced in the chloroplasts and mitochondria. This discovery to concentration gradients of DMS in the atmosphere (Nevitt and
sheds important light on the possible evolution of DMSP produc- Bonadonna, 2005; Nevitt et al., 2008). “Hotspots” of DMS produc-
tion in the dinoflagellates, prymnesiophytes, and diatoms, which tion occur at regions where there are high levels of phytoplankton,
first evolved about 250 MYA (Falkowski et al., 2004). Bullock with accompanying grazing by krill. Higher-order predators like
et al. (2017) hypothesize that the proposed roles for DMSP as an fish and squid will also be attracted to the area, so high DMS levels
antioxidant, osmolyte, cryoprotectant, defensive molecule, and cel- indicate a good place for birds to forage for food. Models can be
lular energy-balancing mechanism may explain the evolution and used to link climatological data linking the distribution of seabirds
dominance of these phytoplankton groups, by enabling their rapid with DMS or types of DMS-producing phytoplankton. Besides
adaptation to extreme environmental changes prevalent at the time birds, many organisms including zooplankton, reef fishes, whale
of their radiation. Bullock et al. also provide evidence for the co- sharks, turtles, and baleen whales have been shown to respond to
evolution of the DMSP-producing phytoplankton and the marine dissolved DMS as a feeding strategy. A fascinating example of this
Roseobacter clade. infochemical effect of DMS on marine ecology and large-scale
processes comes from the research by Savoca and Nevitt (2014)
The role of DMSP and DMS in coral biology. As discussed in who analyzed a 50-year database of Southern Ocean (SO) seabirds
Chapter 10, corals are holobionts consisting of complex commu- to test the hypothesis that DMS emanating from phytoplankton
nities of the coral animal and various microbes, often including mediates a mutualistic trophic interaction, If this is the case, then
Symbiodiniaceae dinoflagellates (zooxanthellae). These produce carnivorous seabirds that are attracted to DMS should specialize
high intracellular levels of DMSP, and the very high densities of in feeding on primary consumers of phytoplankton that produce
zooxanthellae within the tissue of many corals means that the the gas. Indeed, krill comprised 80% of the diet of DMS-tracking
concentration of DMSP produced by corals may be a highly sig- seabirds, whereas non-DMS-tracking species fed mainly on sec-
nificant component of the global sulfur cycle, despite the restricted ondary consumers like fish and squid. Savoca and Nevitt made a
distribution of reefs. Release of large quantities of DMSP in coral connection between phytoplankton, krill, and seabirds, postulating
mucus provides a major link to microbial loop processes in the a three-way trophic mutualism based on the recycling of iron in
nutrient-deficient waters inhabited by most tropical corals. This has surface waters. Krill contains about a quarter of the bioavailable
been investigated on the Great Barrier Reef by scientists from the iron in surface waters of the SO. By feeding on krill, seabirds make
Australian Institute of Marine Science. Raina et al. (2009) isolated large local deposits of iron via their feces, which stimulates fur-
a variety of alpha- and gammaproteobacterial species capable of ther phytoplankton growth. Many other bird species are visually
degrading DMSP from Acropora and Monipora corals. The domi- attracted to krill swarms by the activity of the DMS-sensitive birds.
nance of Spongiibacter correlated with the observations of Bourne Since seabirds only return to land briefly to breed, most of their
et al. (2007), who showed the decline of this bacterium in corals feces will be deposited at sea. Savoca and Nevitt calculated that
during bleaching—when loss of Symbiodinium results in lower there are ~2.5 x 108 seabirds in the SO, with a biomass similar to
DMSP synthesis—and its reappearance as the corals recover and that of blue whales. In a further twist to the DMS story, Savoca et
resume DMSP production. Raina et al. (2010) undertook bioinfor- al. (2016) showed that after less than a month of exposure to sea-
matic analysis of an extensive collection of marine metagenomes water, common plastics acquire a DMS signature at concentrations
and showed that the dmdA gene was consistently found in reef water that attract procellariform seabirds. This may explain why these
samples, supporting the idea that DMSP demethylation is an impor- birds and other animals that use olfactory feeding cues consume so
tant source of sulfur and carbon for microbial communities in coral much plastic, with devastating consequences (Box 13.1), although
reefs. Many of these genes were associated with phage sequences, Dell’Ariccia et al. (2017) question some of the data that Savoca et
and Raina et al. suggest that they may have originated from host al. used to support their hypothesis.
bacteria and are carried by phages as additional information to
252 Chapter 9

IS IT TIME TO
THE PHOSPHORUS CYCLE
? RETIRE THE CLAW
HYPOTHESIS? Phosphorus is often a limiting or co-limiting nutrient
The CLAW hypothesis for regu- Phosphorus is an essential element for all life, being a key constituent of nucleic acids and lip-
lation of the global climate is ids and a crucial component of ATP in energy transfer reactions. Inorganic phosphorus also
named after the authors (Charlson, has a key role in photosynthesis. In the oceans, there are various forms of dissolved inorganic
Lovelock, Andreae, and Warren; phosphorus (DIP), of which orthophosphate PO42− is the most abundant. A large fraction of
Charlson et al., 1987) who pro- the phosphorus in surface oligotrophic waters is present as dissolved organic phosphorus
posed that DMS emissions from (DOP) compounds. The sole input and output of phosphorus to the oceans is via terrestrial
phytoplankton are regulated up runoff and burial in sediments, respectively. Thus, biogeochemical cycling of phosphorus is
and down by the production of closely linked with carbon flux and the availability of phosphorus has a major limiting or co-
cloud condensation nuclei (CCN)
limiting effect on oceanic primary production rates and microbial community composition.
and a consequent albedo effect.
The paper has proved extremely
This phosphorus limitation has been studied extensively in oligotrophic regions of the North
influential (>4340 citations) and Pacific, western North Atlantic (Sargasso Sea), and the Mediterranean Sea, where analyses
has stimulated many technical show that DIP is turned over very rapidly. There is some evidence that climate change over
developments and research investi- the last few decades is leading to stratification of the subtropical Pacific gyre, resulting in an
gations integrating diverse special- increase in nitrogen fixation and changing the system toward limitation by phosphorus. The
isms including microbial ecology, organic and inorganic forms of phosphorus are transformed continuously by microbial activ-
biochemistry, ocean, and atmo- ity. In the surface ocean, phytoplankton assimilate DIP and incorporate it into cellular mate-
spheric chemistry, cloud physics rial, which is ingested by zooplankton or enters the microbial loop via exudation or cell lysis,
and climate dynamics. Quinn and where it is taken up and recycled by heterotrophic bacteria. Heterotrophic and autotrophic
Bates (2011) suggest “it is time to
microbes compete for available phosphorus, which affects productivity. Recycling of DOP
retire the CLAW hypothesis.” They
occurs throughout the water column and regeneration of inorganic phosphorus occurs in
affirm that the hypothesis was
visionary in scope, but they argue deeper waters. Although phosphate seems to be the preferred source of phosphorus for phy-
that the investigations it inspired toplankton, they can also hydrolyze organic phosphorus compounds when phosphate does
have led to discovery of more not meet their demands.
complex processes at work. We
now know that there are multiple
sources of CCN, including sea salt, Marine microbes are adapted to low and
inorganic and organic molecules
from the sea surface microlayer, variable levels of phosphorus
and even living microbes (Amato et As discussed in Chapter 3, planktonic bacteria demonstrate one of two lifestyles. Many are
al., 2017). Also, processes such as
oligotrophic and are adapted to subsist on a relatively fixed diet of constant low levels of
formation of bubbles and aerosols
are affected by complex biologi-
nutrients. This is often reflected in very small cell sizes and streamlined genomes, as exem-
cal interactions, wind speed, and plified by photoautotrophs like Prochlorococcus and chemoheterotrophs like SAR11 bac-
other physical factors (Schiffer et teria. Others are copiotrophic or opportunistic and respond to inputs of increased levels of
al., 2018). nutrients because they have larger genomes and more metabolic versatility. Such organisms
may respond to nutrient starvation by initiating a program of changes in cellular composition
and activity. Since DNA and phospholipids incorporate large amounts of phosphorus, small
cells with reduced genomes and less membrane content require less and are at an advantage
in conditions of phosphorus limitation. In fact, some phytoplankton (both cyanobacteria and
eukaryotes) growing in low-phosphorus environments reduce their demand further by substi-
tuting phospholipids in their membranes with lipids containing sulfur and sugars.

The mechanisms of uptake of phosphorus compounds have been extensively studied in labo-
ratory cultures of bacteria like Escherichia coli, but less so in marine bacteria. Homologs of
PstS, the high-affinity phosphate-binding protein in E. coli, are produced by Synechococcus,
Prochlorococcus, and Trichodesmium, and multiple copies of the gene are present in the
genome sequences of cultured strains. These genes are also abundant in marine metage-
nomes. High-affinity phosphate-uptake genes are also present in the genomes of viruses
infecting cyanobacteria and E. huxleyi, and their expression is upregulated under conditions
of phosphate deficiency—presumably increased phosphate uptake ensures efficient viral rep-
lication. Other phosphate transporter systems have also been identified in marine microbes.
Cell lysis leads to the release of enzymes such as nucleases, lipases, and esterases that degrade
organic compounds to regenerate phosphate.

To obtain phosphorus from organic phosphorus compounds, bacterial cells must hydrolyze
them to orthophosphate. The Pho regulon studied in E. coli contains a number of genes
associated with regulation, uptake, and hydrolysis of phosphorus compounds. Pho gene
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 253

homologs are abundant in genomes and metagenomes of marine bacteria, which appear to
possess several different mechanisms for transporting and hydrolyzing phosphate esters. The
most common type of enzyme is alkaline phosphatase, whose expression is repressed by
high phosphorus levels and is often used as a marker for phosphate stress. These enzymes
are grouped into several families, including PhoA, PhoD, and PhoX, which differ in their
substrate specificity, cellular location, and requirement for metals as cofactors. Based on
bioinformatic analysis of marine metagenomes, it seems that a large proportion of coastal
and oceanic bacteria contain phosphatases within the cytoplasm, suggesting that small DOP
compounds are carried across the cytoplasmic membrane, providing carbon (and possibly
nitrogen) to the cell, as well as phosphorus.

The phosphonates are another group of organic phosphorus compound whose contributions
to marine nutrient cycles have been relatively little studied until recently. Phosphonates are
characterized by a stable C–P bond, rather than the C–O–P bond found in esters, and they
are generally very resistant to chemical and thermal decomposition. Naturally occurring
phosphonates are produced by a range of organisms as membrane components and there-
fore are found in the DOP pool, but interest in these compounds has increased following
the use of synthetic phosphonates as herbicides, insecticides, detergent additives, and other
applications. Tens of thousands of tonnes of these long-lived compounds are released each
year into the environment. Analysis of metagenomic databases for phosphonate utilization
genes (phn) has revealed that they are widespread and abundant among diverse bacterial
phyla that are common in marine bacterioplankton, including members of the Proteobacteria,
Planctomycetes, and Cyanobacteria. In situ studies have shown that expression of these genes
and phosphonate degradation occurs in phosphate-limited regions. Since up to 25% of DOP
may occur as phosphonates, these molecules appear to be a significant phosphorus source for
marine microbes.

THE SILICON CYCLE

Silicon in one of the most abundant elements on Earth. Calcium and magnesium silicate
minerals form the major component of rocks in the Earth’s crust. Weathering of these min-
erals by dissolved CO2 results in the formation of dissolved silicate (DSi, i.e. ortho-silicic
acid, H4SiO4) which is transported through soil and enters the ocean via rivers. Other
inputs include weathering of the seafloor and hydrothermal vents. DSi plays a major role
in the growth of phytoplankton and the export of biogenic Si (BSi) carbon in coastal zones
and deep-sea sediments via the biological pump. A range of marine microbes incorporate
DSi across their cell membrane and control its precipitation in specific patterns as particu-
late hydrated silica (SiO2) glass or opal-A, forming a structural component of their cell
walls. The most important siliceous organisms that require silicon as an essential nutrient
are the diatoms and radiolarians, but some flagellates, choanoflagellates, and cyanobacteria
(as well as non-microbial organisms such as sponges and gorgonian corals) also incorpo-
rate significant quantities of silica. The discussion here focuses on the role of diatoms in
silicon cycling.

Silicification of diatoms is an economic

process for construction of a cell wall
Diatom species differ greatly in their level of silicification of the frustules (p.181). Some spe-
cies, such as Fragilaria, have very thick, strong shells that offer resistance to cracking and
crushing by the mouthparts of mesozooplankton during grazing, meaning that viable cells
may pass through the gut undigested. Other species, such as Thalassiosira, have thin shells
that are easily broken and are subject to heavy grazing pressure, but this can be compensated
by rapid proliferation under favorable light and nutrient conditions. However, there is limited
direct evidence that silicification evolved as an anti-predatory function. Silicification is a
relatively energy-efficient process, and diatoms need to expend a much smaller proportion
of their energy intake making a silica wall than one composed of polysaccharides. Diatoms
that grow slowly under non-limiting silicon conditions have plenty of time to take up silicon
during the reproductive cycle, resulting in thicker cell walls; by contrast, rapidly growing
diatoms are likely to have thinner walls.
254 Chapter 9

WHY DO
Diatom blooms depend on the availability
? SYNECHOCOCCUS
CELLS CONTAIN
of silica in the environment
Large blooms of diatoms typically occur in spring and summer at high latitudes and in
SILICON? eutrophic regions. The average Si:N ratio of diatom cells is ~1:4, so silica is often the
Baines et al. (2012) made the limiting nutrient. Actively growing diatoms remain in the photic zone due to buoyancy
surprising discovery that this cya- through gas vacuoles, but when conditions become unfavourable, they sink rapidly from
nobacterium contains Si at intracel- the surface waters due to aggregation into large masses through the production of muci-
lular levels ~2 x 105 times higher lage or the formation of heavy resting spores. Therefore, there is little recycling of silica
than the surrounding seawater in the photic zone and it becomes exhausted before nitrogen or phosphorus, thus ending
and similar to those in diatoms
the diatom bloom and providing other phytoplankton groups with a competitive advan-
from the same locations. They
tage to proliferate using the remaining nutrients. Areas characterized by intense spring
calculated that the water column
inventory of Si in Synechococcus blooms typically have low Si:N ratios (Figure 9.11). Most sinking cells are transported
could exceed that in diatoms. Tang to the seafloor and will be buried in sediments via the biological pump, but refuge popu-
et al. (2014) found that decompo- lations remain in the water column or the surface of sediments and will seed the next
sition of Synechococcus produces seasonal bloom via vertical mixing. As noted on p.182, bacterial-mediated degradation
“microblebs” of silica, which of the organic matrix covering diatom shells affects the rate of dissolution of BSi and its
could contribute to ballasting and return to the DSi pool. In some regions, a second short bloom can occur in the autumn,
export of picophytoplankton- caused by the breakdown of summer stratification, providing vertical mixing of nutrients
derived POC from the photic zone. while light levels are still sufficiently high to support growth. These bloom cycles mean
Synechococcus possesses genes sim-
that diatoms play a major role in the export of POC to the seafloor via the biological
ilar to the silicon transporter (SIT)
pump—about 40% of oceanic carbon sequestration (~1.5–2.8 Gt (Tg) y-1) can be attrib-
genes of diatoms (Marron et al.,
2016). Based on molecular clock
uted to diatoms. Blooms of diatoms have a significant difference from those of the other
analysis of amino acid substitu- major bloom-forming group, the coccolithophores. Because CO2 is released during the
tions, Conley et al. (2017) suggest formation of calcite coccoliths (p.180), coccolithophore dominance leads to reduction in
that these transporters may have the sequestration of CO2 from the atmosphere to the ocean floor, whereas diatom blooms
evolved >2 BYA in cyanobacteria, increase carbon flux.
at a time when ocean DSi levels
were close to saturation. They
would be essential to transport Si Eutrophication alters the silicon balance in coastal zones
out of the cells to prevent precipi-
tation in the cytoplasm. Most bac- As with other biogeochemical cycles, anthropogenic activities are having a major impact on
teria (and eukaryotes apart from the silica cycle. As noted in earlier sections of this chapter, input of nitrogen and phosphorus
silicifiers) appear to have lost the into rivers and coastal zones from terrestrial runoff of fertilizers has increased greatly, lead-
transporter genes as they became ing to stimulation of algal growth. The construction of dams also leads to longer residence
an unnecessary burden during time in rivers, which results in trapping of DSi and reduced export to the sea. The increased
transition to the low DSi concen- N:Si or P:Si ratios mean that diatom growth is limited, whereas non-siliceous phytoplankton
tration in the modern ocean. Why such as flagellates can flourish. This alters estuarine and coastal food webs, because diatoms
Synechococcus have retained them are replaced by organisms that have less nutritional value or are less acceptable as prey for
is a mystery.
zooplankton. This has a major impact on the productivity of fisheries. Furthermore, blooms
of non-diatom species increasingly lead to harmful algal blooms and anoxic dead zones dis-
cussed above.

Figure 9.11 Monthly mean maps

of global distributions of silica and
nitrogen and distribution of main
phytoplankton groups. Upper panels
show ambient Si: N ratios in June
and December: values <2 (yellow)
represent areas where dissolved silica
becomes depleted before inorganic
nitrogen; values >2 (green) represent
areas where inorganic nitrogen is a
growth-limiting nutrient. Lower pan-
els show the prevailing phytoplank-
ton groups in June and December.
White regions indicate areas with
no data. Reprinted from Pančić and
Kiørboe (2018) with permission by
John Wiley & Sons.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 255

Conclusions
This chapter has illustrated how recent discoveries have led to dramatic shifts in our think-
ing about the role of microbes in the major nutrient cycles and their consequences for global
ocean processes. It is important that we have a clear understanding of underlying mecha-
nisms because of the increased effects of anthropogenic inputs on the composition of our
atmosphere and oceans. Increasing concentrations of CO2 and sulfur from the use of fossil
fuel and industrial activity, inputs of nitrogen and phosphorus nutrients from agriculture
and aquaculture, as well as shifts in silica levels due to deforestation and alteration of river
flows and groundwater are all causing shifts in microbial community composition and alter-
ing the balance of marine microbial nutrient cycles. Indeed, major shifts in the nitrogen
and sulfur nutrient cycles might catalyze further climatic changes in a runaway effect, for
example, through increased output of greenhouse gases like nitrous oxide or depletion of the
albedo effect due to changes in DMS emissions. However, predicting the outcomes of these
changes is very difficult unless we have a clear understanding of underlying processes. In the
last couple of decades, modern techniques have led to the discovery of previously unknown
processes and major new surprises are currently appearing every year. As seen in other chap-
ters, exploring these cycles has also revealed the variability and patchiness of the chemical
and physical properties of the oceans, and the consequent huge diversity and adaptability of
microbes. It is becoming increasingly difficult to categorize microbial taxa according to pos-
session of a particular metabolic function, such as nitrifiers, denitrifiers, aerobes, anaerobes
etc., because there are now so many examples of microbial clades crossing such functional
boundaries. Organisms that appear closely related through their core genome form a diverse
federation due to the presence of variants that possess genes for pathways that allow them to
succeed in particular niches with very different physical and chemical conditions. Networks
of competition and cooperation in the utilization of key nutrients add further layers of com-
plexity to the microbial ecology of the oceans.

Shoenfelt, E.M., Sun, J., Winckler, G. et al. (2017) High particulate iron(II)
References and further reading content in glacially sourced dusts enhances productivity of a model dia-
Iron, ocean fertilization, and other geoengineering tom. Sci. Adv. 3: e1700314.

Bhatia, M.P., Kujawinski, E.B., Das, S.B. et al. (2013) Greenland meltwa- Smetacek, V. (2018) Seeing is believing: Diatoms and the ocean carbon
ter as a significant and potentially bioavailable source of iron to the ocean. cycle revisited. Protist 169: 791–802.
Nat. Geosci. 6: 274–278. Smetacek, V., Klaas, C., Strass, V.H. et al. (2012) Deep carbon export from
Bonnain, C., Breitbart, M., & Buck, K.N. (2016) The Ferrojan Horse a Southern Ocean iron-fertilized diatom bloom. Nature 487: 313–319.
Hypothesis: Iron-virus interactions in the ocean. Front. Mar. Sci. 3: 82. Smetacek, V. & Naqvi, S.W. (2008) The next generation of iron fertiliza-
Boyd, P.W. & Vivian, C.M.G. (eds.) (2019a) High level review of a tion experiments in the Southern Ocean. Philos. Trans. Roy. Soc. A 366:
wide range of proposed marine geoengineering techniques. Rep. Stud. 3957–3967.
GESAMP 98: 144. Strong, A., Chisholm, S., Miller, C., & Cullen, J. (2009) Ocean fertiliza-
Boyd, P.W. & Vivian, C.M.G. (2019b) Should we fertilize oceans or seed tion: Time to move on. Nature 461: 347–348.
clouds? No one knows. Nature 570: 155. Tollefson, J. (2017) Iron-dumping ocean experiment sparks controversy.
National Research Council (2015) Climate Intervention: Carbon Dioxide Nature 545: 393–394.
Removal and Reliable Sequestration. National Academies Press, Trick, C.G., Bill, B.D., Cochlan, W.P. et al. (2010) Iron enrichment stimu-
Washington, DC. lates toxic diatom production in high-nitrate, low-chlorophyll areas. Proc.
Ratnarajah, L., Nicol, S., & Bowie, A.R. (2018) Pelagic iron recycling in Natl. Acad. Sci. 107: 5887–5882.
the Southern Ocean: Exploring the contribution of marine animals. Front. Tzubari, Y., Magnezi, L., Be’er, A., & Berman-Frank, I. (2018) Iron and
Mar. Sci. 5: 109. phosphorus deprivation induce sociality in the marine bloom-forming
Roman, J., Estes, J.A., Morissette, L. et al. (2014) Whales as marine eco- cyanobacterium Trichodesmium. ISME J. 12: 1682–1693.
system engineers. Front. Ecol. Environ. 12: 377–385. Watson, A.J., Boyd, P.W., Turner, S.M. et al. (2008) Designing the next
Roman, J. & McCarthy, J.J. (2010) The whale pump: Marine mammals generation of ocean iron fertilization experiments. Mar. Ecol. Prog. Ser.
enhance primary productivity in a coastal basin. PLoS One 5: e13255. 364: 303–309.

Royal Society (2009) Geoengineering the climate: Science, governance Weier, J. (2010) On the Shoulders of Giants (Obituary, John Martin).
and uncertainty. Online at https://royalsociety.org/topics-policy/publica- NASA Earth Observatory, online at https://earthobservatory.nasa.gov/fea-
tions/2009/geoengineering-climate/ (accessed 5 July 2019). tures/Martin/martin_4.php (accessed 13 Sep 2019).

Rubin, M., Berman-Frank, I., & Shaked, Y. (2011) Dust-and mineral-iron Yoon, J.-E., Yoo, K.-C., Macdonald, A.M. et al. (2018) Reviews and syn-
utilization by the marine dinitrogen-fixer Trichodesmium. Nat. Geosci. theses: Ocean iron fertilization experiments—Past, present, and future
4: 529. looking to a future Korean Iron Fertilization Experiment in the Southern
Ocean (KIFES) project. Biogeoscience 15: 5847–5889.
256 Chapter 9

Nitrogen cycle, oxygen minimum zones Muñoz-Marín, M.D.C., Shilova, I.N., Shi, T. et al. (2019) The transcrip-
tional cycle is suited to daytime N2 fixation in the unicellular cyanobacte-
Aumont, O., Maury, O., Lefort, S., & Bopp, L. (2018) Evaluating the rium “Candidatus Atelocyanobacterium thalassa” (UCYN-A). MBio 10:
potential impacts of the diurnal vertical migration by marine organisms e02495-18.
on marine biogeochemistry. Glob. Biogeochem. Cycles 32: 1622–1643.
Otero, X.L., De La Peña-Lastra, S., Pérez-Alberti, A. et al. (2018) Seabird
Benavides, M., Bonnet, S., Berman-Frank, I., & Riemann, L. (2018) Deep colonies as important global drivers in the nitrogen and phosphorus cycles.
into oceanic N2 fixation. Front. Mar. Sci. 5: 108. Nat. Commun. 9: 246.
Bertagnolli, A.D. & Stewart, F.J. (2018) Microbial niches in marine oxy- Pedersen, J.N., Bombar, D., Paerl, R.W., & Riemann, L. (2018) Diazotrophs
gen minimum zones. Nat. Rev. Microbiol. 16: 723–729. and N2-fixation associated with particles in coastal estuarine waters.
Berube, P.M., Biller, S.J., Kent, A.G. et al. (2015) Physiology and evolution Front. Microbiol. 9: 2759.
of nitrate acquisition in Prochlorococcus. ISME J. 9: 1195–1207. Shiozaki, T., Ijichi, M., Isobe, K. et al. (2016) Nitrification and its influence
Bianchi, D., Babbin, A.R., & Galbraith, E.D. (2014) Enhancement of on biogeochemical cycles from the equatorial Pacific to the Arctic Ocean.
anammox by the excretion of diel vertical migrators. Proc. Natl. Acad. ISME J. 10: 2184–2197.
Sci. USA 111: 15653–15658. Sun, X., Kop, L.F.M., Lau, M.C.Y. et al. (2019) Uncultured Nitrospina-like
Breitburg, D., Levin, L.A., Oschlies, A. et al. (2018) Declining oxygen in species are major nitrite oxidizing bacteria in oxygen minimum zones.
the global ocean and coastal waters. Science 359: eaam7240. ISME J. 13: 2184–2197.
Caputo, A., Stenegren, M., Pernice, M.C., & Foster, R.A. (2018) A short Ulloa, O., Canfield, D.E., DeLong, E.F. et al. (2012) Microbial ocean-
comparison of two marine planktonic diazotrophic symbioses highlights ography of anoxic oxygen minimum zones. Proc. Natl. Acad. Sci. 109:
an un-quantified disparity. Front. Mar. Sci. 5: 2. 15996–16003.
Chan, F., Barth, J.A., Lubchenco, J. et al. (2008) Emergence of anoxia Walworth, N.G., Fu, F.-X., Lee, M.D. et al. (2018) Nutrient-colimited
in the California current large marine ecosystem. Science 319: 920–920. Trichodesmium as a nitrogen source or sink in a future ocean. Appl.
Environ. Microbiol. 84: e02137-17.
Dalsgaard, T., Thamdrup, B., Farías, L., & Revsbech, N.P. (2012)
Anammox and denitrification in the oxygen minimum zone of the eastern Wannicke, N., Frey, C., Law, C.S., & Voss, M. (2018) The response of
South Pacific. Limnol. Oceanogr. 57: 1331–1346. the marine nitrogen cycle to ocean acidification. Glob. Change Biol. 24:
5031–5043.
Delmont, T.O., Quince, C., Shaiber, A. et al. (2018) Nitrogen-fixing popu-
lations of Planctomycetes and Proteobacteria are abundant in surface Zehr, J.P. (2015) How single cells work together. Science 349: 1163–1164.
ocean metagenomes. Nat. Microbiol. 3: 804. Zehr, J.P. & Kudela, R.M. (2011) Nitrogen cycle of the open ocean: From
Farnelid, H., Turk-Kubo, K., Muñoz-Marín, M., & Zehr, J. (2016) New genes to ecosystems. Ann. Rev. Mar. Sci. 3: 197–225.
insights into the ecology of the globally significant uncultured nitrogen- Zehr, J.P., Shilova, I.N., Farnelid, H.M. et al. (2017) Unusual marine uni-
fixing symbiont UCYN-A. Aquat. Microb. Ecol. 77: 125–138. cellular symbiosis with the nitrogen-fixing cyanobacterium UCYN-A.
Foshtomi, M.Y., Braeckman, U., Derycke, S. et al. (2015) The link between Nat. Microbiol. 2: 16214.
microbial diversity and nitrogen cycling in marine sediments is modulated Zerkle, A.L. & Mikhail, S. (2017) The geobiological nitrogen cycle: From
by macrofaunal bioturbation. PLoS One 10: e0130116. microbes to the mantle. Geobiology 15: 343–352.
Gruber, N. (2016) Elusive marine nitrogen fixation. Proc. Natl. Acad. Sci.
USA 113: 4246–4248. Sulfur, DMSP, climate regulation
Gruber, N. (2019) Consistent patterns of nitrogen fixation identified in the Alcolombri, U., Ben-Dor, S., Feldmesser, E. et al. (2015) Identification of
ocean. Nature 566: 191–193. the algal dimethyl sulfide-releasing enzyme: A missing link in the marine
Hagino, K., Onuma, R., Kawachi, M., & Horiguchi, T. (2013) Discovery sulfur cycle. Science 348: 1466–1469.
of an endosymbiotic nitrogen-fixing cyanobacterium UCYN-A in Amato, P., Joly, M., Besaury, L. et al. (2017) Active microorganisms thrive
Braarudosphaera bigelowii (Prymnesiophyceae). PLoS One 8: e81749. among extremely diverse communities in cloud water. PLoS One 12:
Harding, K., Turk-Kubo, K.A., Sipler, R.E. et al. (2018) Symbiotic unicel- e0182869.
lular cyanobacteria fix nitrogen in the Arctic Ocean. Proc. Natl. Acad. Sci. Barak-Gavish, N., Frada, M.J., Ku, C. et al. (2018) Bacterial virulence
USA 115: 13371–13375. against an oceanic bloom-forming phytoplankter is mediated by algal
Henke, B.A., Turk-Kubo, K.A., Bonnet, S., & Zehr, J.P. (2018) DMSP. Sci. Adv. 4: eaau5716.
Distributions and abundances of sublineages of the N2-fixing cyanobacte- Bourne, D., Iida, Y., Uthicke, S., & Smith-Keune, C. (2007) Changes in
rium “Candidatus Atelocyanobacterium thalassa” (UCYN-A) in the New coral-associated microbial communities during a bleaching event. ISME
Caledonian coral lagoon. Front. Microbiol. 9: 554. J. 2: 350–363.
Kuypers, M.M.M., Marchant, H.K., & Kartal, B. (2018) The microbial Bullock, H.A., Luo, H., & Whitman, W.B. (2017) Evolution of dimeth-
nitrogen-cycling network. Nat. Rev. Microbiol. 16: 263–276. ylsulfoniopropionate metabolism in marine phytoplankton and bacteria.
Laverock, B., Gilbert, J.A., Tait, K. et al. (2011) Bioturbation: Impact on Front. Microbiol. 8: 637.
the marine nitrogen cycle. Biochem. Soc. Trans. 39: 315–320. Charlson, R.J., Lovelock, J.E., Andreae, M.O., & Warren, S.G. (1987)
Laverock, B., Smith, C.J., Tait, K. et al. (2010) Bioturbating shrimp alter Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate.
the structure and diversity of bacterial communities in coastal marine Nature 326: 655–661.
sediments. ISME J. 4: 1531–1544. Curson, A.R.J., Liu, J., Bermejo Martínez, A. et al. (2017)
Martiny, A.C., Kathuria, S., & Berube, P.M. (2009) Widespread metabolic Dimethylsulfoniopropionate biosynthesis in marine bacteria and identifi-
potential for nitrite and nitrate assimilation among Prochlorococcus eco- cation of the key gene in this process. Nat. Microbiol. 2: 17009.
types. Proc. Natl. Acad. Sci. USA 106: 10787–10792. Curson, A.R.J., Williams, B.T., Pinchbeck, B.J. et al. (2018) DSYB cataly-
Messer, L.F., Mahaffey, C., Robinson, C. et al. (2016) High levels of het- ses the key step of dimethylsulfoniopropionate biosynthesis in many phy-
erogeneity in diazotroph diversity and activity within a putative hotspot toplankton. Nat. Microbiol. 3: 430–439.
for marine nitrogen fixation. ISME J. 10: 1499–1513.
 Ocean Processes—Nitrogen, Sulfur, Iron, Phosphorus, and Silicon Cycling 257

Dell’Ariccia, G., Phillips, R.A., Van Franeker, J.A. et al. (2017) Comment Seymour, J.R., Simó, R., Ahmed, T., & Stocker, R. (2010) Chemoattraction
on “Marine plastic debris emits a keystone infochemical for olfactory for- to dimethylsulfoniopropionate throughout the marine microbial food web.
aging seabirds” by Savoca et al. Sci. Adv. 3: e1700526. Science 329: 342–345.
Evans, C., Malin, G., Wilson, W.H., & Liss, P.S. (2006) Infectious titers Vila-Costa, M., Pinhassi, J., Alonso, C. et al. (2007) An annual cycle
of Emiliania huxleyi virus 86 are reduced by exposure to millimolar of dimethylsulfoniopropionate-sulfur and leucine assimilating bacte-
dimethyl sulfide and acrylic acid. Limnol. Oceanogr. 51: 2468–2471. rioplankton in the coastal NW Mediterranean. Environ. Microbiol. 9:
2451–2463.
Falkowski, P.G., Katz, M.E., Knoll, A.H. et al. (2004) The evolution of
modern eukaryotic phytoplankton. Science 305: 354–360. Wang, R., Gallant, É., & Seyedsayamdost, M.R. (2016) Investigation of the
genetics and biochemistry of roseobacticide production in the Roseobacter
Howard, E.C., Henriksen, J.R., Buchan, A. et al. (2006) Bacterial taxa that
clade bacterium Phaeobacter inhibens. MBio 7: e02118.
limit sulfur flux from the ocean. Science 314: 649–652.
Yoch, D.C. (2002) Dimethylsulfoniopropionate: Its sources, role in the
Jones, G., Curran, M., Deschaseaux, E. et al. (2018) The flux and emis-
marine food web, and biological degradation to dimethylsulfide. Appl.
sion of dimethylsulfide from the Great Barrier Reef region and potential
Environ. Microbiol. 68: 5804–5815.
influence on the climate of NE Australia. J. Geophys. Res. Atmos. 123:
13835–13856.
Phosphorus
Lenton, T.M., Daines, S.J., Dyke, J.G. et al. (2018) Selection for Gaia
across multiple scales. Trends Ecol. Evol. 33: 633–645. Dyhrman, S., Ammerman, J., & Van Mooy, B. (2007) Microbes and the
marine phosphorus cycle. Oceanography 20: 110–116.
Lenton, T.M. & Latour, B. (2018) Gaia 2.0. Science 361: 1066–1068.
Dyhrman, S.T. (2018) Microbial physiological ecology of the marine phos-
Lidbury, I., Kröber, E., Zhang, Z. et al. (2016) A mechanism for bacterial
phorus cycle. In Microbial Ecology of the Oceans, 3rd edition (Gasol, J.M
transformation of dimethylsulfide to dimethylsulfoxide: A missing link in
& Kirchman, D.L., eds.), Wiley-Blackwell, pp. 377–434.
the marine organic sulfur cycle. Environ. Microbiol. 18: 2754–2766.
White, A. & Dyhrman, S. (2013) The marine phosphorus cycle. Front.
Moran, M.A., Reisch, C.R., Kiene, R.P., & Whitman, W.B. (2012)
Microbiol. 4: 105.
Genomic insights into bacterial DMSP transformations. Ann. Rev. Mar.
Sci. 4: 523–542.
Silicon
Pančić, M. & Kiørboe, T. (2018) Phytoplankton defence mechanisms:
Traits and trade-offs. Biol. Rev. 93: 1269–1303. Baines, S.B., Twining, B.S., Brzezinski, M.A. et al. (2012) Significant sili-
con accumulation by marine picocyanobacteria. Nat. Geosci. 5: 886–891.
Quinn, P.K. & Bates, T.S. (2011) The case against climate regulation via
oceanic phytoplankton sulphur emissions. Nature 480: 51–56. Conley, D.J., Frings, P.J., Fontorbe, G. et al. (2017) Biosilicification drives
a decline of dissolved Si in the oceans through geologic time. Front. Mar.
Raina, J.B., Dinsdale, E.A., Willis, B.L., & Bourne, D.G. (2010) Do the Sci. 4: 397.
organic sulfur compounds DMSP and DMS drive coral microbial associa-
tions? Trends Microbiol.18:101–108. Struyf, E., Smis, A., Van Damme, S. et al. (2009) The global biogeochemi-
cal silicon cycle. Silicon 1: 207–213.
Raina, J.-B., Tapiolas, D.M., Forêt, S. et al. (2013) DMSP biosynthesis
by an animal and its role in coral thermal stress response. Nature 502: Tang, T., Kisslinger, K., & Lee, C. (2014) Silicate deposition during
677–680. decomposition of cyanobacteria may promote export of picophytoplank-
ton to the deep ocean. Nat. Commun. 5: 4143.
Raina, J.-B.B., Tapiolas, D., Willis, B.L., & Bourne, D.G. (2009) Coral-
associated bacteria and their role in the biogeochemical cycling of sulfur. Tréguer, P.J. & De La Rocha, C.L. (2013) The world ocean silica cycle.
Appl. Environ. Microbiol. 75: 3482–3501. Annu. Rev. Mar. Sci. 5: 477–501.
Reisch, C.R., Moran, M.A., & Whitman, W.B. (2008) Dimethy
Other articles
lsulfoniopropionate-dependent demethylase (DmdA) from Pelagibacter
ubique and Silicibacter pomeroyi. J. Bacteriol. 190: 8018–8024. Arrigo, K.R. (2005) Marine microorganisms and global nutrient cycles.
Savoca, M.S. & Nevitt, G.A. (2014) Evidence that dimethyl sulfide facili- Nature 437: 349–355.
tates a tritrophic mutualism between marine primary producers and top Bristow, L.A., Mohr, W., Ahmerkamp, S., & Kuypers, M.M.M. (2017)
predators. Proc. Natl. Acad. Sci. USA 111: 4157–4161. Nutrients that limit growth in the ocean. Curr. Biol. 27: R474–R478.
Savoca, M.S., Wohlfeil, M.E., Ebeler, S.E., & Nevitt, G.A. (2016) Marine Hutchins, D.A. & Fu, F. (2017) Microorganisms and ocean global change.
plastic debris emits a keystone infochemical for olfactory foraging sea- Nat. Microbiol. 2: 17058.
birds. Sci. Adv. 2: e1600395.
Richardson, T.L. (2019) Mechanisms and pathways of small-phytoplank-
Schiffer, J.M., Mael, L.E., Prather, K.A. et al. (2018) Sea spray aerosol: ton export from the surface ocean. Annu. Rev. Mar. Sci. 11: 57–74.
Where marine biology meets atmospheric chemistry. ACS Central Sci. 4:
Schneider, S.H. (2004) Scientists Debate Gaia: The Next Century, MIT Press.
1617–1623.
Chapter 10
Microbial Symbioses
of Marine Animals

In the broadest definition, the term symbiosis can be used to describe any close, long-term
relationship between two different organisms, which can range from commensalism (a loose
association in which one partner gains benefit but does no harm to the host) to parasitism (in
which the parasite benefits at the expense of the host). However, in common usage, the term
symbiosis often refers to a mutualistic relationship, in which both partners benefit through
increased fitness in evolution. We now know that many marine animals obtain their food
directly from the products of photosynthetic or chemosynthetic bacteria living in or on their
bodies. Other microbial symbionts may provide their host with behavioral or reproductive
benefits. These interactions are very widespread and have had a major influence on the evo-
lution of host animals and their associated microbes. This chapter provides examples of
mutualistic interactions with multicellular animals that illustrate their importance in marine
ecology. (Symbioses also occur between different types of microbes—examples of these
are considered in Chapters 4–6.) In addition, we now know that all animals contain diverse
microbes in or on their bodies (the microbiome) and this may have less obvious influences
that we are just beginning to understand. In particular, disruption of the balanced microbi-
ome can lead to disease, and these effects are discussed in Chapter 11.

Key Concepts
• A wide range of marine invertebrates obtain some or all their nutrition from chemo-
synthetic or photosynthetic microbes, which may occur as intracellular endosymbionts
within the tissues or as epibionts on internal or external surfaces.
• Microbial symbioses are fundamental to the function of specific animals as ecosystem
engineers.
• The behavior of some fish and invertebrates depends on bioluminescence produced by
bacterial symbionts that colonize specialized light organs.
• Symbiotic hosts have evolved complex adaptations of recognition, tissue development,
and internal structure in order to maximize the benefits of the relationship.
• Most symbioses in marine animals depend on horizontal transmission, in which free-
living microbes are acquired from the environment by the larvae. In a few cases, sym-
bionts are transmitted vertically from the parent host; this usually leads to an obligate
relationship in which the symbiont undergoes genome reduction.
• Environmental stress often leads to breakdown of the symbiotic relationship.
• Study of marine symbioses has resulted in the recognition that animals and their resi-
dent microbiomes should be regarded as holobionts or meta-organisms. This has led to
models for understanding animal development and immunity, with important applica-
tions in biotechnology and human health.
260 Chapter 10

DIFFERENT TERMS
Symbioses occur in many forms
i FOR THE LOCATION
OF SYMBIONTS
The concept of symbiosis between different organisms developed at the end of the nineteenth
century and had a profound influence on the development of biology and evolutionary theory.
We can consider symbioses in The term symbiosis is derived from the Greek for “living together,” and encompasses a broad
terms of the degree of “intimacy” range of interactions. Our usual definition of symbiosis depends on the concept of mutual
of the association. Some microbes benefit that the partners receive—but benefit can be very hard to assess and is often rather
are intracellular organisms with one-sided. This creates a spectrum of possible interactions. In several of the examples dis-
special adaptations enabling them cussed in this chapter, we will see that the microbial symbionts have been shown to have a
to live inside the host cells—these free-living stage. Therefore, this can be described as a facultative relationship, because the
are endosymbionts (from endo-, symbiosis is not obligate for the microbes as they can clearly live independently. However,
meaning inside). However, this some hosts are completely dependent on their microbes’ activities for their growth and devel-
term is often used to describe opment beyond the initial stages of their life cycle, so for them the relationship is obligate. In
extracellular microbes living inside
other cases, which are less common, the co-evolution of the host and symbiont has led to the
the animal body on the surface of
the internal ducts or body cavities,
symbiont losing the potential for independent existence and the partners are obligately inter-
such as the intestinal tract. These dependent throughout their life cycles. Advances in DNA sequencing and high-resolution
are better described as episymbi- imaging technologies are leading to great insights into the complexities of microbe-animal
onts (from epi-, meaning upon) symbioses. We now recognize that the development, function, health, and evolutionary fit-
to indicate that they are living ness of animals is hugely influenced by microbes, which have been present throughout the
on the surface of the host cells. history of life on Earth. This has led to the realization that we should no longer think about
Episymbionts are also commonly hosts and their microbial partners separately as independent units, but rather as an integrated
found on the external surface of whole—a “holobiont” or “meta-organism.” The study of symbioses in marine invertebrates
many animals or algae, in which provides invaluable models that underpin the holobiont concept and are influencing our
case they can be described as ecto-
understanding of animal development and evolution and the effects of environmental pres-
symbionts (from ecto-, meaning
outside).
sures. Furthermore, some of these models have direct applications for studies relevant to the
advancement of human health.

Chemosynthetic bacterial endosymbionts of

animals were discovered at hydrothermal vents
Chapter 1 introduced the dense undersea communities of giant tubeworms, bivalve molluscs,
and other creatures growing around deep-sea hydrothermal vents (Figure 1.15B), whose dis-
covery at the Galapagos Rift in the Pacific Ocean in 1977 was such a great surprise. How can
such a highly productive ecosystem be sustained? Vent habitats are completely dark, under
enormous pressure, and at such great depth that organic material from the upper layers of
water settles too slowly to support this level of growth. At first, investigators thought that the
vent animals might feed by filtration of planktonic chemosynthetic bacteria that exist in the
DIFFERENT
i TERMS FOR
CHEMOSYNTHESIS
water around the vents, concentrated by local warm-water currents created by the vent activ-
ity. This idea became untenable when equipment was developed for bringing intact animals
to the surface. It was then discovered that the filter-feeding mechanism and digestive tract
Chemosynthesis is a widely used of the giant white clams (named as Calyptogena magnifica, family Vesicomyidae) found at
general term to describe the the vents was greatly reduced. Investigation of the giant tubeworm Riftia pachyptila, family
formation of organic compounds Siboglinidae) showed that it lacked a mouth and gut, so it was assumed at first that it must
which includes chemolithoau- feed by simply absorbing nutrients from the surrounding water. However, the rate of uptake
totrophy and methanotrophy.
of organic material is insufficient to explain the extremely high metabolic rates of these
Chemolithoautotrophs use CO2
as the sole carbon source, using
animals. Thus, it was realized that the nutrition of these animals and the flow of energy
energy derived from the oxida- through the vent food web depends directly on chemosynthetic bacteria within the tissues of
tion of reduced inorganic sub- the animals.
strates such as sulfide, hydrogen,
or ammonium. Bacteria which Various lines of evidence supported this conclusion. First, stable isotope analysis (p.58) of
oxidize hydrogen sulfide, thiosul- the tissue of the clam C. magnifica showed their cellular material does not have its ultimate
phate, and other reduced-sulfur origin in photosynthesis. The ratios of 13C to 12C reflect the efficiency with which different
compounds are also commonly enzymes deal with the different isotopes. The ratio found in these animals is well outside
described as thiotrophic (although the value of photosynthetically fixed carbon, proving that they are not feeding on material
many of these bacteria can also use
derived from CO2 fixed in the upper photic zone.
hydrogen). Methanotrophs oxidize
methane, which acts as a source of
reducing power and as the source When specimens of the tubeworm R. pachyptila were dissected, it was found that the trunk
of carbon. The reactions involved is filled with an organ supplied with many blood vessels (Figure 10.1a). Electron micros-
are described in Chapter 3. copy of this structure (the trophosome) showed that it contained granules of elemental sulfur
 Microbial Symbioses of Marine Animals 261

and large numbers of structures that strongly resembled bacterial cells. The trophosome
tissue was also found to contain large amount of lipopolysaccharide (LPS), indicating that
it is packed with endosymbiotic Gram-negative bacteria. These are contained within host
cells called bacteriocytes. Further biochemical analysis showed the presence of enzymes
associated with sulfur metabolism, including ATP-reductase and ATP-sulfurylase and high
levels of RuBisCO (p.83), providing further proof of endogenous autotrophic metabolism.
Subsequent investigations showed that the products of chemosynthesis are transferred across
the host membrane surrounding the bacteria to the animal tissue for growth. This is a truly
remarkable animal-bacterium symbiosis; there are about 109 bacteria per gram of tissue,
occupying up to a quarter of the volume of the trophosome. Riftia has evolved into a truly
autotrophic animal, with sophisticated physiological mechanisms to ensure the efficient
transport of oxygen and sulfide to the symbionts. The endosymbionts were soon identified by
16S rRNA gene sequences as members of the Gammaproteobacteria and free-living bacteria
corresponding to the symbiont phylotype can be detected in seawater using FISH with 16S
rRNA gene-targeted probes, as far as 100 m from tube worm colonies. Although all attempts
to culture the Riftia symbiont have failed, the high densities of an apparently pure popula-
tion of bacteria within the trophosome has enabled separation of host and symbiont cells and
construction of a composite metagenome. The findings explain why the bacterium is suc-
cessful in the two very different environments, living autotrophically within host cells and
heterotrophically when free-living, as shown in Figure 10.1b. This dual nature is recognized
in the species name given to the symbiont—“Ca. Endoriftia persephone” (from Persephone
in Greek mythology, who was the goddess of fertility as well as queen of the underworld).
Further insights into the acquisition and dissemination of the symbionts are described in
Box 10.1.

In the clam C. magnifica, microscopy revealed that the symbiotic bacteria are contained in
Figure 10.1 Schematic diagrams
large cells of the gills that are exposed to the seawater on one side and the blood supply on
illustrating symbiosis between the
the other, although the anatomical adaptations to ensure efficient transport of nutrients to the
tube worm Riftia pachyptila and
symbionts are not as extensive as those of Riftia. Once again, enzyme assays show the thio-
its endosymbiotic bacterium “Ca.
autotrophic nature of the bacteria. The symbionts of Calyptogena spp. are transmitted verti- Endoriftia persephone.” (a) An exten-
cally. Intuitively, one might expect this to be a common mechanism to ensure that the bacteria sive capillary system transports O2,
are passed on directly to their offspring, but this is one of only a few examples currently H2S, and CO2 bound to hemoglobin
known in marine chemosynthetic invertebrates. The genomes of Calyptogena symbionts are (Hb) to the trophosome where host
very small—just 1.0–1.2 Mb, encoding only about 1000 genes. Genome analysis shows that cells (bacteriocytes) are packed with
the major chemoautotrophic pathways are present, as well as routes for the production of endosymbiotic bacteria. (b) Within
vitamins, co-factors, and amino acids required by the clam. However, many of the genes the bacteriocytes, endosymbionts
oxidize sulfide via the APS pathway
(see Figure 3.3), yielding ATP and
NADPH via an electron transport
system (ETS). CO2 fixation occurs
via the CBB and reductive tricarbox-
ylic acid (r-TCA) cycles (see Figure
3.4). Organic compounds made
by symbionts pass to the host via
translocation of simple organic com-
pounds and digestion of symbiont
cells (open arrows). Nitrogen for the
symbiont’s biosynthesis is derived
from urea excreted by host cells, or
from reduction of nitrate absorbed
from seawater. (c) In their free-living
state, the bacteria are thought to
live heterotrophically by absorbing
dissolved organic compounds in
seawater. These may be respired via
glycolysis and the oxidative (ox-)TCA
cycle. Free-living bacteria may also
use the rTCA cycle for autotrophic
metabolism. Based on data from
Stewart et al. (2005); Harmer et al.
(2008); Robidart et al. (2008).
262 Chapter 10

BOX 10.1 RESEARCH FOCUS

Genomic insights into host–symbiont interactions:

infection, colonization, and dissemination

As we learn more about symbioses of marine invertebrates through Bacteria that had colonized the host were detected using symbiont-
the application of modern techniques, the interactions show many specific FISH probes (p.36). When they first settle, the larvae have
parallels with the interactions between pathogenic bacteria and their an obvious digestive tract, including a mouth and anus. The bac-
hosts. Two areas of active research in this field are discussed here. teria can be detected when the larvae reach a length of about 250
µm. However, even though the larvae feed actively on microbes,
How do symbiotic bacteria infect their tubeworm hosts? The the FISH micrographs showed that the symbionts enter via the skin
bacterial symbionts found in Riftia pachyptila and related sibogli- (where they become enclosed in vacuoles), not via the mouth. Fewer
nid tubeworms have genomes of similar size (~3.4–4.6 Mb) and than 20 bacteria are needed for infection, but these rapidly migrate
composition to their free-living counterparts (Robidart et al., 2008; through the layers of host tissue, multiply, and initiate differentia-
Li et al., 2018) and there is no evidence of co-speciation with their tion of the mesodermal tissue. Their presence then provokes massive
hosts (McMullin et al., 2003). They are acquired from the environ- apoptosis (programmed cell death) of the epidermis, muscles, and
ment at an early stage of larval development. Hydrothermal vents mesoderm, followed by trophosome development (Nussbaumer et
are very ephemeral, and the success of these tubeworms depends on al., 2006). Further insight into the infection and colonization process
larval dispersal over great distances. Since the adult worms depend has been provided by Li et al. (2018), who compared genomes of the
absolutely on the symbionts for their nutrition, it seems a risky strat- symbionts from several vent and seep-dwelling tubeworms. In all
egy for the larvae not to contain symbionts. However, horizontal symbionts, they identified a range of genes that have previously been
transmission may offer more opportunities for the host to acquire associated with infection processes in pathogenic bacteria; these
specific types of bacteria that might be best adapted to local condi- include adhesion, secretion systems, virulence proteins, and over-
tions and this is thought to have played a major role during evolu- coming oxidative stress. In addition, the presence of genes for flagel-
tion (Fisher et al., 2017). To investigate how the bacteria colonize lar motility, chemotaxis, and adhesive pili indicate the role of these
the worms, Nussbaumer et al. (2006) constructed specially designed processes in initial contact with the host larvae, which are thought to
chambers with grooved plates that were placed at a hydrothermal secrete a mucus that attracts potential symbionts (Nussbaumer et al.,
vent using the submersible Alvin. A year later, the tiny (0.2–2 mm) 2006). Li et al. (2018) also identified genes encoding cytolytic pro-
and very fragile bodies of newly settled larvae and juvenile animals teins, chitinase, and collagenase, which probably enable the bacteria
of three different species (Riftia pachyptila, Tevnia jerichonana, to migrate through the trophosome tissue during its early stages of
and Oasisia alvinae) were collected and examined microscopically. development.

How do the symbionts seed the envi-

ronment? To ensure colonization of host
larvae by horizontal transmission, there
must be a mechanism by which the sym-
bionts seed the environment. However,
they are contained inside the bacterio-
cytes deep inside the tissue and there
are no openings to the exterior, through
which they might escape. The tropho-
some is a highly structured organ with an
extremely rapid growth rate (comparable
to cancer or wound healing) in which
symbiont cells within the bacteriocytes
go through a coordinated cell cycle. The
density of symbionts is controlled through
apopotosis and digestion at the periph-
ery and replacement by dividing sym-
bionts at the center (Pflugfelder et al.,
2009). To investigate the possibility that
the symbiont Endoriftia is released from
R. pachyptila when it dies, Klose et al.,
2015 developed high pressure incubation
chambers in which trophosome tissue,
Figure 10.2 Symbiont viability in the course of host degradation of Riftia tissue
dissected from the worms, can be main-
in experimental conditions. Upon host death, symbiont digestion ceases allowing
tained in the laboratory under conditions
escape of viable symbionts before complete autolysis. Adapted from Klose et al.
simulating the deep-sea environment at a
(2016) CC-BY-4.0.
 Microbial Symbioses of Marine Animals 263

BOX 10.1 RESEARCH FOCUS

hydrothermal vent, with controlled supplies of sulfide and oxygen observed in the vesicomyid clam Calyptogena ponderosa, in which
and temperatures simulating the vent fluid (~22°C) or surrounding transmission of the symbiont is vertical, via the eggs. Wentrup and
seawater (−4°C). Escaping bacteria were collected into specially colleagues concluded that this likely occurs via self-infection from
designed porous chambers and identified using FISH. Only living older gill tissues, rather than re-infection with bacteria from the
symbiont cells were detected following host death due to starvation environment and/or co-occurring mussels. Infection of the newly
from lack of sulfide, indicating an active escape process through formed gill tissue results in morphological changes, including the
the skin. Based on the average size of Riftia, Klose and colleagues loss of microvilli. It appears that the bacteria escape the colonized
estimate that a typical clump of tubeworms, dying as result of ces- bacteriocytes without causing lysis of their membranes and infect
sation of vent flow, could release over 109 symbionts within a very neighboring cells.
short period. Scavenging by crabs on dead tubeworms—frequently
observed at vents—could increase the release of symbionts further. Insights into the evolution of symbionts and pathogens. This
In follow-up experiments, (Klose et al., 2016) monitored the death ability to enter the cells, avoid immediate digestion, and have an
and lysis of host cells, using analytical techniques similar to those escape mechanism is reminiscent of that shown by many intracel-
used in forensic investigation of putrefaction. They concluded that, lular pathogens, which have been extensively studied in diseases
following host cell death from starvation, the continued degrada- of humans and other vertebrates. In this case, the mussel sym-
tion of symbionts ceases rapidly (Figure 10.2). Video monitoring of bionts are related to free-living marine bacteria with no known
in situ tubeworm communities shows that tubeworms have a short pathogenic relatives. By comparing the genomes of symbionts
life, typically less than two years, which suggests regular seeding and their free-living relatives, Sayavedra et al. (2015) showed that
of the environment by released Endoriftia to facilitate colonization Bathymodiolus sulfur-oxidizing symbionts have undergone exten-
of nearby vents. However, the mechanism by which Rifta larvae sive genome rearrangements and horizontal gene transfer and that
and their Endorifta symbionts migrate to newly developed vent many of the genes found only in the symbionts are homologous
fields that may be hundreds of kilometers away is something of a to classes of toxin-related genes (TRGs) that are involved in the
mystery. Klose et al. (2015) conclude that the temporary associa- virulence of pathogens. Transcriptomic and proteomic analyses
tion between the symbiont and its Riftia host must provide a fitness showed that these genes are expressed and translated in the gill tis-
benefit, since only 20 or so bacteria are needed to infect the host sue. Sayavedra and colleagues conclude that these genes have prob-
at its larval stage, yet a dying worm will release almost a million ably been acquired via horizontal gene transfer after the divergence
living bacteria. of the symbionts from their free-living relatives. Some of the TRGs
may have beneficial interactions, such as attachment and suppres-
How do symbiotic bacteria infect Bathymodiolus mussels? sion of host rejection or the inhibition of parasites and suggest that
Despite relying on chemosynthetic symbionts for a large part of the Bathymodiolus symbionts have “tamed” the genes for benefi-
their nutrition, Bathymodiolus spp. have a fully functional filtering cial use in the host. Alternatively, they raise the possibility that the
system in their gills and are thus exposed to many different bacteria genes actually evolved as “symbiosis factors … commandeered
in their environment. However, each species appears to harbor a for use in harmful interactions” by pathogens (Sayavedra et al.,
specific phylotype of either a sulfur-oxidizing or methane-oxidiz- 2015). (Interestingly, the Endoriftia genomes analyzed in the study
ing symbiont or have a dual symbiosis of both types (Dubilier et al., mentioned above, contained only a small number of one class of
2008; Duperron et al., 2013). Again, the symbionts are contained TRGs.) In a follow-up study by this group, Sayaveedra et al. (2019)
in bacteriocytes and transmission is horizontal, but the mechanism examined the distribution of TRGs of symbionts in a wider range of
of colonization is unknown. The gills of bivalve molluscs grow Bathymodiolus species and in two deep-sea sponges. Again, they
throughout their life cycle and Wentrup et al. (2014) investigated found a wide array of toxin-related and secretion system genes, but
whether the symbionts are present in undifferentiated growth zones only one toxin class, MARTX, was common to all symbionts and
before new gill filaments develop, or whether the newly differenti- might therefore be essential for recognition, attachment, and sym-
ated gill filaments are freshly infected. Using FISH, immunohis- biont uptake. The authors conclude that the variations observed in
tochemistry, fluorescence, and electron microscopy, they showed the other TRGs may indicate convergent evolution, in which free-
that symbionts were present in the newly-formed filaments but living bacteria adopted different paths in their evolution as intracel-
not in the growth zones, indicating that gill colonization occurs lular symbionts (Sayaveedra et al., 2019).
throughout the host’s lifetime. Interestingly, this pattern was also

encoding processes required for survival outside the host including motility, DNA repair
mechanisms, and stress responses are missing. Such extreme genome reduction is also seen
in intracellular obligate parasites (see p.72) and occurs because of genetic bottlenecks due to
reductions in effective population sizes, with reduced opportunities for recombination and
horizontal gene transfer. In addition, positive selection pressure results in loss of genes for
metabolic functions that are provided by the host. In Calyptogena spp., the symbiotic bacteria
are predominantly passed from one generation to the next via vertical transmission in the
eggs. DNA sequence evolution in the symbiont is closely coupled with the DNA in host mito-
chondria (which are only inherited in the cytoplasm of the eggs), confirming that transmis-
sion is vertical, through the maternal line. However, nothing is known about the development
264 Chapter 10

of the symbiosis in the larvae and how the gill filaments become colonized. The associa-
tion between these clams and their microbial symbionts is relatively young in phylogenetic
terms—less than 50 million years—so it is tempting to think that we are witnessing an early
stage in the evolution of this symbiosis toward full enslavement of the bacteria to become
organelles through further genome reduction. This can be seen in many endosymbionts of
insects and it is how mitochondria and chloroplasts developed in the original evolution of
eukaryotes. However, it is important to note that most marine chemoautotrophic symbionts
provide all, or almost all, of the host’s nutritional requirements, whereas in insects the symbi-
ont often supplies only a few key nutrients or enzymes. Therefore, genome reduction of insect
symbionts can evolve to be much more extensive.

A wide range of other chemosynthetic

endosymbioses occurs in the deep sea
Many other species of siboglinid tubeworms also form dense colonies at hydrothermal vents
and seeps. These include Tevenia, Ridgea, Oasisa, and Lamellibrachia. These may occupy
different sites, according to the local physicochemical properties of the vent or seep fluids,
although the same type of symbiont may occur in different animal species. Some examples
are given below.

Large mytilid mussels (up to 20 cm long) of the genus Bathymodiolus are found in dense
colonies attached to rocks adjacent to hydrothermal vents and also at cold seeps (Fig. 1.16),
which emit fluids containing combinations of methane, other hydrocarbons, and hydrogen
sulfide, in differing proportions depending on geology. Some mussels appear to contain only
ARE MUSSEL a single type of symbiont, either thiotrophic as in B. thermophilus or methanotrophic as in
? EPISYMBIONTS IN
EVOLUTIONARY
B. childressii, while several other species including B. azoricus and B. puteoserpentis have a
dual symbiosis, containing both one (or more) thiotroph(s) and one (or more) methanotroph(s),
TRANSITION? all members of the Gammaproteobacteria. The ratio of the two types of endosymbiont within
the host reflects the relative abundance of sulfide and methane in the site at which the animal
In addition to the gammaproteo- is growing. In some cases, there is an even more remarkable diversity of endosymbionts with
bacterial chemosynthetic endo-
four or more phylogenetically distinct symbionts, providing additional metabolic versatil-
symbionts, Bathymodiolus spp. have
been shown to harbor bacteria
ity. For example, B. heckerae growing in petroleum-rich seeps contain hydrocarbon-degrad-
belonging to the Epsilonbactereota ing bacteria related to Cycloclasticus. There is a well-developed filter-feeding apparatus in
phylum (recently reclassified, see Bathymodiolus spp., enabling them to gather food particles from the water column, so they
p.129). This was revealed by the probably obtain their nutrition from a mixture of autotrophic and heterotrophic routes. This,
presence of characteristic gene combined with their greater mobility, explains the fact that these mussels can tolerate a wider
sequences in metagenomic librar- range of habitats at the vent sites than the tubeworms and vesicomyid clams.
ies from two distantly related and
geographically separated species Snails of the genus Alvinoconcha, found at vents in the Indian and south-western Pacific
(Assié et al., 2016). Subsequent Oceans, harbor dense populations of intracellular bacteria in their enlarged gill tissue and
analysis of DNA from seven species
are thought to provide the bulk of the animal’s nutrition via chemosynthesis. Most of the five
around the world and transcrip-
known species of Alviniconcha contain dominant populations of bacteria from either the
tomic analysis of mussel showed
that they all contained members of Gammaproteobacteria or Epsilonbactereota, but one species holds approximately equal pop-
a novel group of Epsilonbactereota. ulations of two distinct Gammaproteobacteria. Some specimens have also been found to con-
FISH, in conjunction with electron tain smaller numbers of heterotrophic bacteria provisionally identified as Endozoicomonas,
microscopy, identified these as fila- belonging to the Oceanospirillales, an order whose members are known for their ability to
mentous epibionts on the surface degrade complex organic material (p.127). These bacteria are contained in vacuoles; how-
of the gill. They are similar to epi- ever, it is not known whether they have a mutualistic function as a secondary symbiont,
biotic bacteria found in hydrother- perhaps involved in providing carbon compounds or in the cycling of sulfur from organic
mal vent crustaceans (see p.265), compounds. A different species, Ifremeria nautilei, is one of the most abundant snails found
so Assié and colleagues suggest at hydrothermal vents in the Western Pacific. It contains gammaproteobacterial symbionts
that may also be sulfur-oxidizers.
(one methanotroph and one thiotroph), as well as two alphaproteobacterial phylotypes of
Interestingly, the epibionts were
not found in mussels which lacked
unknown function.
sulfide-oxidizing endosymbionts,
leading to speculation on the pos- Other deep-sea symbiotic associations have been described in which episymbiotic bacteria
sibility that the epibiotic symbionts occur in addition to endosymbionts. The scaly foot snail (Chrysomallon squamiferum),
are in a phase of evolutionary found at hydrothermal vents in the Indian Ocean at ~2500 m deep, is encased in an unusual
transition towards endosymbiotic shell armored with iron sulfides (greigite Fe3S4 and pyrite FeS2) and the foot is covered in
integration (Assié et al., 2016). overlapping scales (sclerites) of proteinaceous material containing sulfide (Figure 10.3).
 Microbial Symbioses of Marine Animals 265

Figure 10.3 The scaly-foot snail

Chrysomallon squamiferum. Credit:
Ching Chen, JAMSTEC.

This is thought to provide further protection against predators attempting to attack its foot,
enabling the snail to stay firmly attached to vertical rocks with good feeding locations,
rather than retracting into its shell. The sclerites are covered by various epibiotic bacteria
dominated by sulfate-reducing bacteria in the Deltaproteobacteria and Epsilonbactereota
classes, which are thought to be responsible for their mineralization with iron sulfide.
The snail may secrete mucus secretions which stimulate colonization by these bacte-
ria. The snail has a highly reduced digestive tract and appears to obtain all its nutrition
from another set of chemosynthetic bacteria, this time endosymbionts belonging to the
Gammaproteobacteria, contained within bacteriocytes in an esophageal gland, reminis-
cent of the Riftia trophosome.

Another vent animal that has been the subject of intensive research is Alvinella pompejana,
an oligochaete worm about 9 cm long and 2 cm wide, which lives in dense masses in the walls
of black smoker chimneys on the East Pacific vents. This animal is known as the “Pompeii
COULD THE SCALY-
?
worm” in recognition of the association with volcanic activity and its remarkable temperature
gradient across its body; the worm’s posterior end is in galleries within the chimney wall FOOT SNAIL INSPIRE
(~85°C), while the anterior end projects into the surrounding cooler water (~20°C). Hair-like NEW FORMS
projections secreted from mucous glands on the dorsal surface are covered in episymbiotic OF ARMOR?
bacteria affiliated with members of the Epsilonbactereota, forming a dense fleece-like struc- The protective shell of the
ture. They are chemoautotrophic, possessing genes for sulfide oxidation and denitrification, scaly-foot snail discovered by
and fix CO2 via the reductive TCA cycle. It is not clear whether the host obtains any nutri- Warén et al. (2003) and named
tional benefit. The main benefit to the worm appears to be protection against extremes of Chrysomallon squamiferums by
temperature and detoxification of metals in the vent fluids, rather than nutrition. The bacterial Chen et al. (2015) has a unique
proteins predicted from the metagenomic sequences appear to have special adaptations to three-layered shell. Unlike most
variable temperature and chemical conditions. snail shells that are composed
solely of calcium carbonate, the
shell of C. squamiferum is fortified
Crustaceans have also been shown to host chemosynthetic episymbionts. When hydrother-
with iron compounds and has very
mal vents on the mid-Atlantic Ridge were first investigated in 1985, large populations of a unusual mechanical properties that
type of shrimp (Rimicaris exoculata) were discovered. They occur in large swarms, with dissipate energy, as shown experi-
rapid and continuous movement, appearing to graze the walls of the vent (Figure 10.4B). mentally by Yao et al. (2010). They
These animals have specially adapted mouthparts and an enlarged gill chamber, contain- used diamond-tipped indenters to
ing a large mixed population of up to 107 episymbiotic bacteria dominated by members measure the enormous mechani-
of the Gammaproteobacteria and Epsilonbactereota, with lesser numbers of Delta- and cal strength of the shell. Materials
Zetaproteobacteria. The varied metabolic capabilities of the mixed symbiont community scientists and engineers are being
allow hydrogen, sulfide, methane, or iron to be used as reductants for chemosynthesis. The inspired by such examples from
symbionts have a free-living stage and the same phylotypes of bacteria are also present in the natural world that may lead to
new forms of body armor, sport-
large numbers on and around the vent chimneys. It seems likely that the shrimps obtain much
ing equipment, and protective
of their nutrition by feeding on their episymbionts. The shrimps provide the bacteria with the surfaces.
optimum concentrations of sulfide and oxygen needed for chemosynthesis by scraping metal
266 Chapter 10

Figure 10.4 A. The “yeti crab”

Kiwa hirsute. B. Swarms of Rimicaris
shrimps grazing on a rock at a hydro-
thermal vent. Credits: A. Fifis/Ifremer
(CC-BY-3.0 via Wikimedia Commons).
B. NOAA Okeanos Explorer, Mid-
Cayman Rise Expedition, 2011.

sulfides from the chimney wall and by fanning water currents over the bacteria with their
rapid movements.

Kiwa hirsuta is a species of decapod crustacean discovered in 2005 at a 220 m deep hydro-
thermal vent on the Pacific-Atlantic Ridge. It was nicknamed the “yeti crab” because it is
covered in silky blond filaments (setae), resembling fur (Figure 10.4A). Molecular phyloge-
netic analysis revealed the setae to be covered by dense populations of bacteria in the phyla
Gammaproteobacteria, Epsilonbactereota, and Bacteroidetes. Key enzymes for sulfide oxi-
dation and carbon dioxide fixation indicate a thioautotrophic lifestyle.

Chemosynthetic symbioses are also

widespread in shallow-water sediments
The discovery of chemosynthetic symbionts in animals from hydrothermal vents was quickly
DANCING CRABS
i
followed by their detection in methane cold seeps. This then led biologists to search for exam-
FARM THEIR ples in other environments, such as coastal sediments with sources of methane or sulfide
BACTERIA provided by the activities of sediment sulfate-reducing bacteria and methanogenic archaea.
A new species similar to the yeti (see p.21) It was quickly realized that chemosynthetic symbioses are not restricted to a few
crab was discovered at a methane “exotic” animals in deep-sea habitats, but occur in almost all sediments where reduced com-
seep off the coast of Costa Rica by pounds able to act as an electron donor (especially sulfides, methane, or hydrogen) are avail-
Thurber et al. (2011). Kiwa puravida able. Such habitats include seagrass beds, coral reefs, marshes, and mangrove swamps where
was shown to have similar thiotro- there is a redox gradient between the top oxygenated layer of sediment and the anaerobic
phic epibiont communities to those bottom layers.
found by Goffredi et al. (2008) in
K. hirsuta. Thurber and colleagues
Chemosynthetic symbioses have now been described in hundreds of species in at least seven
provide powerful evidence that
phyla of invertebrates. A few examples of these are discussed below.
K. puravida obtains its nutrition
by “farming” the chemosynthetic
bacteria on its surface. Analysis of Investigation of different hosts soon revealed that many—probably most—involve interactions
stable isotopes and fatty acid bio- with more than one partner. This enhances the advantage to the host through the acquisition
markers in the crab tissue indicated of different metabolic capabilities. Perhaps the most remarkable example of multiple symbi-
that the epibiotic bacteria are the onts discovered so far occurs in the small oligochaete worm Olavius algarvensis and related
main source. Thurber et al. (2011) species in the family Siboglinidae. This worm contains both sulfate-reducing and sulfur-oxi-
also provided video footage of dizing bacteria, members of the Deltaproteobacteria (see note on reclassification, p.129) and
the crabs’ behavior, filmed during Gammaproteobacteria respectively. These tiny worms, only 0.1–0.2 mm in diameter and up
submersible dives to the methane to 4 cm long (Figure 10.5A), are completely devoid of a gut system and rely almost entirely
seep. They can be seen to wave
on the production of nutrients by the thioautotrophic bacteria. Unlike the previous examples,
their chelipeds (front legs) back
the bacteria are not contained within specialized cells, but occur extracellularly just below the
and forth in a rhythmic fashion.
This is assumed to increase the cuticle of the worm (Figure 10.5B). The bacteria show cooperative metabolism (syntrophy): the
supply of oxygen to the symbionts deltaproteobacterial symbionts reduce sulfate—in the seawater permeating the sediment and
needed for oxidation of sulfide. absorbed through the worm’s cuticle—to sulfide, which is used by the gammaproteobacterial
The crab also has specialized symbionts as an electron donor for the autotrophic fixation of CO2. Because the worm’s habitat
structures on its mouthparts that has very low external concentrations of sulfide, the internal generation of sulfide by the delta-
it uses to scrape the bacteria from proteobacterial symbionts is essential for sustaining the autotrophy by the gammaproteobacte-
the setae covering its body and rial symbionts (Figure 10.5C). When first described, it was thought that the worm contained
transfers them to its mouth. Thus, one population of each type, but metagenomic analysis has revealed that there are two geneti-
K. puravida has both behavioral cally different types of symbionts affiliated with Deltaproteobacteria and two types affiliated
and anatomical adaptations that
with Gammaproteobacteria. Some individuals also contain up to five other symbionts from
enable it to farm and harvest its
epibionts for food.
different bacterial groups, with unknown metabolism. Analysis of the metagenome shows that
one of the symbionts possesses an alternative mechanism for production of electrons for CO2
 Microbial Symbioses of Marine Animals 267

Figure 10.5 A. An adult gutless

worm Olavius algarvensis isolated
from sediment, Isle of Elba, Italy. B.
Model showing syntrophic cycling
of oxidized (Sox) and reduced (Sred)
sulfur compounds between SOB
(green) and SRB (red) endosymbi-
onts within the tissue of the worm.
Credits: A. Alexander Gruhl, MPI
Marine Microbiology, Bremen. B.
Redrawn from Dubilier et al. (2001)
with permission of Springer Nature.

BUCKETS OF MUD
i AND “STONE AGE”
SEQUENCING
The herculean task in unraveling
the interactions in the multi-
partner consortium of the worm
O. algarvensis is only partly evident
in the paper by Woyke et al.
(2006). However, in a follow-up
paper revisiting this investigation,
Kleiner et al. (2011) reveal some of
the complexities of their work. This
involved collaboration between
teams led by Nicole Dubilier of the
MPI for Marine Microbiology and
Eddy Rubin of the Joint Genome
Institute, at a time when metage-
nomic studies were just begin-
ning. For lead author Tanja Woyke,
fieldwork to collect fresh worms by
fixation, using nitrate or fumarate in the absence of O2. The deltaproteobacterial symbionts pos- scuba diving off the Mediterranean
sess genes for CO2 fixation via the reductive tricarboxylic acid (rTCA) cycle, as well as numer- Isle of Elba must have seemed
appealing, but the reality involved
ous genes for heterotrophic metabolism of organic compounds in aerobic conditions. Thus, as
collection and sifting of bucket-
the worm moves through the oxic and anoxic regions of the coastal sediments that it inhabits, loads of sediment searching for the
electron donors will be available at all levels and the symbiotic consortium provides the worm tiny microlitre-volume worms. It
with an optimal energy supply. As in Riftia, it is likely that Olavius digests the bacteriocytes took many weeks to collect enough
to obtain fixed carbon compounds. Olavius is also thought to have an adapted hemoglobin that material—about 1000 worms—for
transports oxygen and sulfide for metabolism by the thiotrophic symbionts; there are also large DNA sequencing. The team had to
amounts of a storage protein for oxygen that ensures a supply of oxygen in anoxic environ- construct large-insert clone librar-
ments. Furthermore, the symbionts also recycle the host’s waste products of ammonia and urea, ies and take special steps to enrich
which explains why the worms can live with a completely reduced excretory system. Further the proportion of microbial DNA
complexities have been revealed by metaproteomic analyses, showing that the SRB can use before developing novel algorithms
to assign sequences to the differ-
hydrogen as an energy source and that three of the symbionts use carbon monoxide (CO) as an
ent symbionts, depending on the
energy source—this is very surprising given its toxicity to animal life. Examination of related
relative frequency of nucleotides.
species from other parts of the world reveals similar symbiotic consortia. The relationship is These binned sequences were
clearly obligate for the host’s nutrition, but how the worms acquire the symbionts is still unclear. then assembled and annotated
There is some evidence for vertical transmission, in which the symbionts may be transferred to to construct the genome of four
the eggs via “smearing” during egg-laying, as occurs in some insects. symbionts, allowing their meta-
bolic potential to be predicted.
Flatworms in the genus Paracatenula (phylum Platyhelminthes) are also completely devoid Kleiner et al. (2011) comment that,
of a digestive system and rely on mutualistic symbiosis with the bacteria identified as “Ca. at the time (2006), this was a very
Riegeria santandreae” (family Rhodospirillaceae). To date, this is the only known chemo- large-scale project, while mod-
synthetic symbiont in the Alphaproteobacteria, containing genes encoding sulfur oxidation ern metagenomic analyses easily
provide gigabases of data—“our
and CO2 fixation pathways for chemolithoautotrophy. Despite possessing one of the most
study … already belongs to the
highly reduced genomes known in marine bacteria (1.34 Mb, see Table 3.1), it has retained stone age of Sanger sequencing.”
genes for a versatile carbon metabolism, including a complete TCA cycle, which allow it to
268 Chapter 10

use sugars and fatty acids an energy source. However, these are not used for heterotrophic
metabolism, but for internal recycling and storage, as shown by proteomic and transcriptomic
analysis. Remarkably for a chemoautotroph with such a reduced genome, the symbiont has
pathways for the synthesis of the storage carbohydrates polyhydroxyalkonoates, trehalose,
and glycogen, which accumulate as inclusion bodies and make up a large part of the volume
of bacterial cells (Figure 10.6). For the bacterium, these reserves provide an energy sink and
carbon store when the worm moves into anoxic regions of the sediment (necessary to acquire
sulfide, as discussed above for Olavius), but they also serve as the main source of nutrition for
the host. The symbiont does not possess sufficient transport mechanisms for translocation of
nutrients to its host, and Paracatenula does not digest the symbionts by lysosomal fusion as
occurs in many other nutritional symbioses (including Bathymodiolus mussels and insects).
Indeed, Paracatenula does not produces lysozyme enzymes—except under conditions of
prolonged starvation—which would be necessary for degradation of the bacterial peptido-
glycan cell wall. Instead, the symbiont builds up a stockpile of substantial stores of nutrient
reserves which provide the primary energy storage for the bacterium themselves and for the
host, by transfer in outer membrane vesicles, as shown by gene expression and ultrastructural
studies (Figure 10.6). This fascinating symbiosis also provides valuable insights into the
evolution of mutualistic symbioses. Phylogenetic analysis indicates that Paracatenula and
its Riegeria symbionts have been co-evolving for ~500 million years. Although the bacte-
rium has undergone extensive genome reduction, it has retained many essential functions
and shows no signs of the continued gene loss that would result in it becoming a functional
organelle of the host, as has occurred with many insect endosymbionts, over much shorter
periods. This can be explained by the fact that Paracatenula relies entirely on the symbiont
for all its nutrition, whereas symbionts of insects often only supplement the host diet with
one or a few nutrients, such as vitamins or essential amino acids. This is often complemented
by secondary endosymbiosis, so that restrictions on continued gene loss in the bacterium are
removed and it eventually becomes “enslaved” as an organelle.
Figure 10.6 Symbiosis of “Ca.
Riegeria santandreae” with the
flatworm Paracatenula. (A) Habitus
of Paracatenula from Sant’Andrea
Bay, Elba. The white trophosome
contains endosymbionts, while the
anterior and transparent part of the
worm (rostrum) is bacteria-free. B.
Differential interference contrast
image of Riegeria symbionts indicat-
ing multiple intracellular inclusions.
C. Confocal laser-scanning image of
CARD-FISH combined with lipophilic
staining (Nile Red) on a transverse
section of a Paracatenula specimen in
the symbiont-bearing region. Overlay
of DAPI signal (blue), Nile Red (red),
and probe targeting the symbionts
(green, EUB I–III) is shown. D. Trace
illustration of TEM of the trophosome
region with the status of the intracel-
lular bacterial symbiont indicated
in dark blue (OMVs detected), light
blue (no OMVs detected), and orange
(digested—lysosomal bodies). E-G.
Representative image outcrops of the
OMV-secreting symbiont population.
CR, “Ca. R. santandreae”; hc, host
cytosol; om, OMVs; ph, storage inclu-
sion (PHA). H. A symbiont cell that has
undergone lysosomal digestion (ly) in
a starved worm. The enclosed struc-
ture in the lysosome resembles a PHA
inclusion (ph). Reprinted from Jäckle
et al. (2019), CC-BY-NC-ND 4.0.
 Microbial Symbioses of Marine Animals 269

Figure 10.7 A. Part of the skeleton

of a gray whale carcass discov-
ered in 2002 in Monterey Canyon
at >3000 m deep, densely covered
with Osedax worms. B. Close-up
of Osedax trunk and palps emerg-
ing from a whalebone. Credits: A.
Copyright MBARI, reprinted with
permission. B. Nick Higgs, University
of Plymouth.

Animals colonizing whale falls depend on

autotrophic and heterotrophic symbionts HOW LONG HAVE
Large whales weigh 30–160 tons, so the amount of organic carbon from a single carcass ? OSEDAX WORMS
BEEN DEVOURING
reaching the seafloor can be the equivalent of thousands of years of marine snow falling from
BONES?
the surface. The occasional, but huge, pulse of organic matter to the ocean floor supports
diverse animal communities. After initial removal of soft tissue by sharks, hagfish, crabs, and The exceptional diversity of Osedax
other scavengers, followed by colonization of the surrounding sediments, the whale skeleton worms and its wide geographic
can support a rich population of animals sustained by thioautotrophic microbes oxidizing distribution at depths from 21
sulfur obtained from the bones. Many of these associations resemble those found at hydro- to 4000 m suggests a very long
thermal vents and cold seeps. Some scientists have speculated that whale carcasses provide evolutionary history and poses
something of a mystery. Using
“evolutionary stepping stones” for dispersal of animals harboring autotrophic symbionts as
micro-computed tomography
they have radiated from shallow- to deep-water habitats, but other researchers do not support (CT), paleontologists can detect
this idea. the characteristic traces of penetra-
tion by Osedax roots in fossilized
In 2002, scientists from the Monterey Bay Aquarium Research Institute who were following bones of early whale species from
the decay of a gray whale carcass that had sunk to a depth of nearly 3000 m off the coast ~30 MYA, suggesting that they
of California, discovered Osedax, a new genus of gutless sibliognid worms (the name is have been living on whale skele-
derived from the Latin for “bone devourer”). The animals covered the skeleton, penetrating tons since the first whales evolved.
the bones with a root system developed from an enlarged egg sac and richly supplied with Danise and Higgs (2015) also found
blood vessels (Figure 10.7). In electron micrographs, large bacterial cells were observed these traces in fossil bones from
100 MY-old plesiosaurs. Therefore,
inside bacteriocytes in the host cells; 16S rRNA gene sequencing showed these to be very
before the evolution of whales,
different to the chemoautotrophic bacteria found in other gutless worms like Riftia. The
marine reptile carcasses played a
dominant symbionts belong to the Oceanspirillales group (p.127) which are heterotrophs that key role in the evolution and dis-
break down the collagen, cholesterol, and lipids, present in large amounts in the whale bone. persal of Osedax and its symbiotic
The host derives all its nutrition from the bacteria, probably by digesting the bacteriocytes. bacteria. The fossil traces found
Analysis of lipids in the worm tissue confirms that they are derived from bacterial metabo- in this study also showed that the
lism. The animals are biologically very unusual since the males are contained within the whole family of gutless worms is
female body; little is known about the nutrition of the males. Free-living Oceanospirillales much older than previous esti-
bacteria occur in the bone tissue and are acquired by the larvae during settlement. It is likely mates. The Cretaceous period was
that signaling between the bacteria and host triggers the development of the root system of thus a key period for the evolution
the worm, but nothing is currently known about this. Presumably, the symbionts benefit by of symbiotic associations.
270 Chapter 10

access into the bone via the worm’s roots, receiving plentiful supplies of oxygen for aerobic
breakdown. The wide geographic distribution of Osedax is something of a mystery, given
the rarity of carcasses of whales since their massive depletion by whaling in the nineteenth
century. However, a large whale carcass should sustain a community for many years and,
based on calculation of natural mortality rates, significant deposition of carcasses could still
occur, especially along migration routes. Thus, whale falls still play a major role in the ecol-
ogy of deep-sea benthic processes, sustained by long-term microbial activity. We now know
that there are numerous species of Osedax and many of these are found in much shallower
habitats, where they feed on bones of other vertebrates, as well as whales.

Large inputs of organic material to the floor of the deep sea are also provided in the form
of sunken logs, masses of kelp or large accumulations of dead fish or invertebrates. Wood-
boring bivalves (Xylophaginae) containing symbiotic bacteria digest the wood. Because of
the difficulties of study, little is known about the nature and diversity of the symbionts, It
might be expected that they are like the heterotrophs found in the shallow-water bivalves
known as shipworms (see p.358), which produce enzymes for the degradation of cellulose and
other carbohydrates, as well as providing fixed nitrogen to their host. Thiotrophic symbionts
may also be involved, using sulfides generated by sulfate-reducing bacteria in the anoxic
sediments surrounding the wood.

Sea squirts harbor photosynthetic bacteria

In their adult stage, ascidians (commonly known as sea squirts) are colonial sessile filter-
feeding invertebrates belonging to the subphylum Tunicata, which are characterized by a
tough polysaccharide coat or “tunic.” Ascidians have assumed special importance because
they are transmitted over great distances by attachment to ship’s hulls and floating marine
debris and non-native species have appeared as invasive species in harbors and marinas all
over the world, where they quickly become established as invasive species that can affect the
IT’S NOT EASY TO
i SINK A WHALE ecology of sub-tidal waters or cause damage to docks or other floating structures. This occurs
despite their sedentary nature and limited mechanisms of dispersal. The adult stage of some
Although some knowledge of the species harbors the photosynthetic cyanobacterium Prochloron, and cells of the symbionts
biology of whale falls came from are transmitted vertically during late stages of embryonic development of the larvae. Some
chance discovery of carcasses, species of ascidians also harbor a diverse community of bacteria, including some types that
most of our knowledge of the
carry out aerobic anoxygenic phototrophy (AAnP, p.77). This unexpected finding indicates
succession of events in the decay
of whale skeletons comes from
a new type of light-based symbiosis very different from the oxygen-generating activities of
manipulative experiments. Craig zooxanthellae and symbiotic cyanobacteria (discussed below). It will be interesting to find out
Smith (University of Hawaii) devel- whether the host derives nutritional benefit from the association and, if so, the mechanisms
oped methods for towing carcasses involved. As occurs in many other marine invertebrates, Endozoicomonas may form a fac-
of beached whales to drop sites ultative association with tunicates, with a key role in degradation of mucus. It appears that
(mapped precisely with GPS coor- different ascidians contain stable bacterial communities with a high degree of host specific-
dinates) and sinking them to the ity, and this may be a factor in their ability to adapt to new environments, explaining their
seafloor (Smith and Baco, 2003). success as invasive species.
Apart from the logistic problems
of arranging transport and support
at short notice, sinking a whale Endosymbionts of bryozoans produce compounds
carcass bloated with gases after
many days of towing behind a ship that protect the host from predation
is a difficult and unpleasant task. Bryozoans are tiny colonial animals that resemble moss coating the surface of rocks. One
Many such “whale-drops” have
species, Bugula neritina, has been the subject of much interest because it produces complex
now been conducted, especially
polyketides (bryostatins), which is effective in the treatment of some cancers (p.394). We
by scientists from MBARI. Manned
submersibles and ROVs are used know now that an endosymbiotic bacterium synthesizes bryostatin and the genes responsible
for long-term video recording and have been identified. The bryostatin molecules coat the surface of the larvae at high concen-
sampling of the carcass over many trations and seem to function in protecting the larvae from predation, as they are unpalatable
years. Deliberate creation of sunken to fish. They may also play a role in attracting conspecific larvae to settle to form colonies.
wood falls and long-term monitor- The bacterium has never been cultivated, but genetic analysis shows that it is a gammapro-
ing of their decay and colonization teobacterium, and it has been designated “Ca. Endobugula sertula.” Several other bryozoan
may be slightly less challenging, symbionts, including an alphaproteobacterial example in Watersipora spp., have now been
but still requires careful logistical described. Since the Bryozoa show extensive diversity in structure and ecology, it is likely
planning, as described by Ristova that symbiosis has evolved independently and may have different physiological functions in
et al. (2017).
different groups.
 Microbial Symbioses of Marine Animals 271

BOX 10.2 RESEARCH FOCUS

Chemosynthetic bacteria fuel ecosystem-engineering bivalves

We know now that chemosynthetic symbioses occur in many organic material. This results in production of large amounts of
shallow-water habitats, as well as the deep sea. However, the eco- sulfide, which is highly toxic to the seagrass roots and to ani-
logical role of these interactions has been poorly understood until mals. The activity of the thiotrophic symbionts in the lucinids
recently. Genomic and proteomic approaches are revealing the reduces this toxic effect, resulting in enhanced seagrass produc-
importance of chemotrophic symbionts in fixation of carbon and tion, while the clams and their symbionts benefit from increased
nitrogen, underpinning the ecology and productivity of coastal organic matter and oxygen from the seagrass roots—a three-
ecosystems. stage symbiosis (Heide et al., 2012). The symbiont cannot be
cultured, so König et al. (2016) separated bacterial DNA from
Lucinid clams detoxify seagrass sediments and fix nitrogen. C. orbicularis tissue and carried out genomic and proteomic
Molluscs belonging to the family Lucinidae inhabit various analysis. As expected, the appropriate genes encoding enzymes
habitats, including mangroves, coral reef sediments, and the in the sulfide oxidation and Calvin–Benson–Bassham (CBB)
dense beds (meadows) of seagrasses—flowering plants attached carbon fixation pathways were all present. Unexpectedly, they
to sand or mud bottoms in shallow coastal waters. There are also found that the genome contained a complete nif gene cluster
numerous seagrass species found throughout the world, and they for nitrogen fixation, including the nitrogenase complex, which
have great ecological importance as “ecosystem engineers” due they showed to be functionally active using an acetylene reduc-
to their high production from photosynthesis and the trapping of tion assay (see p.242). By using an antibody detection method,
sediment and reduction of water movement and coastal erosion. König and colleagues confirmed the presence of the protein
They harbor highly diverse and abundant animal communities NifH in the bacterial endosymbionts inside the bacteriocytes of
and they provide nursery areas for fish and a direct food source clam’s gill tissue. Seagrass bed sediments in which the C. orbi-
for turtles, dugongs, manatees, seabirds, and other animals. The cularis grow have low levels of NH4+, so nitrogen fixation by the
lucinid clam Codakia orbicularis, which lives in the sediment of symbionts would be advantageous to the clams, by supplement-
Thalassia testudinium seagrass beds in the Caribbean, contains ing the supply of nitrogen compounds. The symbiont belongs
a single species of thioautrophic symbiont in its gills (Figure to the class Gammaproteobacteria and has been named “Ca.
10.8). The bacteria are housed in bacteriocytes and appear to be Thiodiazotropha endolucinda” and appears similar to several
digested, although this may only be significant during periods of other uncharacterized symbionts from lucinid clams. In a sepa-
starvation when normal filter feeding is insufficient (König et al., rate study, Petersen et al. (2017) showed the presence of NifH
2015). However, in this case a more important ecological role for genes in genome sequences obtained from the thiotrophic sym-
the symbionts is detoxification of sediments by removal of sul- biont of another lucinid species, Loripes lucinales. Again, they
fide. The high productivity of seagrass beds is highly dependent identified a member of the Gammaproteobacteria, which they
on nitrogen fixation in the sediment, which is largely carried out named “Ca. Thiodiazotropha endoloripes”. PCR analysis of L.
by anaerobic sulfate-reducing bacteria during the breakdown of lucinales specimens from around the world provided evidence

Figure 10.8 Model of interactions in a seagrass bed between Thalassia testudinum, Codakia orbicularis, and its symbionts.
(a) Seagrass roots are inhabited by a complex microbial community, including diazotrophic SRB. (b) SRB produce sulfide
from the decay of dead seagrass and other organic detritus. Bacterial endosymbionts and free-living sulfide-oxidizing
bacteria detoxify sulfide to the benefit of plants and animals. Reprinted from König et al. (2017) CC-BY-4.0.
272 Chapter 10

BOX 10.2 RESEARCH FOCUS

of genetically similar nifH sequences. Petersen and colleagues it is not completely dependent on its chemosynthetic symbionts.
used metatranscriptomic analysis of clam gill tissue to show Batstone et al. (2014) found wide differences in the structure of
the expression of these genes. Further to the suggestion that the gills and the presence of symbionts, even within the same
seagrasses grow better when they contain large populations of species, which probably reflect local differences in sulfide con-
lucinid clams (van der Heide et al., 2012), Petersen et al. (2017) centration. In early microscopic studies, dark intracellular inclu-
suggest that excess fixed nitrogen released into the sediments sions had been seen inside the gill cells and these were assumed
could be taken up by the seagrass roots. to be viruses. However, using specialized electron microscopy
techniques, Dufour et al. (2014) showed that these inclusions
Chemosynthetic products feed lobsters. Higgs et al. (2016) are magnetosomes (Figure 10.9A). They form octahedral bod-
showed that, besides its obvious ecological importance, this sym- ies containing iron, characteristic of magnetotactic bacteria
biosis provides a rare example of direct input of chemosynthetic (p.123), and rRNA gene sequencing confirmed their relationship
production into marine food webs. The clams are a major food to this group. It seems strange that such bacteria are symbiotic,
source for the Caribbean spiny lobster, supporting commercial fish- as magnetotactic behavior can have no advantage inside the host.
eries of high economic worth in the Caribbean. By analyzing the C, Dufour and colleagues suggest that in their free-living state these
N, and S stable isotope ratios of spiny lobsters, Higgs et al. (2016) bacteria possess flagellar motility and use magnetotaxis in con-
concluded that they obtain 20–35% of their nutrition from organic junction with aerotaxis to track the oxic-anoxic interface. They
material produced by chemosynthesis. would thus be attracted to the Thyasira burrows, which provide
optimal conditions for chemolithotrophic sulfur oxidation; they
Thyasirid clams oxygenate marine sediments with help from could then become trapped in mucus and accumulate in the gills.
magnetotactic, thiotrophic symbionts. Members of the diverse Once inside the host, they lose their flagella. Do these symbi-
bivalve family Thyasiridae are widely distributed in seeps and onts provide advantage to their host—are the hosts “gardening”
coastal sediments. Although small (usually <1 cm), they create their symbionts? We know that the bacteria are endocytosed by
extensive ramifying burrows, oxygenating the sediment (Dufour the host, but the iron inclusions are not digested and accumulate
and Felbeck 2003). This bioturbation has large impacts on the in the gill tissue (Figure 10.9B). Using a model to estimate the
ecology of the sediments that they inhabit. Many contain thio- dynamic energy budget of Thyasira, Mariño et al. (2018) provide
trophs in their gills, but in this case the bacteria are extracellular. evidence that the presence of the symbionts constitutes an adapta-
The molluscs promote sulfide oxidation by their deep burrowing, tion to buffer fluctuations in the availability of food obtained by
“mining” the sediment for sulfide, and provide active oxygen- feeding on organic particles. Therefore, the host does not need to
ation by villi in the gill filaments. The host obtains some nutri- build up large energy reserves, which improves its maturation and
tion by engulfing and digesting the extracellular symbionts, but reproductive success.

Figure 10.9 TEMs of a gill-associated symbiont of Thyasira cf. gouldi. A. Symbiont showing a chain of magnetosomes,
located close to the microvillar boundary (mv) of the bacteriocyte. B. View of symbionts in an extracellular ‘pocket’ (p), with
some undergoing endocytosis (e). A large aggregate of lysed products of symbiont digestion (l), including magnetosomes, is
visible in the host cytoplasm (cy). Dark inclusions may be resistant particles of iron. Reprinted from Dufour et al. (2014), with
permission of Springer Nature.
 Microbial Symbioses of Marine Animals 273

Sponges contain dense communities of specific microbes EXCEPTIONAL

Sponges (Phylum Porifera) are the oldest group of multicellular animals, comprising more
than 8000 species in a wide variety of tropical, temperate, and cold-water marine habitats,
i DIVERSITY OF
SPONGE MICROBES
with a smaller number of freshwater examples. They have a simple three-layered body struc- Various investigations have been
ture with an internal system of pores and channels lined by flagellated cells (choanocytes, consolidated into the Global
Figure 6.4) that pump water through the body. They feed by filtering bacteria and microalgae Sponge Microbiome Project, as
from the surrounding water and digesting them with specialized phagocyte cells. Sponges part of the Earth Microbiome
filter huge volumes of water to concentrate food particles from the surrounding water—a initiative (www.earthmicrobi-
modest-sized sponge filters tens of thousands of liters of water a day—and they can remove ome.org/). Building on previous
over 90% of the bacteria in the inhalant water. The sponge body (mesohyl) is a gelatinous large-scale analyses of sponge
matrix strengthened by fibers and spicules of calcium carbonate or silica. In most sponges, microbiota (Schmitt et al., 2012;
the mesohyl is packed with a diverse community of many different microbes, which can con- Taylor et al., 2013), this collab-
orative venture between different
stitute up to half of the volume of the sponge. Most microbes in the mesohyl are extracellular,
laboratories used standardized
but intracellular bacteria and microalgae are sometimes found. methods of DNA extraction and
amplification of the V4 region of
Original interest in the study of the microbial content of sponge tissue was largely driven the 16S rRNA genes from differ-
by the interest in natural products such as antibiotics and antitumor compounds, which ent sponge species, as well as
they often contain. Only a tiny fraction of sponge microbes has been cultured, but sus- seawater and sediment controls.
picion that the microbes (rather than the host’s own metabolism) produce a wide range In addition, standardized bioin-
of bioactive compounds has been confirmed by recognition of microbial genes for their formatics protocols for identifying
biosynthesis. This offers new possibilities for biotechnological exploitation, which is dis- and clustering retrieved sequences
cussed in Chapter 14. were employed (Thomas et al.,
2016; Moitinho-Silva et al., 2017).
The technical standardization has
In addition to this biotechnological focus, there is inherent interest in the interaction
resulted in large publicly available
between microbes and sponges from a biological and evolutionary perspective. Sponges datasets that can be used reliably
have a major role in marine ecosystem functioning, including the creation and stabilization and consistently for research into
of reef communities and the coupling of pelagic and benthic processes. Recent efforts to sponge host-microbe specificity
understand the microbiota of sponges have been stimulated by the need to understand the and the effects of environmental
effects of pollution, climate change, ocean acidification, and other pressures on sponge factors on microbiome structure.
health and disease. In the past few years, there has been intensive application of molecular Thomas et al. (2016) concluded
methods to investigate sponge microbes, and many thousands of rRNA gene sequences that sponge-associated communi-
have been described. ties are a major contribution to
the total microbial diversity of the
oceans.
These studies reveal that sponge microbes are very diverse, with at least 40 different
phyla or candidate bacterial or archaeal phyla identified. All sponge species examined
contain members of numerous different phyla, with Gamma- and Alphaproteobacteria
dominant in most sponge species (Figure 10.10). A wide diversity of fungi has also been
described; most of these are ascomycetes related to terrestrial fungi, but some fungal spe-
cies may be specialist associates of particular sponge species and can be identified at dif-
ferent geographical locations. The variability in the richness of their microbiota between
hosts of the same species is generally low, indicating that there is species selectivity for
particular microbial partners. Some microbes can be detected in even distantly related
sponges isolated from a wide geographic range while others appear to be specialists that
are present in only a few sponge species. There is evidence of a stable core microbiome,
where particular microbial types are abundant in all members of the sponge species,
while others are much more variable. One explanation for this is that specific microbes
became associated with sponges very early in their evolution, more than 600 million
years ago, and remained associated with the sponges as they underwent evolutionary
radiation. This hypothesis requires that the microbes are passed vertically from one
generation to the next, and there is now considerable evidence (largely from microscopic
studies) that this occurs during sexual reproduction, with bacteria and yeasts observed
in the eggs, embryos, or larvae of many species. During asexual reproduction, microbes
could be passed via tissue buds.

The almost ubiquitous occurrence of symbiotic microbes and their probable presence
throughout evolution of sponges indicates strongly that these are mutualistic associations.
Most sponges are heterotrophic, consuming microbes harvested from the seawater. Some
sponge species rely on photosynthetic cyanobacteria, dinoflagellates, or diatoms for up to
half of the sponge’s energy requirements and some species of sponges growing at methane
274 Chapter 10

Figure 10.10 Microbial diversity

(OTU richness) in sponge-associated
microbial communities at phylum
level detected by the Global Sponge
Microbiome project. Redrawn from
Pita et al. (2018), CC-BY-4.0.

seeps contain methanotrophic symbionts, which undoubtedly provide nutritional benefit

COULD SPONGES
?
to their host. Many sponge-associated microbes are active degraders of complex carbo-
FILTER RARE hydrates such as algal polysaccharides. In some sponges, ingestion and breakdown of the
BACTERIA FROM bacteria in the mesohyl has been observed, leading some scientists to suggest that sponges
SEAWATER? cultivate bacteria as a food source. Bacteria and archaea certainly seem to play a key role in
Although the idea of a very ancient nitrogen cycling within sponges, particularly by conversion of toxic ammonia—produced
symbiosis with specific microbes is in large amounts by the sponge tissue—by oxidation to nitrate by nitrification. Anaerobic
favored by most sponge microbi- processes of denitrification or anammox can occur when sponge tissue becomes anoxic
ologists, alternative explanations when pumping activity stops temporarily. Nitrogen fixation by cyanobacteria may also
involving acquisition of some be important in some sponges. The production of large amounts of polysaccharides by
organisms from the environment the mesohyl bacteria is also thought to contribute to the structural integrity of the tissue.
are possible. One of the main Sponge symbionts have numerous genes involved in synthesis of vitamins and vital amino
reasons given in favor of specific
acids, which complement the host’s metabolism. One the most important beneficial func-
associations is that many types
tions of the microbiome is the production of bioactive compounds that may protect the
of sponge microbes can never be
detected in the environment and sponge against the harmful effects of ultraviolet radiation or oxidative stress, or act as
these have been assigned to a defense compounds to prevent predation.
candidate phylum “Poribacteria”
(Fieseler et al., 2004). However, Recognition of the association between the sponge organisms and their specific microbial
because a sponge can filter tens of inhabitants allows us to consider how the activities of these holobionts becomes integrated
thousands of liters each day, even if into larger communities and the ecosystem. Thus, activities of particular microbes can have
a very rare bacterium was pres- ecosystem-wide effects. The sponge microbiome carries out functions that are amplified to
ent at only one cell per liter—and affect the productivity, nutrient cycles, and food webs of the sponge habitat, as illustrated in
therefore unlikely to be detected Figure 10.11.
by commonly used methods such
as PCR amplification of genes in
Given that sponges filter-feed on microbes, how is it that most of the bacteria can live
water samples—a sponge could
still acquire tens of thousands of within the tissues, adjacent to phagocytic cells? It is likely that the symbionts possess
this type every day. Indeed, later surface properties such as production of grazing deterrents, altered cell walls, or defen-
studies using deep sequencing sive proteins that protect them from recognition and ingestion by the phagocytes. As yet,
showed that many of the previously firm conclusions about the origin of the association between sponges and their resident
described “sponge-specific bacte- microbes are not possible. Further application of metagenomics, metatranscriptomics,
ria” gene clusters are widespread metaproteomics, and other advanced techniques will aid our understanding of the speci-
in seawater, albeit at very low abun- ficity and interdependence of the host–symbiont interactions, which is such a fascinating
dances (Taylor et al., 2013). aspect of sponge microbiology.
 Microbial Symbioses of Marine Animals 275

Figure 10.11 Ecosystem functions

of the sponge holobiont. Key func-
tions carried out by the microbiome
(colored arrows) influence holobiont
functioning, which cascade to influ-
ence community structure and eco-
system functioning. Environmental
factors act at multiple scales to alter
microbiome, holobiont, commu-
nity, and ecosystem scale processes.
DOM, dissolved organic matter;
POM, particulate organic matter;
DIN, dissolved inorganic nitrogen.
Reprinted from Pita et al. (2018),
CC-BY-4.0.

Many marine invertebrates depend on

photosynthetic endosymbionts
Biologists first recognized associations between marine invertebrates and photosynthetic
microbes more than a century ago and introduced several descriptive terms at this time, based
largely on the coloration of the tissue due to pigments from the symbionts. The best known
and most important group is the zooxanthellae—the term refers to the golden-brown color
(Figure 10.12)—which are dinoflagellates occurring in a wide range of Cnidaria (corals, anemo-
nes, jellyfish, and zoanthids), as well as some molluscs, sponges, and flatworms. Green zooch-
lorellae are members of the Chlorophyceae, occurring mainly in sponges, coelenterates, and
flatworms, while blue–green zoocyanellae belong to the Cyanobacteria (including Prochloron)
found in some sea squirts and molluscs. Many tropical sponges also rely on cyanobacteria for
more than half of their energy requirements. Zooxanthellae also occur in the protist group
foraminifera.

Zooxanthellae (Symbiodiniaceae) show extensive

genetic diversity and host specificity
Zooxanthellae are members of the family Symbiodiniaceae (class Dinophyceae). Although
differences in structure, physiology, biochemistry, and behavior were recognized, until
the 1970s all zooxanthellae were considered members of the single species Symbiodinium

Figure 10.12 Zooxanthellae

isolated from tentacles of upside-
down jellyfish, Cassiopea sp. show-
ing the distinctive golden-brown
color. Individual cells are ~6-10 µm
diameter. Credit: Todd LaJeunesse,
Pennsylvania State University.
276 Chapter 10

microadriaticum. However, molecular analysis has revealed a very high level of phylogenetic
diversity. Symbiodinium has been divided into nine major groupings (clades A–I) based on
sequence analysis of 18S rRNA genes. However, more discriminating methods based on
gene sequencing of the variable internal transcribed (ITS) region show that Symbiodinium
has significant intraclade diversity, and there are several hundred genetically distinct types,
of which some have been formally described as named Symbiodinium species. However, the
genetic differences between clades of the genus is much greater than that between higher
dinoflagellate taxa, suggesting that they should belong to different genera. Whole genome
analysis has proved difficult due to the very large size of dinoflagellate genomes and only a
few genomes have been assembled, but multigene analysis of key markers has yielded valu-
able insights. Using a variety of morphological, physiological, and genetic criteria (including
molecular clock analysis to determine evolutionary relationships), seven of the clades have
now been formally named as genera within the family Symbiodiniaceae. Only members
of clade A have been retained in the genus Symbiodinium (Table 10.1). These taxonomic
changes should become quickly accepted, although readers should be aware that the des-
ignation Symbiodinium as a general descriptor may continue to be used by some authors.
Correlative studies of distribution and depth zonation of different clade types, together with
numerous experimental studies, have shown that different clade types differ in their physi-
ological characteristics, especially sensitivity of photosynthesis to light and temperature.
Some Symbiodinaceae species associate with a broad range of hosts, while others are found
in only a few host species.

Table 10.1 shows that Symbiodinaceae have been described in five different animal phyla. In
their hosts, zooxanthellae usually form coccoid cells surrounded by a series of algal mem-
branes, within a host vacuole called a symbiosome; however, in clams the zooxanthellae lie
extracellularly in specialized channels in the gill mantle. Many symbiotic dinoflagellates can
be isolated and maintained in culture. The morphology and life cycle of the free-living and
symbiotic forms are very different; in particular, the dinoflagellate loses its flagella inside
the host. The symbionts carry out photosynthesis, harvesting light energy via a complex
of chlorophyll a, chlorophyll c2, and the protein peridinin. Light energy is used to fix CO2
which occurs mainly via the C3 pathway (CBB cycle) using the enzyme RuBisCO. This can
be shown by enzyme assays and by incubating with radiolabeled bicarbonate and measuring
the incorporation of 14C into the tissues. Some species may also use a C4 route employing
phospho-enol pyruvate carboxylase as the key enzyme. Animals that contain zooxanthellae
clearly derive nutritional benefit from the relationship in the form of photosynthetically fixed
carbon compounds, although views differ on the exact contribution that each partner makes
to the acquisition of other nutrients—especially nitrogen and phosphorus. Radiolabeling
experiments show that the zooxanthellae release a high proportion of photosynthetically
fixed carbon as small molecules, including glycerol, glucose, and organic acids. It is possible
that some essential amino acids are also provided by the zooxanthellae. The density of zoo-
xanthellae varies widely in different hosts; high numbers are often associated with a reduced
dependence on feeding by capture of plankton, manifested by smaller tentacles or reduced
digestive systems.

We know little about how these intimate relationships came about, but fossil evidence
shows that coral reefs first evolved in the mid-Triassic period (~250 MYA). In all the ani-
mal phyla that harbor dinoflagellates, the final step in digestion is intracellular. Cells of
the animal’s digestive tract may have retained algal cells that were resistant to digestion,
and subsequent evolution has refined this association. A key step in establishment of the
symbiosis appears to be due to the zooxanthellae arresting the maturation of phagosomes.
The importance of the association to the host is emphasized by anatomical and behavioral
adaptations observed in the various animal groups, which have evolved in order to expose
the zooxanthellae to light.

Many corals are dependent on zooxanthellae for nutrition

Corals and anemones belong to the Anthozoa class of the phylum Cnidaria. Corals typically
form colonies containing numerous identical (clonal) polyps. They feed on prey—varying
from small plankton to small fish—using tentacles and stinging cells for capture, but in many
 Microbial Symbioses of Marine Animals 277

Table 10.1 Lineages of Symbiodiniaceae, showing redesignation of Symbiodinium clades to new genera

Clade Host taxaa New genus designation

and key featuresb
Foraminifera Ciliophora Porifera Cnidaria Platyhelminthes Mollusca

A + + + Symbiodinium. Adapted to
variable or high light
conditions.

B + + + Breviolum. Small cells.

Primarily associated with
cnidarians. Prevalent in
Caribbean.

C + + + + + Cladocopium. Large number

of species. Broad host
range. Ecologically
abundant and broad
geographic distribution.

D + + + + Durusdinium. Adaptations
to survive fluctuations in
temperature or turbidity;
more resistant to
environmental disturbance
(e.g. coral bleaching).

E Effrenium. Exclusively
free-living. Grazes on
bacteria and unicellular
eukaryotes.

F + + Fugacium (Subclade F5).

Found in foraminifera.
Several non-symbiotic
species. Occur as transient,
low-abundance densities in
cnidarians. Other subclades
of clade F not yet fully
characterized.

G + + + Gerakladium. Basal lineage

of family. Ecologically rare.
Certain species form
association with excavating
sponges and black corals.

H + Not yet fully characterized.

I + Not yet fully characterized.

aData from Pochon et al. (2014). Representatives of most clades have also been detected in the free-living environment; some sub-clades/species
may be exclusively free-living. b Data from LaJeunesse et al. (2018). Only clade A is retained in the genus Symbiodinium.

species, much of their nutrition comes from high densities (~106 cm−2) of the endosymbiotic
zooxanthellae (Symbiodiniaceae). This is especially true of the best known and most-studied
scleractinian corals in shallow tropical and sub-tropical waters, which are responsible for
building coral reefs through formation of a calcium carbonate skeleton. The zooxanthellae
occur in the gastroderm (innermost cell layer) of the tissue (Figure 10.15). Reef-building
(hermatypic) corals lay down their skeleton over a very fine organic matrix and field observa-
tions show that a high content and activity of zooxanthellae is essential for the rapid building
of the reef structure. Under optimum growth conditions, zooxanthellae may incorporate up
to 100 times more photosynthate than they need for their own growth and reproduction—
most of this excess is transferred to the coral, where it is mainly respired. Excess carbon is
also secreted in the mucus, which provides a major source of nutrients sustaining the micro-
bial loop and benthic processes, explaining why coral reefs form highly productive “oases”
278 Chapter 10

surrounded by nutrient-poor waters. The coral produces signal molecules, which alter the
control of photosynthesis and stimulate the release of organic compounds from the zooxan-
thellae by altering membrane permeability. When photosynthesis is inhibited, for example,
by restriction of light, the uptake of calcium and subsequent secretion of calcium carbonate
is reduced. Excretion of lipids by the zooxanthellae is also important in construction of the
skeleton.

As in other symbioses, a key question is how the host acquire their symbiotic partners. The
larvae of most corals do not contain zooxanthellae (they are described as aposymbiotic), and
they must obtain zooxanthellae from the environment. In almost all cases, this occurs by
horizontal transmission in the early larval stages, but it can also occur in adults, especially
during adaptation to environmental change (see below). Brooding corals show vertical trans-
mission, transferring zooxanthellae in the eggs or brooded planula larvae. Mechanisms by
which the initial selection and uptake of specific symbionts occurs is still unclear, although
this undoubtedly involves recognition of microbe-associated patterns on the zooxanthellae
by host cell receptors. Free-living motile Symbiodiniaceae are thought to be attracted to the
coral coelenteral mouth and are then ingested by the gastrodermal cells. Juvenile colonies
appear to be relatively non-specific, with uptake of a mixture of symbiont types, which may
reflect the status of the host immune system. Later, a “winnowing” process ensures that one
symbiont type increases in abundance, so that particular clades can be characteristic of cer-
tain coral species. At this stage, the host somehow marks the less desirable Symbiodiniaceae,
so that they are recognized by lysosomes and digested. However, some corals maintain stable
relationships with multiple phylotypes (evident at clade or sub-clade level) present within the
colony. This is often considered to be an adaptive feature, as it means that the host “culti-
WHAT’S IN vates” the symbiont type with the maximal physiological performance under the prevailing
? IT FOR THE
ZOOXANTHELLAE?
environmental conditions.

Is the association between corals Sunlight is obviously essential for maximal photosynthesis, but the high UV irradiation
and their Symbiodiniaceae partners in clear, shallow waters is highly damaging. To minimize this damage, corals accumulate
really of mutual benefit? The mycosporine-like amino acids, which act as sunscreens. Within the symbiosome, the intra-
nutritional benefits for the host cellular zooxanthellae depend on a supply of CO2 to carry out photosynthesis. The coral host
are obvious, but the benefits that has efficient ATP-dependent mechanisms for capturing CO2 from the surrounding seawater.
the zooxanthellae receive in return Thus, the symbionts must “pay” the host with photosynthetic products in order to receive a
are less clear. It is often stated that continuing supply of CO2 for carbon fixation. The host has several other controls for tight
the zooxanthellae within the host regulation of photosynthesis by the zooxanthellae, including fluorescent pigments that limit
have stable access to a favorable photoinhibition of photosystem II by high light levels.
light regime and receive a reliable
source of nitrogen compounds
produced by the host and associ- Coral bleaching occurs when the host–symbiont
ated microbes. However, free-living
Symbiodiniaceae “make big sacri- interactions are uncoupled
fices”—loss of cell wall and flagella,
The importance of zooxanthellae to tropical corals is shown dramatically by the phenomenon
reduction in growth rate, and
of coral bleaching, due to a breakdown in the tight coupling of host–symbiont processes. The
tight controls on their metabolism
and reproduction—when they are combined effects of elevated sea temperature and high solar irradiation are the major triggers
acquired by the host to become for mass coral bleaching. Degradation of the photosynthetic pigments and expulsion of the
zooxanthellae. And when things go zooxanthellae makes the coral tissue transparent, revealing the white skeleton (Figure 10.13).
wrong (in bleaching), the zooxan- Loss of zooxanthellae does not necessarily lead to immediate death of the corals, because
thellae are ejected. Wooldridge they can still obtain some nutrition by capture of plankton. During bleaching of some cor-
(2010) argues that regarding this als, the activity of endolithic algae harbored in the skeleton increases, and they may provide
as mutualism is illogical as it is the coral with significant nutrition from photoassimilates via this source. Bleached corals
like saying that “dairy cattle and are able to regain a fresh population of zooxanthellae when conditions improve. However,
humans have a comparable rela- the health of the colony is severely impaired, and the corals quickly become susceptible to
tionship, because we provide them
disease or overgrowth by algae, especially when prolonged high temperatures prevail in suc-
with food and shelter and regulate
their population before we harvest
cessive years.
(exploit) their milk production.”
He proposes that the relationship Because of its importance, numerous investigations of the mechanisms of coral bleaching
would be better regarded as a have been undertaken, revealing complex interactions and diverse physiological factors that
“controlled parasitism in which the affect the process. The most favored explanation for expulsion of the zooxanthellae is that it
functioning of symbionts is con- results from photo-oxidative stress under elevated temperature, which damages the photosyn-
strained and manipulated.” thetic system and leads to leakage of reactive oxygen species into the cytoplasm of the host.
 Microbial Symbioses of Marine Animals 279

Figure 10.13 Coral bleaching.

Upper panel shows early (left) and
late (right) stages of bleaching on
the Great Barrier Reef, resulting in
the loss of pigmented zooxanthellae.
Images courtesy of Mary Wakeford,
Australian Institute of Marine Science.
Lower panel shows a representa-
tion of internal carbon cycling in
the coral-zooxanthellae symbiosis.
Photosynthesis takes place within
the algal chloroplast, with the ‘light
reactions’ occurring in the thylakoid
membranes, and the ‘dark reactions’
(CBB cycle) in the stroma. Typically,
about 95% of the photosynthates
[(CH2O)n] are transferred to the coral
host. Breakdown of the symbiosis
(zooxanthellae expulsion) is triggered
by a limitation of CO2(aq) substrate
for the dark reactions of zooxanthellae
photosynthesis, resulting in inability
to turn over ATP and NADPH (indi-
cated by red crosses). As a result, the
photosynthetic electron transport
chain becomes blocked, which dam-
ages the light-sensitive photosystems
and generates damaging reactive
oxygen species (O2–). Reprinted from
Wooldridge (2010), with permission
of John Wiley & Sons.

However, recent research indicates that nutritional mechanisms also regulate bleaching.
Nutrient availability, especially the forms and ratios of N and P can shift the relationship
between Symbiodiniaceae and the host from mutualistic to parasitic. Stable metabolic com-
patibility between the coral host and algal symbiont can ameliorate bleaching and increase
resilience to environmental stress.

It is still unclear how the damaged zooxanthellae are removed. Possible mechanisms include
expulsion of zooxanthellae by exocytosis; induction of apoptosis, in which programmed cell
death is initiated; and symbiophagy, in which the vacuole containing the zooxanthellae is
transformed into a digestive organelle. Infection by bacteria and viruses has also been impli-
cated as the cause of some types of coral bleaching, as discussed in Box 11.1.

It is possible that the coral holobiont can adapt to changing environmental conditions by “shuf-
fling” or reassortment of the mixture of symbionts already present, or by acquisition of new
types of symbiont better suited to the new situation. This “adaptive bleaching hypothesis,”
280 Chapter 10

was first proposed by Buddemeier and Fautin in 1993, and there is now considerable experi-
mental evidence to support many aspects of the concept. For example, aquarium experiments
have shown that Symbiodinium clade D (now reclassified as the genus Durusdinium) has a
better quantum yield of photosynthesis than other clades, but only at elevated temperatures.
Thus, the physiological cost of harboring a thermotolerant clade may override the apparent
benefits. Field experiments, such as transplantation of corals from cooler to warmer locations
on the Great Barrier Reef support the idea that adult corals may acquire more thermotolerant
zooxanthellae that enable them to be more resistant to bleaching.

Although zooxanthellae are vitally important in the reef-building corals, it is important to

bear in mind that most corals obtain much of their nutrition by feeding on zooplankton
via their tentacles, and some do not contain any symbionts (these are described as azoo-
xanthellate). For example, an azooxanthellate variant of the Mediterranean coral Oculina
patagonica occurs in undersea caves. In particular, thousands of species of azooxanthellate
cold-water corals, such as Lophelia pertusa, form massive colonies in the deep ocean, hun-
dreds of meters below the photic zone.

The coral holobiont contains multiple microbial partners

In addition to the zooxanthellae, the tissues and secreted mucus layer of healthy corals sup-
CAN WE HELP
? CORALS ADAPT TO
CLIMATE CHANGE?
port a diverse and dynamic community of other microbes, including other protists, bacteria,
archaea, fungi, and viruses. These contribute to coral nutrition and the health of the coral.
Bacteria have been isolated from corals since the 1970s, but rapid advances in this field
Some coral biologists have pre- did not take off until 2002 with the application of culture-independent DNA-based meth-
dicted that major reef systems will ods. Numerous studies of multiple coral species in different geographic regions have now
be irreversibly damaged during
revealed a high diversity (hundreds or thousands of phylotypes) and abundance of coral-
this century (e.g. Hoegh-Guldberg
associated bacteria.
et al., 2009). Rohwer and Youle
(2010) remind us that corals have
survived previous major environ- The surface mucus layer supports particularly high populations of diverse bacteria (106 –108
mental upheavals over millennia cells mL−1) with numerous extracellular enzymes for digestion of the rich proteins and car-
and inject a note of optimism bohydrates secreted by the coral. Their activities probably benefit the host through provi-
with the view that “The possibil- sion of vitamins, co-factors and other essential nutrients. This breakdown of coral mucus
ity of adaptation by the holobi- also provides a major contribution to the productivity of the reef, by making released DOM
ont, thanks to its algal, viral and available to planktonic microbes (p.25) Many of these bacteria are also important in pro-
microbial partners, is a source of tecting the coral from colonization by opportunist pathogens, although shifts in commu-
hope.” Madeleine van Oppen of nity composition can lead to dysbiosis and disease, as discussed in Chapter 11. In some
the Australian Institute of Marine
coral species, microscopy of thin sections of coral tissue reveals colonization of the epithe-
Science and the late Ruth Gates
lial and gastrodermal tissues by bacteria-like structures which are sometimes aggregated
of the Hawaii Marine Laboratory
championed the proposal that we into closely packed groups (illustrated in Figure 10.14), although it is not clear if they are
might “assist evolution” by breed- contained within a host membrane like the bacteriocytes found in other endosymbioses.
ing corals that are more resilient to Application of FISH with bacteria-specific probes is the ideal technique to confirm the
future climate change. Although identity of these structures; however, this has been difficult due to background autofluores-
this requires a careful risk–benefit cence and non-specific binding by various structures within coral tissue. Refinements of
analysis, van Oppen et al. (2015) techniques to overcome these limitations are now yielding new insights into these aggre-
compare this approach to the suc- gates. The bacteria most commonly found belong to the Alpha- and Gammaproteobacteria,
cessful use of genetic and epigen- Cyanobacteria, Bacteroidetes, and Actinobacteria. Besides bacteria, various other microbes
etic modification in commercial are found. Archaea have been less well studied, but members of the Crenarchaeota and
agriculture and aquaculture. Torda
some other groups have been found to have important roles in some corals. Endolithic fungi
et al. (2017) and Donelson et al.
(2018) set out the experimental
are regularly identified in the skeleton, and ciliates and other alveolate protists often occur
approaches needed to determine in the mucus and surface tissue. Distinguishing the normal inhabitants of healthy corals
whether such “transgenerational and those associated with the onset of disease can be very difficult, as discussed in the
plasticity” can protect corals over next chapter. Many early investigations showed a high specificity of bacterial associations
several generations. Many groups with particular coral species, across broad geographic distances. However, as with the per-
are now working on different ceived host specificity of particular Symbiodiniaceae clades, more focused examination of
approaches, and guidelines to coral colonies shows that this is not so straightforward. The diversity and relative abun-
evaluate the risks and benefits of dance of different bacterial phylotypes can be affected by the physiological status of the
novel ecological, genetic, and envi- coral host, location within the individual polyps or colony, and environmental conditions.
ronmental interventions to help
Therefore, recent analyses have attempted to identify the “core microbiome,” which can be
corals survive have been agreed
defined as the phylotypes that appear to be stable and consistent in all cases, irrespective of
(NAS, 2019).
whether they are dominant or rare, discussed in Box 10.3. Identifying the core microbiome
 Microbial Symbioses of Marine Animals 281

is particularly important to understanding the metabolic roles and core functions provided
THE HOLOGENOME
by the microbes in the holobiont.
i CONCEPT OF
EVOLUTION IS
Recent metagenomic analyses have confirmed a very high diversity of metabolic functions
CONTROVERSIAL
associated with coral microbes, including nitrogen cycling, sulfur cycling, photosynthesis,
breakdown of complex proteins, and polysaccharides. Genes for stress responses, virulence, Margulis (1991) first used the term
DNA repair, and antibiotic resistance also seem to be important. The chemical properties of holobiont to describe an assem-
coral mucus, as well as intercellular signaling (such as quorum sensing, p.102) and antagonis- blage of different species of organ-
tic interactions between the microbes themselves, are important factors in the establishment isms that function collectively as an
of specific microbiota, although the protective function is easily disturbed, leading to disease ecological unit. The term was little
used until Rohwer et al. (2002)
(see Box 11.1). Of particular significance is the discovery that a diverse group of resident
used it for the association of coral
nitrogen-fixing bacteria (including Cyanobacteria and vibrio-like Gammaproteobacteria) are hosts with diverse microbes and it
present in many corals. Provision of nitrogen compounds to the zooxanthellae may help to quickly became widely accepted.
explain the high productivity of corals in the apparently nitrogen-limited environments that The hologenome concept also
they inhabit. The role of coral bacteria in the global cycling of sulfur has also recently been emerged from coral microbiology
shown to be significant (see Box 9.2). Although present in low abundance, archaea seem to research, as an extension of the
be especially important in recycling waste products in the holobiont, by nitrification and coral probiotic hypothesis (see
denitrification of ammonia in the mucus layer. Study of cold-water, deep-sea corals such as p.294). Rosenberg et al. (2007)
Lophelia pertusa is much more difficult, but it appears that they also have specific micro- proposed that all the genetic com-
biota, including archaea. ponents of a host and its microbial
partners must be considered as a
unit of natural selection in evolu-
Fungi are also an important component of the coral holobiont, being particularly associated
tion. This has provoked consider-
with the surface layer and growing within the calcium carbonate skeleton of hard corals. able differences of opinion. For
These endolithic fungi are mostly related to terrestrial fungi and metagenomic studies have example, Moran and Sloan (2015)
revealed a high diversity of chytridiomycetes, ascomycetes, and basidiomycetes. Some stud- and Douglas and Werren (2016)
ies have identified obligate marine ascomycetes. Heterokont microalgae, labyrunthulids, and point out the confusion caused by
thraustochytrids are also common and grow profusely in coral mucus. inconsistent interpretations and
the importance of mutualistic and
Viruses are also abundant in corals, with over 60 different virus families detected in coral antagonistic interactions (fitness
viromes. Many of the viruses are phages that must play a key role in healthy corals by regu- conflicts) that undermine the
lating the bacterial population, while others have been found associated with the zooxan- concept. However, Rosenberg and
Zilber-Rosenberg (2018) cite con-
thellae and coral animal tissue. There is also evidence for widespread integration of phages
siderable theoretical and experi-
into the genomes of their hosts (prophages, see p.203). Environmental changes induce major mental evidence in support of their
shifts in the abundance and activity of viruses, and this is often associated with the onset of theory from microbiome research
disease as discussed in Chapter 11. in numerous animals. Morris (2018)
reviews the ongoing controversy
between proponents and critics
Zooxanthellae boost the growth of giant clams of the concept, considering that it
is stimulating the development of
Anyone who has dived or snorkeled on the Great Barrier Reef or Pacific Islands will have
testable hypotheses “to produce
marveled at the size and beautiful colors of the shell mantles of the giant clams such as a way of looking at life which is
Tridacna gigas and related species, which can reach enormous sizes up to 300 kg. How simultaneously exciting, confusing,
can these animals grow so rapidly when the waters they inhabit are very poor in nutrients? and challenging.”
The siphon tissue around the mantle is packed with endosymbiotic zooxanthellae. In this
case, there are additional pigments of many different colors that provide protection to the
zooxanthellae from the high ultraviolet irradiation in the clear waters that the clams inhabit.
Tridacna has evolved remarkable anatomical and behavioral modifications to optimize the
benefits from their photosynthetic symbionts. This type of clam is unusual among bivalve
molluscs because it lives above the sediment. During evolution, the orientation of the internal
organs has twisted through 180°, allowing the siphon to be at the top of the body so that the
maximum surface area colonized by the zooxanthellae is exposed to the light. Photosynthesis
is further enhanced by specialized hyaline structures, which focus light onto the zooxanthel-
lae. Experimental studies have shown that, under the right conditions, the zooxanthellae can
provide all the clam’s carbon requirements through the release of small molecules such as
glycerol and organic acids. However, as in many corals, acquisition of nutrients by hetero-
trophic feeding is also important. Tridacnid clams have particularly efficient filter-feeding
mechanisms and can extract significant quantities of plankton from the water, even though
the plankton is in low concentrations because of the oligotrophic nature of the environment.
Thus, it seems that under natural conditions, the clams rely on both autotrophic and hetero-
trophic sources for their nutrition, with between 35% and 70% of the carbon requirements
coming from photosynthetic activity of the zooxanthellae.
282 Chapter 10

BOX 10.3 RESEARCH FOCUS

Do corals have a core microbiome?

Diversity of the coral holobiont is revealed by high through- colonization, as shown in Figure 10.14. Due to exposure to very
put sequencing (HTS). We now know that corals associate with different environmental conditions of light, temperature, or water
hundreds or thousands of different kinds of highly diverse bacteria flow, variability occurs between individual colonies and different
(and to a lesser extent, archaea, fungi, and protists). There is strong parts of a colony, as well as between the different compartments
evidence that some of these microbes play critical functional roles of individual polyps. Ainsworth et al. (2015) extracted DNA from
in the coral holobiont, through functions including photosynthesis, replicate samples of three species of Pacific corals, using PCR
protection against pathogens, and nitrogen and sulfur cycling. In amplification and 454 tag sequencing (p.51) of 16S rRNA genes.
recognition of the fact that the microbiome plays a key role in the Laser microdissection was used to separate the niche microhabitats
health of corals and in their resistance and resilience to environ- of host tissues, and significant differences were evident between
mental pressures, numerous research groups have used HTS to the whole coral community (holobiont), the skeleton-associated
catalogue the microbes associated with diverse coral species. Only tissue, and the community in cells containing endosymbiotic
some of these have been cultured, so diversity is usually assessed Symbiodiniaceae. In all corals sampled, the relative abundance of
by grouping related sequences into OTUs—often interpreted as the phylotypes within the different niche habitats differed significantly.
equivalent of a species (see p.129). Recently, refinement of sequenc- Results for the coral Acropora granulosa showed the communities
ing methods allows finer resolution and there are proposals to aban- to be very diverse, with 1508 bacterial phylotypes (from 16 bacte-
don OTUs in gene surveys in favor of exact or amplicon sequence rial families) found in the whole community. Of these, 159 were
variants (ASVs, Callahan et al., 2017). A key aim is to identify identified as the core microbiome (defined as present in >30% of
a core set of microbes that is always present and likely to have a all samples), with 76 and 71 phylotypes identified in the symbiotic
particularly critical functional role. However, Blackall et al. (2015) tissue and endosymbiont microbiomes, respectively. Only 15 phylo-
and Hernandez-Agreda et al. (2017) emphasize the difficulties in types were found in all three microbiomes, but 41 were found exclu-
interpreting results due to a huge disparity in methods of sampling, sively in the endosymbiont microbiome. Further analysis showed
replication, and DNA amplification and sequence analysis. that most of the 159 phylotypes in the A. granulosa microbiome
had very low relative abundance. With higher cut-off values used to
Corals contain niche habitats for microbial colonization. Many define the microbiome, only 0.09% of all phylotypes identified were
profiling studies have not fully recognized the fact that a coral col- present in 90% of the coral hosts, 0.5% in 75% were in 50% of the
ony provides a variety of very different niche habitats for microbial hosts, and 2.3% were in 50% of the hosts. Thus, the vast majority

Figure 10.14 Diagrammatic representation of a coral polyp (left) with cross-section showing the habitats of microbes
within the mucus and tissues (not to scale). Zooxanthellae are represented as large gold cells within the gastroderm.
Aggregates of bacteria observed in some corals are shown as black clusters.
 Microbial Symbioses of Marine Animals 283

BOX 10.3 RESEARCH FOCUS

of phylotypes amplified from individual corals were not associated pistillata) sampled across seven major geographical regions. This
with all colonies. Further analysis of the data showed that different intimate association in 85% and 79%, respectively, of these cor-
phylogenetic groups are restricted to particular habitats within the als led Neave and colleagues to consider it a member of the core
holobiont. In all three corals examined, members of the Rhizobiales, microbiome. They identified fine-scale specificity of association
Caulobacterales, and Burkholderiales—groups that contain known with different Endozoicomonas types and speculated that this
symbionts in other metazoans and plants—were associated with the bacterial-coral association might resemble the Symbiodiniaceae-
symbiotic microbiome, but not the whole coral community. In con- coral partnership, in which shuffling—or acquisition of different
trast, bacteria belonging to the Rickettsiales and Rhodobacterales symbiont strains—is associated with response to different envi-
were amplified from the whole coral microbiome, but not from sym- ronmental conditions. Surprisingly, in their study Ainsworth et al.
biotic tissue. The authors identified just seven bacterial phylotypes (2015) amplified sequences affiliated with Endozoicomonas in the
that were universally present in the core microbiome of the three holobiont genome, but not in the internal tissues, implying that the
corals analyzed, even though the samples spanned a wide range of bacterium resides in the surface mucus or skeleton. In their study of
depth and geographic area. Another study by Hernandez-Agreda et Corallium, van de Water et al. (2016) found that Endozoicomonas
al. (2016) found a core of eight persistent phylotypes in the widely only accounted for 3.4% of the core microbiome. In a meta-anal-
distributed Pacific coral Pachyseris speciosa at different scales and ysis of coral-associated bacterial sequences from numerous stud-
depth gradients, using an 80% cut-off. This core contained mem- ies, Blackall et al. (2015) found that the relative abundance of
bers of the classes Actinobacteria, Alpha-, Delta-, Epsilon-, and Endozoicomonas varied substantially, from below detection to 43%
Gammaproteobacteria, and the phylum Bacteroidetes. The study of the community, confirming that its presumed role as universal
by van de Water et al. (2016) found a core of 12 bacterial OTUs member of the core microbiome of corals is dubious. It appears
accounting for 95% of the overall community that were consis- to be rare or undetectable in deep-sea corals. The functional role
tently associated with the gorgonian Corallium rubrum from five of Endozoicomonas is unknown, although various possibilities
reef sites persisting over a three-month period. In Corallium the include nutrition, sulfur cycling, and production of protective
core was dominated by members of the Spirochaetales and a rela- compounds.
tively unknown family in the Oceanospirillales. Ainsworth et al.
(2015) and Hernandez-Agreda et al. (2017) discuss the precautions Co-evolution of the host genome and microbiome. The huge
needed when choosing the parameters that define the core micro- diversity of corals and their associated microbiomes presents a
biome. Using high or low relative abundance, without considering major challenge to understanding how these relationships evolved.
persistence, can be greatly affected by methodological and biologi- How are the microbial communities in the different coral anatomi-
cal factors. cal compartments shaped by phylogeny and functional traits of the
hosts and by environmental variables? Is there evidence of phylo-
From their study of P. speciosa Hernandez-Agreda et al. (2016) symbiosis, i.e. a correlation between host phylogenetic relationships
concluded that this generalist coral, which occurs in many differ- and composition of the microbial community—does the microbi-
ent reef habitats, can be associated with a vast number of distinct ome recapitulate the host’s evolution? To answer these questions,
bacterial phylotypes. Interestingly, they found that samples from Pollock et al. (2018) used standardized sampling, sequencing, and
mesophotic reefs (60–80 m depth) consistently associated with a complex computational methods to analyze 16S rRNA gene librar-
higher number of bacterial phylotypes. They proposed that there ies from the mucus, tissue, and skeletons of diverse Australian cor-
is a ubiquitous core microbiome of a very small number of symbi- als across a broad geographic range and to compare them to host
otic host-selected bacteria, a microbiome of <100 phylotypes that phylogenetic trees based on mitochondrial genes. They found strong
fills specific functional niches, and a highly variable community of evidence for phylosymbiosis between corals and their microbiomes
thousands of phylotypes. reflecting the speciation of modern corals that began 25–65 MYA.
Surprisingly, the endolithic skeleton communities were richer in
Is Endozoicomonas a core associate? A bacterium of particular microbial diversity than the tissue and showed the strongest signals
interest in microbiome studies is Endozoicomonas, because numer- of long-term phylosymbiosis. Host specificity in the mucus micro-
ous previous studies have identified this as a potential core symbi- biome seemed to be due to relatively recent divergences. Further
ont of different corals, as well as many other marine invertebrates analysis of a wider, global range of corals in different habitats,
(Neave et al., 2017). Neave and colleagues showed the presence of including deep-sea azooxanthellate species, and refinement of the
Endzoicomonas by CARD-FISH deep within the tissues of two analytical methods will reveal further insights into the evolution of
different species of coral (Pocillopora verrucosa and Stylophora coral holobionts.
284 Chapter 10

Some fish and invertebrates employ

symbiotic bacteria to make light
Another type of cooperative association exists between bacteria and animals, in which the
DEEP-SEA CORALS
i ALSO CONTAIN
MICROBIAL
activities of the bacteria confer a behavioral and ecological benefit to the host, rather than a
direct nutritional one. Bioluminescence is the emission of blue or green light from oxygen-
ASSEMBLAGES utilizing reactions via the luciferase enzymes (Figure 3.17). Bioluminescence occurs very
widely in the oceans, particularly in animals that inhabit deep waters where no sunlight
Many coral species do not rely penetrates or in animals that are active in shallow waters at night. Bioluminescent animals
on zooxanthellae. Some occur in use light emission for avoidance or escape from predators, attracting prey, or as a means of
cold, dark conditions up to 2000
communication (such as mate recognition). Bioluminescence seems to have multiple evo-
m deep. They occur worldwide,
with the best studied locations
lutionary origins, as there is a very wide diversity of enzymes and substrates for the biolu-
being large mounds in the North minescent reaction mechanism. Apart from light emission, the only common feature is the
Atlantic, where they form a criti- requirement for oxygen, and it is possible that bioluminescence evolved originally as an
cal, threatened habitat for many antioxidative mechanism. Bioluminescent animals often possess complex structures such
organisms. Because of the difficulty as lenses, filters, and shutters to control or modify the light emitted. In most bioluminescent
of sampling, much less is known animals, the light is generated by the animal itself within specialized cells of the animal.
about their associated microbes However, in some invertebrates and fish, the light is produced by symbiotic bacteria living
than other corals, but genomic within specialized organs.
analyses have revealed their diver-
sity and function (e.g. Penn et al.,
In some cases, these bioluminescent bacteria can be isolated and will emit light in culture,
2006; Kellogg et al., 2009). Using
13C and 15N tracer techniques, but some have not yet been cultured and have been identified only by their genetic sequences.
Middelburg et al. (2015) dem- Another clue to the bacterial origin of bioluminescence is that bacteria usually emit light
onstrated chemoautotrophy, N2 continuously over a period, whereas light produced by eukaryotic enzymes tends to occur
fixation, and recycling by microbial as brief flashes. Final proof of the role of bacteria in light emission by animals is the use of
symbionts in one of the most-stud- an assay for luciferase in the presence of reduced flavin mononucleotide, as this reaction is
ied species, Lophelia pertusa. Gene unique to bacterial bioluminescence. Bacterial bioluminescence occurs in a wide range of
analysis provided further evidence phylogenetically distinct animals, and the associations range from relatively unspecialized
that these functions are carried out facultative colonization of the intestinal tract or organs derived from it, through to obligate
by a core microbiome (Kellogg et interactions with specialized external light organs.
al., 2017). Some deep-water corals
found in the Gulf of Mexico and
The bacteria associated with symbiotic bioluminescence have probably evolved via adaptive
Red Sea contain hydrocarbon-
degrading bacteria which may
radiation with at least 17 origins in fish and two origins in squid over the last few hundred
contribute to coral nutrition by million years. Over 460 host species are known to harbor bioluminescent bacterial sym-
provision of organic compounds bionts. In all cases, dense communities of extracellular bacteria occur in specialized light
(Kellogg et al., 2017; Röthig et al., organs contained in tubules that communicate with the intestine or the external environ-
2017). ment. Release of bacteria is an important component in the control of the populations of
these symbionts. All known types of bacteria in bioluminescent symbioses are members
of the Vibrionaceae. The best known are Photobacterium phosphoreum, Photobacterium
leiognathi, and Aliivibrio fischeri (formerly Vibrio fischeri; see p.103), which are facultative
symbionts and are readily culturable. They occur as free-living bacteria in numerous habitats
and are transmitted horizontally, being acquired by host larvae from the environment by
each generation. In the animal host, these bacteria grow more slowly than in culture. It is in
the host’s interest to maximize bioluminescence but to minimize diversion of nutrients to
the bacteria; restriction of oxygen supply or iron limitation may be important factors in this
process. In other cases (see below), the bacterial symbionts have never been cultivated and
are probably obligate symbionts.

Bacterial light organs in flashlight fishes (members of the family Anomalopidae) are the
largest in any fish relative to body size. The light organs in these strictly nocturnal tropical
reef fish are located below the eyes and emit a constant blue light (Figure 10.16). In very
dark conditions, the forward illumination is bright enough to allow the fish to seek out their
zooplankton prey. Rapid blinking, which is affected by the ambient light, may serve a func-
tion in communication with other members of the species (possibly for group interactions or
mating).

Anglerfishes are a diverse group of species belonging to nine families in the suborder
Ceratioidei. Most are solitary animals found in the deep sea. The females have a light organ
(esca), encased in a crystalline reflector, at the end of a “fishing rod” projecting from a modi-
fied dorsal fin ray on the head (Figure 10.15). This acts as a lure to attract prey near to the
 Microbial Symbioses of Marine Animals 285

BOX 10.4 RESEARCH FOCUS

Genomic insights into the evolution of bioluminescent symbionts of fish

Flashlight fish symbionts have a free-living phase. It has long daylight. This provides a possible explanation for evolution of symbi-
been known from 16S rRNA gene sequences, that the biolumines- ont gene loss due to population bottlenecks resulting from symbiont
cent symbionts of anglerfishes and flashlight fishes are members of co-divergence at the level of the population rather than the individual.
the Vibrionaceae. However, repeated attempts to culture these sym-
bionts have failed. Does this mean that they are obligate symbionts? Angler fish symbionts. Similarly, numerous attempts to culture
Hendry and Dunlap (2011) sought to answer this question by extract- the symbiotic bacteria of ceratioid deep-sea anglerfish have been
ing bacterial DNA from the light organs of Anomalops katopraton unsuccessful. Hendry et al. (2018) isolated bacterial DNA from the
(Figure 10.15). The bacterium was identified as a clade nested within light organs of specimens of two distantly related species of angler-
the Vibrionaceae, with high divergence from other members and fish, Melanocetus johnsonii (Figure 10.16) and Cryptopsoras coe-
named “Ca. Photodesmus katoptron.” Next, the authors assembled its sii. Comparison of the symbionts’ genome sequences placed them
draft genome and found that it was only 1 Mb, about a fifth of the size within the Enterovibrio clade, and they were given the Candidatus
of other Vibrionaceae genomes and lacking almost all genes for amino names “Enterovibrio luxaltus” and “Enterovibrio escacola,” respec-
acid synthesis and energy-yielding pathways (Hendry and Dunlap, tively. Hendry and colleagues noted that these species are sepa-
2014). This supports the idea that that it is an obligate symbiont that rated from other taxa by long branches, indicating that they may
has become dependent on its host for nutrition. Surprisingly, however, be evolving at a rate several times faster than their relatives. This
the genome retains almost all the genes for synthesis of a robust cell was supported by molecular clock analysis of genes and predicted
wall, flagellar motility, and chemotaxis. These are almost always lost amino acid sequences. The total genome sizes (2 and 2.6 Mb) of
in previously described obligate symbionts and pathogens. It seems both anglerfish symbionts are about 50% smaller than those of their
unlikely that these gene pathways would be retained, while so many free-living relatives, but if only predicted functional protein-coding
others have been lost, unless they are required for part of the bacte- genes are considered, the difference is even more striking. As found
rium’s lifestyle. The idea that they are used outside the host fits with in the flashlight fish symbiont, Hendry et al. (2018) conclude that this
previous observations that the symbionts are shed from pores in the genome reduction is due to relaxed selection and high genetic drift
light organ and can be found for a short time in seawater. Although the due to the close association of the bacteria within the light organ
absence of so many metabolic genes would prevent the released bac- of their hosts. Also, as seen with the “Photodesmus” flashlight fish
teria from growing and multiplying, the robust cell wall would permit symbiont, the anglerfish symbionts have a highly reduced comple-
survival in the seawater and motility and chemotaxis could be used ment of genes encoding amino acid synthesis and energy metabo-
to seek and infect larvae to establish a new symbiosis. Hendry and lism, membrane transport, and cell signaling, but have retained a
Dunlap (2014) review the knowledge of the life history and behavior large complement of genes for cell wall synthesis, motility, and
of Anomalops gained from earlier studies by Haygood (1993) and oth- chemotaxis. Both symbiont genomes contain a very large number
ers, which suggest a pseudovertical transmission due to proximity of of pseudogenes and show large proliferation of transposable ele-
eggs, larvae, and adults when the fish group together in caves during ments that appear to have occurred at different times. Based on this,
Hendry et al. (2018) concluded that genome reduction is still going
on in these symbiont genomes. It is surprising that this rapid evolu-
tion towards tighter relationship is still occurring, even though the
fish themselves evolved about 100 MYA and the bacteria are not
constrained by a purely host-associated, intracellular lifestyle. How
the symbionts are transmitted between generations remains a mys-
tery—the pseudovertical transmission suggested for the flashlight
fish symbiont seems untenable, since anglerfishes live a very solitary
existence in the deep sea.

Figure 10.15 (a) Side view of Anomalops katoptron showing

position of the light organ. This appears as a white patch
because of the guanine crystal reflector on the backside
of the light organ and flash photography. (b) Enlarged
view of light organ. (c) Front view of both sub-ocular light
organs of showing bioluminescence during the night.
Scale bars = 5 mm. Reprinted from Hellinger et al. (2017), Figure 10.16 The anglerfish Melanocetus johnsonii. Image
CC-BY-4.0. courtesy of Edie Widder, ORCA.
286 Chapter 10

jaws, and light emission is controlled by the supply of oxygenated blood to the esca. Some
LURING MALES
i INTO LIFE-LONG
ATTACHMENT
species have several escae and filaments. In these groups of fishes, the symbionts have never
been cultured and appear to be obligate symbionts, although recent work suggests the bacte-
ria are released into the seawater and in some cases are transmitted to larvae via a pseudover-
Anglerfishes show an unusual sexual tical mode, as discussed in Box 10.4).
dimorphism. Only the females pos-
sess the bioluminescent organ. The
males are dwarfs that reach sexual The bobtail squid uses bacterial
maturity soon after the larval stage bioluminescence for camouflage
and they probably never feed. They
have large eyes and a large olfactory The Hawaiian bobtail squid (Euprymna scolopes) inhabits shallow reefs, hiding in the
organ, so it is assumed that they sediment during the day and emerging to feed at night (Figure 10.17A). The squid emit
locate the females by a combination light from their ventral surface, adjusted to match the intensity of moonlight. This coun-
of looking for light from the flashing ter-illumination camouflage helps them to be less visible to predators from below. A
esca and detection of pheromones.
highly specific association with a certain strain of bioluminescent Aliivibrio fischeri is
It is thought that the male may
responsible. (Note that many papers on this topic still use the original classification Vibrio
detect the shape and flash patterns
of the esca to ensure he has found fischeri.) Newly hatched squid are aposymbiotic and acquire this particular bacterium
a female of the same species. The exclusively from the surrounding seawater within a few hours, even though it is present
male bites the underside of the there at very low densities—just a few hundred cells per mL—comprising just ~0.01%
female; his body then fuses with hers of the total bacterial population. During respiration and movement, ventilation by the
and all organs apart from the testis tiny squid flushes a miniscule amount of seawater—just 1 µL or so—in and out of their
degenerate. The female now carries mantle cavity about twice per second, and this brings bacteria into contact with epithelial
a “sexual parasite” with a supply of fields covered in cilia. The host responds to the presence of bacteria by the production
sperm for fertilization of her eggs. of mucus which leads to an aggregation of cells over three tiny pores leading into each
side of the nascent light organ (Figure 10.17B). Motile A. fischeri cells out-compete other
members of this mixed population and swim through a duct against the outward flow of
water and mucus produced by ciliated cells before reaching the crypts of the light organ.
Elucidation of the full genome sequence of A. fischeri and subsequent mutation studies
have revealed the mechanisms involved in this critical stage of colonization. Of special
importance is the regulation of expression of syp genes associated with polysaccharide
Figure 10.17 The Euprymna synthesis, biofilm formation, and aggregation outside the squid light organ. A complex
scolopes-Aliivibrio fischeri symbiosis. network of interconnected regulatory systems controls colonization, with multiple inter-
A. The adult host. B. Outline of the actions between genes for structural (flagella, pili, polysaccharides) and regulatory (two-
host’s body, superimposed over a component regulators, quorum sensing, and other signaling) processes in the formation
laser-scanning confocal micrograph
(LSM) of the nascent light organ,
indicating its relative size and posi-
tion within the mantle cavity. The
organ is circumscribed by the poste-
rior portion of the excurrent funnel
(dotted white lines). Ventilatory
movements of the host draw ambi-
ent seawater (blue arrows and lines)
containing A. fischeri cells into the
mantle cavity. The water travels
into the funnel where, before being
vented back into the environment,
it encounters complex ciliated fields
(bright green) on the lateral sur-
faces of the organ. The fields entrain
water into the vicinity of pores on
the light organ surface. C. Higher-
magnification LSM of one side of a
hatchling light organ, showing the
location of the three pores (arrows)
that lie at the base of the append-
ages of each ciliated field. Credits:
A. Reprinted from McFall-Ngai
(2014), CC-BY-4.0. B, C. Reprinted
with permission from Nyholm et al.
(2000). Copyright (2000) National
Academy of Sciences, USA.
 Microbial Symbioses of Marine Animals 287

of biofilms. Genome analysis has revealed the importance of a single gene encoding a
LITTLE SQUID—
master regulator of these pathways in determining the host specificity of the particular
strains of A. fischeri that colonize Euprymna. Production of nitric oxide, reactive oxygen i BIG IMPACT
species, and other antimicrobial responses of the host also play a key role in this selection The pioneering study of the squid–
process. In the crypt spaces, peptidoglycan and lipopolysaccharide from the bacterial cell vibrio symbiosis has been car-
wall initiate a series of changes to the development of the growing light organ, starting ried out since the 1990s through
with apoptosis of the ciliated epithelium. The monoculture of A. fischeri within the larval partnership between Ned Ruby,
host is established within as little as 2 h from hatching. After adhesion to the crypt cells, who works on the bacterium, and
Margaret McFall-Ngai, who studies
the bacteria grow rapidly (about three doublings per hour) for the first 10–12 h until they
the developmental biology of the
reach a density in the crypt fluid of about 1011 bacteria mL −1. Once this critical density
host. Their many graduate students
is reached, an autoinducer of bioluminescence accumulates, leading to initiation of light have gone on to form their own
emission via quorum-sensing regulation of the lux operon (see Figure 3.17). Once estab- research teams investigating this
lished, the bacteria grow more slowly (about 0.2 doublings per hour). The host is able to topic—described as the “F1 and
detect the production of light by its symbionts and experimental infection with non-light- F2 generations” by McFall-Ngai
emitting mutants results in their rejection by the host. Even more surprising, however, is (2014). This exceptional network
the finding that the bioluminescent strains of A. fischeri isolated from E. scolopes do not of collaboration has yielded many
emit light outside their host, raising the possibility that the host has some mechanism for remarkable results, and hundreds
actively selecting for these “dim” strains. of research papers have been
published on diverse aspects of
the host–symbiont interactions. It
The established symbiosis in the adult squid has a distinct circadian rhythm. Each morn-
has provided an ideal model for
ing, in response to daylight, the squid squeezes out the contents of the light organ—this
the study of symbiosis because
is a thick paste of mucus, bacteria, and macrophage-like cells—so that over 90% of the the host is easily maintained in
bacteria are expelled. This behavior of the squid ensures that it maintains a “fresh” active culture and the symbiont can be
culture of A. fischeri that will build up to the high density required for bioluminescence, readily analyzed and manipulated
coinciding with the onset of darkness when the squid emerges from its hiding place. It also by genetic techniques. This little
provides regular seeding of the environment with the specific strain of A. fischeri, which squid and its bacterial partner have
ensures horizontal transmission of the symbiont to juvenile squid. This has a clear competi- provided a model that has tran-
tive advantage for the bacterium, which can maintain larger population sizes than if it were scended marine biology and led to
entirely free-living. major changes in the way that we
think about the broader context of
the role of microbes in health, dis-
We could say that the adult squid host “tolerates” the presence of A. fischeri for a while, but
ease, development, and evolution
the daily expulsion of the bacteria from the light organ indicates that a process akin to immu- of animals, including humans.
nological responses to the presence of a pathogen takes over. Numerous A. fischeri genes
closely resemble those encoding toxins and surface-associated virulence factors associated
with pathogenicity in other vibrios, including Vibrio cholerae, V. parahaemolyticus, and V.
vulnificus (see Chapter 12). Studies of gene expression show that a key set of virulence-like
bacterial genes are switched on just before dawn, and this correlates with the appearance in
the host tissue of blebbing and effacement of the light organ crypt epithelial surface, very
similar to what is seen during infection of the human gut by some pathogens. The bacteria
vary the expression of genes associated with utilization of different substrates throughout the
diel cycle. After the dawn expulsion, the remaining bacteria regrow, upregulating genes for
the anaerobic metabolism of glycerol derived from the host membranes. After 12 h, genes
associated with fermentation of chitin are upregulated.

Conclusions
Since the discovery of hydrothermal vent communities in 1977 and the realization that the
animals found there depend for their nutrition on chemosynthetic endosymbiotic bacteria,
knowledge of the range and diversity of mutualistic associations between microbes and ani-
mals has expanded enormously. The examples provided here should convince the reader
that such symbioses, far from being an exceptional occurrence in specialized habitats, occur
in diverse habitats and in many animal phyla. Frequently, multiple microbial partners are
present, providing the host with fixed carbon and nitrogen or performing other essential
functions. As the use of genomic techniques to explore these relationships is expanded, we
are gaining deep insight into their molecular basis and the evolutionary adaptations of both
the hosts and their microbial partners, as well as the importance of these animal–microbe
associations in the function of ecosystems.
288 Chapter 10

References and further reading Goffredi, S.K., Jones, W.J., Erhlich, H. et al. (2008) Epibiotic bacteria
associated with the recently discovered Yeti crab, Kiwa hirsuta. Environ.
Books and general reviews Microbiol. 10: 2623–2634.
Apprill, A. (2017) Marine animal microbiomes: Toward understanding Grzymski, J.J., Murray, A.E., Campbell, B.J. et al. (2008) Metagenome
host–microbiome interactions in a changing ocean. Front. Mar.Sci. 4: 222. analysis of an extreme microbial symbiosis reveals eurythermal adaptation
and metabolic flexibility. Proc. Natl. Acad. Sci. USA 105: 17516–17521.
Bright, M. & Bulgheresi, S. (2010) A complex journey: Transmission of
microbial symbionts. Nat. Rev. Microbiol. 8: 218–230. Harmer, T.L., Rotjan, R.D., Nussbaumer, A.D. et al. (2008) Free-living
tube worm endosymbionts found at deep-sea vents. Appl. Environ.
Douglas, A.E. & Werren, J.H. (2016) Holes in the hologenome: Why host-
Microbiol. 6: 3895–3898.
microbe symbioses are not holobionts. mBio 7: e02099-15.
Heide, T. van der, Govers, L.L., de Fouw, J. et al. (2012) A three-stage
Duperron, S. (2017) Microbial Symbioses. ISTE Press/Elsevier, Oxford.
symbiosis forms the foundation of seagrass ecosystems. Science 336:
Fisher, R.M., Henry, L.M., Cornwallis, C.K. et al. (2017) The evolution of 1432–1434.
host-symbiont dependence. Nat. Commun. 8: 15973.
Higgs, N.D., Newton, J., & Attrill, M.J. (2016) Caribbean spiny lobster
Hurst, C.J. (ed.) (2016) The Mechanistic Benefits of Microbial Symbionts. fishery is underpinned by trophic subsidies from chemosynthetic primary
Springer International Publishing, Switzerland. production. Curr. Biol. 26: 3393–3398.
Margulis, L. (1991) Symbiosis as a source of evolutionary innova- Jäckle, O., Seah, B.K., Tietjen, M. et al. (2019) Chemosynthetic symbi-
tion: Speciation and morphogenesis. In: Margulis, L., Fester, R., eds. ont with a drastically reduced genome serves as primary energy storage
Symbiogenesis and Symbionticism. MIT Press, Cambridge, pp. 1–14. in the marine flatworm Paracatenula. Proc. Natl. Acad. Sci. USA 116:
Moran, N.A. & Sloan, D.B. (2015) The hologenome concept: Helpful or 8505–8514.
hollow? PLoS Biol. 13: e1002311. Kleiner, M., Woyke, T., Ruehland, C., & Dubilier, N. (2011) The Olavius
Morris J.J. (2018) What is the hologenome concept of evolution? algarvensis metagenome revisited: Lessons learned from the analysis
F1000Research 7: 1664. of the low-diversity microbial consortium of a gutless marine worm. In
Handbook of Molecular Microbial Ecology II. John Wiley & Sons, Inc.,
O’Brien, P.A., Webster, N.S., Miller, D.J., & Bourne, D.G. (2019) Host- Hoboken, NJ, pp. 319–333.
microbe coevolution: Applying evidence from model systems to complex
marine invertebrate holobionts. MBio 10: e02241-18. Klose, J., Aistleitner, K., Horn, M. et al. (2016) Trophosome of the Deep-
sea tubeworm Riftia pachyptila inhibits bacterial growth. PLoS One 11:
Raina, J.B., Eme, L., Pollock, F.J. et al. (2018) Symbiosis in the microbial e0146446.
world: From ecology to genome evolution. Biol. Open 7: bio032524.
Klose, J., Polz, M.F., Wagner, M. et al. (2015) Endosymbionts escape dead
Rosenberg, E., Koren, O., Reshef, L. et al. (2007) The role of microor- hydrothermal vent tubeworms to enrich the free-living population. Proc.
ganisms in coral health, disease and evolution. Nat. Rev. Microbiol. 5: Natl. Acad. Sci. USA. 112: 11300–11305.
355–362.
König, S., Gros, O., Heiden, S.E. et al. (2016) Nitrogen fixation in a chemo-
Rosenberg, E. & Zilber-Rosenberg, I. (2018) The hologenome concept of autotrophic lucinid symbiosis. Nat. Microbiol. 2: 16193.
evolution after 10 years. Microbiome 6: 78.
König, S., Le Guyader, H., & Gros, O. (2015) Thioautotrophic bacterial
Schwartzman, J.A. & Ruby, E.G. (2016) Stress as a normal cue in the sym- endosymbionts are degraded by enzymatic digestion during starvation:
biotic environment. Trends Microbiol. 24: 414–424. Case study of two lucinids Codakia orbicularis and C. orbiculata. Microsc.
Res. Tech. 78: 173–179.
Chemosynthetic symbioses Li, Y., Liles, M.R., & Halanych, K.M. (2018) Endosymbiont genomes yield
Assié, A., Borowski, C., van der Heijden, K. et al. (2016) A specific and clues of tubeworm success. ISME J. 12: 2785–2795.
widespread association between deep-sea bathymodiolus mussels and Mariño, J., Augustine, S., Dufour, S.C., & Hurford, A. (2018) Dynamic
a novel family of epsilonproteobacteria. Environ. Microbiol. Rep. 8: Energy Budget theory predicts smaller energy reserves in thyasirid
805–813. bivalves that harbour symbionts. J. Sea Res.143: 119–127.
Batstone, R.T., Laurich, J.R., Salvo, F., & Dufour, S.C. (2014) Divergent Mcmullin, E.R., Hourdez, S., Schaeffer, S.W., & Fisher, C.R. (2003)
chemosymbiosis-related characters in Thyasira cf. gouldi (Bivalvia: Phylogeny and biogeography of deep sea vestimentiferan tubeworms and
Thyasiridae). PLoS One 9: e92856. their bacterial symbionts. Symbiosis 34: 1–41.
Chen, C., Rogers, A.D., Linse, K. et al. (2015) The “scaly-foot gastro- Nussbaumer, A.D., Fisher, C.R., & Bright, M. (2006) Horizontal endosym-
pod”: A new genus and species of hydrothermal vent-endemic gastropod biont transmission in hydrothermal vent tubeworms. Nature 441: 345–348.
(Neomphalina: Peltospiridae) from the Indian Ocean. J. Mollusc. Stud.
81: 322–334. Petersen, J.M., Kemper, A., Gruber-Vodicka, H. et al. (2017)
Chemosynthetic symbionts of marine invertebrate animals are capable of
Dubilier, N., Bergin, C., & Lott, C. (2008) Symbiotic diversity in marine nitrogen fixation. Nat. Microbiol. 2: 16195.
animals: The art of harnessing chemosynthesis. Nat. Rev. Microbiol. 6:
725–740. Pflugfelder, B., Cary, S.C., & Bright, M. (2009) Dynamics of cell prolif-
eration and apoptosis reflect different life strategies in hydrothermal vent
Dubilier, N., Mulders, C., Ferdelman, T. et al. (2001) Endosymbiotic sul- and cold seep vestimentiferan tubeworms. Cell Tissue Res. 337: 149–165.
phate-reducing and sulphide-oxidizing bacteria in an oligochaete worm.
Nature 411: 298–302. Robidart, J.C., Bench, S.R., Feldman, R.A. et al. (2008) Metabolic versa-
tility of the Riftia pachyptila endosymbiont revealed through metagenom-
Dufour, S.C. & Felbeck, H. (2003) Sulphide mining by the superextensile ics. Environ. Microbiol. 10: 727–737.
foot of symbiotic thyasirid bivalves. Nature 426: 65–67.
Sayavedra, L., Ansorge, R., Rubin-Blum, M. et al. (2019) Horizontal
Dufour, S.C., Laurich, J.R., Batstone, R.T. et al. (2014) Magnetosome- acquisition followed by expansion and diversification of toxin-related
containing bacteria living as symbionts of bivalves. ISME J. 8: 2453–2462. genes in deep-sea bivalve symbionts. BioRxiv 505386.
Duperron, S., Gaudron, S.M., Rodrigues, C.F. et al. (2013) An overview Sayavedra, L., Kleiner, M., Ponnudurai, R. et al. (2015) Abundant toxin-
of chemosynthetic symbioses in bivalves from the North Atlantic and related genes in the genomes of beneficial symbionts from deep-sea hydro-
Mediterranean Sea. Biogeoscience 10: 3241–3267. thermal vent mussels. eLife 4: e07966.
 Microbial Symbioses of Marine Animals 289

Stewart, F.J., Newton, I.L.G., & Cavanaugh, C.M. (2005) Chemosynthetic Zooxanthellae and the coral holobiont
endosymbioses: Adaptations to oxic-anoxic interfaces. Trends Microbiol.
13: 439–448. Ainsworth, T.D., Krause, L., Bridge, T. et al. (2015) The coral core micro-
biome identifies rare bacterial taxa as ubiquitous endosymbionts. ISME J.
Thurber, A.R., Jones, W.J., & Schnabel, K. (2011) Dancing for food in the 9: 2261–2274.
deep sea: Bacterial farming by a new species of yeti crab. PLoS ONE 6:
e26243. Blackall, L.L., Wilson, B., & Oppen, M.J.H. van (2015) Coral—The
world’s most diverse symbiotic ecosystem. Mol. Ecol. 24: 5330–5347.
Warén, A., Bengtson, S., Goffredi, S.K., & Van Dover, C.L. (2003) A hot-
vent gastropod with iron sulfide dermal sclerites. Science 302: 1007. Bourne, D.G., Morrow, K.M., & Webster, N.S. (2016) Insights into the
coral microbiome: Underpinning the health and resilience of reef ecosys-
Wentrup, C., Wendeberg, A., Schimak, M. et al. (2014) Forever compe- tems. Annu. Rev. Microbiol. 70: 317–340.
tent: Deep-sea bivalves are colonized by their chemosynthetic symbionts
throughout their lifetime. Environ. Microbiol. 16: 3699–3713. Callahan, B.J., McMurdie, P.J., & Holmes, S.P. (2017) Exact sequence
variants should replace operational taxonomic units in marker-gene data
Woyke, T., Teeling, H., Ivanova, N.N. et al. (2006) Symbiosis insights analysis. ISME J. 11: 2639.
through metagenomic analysis of a microbial consortium. Nature 443:
950–955. Davy, S.K., Allemand, D., & Weis, V.M. (2012) Cell biology of cnidarian-
dinoflagellate symbiosis. Microbiol. Mol. Biol. Rev. 76: 229–61.
Yao, H., Dao, M., Imholt, T. et al. (2010) Protection mechanisms of the
iron-plated armor of a deep-sea hydrothermal vent gastropod. Proc. Natl. Donelson, J.M., Salinas, S., Munday, P.L., & Shama, L.N.S. (2018)
Acad. Sci. USA 107: 987–992. Transgenerational plasticity and climate change experiments: Where do
we go from here? Glob. Change Biol. 24: 13–34.
Deep-sea wood and whale falls Fine, M. & Loya, Y. (2002) Endolithic algae: An alternative source of pho-
toassimilates during coral bleaching. Proc. Roy. Soc. B 269: 1205–1210.
Danise, S. & Higgs, N.D. (2015) Bone-eating Osedax worms lived on
Mesozoic marine reptile deadfalls. Biol. Lett. 11: 20150072. Hernandez-Agreda, A., Gates, R.D., & Ainsworth, T.D. (2017) Defining
the core microbiome in corals’ microbial soup. Trends Microbiol. 25:
Little, C.T. (2017) The prolific afterlife of whales. Sci. Am. 26: 14–19. 125–140.
Ristova, P.P., Bienhold, C., Wenzhöfer, F. et al. (2017) Temporal and spa- Hernandez-Agreda, A., Leggat, W., Bongaerts, P., & Ainsworth, T.D.
tial variations of bacterial and faunal communities associated with deep- (2016) The microbial signature provides insight into the mechanistic basis
sea wood falls. PLoS ONE 12: e.0169906. of coral success across reef habitats. MBio 7: e00560-16.
Smith, C.R. & Baco, A.R. (2003) Ecology of whale falls at the deep-sea Hoegh-Guldberg, O., Hughes, T., Antoiony, K. et al. (2009) Coral reefs
floor. Oceanogr. Mar. Biol. 41: 311–354. and rapid climate change: Impacts, risks and implications for tropical soci-
eties. IOP Conf. Ser.: Earth Environ. Sci. 6: 302004.
Ascidians, bryozoans and sponges
Hughes, T.P., Barnes, M.L., Bellwood, D.R. et al. (2017) Coral reefs in the
Anderson, C.M. & Haygood, M.G. (2007) α-proteobacterial symbionts of Anthropocene. Nature 546: 82–90.
marine bryozoans in the genus Watersipora. Appl. Environ. Microbiol.
Hughes, T.P., Kerry, J.T., Baird, A.H. et al. (2018) Global warming trans-
73: 303–311.
forms coral reef assemblages. Nature 556: 492–496.
Evans, J.S., Erwin, P.M., Shenkar, N., & López-Legentil, S. (2018) A
LaJeunesse, T.C., Parkinson, J.E., Gabrielson, P.W. et al. (2018) Systematic
comparison of prokaryotic symbiont communities in nonnative and native
revision of Symbiodiniaceae highlights the antiquity and diversity of coral
ascidians from reef and harbor habitats. FEMS Microbiol. Ecol. 94: fiy139.
endosymbionts. Curr. Biol. 28: 2570–2580.
Fieseler, L., Horn, M., Wagner, M., & Hentschel, U. (2004) Discovery
Morris, L.A., Voolstra, C.R. Quigley, K.M. et al. (2019) Nutrient avail-
of the novel candidate phylum “Poribacteria” in marine sponges. Appl.
ability and metabolism affect the stability of coral-Symbiodiniaceae sym-
Environ. Microbiol. 70: 3724–3732.
bioses. Trends Microbiol. 27: 678–689.
Karagodina, N.P., Vishnyakov, A.E., Kotenko, A.L. et al. (2018)
NAS (2019) National Academies of Sciences Engineering and Medicine.
Ultrastructural evidence for nutritional relationships between a marine
A Research Review of Interventions to Increase the Persistence and
colonial invertebrate (Bryozoa) and its bacterial symbionts. Symbiosis 75:
Resilience of Coral Reefs. The National Academies Press, Washington,
155–164.
DC.
López-Legentil, S., Turon, X., Espluga, R., & Erwin, P.M. (2015) Temporal
Neave, M.J., Rachmawati, R., Xun, L. et al. (2017) Differential specificity
stability of bacterial symbionts in a temperate ascidian. Front. Microbiol.
between closely related corals and abundant Endozoicomonas endosymbi-
6: 1022.
onts across global scales. ISME J. 11: 186–200.
Moitinho-Silva, L., Nielsen, S., Amir, A. et al. (2017) The sponge micro-
Pochon, X., Putnam, H.M., & Gates, R.D. (2014) Multi-gene analysis of
biome project. GigaScience 6: 1–7.
Symbiodinium dinoflagellates: A perspective on rarity, symbiosis, evolu-
Pita, L., Rix, L., Slaby, B.M. et al. (2018) The sponge holobiont in a chang- tion. PeerJ 2: e394.
ing ocean: From microbes to ecosystems. Microbiome 6: 46.
Pollock, F.J., McMinds, R., Smith, S. et al. (2018) Coral-associated bacte-
Schmitt, S., Tsai, P., Bell, J. et al. (2012) Assessing the complex sponge ria demonstrate phylosymbiosis and cophylogeny. Nat. Commun. 9: 4921.
microbiota: Core, variable and species-specific bacterial communities in
Raghukumar, S. (2017) Fungi in Coastal and Oceanic Marine Ecosystems:
marine sponges. ISME J. 6: 564–576.
Marine Fungi. Springer International.
Schreiber, L., Kjeldsen, K.U., Funch, P. et al. (2016) Endozoicomonas are
Rohwer, F., Seguritan, V., Azam, F., & Knowlton, N. (2002) Diversity and
specific, facultative symbionts of sea squirts. Front. Microbiol. 7: 1042.
distribution of coral-associated bacteria. Marine Ecol. Prog. Ser. 243:
Taylor, M.W., Tsai, P., Simister, R.L. et al. (2013) “Sponge-specific” bac- 1–10.
teria are widespread (but rare) in diverse marine environments. ISME J.
Rohwer, F. & Youle, M. (2010) Coral Reefs in Microbial Seas. Plaid Press,
7: 438–443.
San Diego.
Thomas, T., Moitinho-Silva, L., Lurgi, M. et al. (2016) Diversity, structure
Torda, G., Donelson, J.M., Aranda, M. et al. (2017) Rapid adaptive
and convergent evolution of the global sponge microbiome. Nat. Commun.
responses to climate change in corals. Nat. Clim. Change 7: 627.
7: 11870.
290 Chapter 10

van de Water, J.A.J.M., Allemand, D., & Ferrier-Pagès, C. (2018) Host- Röthig, T., Yum, L.K., Kremb, S.G., et al. (2017) Microbial community
microbe interactions in octocoral holobionts - recent advances and per- composition of deep-sea corals from the Red Sea provides insight into
spectives. Microbiome 6: 64. functional adaption to a unique environment. Sci. Rep. 7: 44714.
van de Water, J.A.J.M., Melkonian, R., Junca, H. et al. (2016) Spirochaetes
dominate the microbial community associated with the red coral Corallium Bioluminescent symbioses
rubrum on a broad geographic scale. Sci. Rep. 6: 27277. Haygood, M.G. (1993) Light organ symbioses in fishes. Crit. Rev.
van Oppen, M.J.H., Oliver, J.K., Putnam, H.M., & Gates, R.D. (2015) Microbiol. 19: 191–216.
Building coral reef resilience through assisted evolution. Proc. Natl. Hellinger, J., Jägers, P., Donner, M. et al. (2017) The flashlight fish
Acad. Sci. 112: 2307–2313. Anomalops katoptron uses bioluminescent light to detect prey in the dark.
Wada, N., Ishimochi, M., Matsui, T. et al. (2019) Characterization of coral- PLOS ONE 12: e0170489.
associated microbial aggregates (CAMAs) within tissues of the coral Hendry, T.A. & Dunlap, P.V. (2011) The uncultured luminous symbiont
Acropora hyacinthus. Biorxiv, BioRxiv: 576488. of Anomalops katoptron (Beryciformes: Anomalopidae) represents a new
Wada, N., Pollock, F.J., Willis, B.L. et al. (2016) In situ visualization bacterial genus. Mol. Phylogenet. Evol. 61: 841–843.
of bacterial populations in coral tissues: Pitfalls and solutions. PeerJ 4: Hendry, T.A. & Dunlap, P. V (2014) Phylogenetic divergence between the
e2424. obligate luminous symbionts of flashlight fishes demonstrates specificity
Webster, N.S. & Reusch, T.B.H. (2017) Microbial contributions to the per- of bacteria to host genera. Environ. Microbiol. Rep. 6: 331–8.
sistence of coral reefs. ISME J. 11: 2167–2174. Hendry, T.A., Freed, L.L., Fader, D. et al. (2018) Ongoing transposon-
Wegley, L., Edwards, R., Rodriguez-Brito, B. et al. (2007) Metagenomic mediated genome reduction in the luminous bacterial symbionts of deep-
analysis of the microbial community associated with the coral Porites ast- sea ceratioid anglerfishes. MBio 9: e01033-18.
reoides. Environ. Microbiol. 9: 2707–2719. McAnulty, S.J. & Nyholm, S.V. (2017) The role of hemocytes in the
Wooldridge, S.A. (2010) Is the coral-algae symbiosis really “mutually Hawaiian bobtail squid, Euprymna scolopes: A model organism for study-
beneficial” for the partners? BioEssays 32: 615–625. ing beneficial host-microbe interactions. Front. Microbiol. 7: 2013.
McFall-Ngai, M. (2014) Divining the essence of symbiosis: Insights from
Deep-sea corals the squid-vibrio model. PLoS Biol. 12: e1001783.
Kellogg, C.A., Goldsmith, D.B., & Gray, M.A. (2017) Biogeographic Nawroth, J.C., Guo, H., Koch, E. et al. (2017) Motile cilia create fluid-
comparison of Lophelia-associated bacterial communities in the western mechanical microhabitats for the active recruitment of the host microbi-
Atlantic reveals conserved core microbiome. Front. Microbiol. 8: 796. ome. Proc. Natl. Acad. Sci.114: 9510–9516.
Kellogg, C.A., Lisle, J.T., & Galkiewicz, J.P. (2009) Culture-independent Norsworthy, A.N. & Visick, K.L. (2013) Gimme shelter: How Vibrio
characterization of bacterial communities associated with the cold-water fischeri successfully navigates an animal’s multiple environments. Front.
coral Lophelia pertusa in the northeastern Gulf of Mexico. Appl. Environ. Microbiol. 4: 356.
Microbiol. 75: 2294–2303. Nyholm, S.V. & McFall-Ngai, M.J. (2004) The winnowing: Establishing
Middelburg, J.J., Mueller, C.E., Veuger, B. et al., (2015) Discovery of sym- the squid-vibrio symbiosis. Nat. Rev. Micriobiol. 2: 632–642.
biotic nitrogen fixation and chemoautotrophy in cold-water corals. Sci. Nyholm, S.V., Stabb, E.V., Ruby, E.G., & McFall-Ngai, M.J. (2000)
Rep. 5: 17962. Establishment of an animal–bacterial association: Recruiting symbiotic
Penn, K., Wu, D., Eisen, J.A., & Ward, N. (2006) Characterization of bac- vibrios from the environment. Proc. Natl. Acad. Sci. USA 97: 10231–10235.
terial communities associated with deep-sea corals on Gulf of Alaska sea-
mounts. Appl. Environ. Microbiol. 72: 1680–1683.
Chapter 11
Microbial Diseases of
Marine Organisms

Marine biologists have become increasingly aware of the effects of infectious microbial dis-
eases on populations and communities of marine organisms, which can sometimes result in
major changes at the ecosystem level. Infectious diseases caused by viruses, bacteria, fungi,
and protists have been characterized in marine mammals, fish, and turtles, as well as in a
wide range of invertebrates, including molluscs, crustaceans, sponges, echinoderms, and cor-
als. Infections of multicellular algae and seagrasses also have important ecological and eco-
nomic effects. Toxins produced by diatoms and dinoflagellates also cause disease in marine
organisms. This chapter explores the various types of microbial diseases and their impor-
tance in natural ecosystems and aquaculture, with special consideration of the impact of
climate change, pollution, global transport, and other anthropogenic factors. Aspects of coral
disease link closely with the discussion of the microbiome in Chapter 13. Biotechnological
aspects of disease control are further considered in Chapter 14.

Key Concepts
• The development of disease depends on complex interactions between the host (and its
associated microbiota), the pathogen, and environmental factors.
• It is not possible to define single causative agents for many marine diseases, and dis-
turbance of the microbiome is a major factor allowing development of opportunistic
pathogens.
• Incidence of disease in many groups of organisms has increased in natural marine
ecosystems in the past few decades, with global climate change, pollution, and other
anthropogenic factors being contributory factors.
• Diseases of foundation species, keystone predators, and ecosystem engineers can lead
to serious alterations to marine ecosystems.
• Development of intensive marine aquaculture has led to increased levels of bacterial
and viral diseases in molluscs, crustaceans, and fish; this has major economic effects
in the industry as well as impacts on natural ecosystems.
• Bacteria (especially vibrios) are strongly associated with diseases of corals, molluscs,
crustaceans, and fish.
• Viruses are responsible for major epizootics in molluscs, crustaceans, fish, reptiles,
and mammals, some of which seriously affect natural populations.
292 Chapter 11

Diseases of marine organisms have major

ecological and economic impact
There is an enormous diversity and abundance of marine parasites and pathogens, which
are a natural part of the life cycle of all organisms, regulating food webs and shaping the
structure and function of ecosystems. In the last few decades, there has been increasing
concern about the spread of infectious diseases and outbreaks of mass mortality in wild
marine organisms. In some cases, disease outbreaks in foundation species, keystone species,
or ecosystem engineers lead to substantial long-term changes in ecosystem structure and
function. With “emerging diseases,” it is always difficult to decide whether there is a genu-
ine increase in incidence, or whether it simply reflects more intensive observation. There
is no doubt that interest in monitoring the health of marine habitats has increased greatly
and there is now a large community of sub-aqua diving scientists, which partly explains the
increased reporting of disease. However, it is generally accepted that diseases are increasing
as a result of recent disturbance of balanced marine ecosystems caused by human activities.
Pollution, destruction of habitats, climate change, global sea traffic, and intensive aquacul-
ture all affect the complex interactions between hosts, pathogens, and the environment in
the disease process (Figure 11.1).

While epidemiology of diseases in humans and terrestrial plants and animals is a well-devel-
oped science, the dynamics of the origins, spread, and cycles of disease in the oceans are less
well understood. In part, this is due to the greater host and pathogen diversity, but it must also
be noted that many marine organisms such as corals, sponges, and ascidians are colonies of
genetically identical individuals, which facilitates the spread of disease. Furthermore, marine
systems are more open, allowing rapid dispersal of pathogens over large distances and oppor-
tunities to survive in reservoirs allowing reinfection. It is vital that we develop a better under-
standing of the role of microbes in the ecology of marine infectious diseases. Thus, research
efforts are directed towards developing new epidemiological models to explain, predict, and
manage marine infectious diseases. The direct costs to animal health in aquaculture are
considerable and quantifiable, but disease also has considerable economic impacts in natu-
ral ecosystems due to adverse direct and indirect effects on fisheries, food security, coastal
management, and tourism. Of course, this is in addition to obvious concerns about the loss of
biodiversity and the short- and long-term ecological impact of these events.

DISEASES OF CORALS, SPONGES, AND ECHINODERMS

Infectious diseases threaten the survival of corals
As discussed in Chapter 10, coral-associated microbes play a vital role in the health of cor-
als and the reef ecosystem, through provision of fixed nitrogen and carbon, recycling of
waste products, antibiotic defense, and contribution to food webs. However, the balanced

HOST ENVIRONMENT
Nutrition Pollution
Disease resistance Temperature
Normal microbiota UV irradiation
Physical damage Ocean acidification
Immune status Population density

PATHOGEN
Transmission
Infectivity
Virulence
Figure 11.1 The disease process in
Reservoirs
marine organisms depends on com- Survival
plex interactions of the host, patho-
gen, and environmental factors.
 Microbial Diseases of Marine Organisms 293

microbiome in healthy corals is easily disturbed by shifts in environmental conditions. Many

HOW CAN WE BE
of the world’s tropical and subtropical reefs are at severe risk from a multitude of threats,
including overfishing, coastal development, physical damage from boating and diving activi- ? SURE THAT MARINE
DISEASES ARE TRULY
ties, pollution, ocean acidification, and temperature-induced bleaching. There is now increas-
INCREASING?
ing recognition of the importance of coral diseases associated with microbial infection as a
factor in their decline. Coral diseases were first studied intensively in the Caribbean, where In the late 1990s, concerns were
successive outbreaks of disease have contributed to the decline of the major reef-building raised about dramatic increases
species, especially Acropora, resulting in major ecosystem changes. Diseases have now been of diseases in the marine environ-
described in all the major reef systems of the world, involving several hundred different spe- ment. Is there truly an increase in
disease rates, or is it simply that
cies of coral.
more scientists are conducting
studies in this field? To answer
One of the problems in evaluating the historical importance of coral diseases is that the this question, Ward and Lafferty
early literature contains many poorly documented reports of disorders for which details of (2004) analyzed 5900 papers
pathology and evidence of specific causation are incomplete. Diagnosis of disease in corals is (1970–2001) reporting disease
primarily based on characteristics such as the loss of tissue, change in color, and exposure of in nine taxonomic groups. They
coral skeleton. These limited macroscopic descriptions lead to imprecise names for the con- used a sophisticated normalization
ditions—for example, in the Caribbean, the terms “white plague,” “white disease,” “white method to determine the propor-
pox,” and “white band” have all been used to represent different syndromes, without clear tion of disease reports about a
definitions—and this results in inaccurate subsequent diagnosis. The term “white syndrome” given taxonomic group relative to
is a collective term that encompasses the range of coral pathologies observed in the Indo- the total number of reports about
any subject in that group. The data
Pacific in which a spreading zone of tissue loss exposes the white coral tissue. It is likely
were also adjusted to remove the
that there are multiple diseases involving different pathogens, but methods are currently not effect of multiple reports of the
robust enough to distinguish them. same outbreak, or the effect of
particularly prolific authors investi-
Although several studies have reported fulfilment of Koch’s Postulates, which are the usual gating disease. Reports of disease
criteria needed to prove the etiology of disease (see information box on p.305), there is usu- were increased for all groups over
ally no evidence specifically linking cellular responses of the host to the presence of the this period, but after removing the
pathogen and development of the disease. Thus, there is only a small number of diseases skewing effects, Ward and Lafferty
that have been sufficiently well characterized for us to be confident of a clear and possibly concluded that there was a clear
causative association with specific bacteria, fungi, or protistan microbes (Table 11.1). Even increase in disease among turtles,
so, the situation is further complicated by the fact that the same pathogen may cause differ- corals, mammals, urchins, and mol-
luscs. Reports of diseases in fishes
ent signs in different hosts; alternatively, different pathogens may cause the same signs in a
actually decreased, which they
particular host. Furthermore, organisms established as the causative pathogen of a disease suggested was caused by drastic
are often not found in all cases and may change with time. For most diseases, knowledge of reductions in population density
the mechanisms of pathogenesis and the host responses remains very limited and some coral due to overfishing, presenting
biologists remain skeptical about the role of microbes as primary causative agents of disease fewer opportunities for transmit-
in uncompromised hosts (see Box 11.1). As we saw in Chapter 10, it is increasingly important ting infection.
to consider hosts and their associated microbes together as holobionts, and disease must be
considered within this context.

Vibrios are associated with many coral diseases

It can be seen from Table 11.1 that Vibrio spp. are particularly prominent as possible caus-
ative agents of well-recognized coral diseases. Furthermore, other members of the family
Vibrionaceae, including Enterovibrio and Photobacterium, form a significant proportion of
the microbiota of “healthy” corals (i.e. lacking any obvious disease signs) and some contrib-
ute to nitrogen fixation, defense against pathogens, and other functions. The Vibrio spp. in
Table 11.1 possess an array of enzymes, toxins, and other virulence factors, which supports
them being primary pathogens, but in many cases bacteria must be regarded as “opportunis-
tic” pathogens that are only able to overwhelm the natural defenses of the host under certain
conditions. Environmental factors can alter the immune status of the host or the probiotic
effects of resident microbiota. All vibrios are versatile heterotrophs capable of rapid growth
when nutrients are, so they can quickly take advantage of these changes and digest damaged
tissue, leading to signs of disease.

As discussed in Chapter 10, bleaching is a highly damaging process affecting coral reefs.
It is generally perceived to be due to physiological disturbance resulting in disruption of the
symbiosis between the host and the photosynthetic zooxanthellae, most often as the result of
increased seawater temperatures. The concept that some types of bleaching might be due to
bacterial infection stems from the comprehensive studies of the research group led by Eugene
294 Chapter 11

Table 11.1 Diseases of corals for which there is a strong association with microbial infection. (Data from
Sweet et al., 2011; Mera and Bourne, 2018). Changes in our understanding of the causative agents of coral diseases
may occur because more than one pathogen causes the same signs on different hosts, and etiologies may change
through time (the “moving target hypothesis” of Sutherland et al., 2016).

Disease Associated pathogen(s) Commonly affected species Locationa

Aspergillosis Aspergillus sydowii Gorgonia ventalina and other octocorals CS

Black band disease Mixed consortium; Phormidium Wide range of corals including CS, IP, MS
corallyticum, Roseofilum reptotaenium, Acropora spp. Montipora spp., Favia spp.,
Trichodesmium erythraeum, Porites spp., Siderastrea spp.
Desulfovibrio spp. and Beggiatoa spp. are
key components

Bleaching Vibrio shilonii (originally classified as Oculina patagonica RS.MS

V. shiloi)

Bleaching and tissue lysis Vibrio coralliilyticus and related Vibrio spp. Pocillopora damicornis, MS, IP

White band Vibrio carchariae (= V. harveyi) and other Acropora spp. CS

Vibrio spp.

White plague-like disease Thalassomonas loyana Favia favus RS

White plague type II Aurantimonas coralicida Montastrea spp. and other CS

scleractinian corals

White pox Serratia marcescens Acropora palmata CS

White syndrome Vibrio spp. including V. coralliilyticus, V. Acropridae, Pocilloporidae, Montipora IP

owensii, V. harveyi: Rhodobacteriaceae- capitata
associate; Philaster lucinda

Yellow blotch/band Consortium of Vibrio spp.: V. rotiferianus, V. Acropora spp., Orbicellab spp., CS, IP
harveyi, V. alginolyticus, V. proteolyticus Diploastrea spp., Fungia spp.
a CS, Caribbean Sea; IP, Indo-Pacific Ocean; MS, Mediterranean Sea; RS, Red Sea. b Formerly Montastrea.

Rosenberg of Tel Aviv University. In the 1990s, they showed that Vibrio shiloi (subsequently
renamed as V. shilonii) is the causative agent of summer bleaching in the invasive coral
Oculina patagónica on the Israel coast of the Mediterranean Sea. In a series of experiments,
researchers showed that when O. patagonica was infected with V. shilonii at 29°C, bleaching
occurred rapidly, whereas at 20 and 25°C, the rate of bleaching was slower and less complete
and none was observed at 16°C or in uninoculated controls at any temperature. Also, addi-
tion of antibiotics prevented bleaching. Virulence was attributed to temperature-regulated
production of an adhesin to a carbohydrate receptor on the coral surface and the production
of a peptide toxin that inhibits photosynthesis in the zooxanthellae, leading to bleaching.
Subsequently, the marine fireworm Hermodice carunculata was implicated as a winter res-
ervoir and vector of the pathogen. Extrapolating from these findings to a general hypothesis
that bacteria cause bleaching provoked considerable controversy at the time, because the
temperature shifts involved are extreme (16‒29°C), whereas mass bleaching events in the
Caribbean, Pacific, and Indian Oceans typically involve rises of just 1‒2°C above normal sea-
surface temperatures. The O. patagonica–V. shilonii interactions became an elegant model
system in Rosenberg’s laboratory and, until 2002, all strains of pathogenic V. shiloi held in
the laboratory caused bleaching in aquarium experiments. However, the same strains are now
incapable of infecting the coral. Because coral lack adaptive immunity, the “Coral Probiotic
Hypothesis” was proposed as an explanation for this phenomenon, which was later developed
as the “Hologenome Hypothesis” (see p.281). The deliberate manipulation of the coral micro-
biome by introduction or enhancement of beneficial microbes as “probiotics” has recently
been proposed as a method of treating diseased corals or preventing the spread of disease.
This idea is considered in Chapter 14.

Further support for the role of bacteria in coral bleaching was provided by the discovery of
a new species, Vibrio coralliilyticus, first isolated from the coral Pocillopora damicornis
in the Indian Ocean. Subsequently, strains of V. coralliilyticus have been associated with
 Microbial Diseases of Marine Organisms 295

a variety of disease syndromes in corals and other invertebrates in many different regions.
DEVELOPMENT
Recent research on the virulence of this versatile marine pathogen is discussed in Research
Focus Box 11.1. i OF THE CORAL
PROBIOTIC
HYPOTHESIS (CPH)
The fungus Aspergillus sydowii caused mass After 2002, V. shilonii no longer
mortality of sea fans in the Caribbean Sea infected O. patagonica. Reshef
et al. (2006) proposed the CPH
Beginning in 1995, one of the most devastating epizootics yet seen in the marine environ- to explain this: “the coral animal
ment occurred in sea fans (Gorgonia ventalina) in the Caribbean Sea. A new fungal spe- lives in a symbiotic relationship
cies, Aspergillus sydowii, was identified as the causative agent responsible for massive tissue with a diverse metabolically active
destruction. The disease was shown to be transmissible through contact with healthy cor- population of microorganisms …
als. Since 1998, the severity of the disease has declined and it now occurs only sporadi- When environmental conditions
cally, although many of the largest sea fans have been killed, especially in the Florida Keys. change … the relative abundance
Investigations reveal that there is a complex relationship between temperature, growth of the of microbial species changes [to]
pathogen, and host resistance due to production of antifungal compounds. The host attempts allow the coral holobiont to adapt
to the new condition.” We cannot
to limit spread of the pathogen by encapsulation of necrotic tissue, resulting in the character-
go back in time to investigate
istic purple lesions (Figure 11.3). Other microbes associated with surface mucus also differ what changes may have occurred
between healthy and diseased sea fans and between healthy and diseased parts of the same in O. patagonica, but one clue
individual. A. sydowii has also been implicated as the causative agent of the virtual eradi- comes from the fact that several
cation of sea urchins (Diadema antillarum) from parts of the Caribbean Sea in the 1980s, bacteria isolated from the coral in
although there is no direct evidence for this. There is recent evidence that the fungus can be 2008 show strong activity against
isolated from apparently healthy G. ventalina and other species of octocorals, suggesting that V. shilonii (Nissimov et al., 2009).
it is an opportunist pathogen that causes disease only under certain conditions. Alternatively, the pathogen may
have changed; analysis indicates
that modern strains lack genomic
Black band disease of corals is a islands that might encode viru-
lence factors (Reshef et al., 2008).
disease of corals worldwide Seasonal bleaching of O. pata-
Black band disease (BBD) was the first coral disease to be described. It takes its name from gonica still occurs and Ainsworth
the characteristic black band (from about 10 mm to several cm wide), which is a thick micro- et al. (2008) questioned whether
V. shilonii was ever the primary
bial mat that migrates over the coral colony by as much as 2 cm a day (Figure 11.4). Healthy
cause of bleaching. However,
coral tissue is killed, and the band moves on leaving the exposed skeleton, which usually
Rosenberg et al. (2009) provide
becomes overgrown by turf algae. BBD usually occurs in the summer when water tempera- strong counter-arguments to this
tures rise. It can arise on apparently pristine reefs but is more often associated with reefs and Mills et al. (2013) show that
receiving sewage and runoff from the coast. BBD can have severe effects because it often other bacteria may currently play
attacks large corals such as Orbicella (formerly Montastrea), which are important in build- important roles in both causing
ing the framework of reefs. Infection can be transmitted via contact, and damaged colonies and preventing bleaching.
are more susceptible. Since its first description in the Caribbean in the 1970s, BBD has been
observed in a wide range of coral species throughout the world.

BBD has been intensively studied but establishing the etiology of the disease has been
extremely difficult. Microscopic examination of diseased tissue consistently shows the
presence of large, gliding, filamentous cyanobacteria, including Phormidium corallyticum,
Leptolyngbya spp., Geitlerinema spp., and Pseudoscillatoria coralii, together with numerous
other heterotrophic bacteria, sulfide-oxidizing bacteria (SOB), and sulfate-reducing bacteria
(SRB, especially Desulfovibrio spp.); the SRB are responsible for the characteristic black
pigment. Protists and members of the Archaea have also been identified. It is generally rec-
ognized that BBD is a polymicrobial disease caused by a succession of microbial virulence
factors in microenvironments produced by activities of the mixed consortium. Anoxic and
sulfide-rich conditions at the base of the band result in death of the coral tissue as it spreads
across the coral. Molecular analysis of the lesions as they develop in the field has yielded new
information about the causes and progression of the disease.

Increases in seawater temperature, water quality, presence of nutrients from sewage, damage
from divers or boat anchors, and other factors have all been linked to the incidence of BBD.
Pathogens are transmitted to other corals in the reef via water currents and vectors such as
snails, worms, and reef fishes. A similar condition called red band occasionally affects hard
star and brain corals in the Caribbean and GBR. It also seems to involve similar cyanobacte-
ria and other members of the microbial consortium; the reasons for difference in pigmenta-
tion are not clear.
296 Chapter 11

BOX 11.1 RESEARCH FOCUS

The pathogenicity of Vibrio coralliilyticus

Discovery of a versatile marine pathogen. The discovery by Ben- the hemagglutinin/protease genes described as major virulence
Haim and Rosenberg (2002) of V. coralliilyticus as the cause of factors in other vibrios, including V. cholerae, V. splendidus, and
temperature-dependent bleaching and necrosis in Pocillopora gave V. tubiashii. Surprisingly, there was no difference in pathogenicity
support to the bacterial bleaching hypothesis, because this coral to zooxanthellae and two model animal systems of a wild type and
is widely distributed and susceptible to bleaching throughout the a mutant strain (in which vcpA was deleted). This led Santos et
world. Laboratory experiments by Ben-Haim et al. (2003) showed al. (2011) to conclude that “the pathogenicity of vibrios in marine
that infection and tissue lysis proceed rapidly at 29°C, but no tissue animals is a complex interplay of multiple genetic factors and
damage occurs at 25°C or lower. We now know that V. corallii- unlikely the result of one determinant.” Some caution is perhaps
lyticus shows global distribution, and numerous strains have been necessary in the interpretation of such model systems for experi-
isolated in association with a wide range of diseased hard and soft mental determination of pathogenicity; for example, V. coralliilyti-
corals. Besides corals, V. coralliilyticus also infects a range of mol- cus and its extracellular products were previously shown to induce
luscs, crustaceans, and fish, although caution is needed in interpre- high mortalities of artemia and fish, even when cultured at 18°C
tation of some experimental “model systems” used to investigate (Austin et al., 2005). In their study of the Vc450 genome, Kimes
pathogenicity by Austin et al. (2005). Natural infections of bivalve et al. (2012) showed that a number of virulence factors involved
larvae by V. coralliilyticus and the closely related V. tubiashii are in motility, host degradation, secretion, antimicrobial resistance,
responsible for high mortalities in oyster hatcheries and exten- and transcriptional regulation factors are upregulated at 27°C, cor-
sive cultivation (Genard et al., 2013). Whole-genome sequencing relating with the induction of bleaching and tissue damage previ-
has revealed that several strains associated with disease in bivalve ously observed with this strain. Interestingly, temperature (23°C
larvae and originally identified as V. tubiashii are in fact V. cor- or 27°C) did not affect virulence of a different strain, ON008,
alliilyticus, confirmed by PCR assays based on detection of the shown to be responsible for acute WS in Montipora capitata in
gene vcpA encoding a zinc metalloprotease implicated in virulence Hawaii despite this strain sharing many temperature-regulated
(Wilson et al., 2013). genes with strain BAA-450 (Ushijima et al., 2014). However, the
true cause (or causes) of WS remains elusive. Pollock et al. (2017)
Role in coral white syndromes. Sussman et al. (2008) used a concluded that WS in acroporids on the GBR is not due to infection
combination of molecular and culture-based techniques to isolate by vibrios, based on an 18-month field study. They found no sig-
putative pathogens from various species of corals during epizoot- nificant differences in the abundance of Vibrio sequences between
ics of white syndrome (WS) on the Great Barrier Reef (GBR) and healthy and diseased samples and found no sign of these bacteria
other areas of the Indo-Pacific. They showed that two novel strains in the winter, despite year-round disease progression. Pollock et
of V. coralliilyticus and four closely related Vibrio spp. are caus- al. found elevated levels of Rhodobacteriaceae sequences in WS
ative agents of WS. Virulence is highly dependent on temperature, lesions and suggested that these bacteria are opportunists, which
with the most important factor being a powerful zinc metallopro- are normally regulated by antibiotic activities of other members of
tease (MP). Sussman et al. (2009) tested the effects of the metal- the microbiome.
loprotease on photosynthetic activity in zooxanthellae isolated
from different corals which had varying degrees of sensitivity to Motility and chemotaxis. Meron et al. (2009) first demonstrated
WS. The MP inactivated photosystem II of zooxanthellae isolated the importance of flagellar motility and chemotaxis of V. coral-
from hosts susceptible to WS, followed by spreading lesions lead- liilyticus in virulence, by showing that a mutant strain defective
ing to death of the coral tissue. Despite the strong link between in the flagellar gene fhlA was unable to swim towards mucus and
MP and the development of disease signs, Sussman et al. (2009) adhere to the tissue of P. damicornis. Further advances in studying
were careful to point out that WS is a multifactorial disease. In the behavior of V. coralliilyticus have been made by the applica-
culture, protease production is highly dependent on bacterial cell tion of microfluidics, in which very small volumes of fluid (micro-
density, and accumulation of sufficient levels of protease to cause liters to picoliters) can be controlled within micro channels, often
cell damage in the field would require similar high densities; these mounted on a microscope slide. Garren et al. (2014) used a micro-
might occur at the band that is the progressing interface between fluidics device to create a 400 µm thick layer of mucus adjacent
dead and living tissue, characteristic of WS. The genomes of two to a 1 mm thick suspension of V. coralliilyticus in seawater. The
V. coralliilyticus (strains P1 and Vc450, from the GBR and near behavior of the bacteria in the diffusive gradients was measured
Zanzibar, respectively) show many similarities, with gene dis- by high-speed video microscopy. V. coralliilyticus showed a very
tribution on two chromosomes typical of that observed in other rapid and intense response to the presence of mucus, and further
vibrios, the presence of large plasmids, and a large proportion of analysis showed that the key component that initiates this response
shared genes (Santos et al., 2011; Kimes et al., 2012). However, is DMSP, produced by zooxanthellae and coral cells (p.253). By
there are variations in genes, including important virulence fac- tracking the movement of individual bacteria, it was found that
tors, pathogenicity islands, and phage genes that could explain dif- V. coralliilyticus demonstrates both chemotaxis (biased swim-
ferent physiological characteristics, suggesting that the two strains ming direction in response to a chemical gradient) and chemoki-
could be regarded as distinct ecotypes or subspecies. Santos et al. nesis (increased swimming speed when moving up an attractant
(2011) found the vcpA metalloprotease gene to be very similar to gradient) in response to DMSP. Garren et al. surmised that the
 Microbial Diseases of Marine Organisms 297

BOX 11.1 RESEARCH FOCUS

rare presence of these two processes is an adaptation that maxi- coral fragments allowed biochemical and microbial analyses to
mizes the likelihood of V. coralliilyticus reaching the coral sur- be performed in real time, alongside video microscopy to follow
face and remaining close to it. Surprisingly, V. coralliilyticus disease progression. This experimental infection sytem has been
does not appear to utilize DMSP as a nutrient source, but the used to challenge coral fragments with V. coralliilyticus (labeled
pathogen could be using it as an infochemical to target stressed with red-fluorescent protein for microscopic tracking). Gavish et
hosts, because DMSP levels are higher in heat-stressed corals. al. (2018) showed that soon after infection, the bacteria accumu-
In follow-up experiments, Garren et al. (2016) recorded the indi- lated in the coral pharynx. Over the next few hours, the polyps
vidual swimming behavior of thousands of individual V. coral- released large amounts of mucus containing the pathogen. About
liilyticus cells at different temperatures. At temperatures >23°C, two-thirds of the infected fragments showed distinct pathol-
chemotaxis towards coral mucus increased by >60% and at 30°C ogy, with tearing of the coenosarc and subsequent loss of colony
chemokinesis increased by >57%. Q-PCR analysis of shifts in bac- integrity and necrosis, accompanied by pathogen multiplication
teria associated with P. damicornis during heat shift experiments and increased MP. If the coral tissue had previous damage, accu-
showed dramatic increases of Vibrio spp., especially V. corallii- mulation of V. coralliilyticus occurred at the lesion edges result-
lyticus (Tout et al., 2015). Thus, rising temperatures affect both ing in tissue necrosis. The control coral colonies (inoculated with
the behavior of V. coralliilyticus and the chemical ecology of the only seawater or non-pathogenic V. fischeri) were unaffected. In
coral, and the interaction between these processes favors coloniza- follow-up experiments, Gibbin et al. (2018) inoculated P. dami-
tion of heat-stressed hosts by the pathogen. cornis with 15N-labeled V. coralliilyticus and used NanoSIMS
and TEM to visualize penetration and dispersal of the bacteria
Studying early stages of bacterial bleaching in real time. and their products. They identified that the mesentery filaments
Despite the presence of powerful tissue-degrading enzymes and had hotspots of 15N enrichment, suggesting that phagosomal cells
rapid disease progression, it is rather surprising that experimental in these tissues collect and digest pathogenic bacteria. Gibbin
infection of corals in aquaria requires a high inoculum. However, et al. (2019) then used 13C-labeled seawater to determine if the
it is not known what concentration is required for natural infection presence of the pathogen affects carbon assimilation in the zoo-
and whether physical injury or a specific environmental stressor xanthellae and its transfer to the host. Infection led to reduced
is necessary (Ushijima et al., 2014). Part of the problem in elu- 13C-assimilation in zooxanthellae and increase in the host.

cidating the infection mechanism is the complexity of the multi- 13C-turnover was reduced in zooxanthellae by Vibrio infection

partner coral holobiont and the difference in scale of the partners. at night, but host 13C turnover was not affected. Gibbin and col-
Also, traditional cellular pathology methods are only useful at leagues suggest that the coral “under attack” needs extra energy
advanced stages of the infection process. Further insights into the from this translocated carbon for the mucus spewing defensive
pathogenicity of V. coralliilyticus in bleaching of P. damicornis response, and that the zooxanthellae might increase their rate of
are emerging from use of a “coral on a chip” system, combining carbon fixation to provide this. The authors emphasize the need
microfluidics, live-imagining microscopy, and NanoSIMS (p.59) for measurement of other parameters such as photosynthesis, res-
to connect macro-scale behavioral responses of the coral with the piration, and symbiont density, but these micro-scale techniques
micro-scale metabolic interactions between the host and its sym- offer clear benefits for following the coral’s immune, behavioral,
biotic partners (Figure 11.2). Micro-propagated coral fragments and nutritional responses to infection. and determining the “tip-
are placed in a microchip with 250 µL chambers with individual ping point” between health and disease.
inlet and outlet tubes. Continuous collection of exudates from the

Figure 11.2 Schematic illustration of the “coral on a chip” microfluidics system. Reprinted from Shapiro et al. (2016).
298 Chapter 11

Figure 11.3 Gorgonia ventalina

infected with Aspergillus sydowii,
showing multifocal purple annular
lesions are indicative of infection.
Credit: Ernesto Weil, reprinted from
Mydlarz et al., (2008), CC BY 2.0.

THE WIND IS IN
? FROM AFRICA … DID
IT BRING THE SEA
FAN PATHOGEN? As is also observed in microbial mats the distribution in the band of sulfide (produced by
One long-held explanation for SRB) and oxygen (produced by cyanobacterial photosynthesis) varies according to light lev-
the sudden emergence of the els. The SOB migrate vertically with this diurnal variation to locate the optimal S:O2 inter-
Aspergillus sydowii epizootic in sea face for their metabolism. Field studies of cyanobacterial patches on corals show there is a
fans is that the fungus is of ter- succession of events; formation of anoxic conditions, proliferation of SRB, and deep anoxia
restrial origin and is associated with and sulfide-rich conditions produced by degradation of the host tissue. Metagenomic and
airborne dust transported from
transcriptomic analysis of the changes in microbial types and the activity of metabolic genes
North Africa and deposited in the
has confirmed this sequence of events, as illustrated in Figure 11.5.
Caribbean. It seems unlikely that
this is the only source, as Rypien
(2008) failed to find A. sydowii in
samples of dust and sediments
White plague and white pox are major
from Africa, the Caribbean, or Cape diseases affecting Caribbean reefs
Verde Islands. Rypien et al. (2008)
also used molecular markers to
White plague disease (WPD) of corals was first described in the 1970s as a disease of large
study strains of A. sydowii collected encrusting and branching corals in the Florida Keys. This form (subsequently designated
from many different regions, hosts, WPD Type I) did not appear to cause major problems, but a new virulent outbreak in 1995
and environments and concluded caused high mortalities in many species of coral, including the dominant reef-building
that a single source introduction Orbicella species, and by 2001 outbreaks had spread throughout the Caribbean. The more
of the fungus is unlikely. Rypien virulent form was designated WPD Type II. It starts at the base or edge of the coral and pro-
and Baker (2009) found evidence gresses up to 2 cm per day, with the destruction of tissue (Figure 11.6). Type III of the disease
from isotopic labeling studies that spreads at >2 cm per day. Unlike BBD, there seems to be a sharply defined line between
a reef snail Cyphoma gibbosum may healthy and diseased tissue, suggesting that toxins or enzymes are excreted by the advancing
act as a vector. The fungus is salt
microbial population. The bacterium Aurantimonas coralicida was named as the causative
tolerant and can grow in the sea.
Interestingly, a massive dust storm
agent of WPD-II in the coral Dichocoenia stokesii, but the pathogen has not been detected in
in Australia in 2009 deposited huge other WP infected corals and, conversely, it has been detected in healthy but not WP-diseased
amounts of spores and hyphae
in coastal waters, but no infec-
tions were observed (Hallegraeff
et al., 2014). A. sydowii has also
been detected in gorgonians in
Ecuadorian Pacific coral, although
no disease signs were found (Soler-
Hurtado et al., 2016).

Figure 11.4 A colony of sym-

metrical brain coral, Diploria strigosa,
affected by black-band disease
(BBD), Florida Keys. Credit: Christina
Kellogg, USGS.
 Microbial Diseases of Marine Organisms 299

Figure 11.5 Concept model of BBD

onset showing key microbial driv-
ers according to sequence-based
evidence. A transition from a benign
cyanobacterial population (Cya1)
to a virulent consortium (Cya2).
Increased photosynthesis by Cya2
drives organic material synthesis and
oxygen-consuming respiration of
bacterial heterotrophs, creating an
anaerobic niche. Here, cyanobacte-
rial photosynthates fuel SRB metabo-
lism, producing sulfide with possible
further contributions from desulfura-
tion of degrading coral tissue and
mucus resulting in anoxic, sulfide-
rich conditions that contribute to
increased pathogenicity to underly-
Orbicella annularis corals. In 1999, another form of the disease, designated WPD Type 3,
ing coral tissue. Reprinted from Sato
which spreads at >2 cm per day, appeared in the northern Florida Keys. WPD-2 often causes
et al. (2017), CC BY 4.0.
mass mortality after temperature-induced bleaching, with which it can be confused. Electron
microscopy and metagenomics studies have suggested a possible role for viruses in WPD,
with a group known as small circular ssDNA viruses being consistently associated with dis-
eased O. annularis. However, no evidence for a causal relationship has been found. The role
of viruses in coral health and disease is further considered below. Further uncertainty arises
because a different bacterial pathogen, Thalassomonas loyana, was shown to be the caus-
ative agent of WP-like disease in the Red Sea, and bacterial etiology was supported by the
i
USING BETTER
successful use of phage therapy to control the disease (see Box 14.1). METHODS TO
DIAGNOSE CORAL
Since the mid-1990s, the disease termed white pox (WPX) has had a devastating effect on DISEASES
the ecology of the shallow-water elkhorn coral, Acropora palmata, in the Florida Keys and
Some leading coral biologists
throughout the Caribbean. It is now considered an endangered species in the USA, with criticize the experimental basis
losses approaching 90% in living cover in the Florida Keys; its function as a foundation for claims of specific bacterial
reef-builder has been lost. Rapid tissue loss resulting in patches of coral skeleton occurs etiology, concluding that “most
in the summer. Disease signs (Figure 11.6) are very different from the tissue loss occur- common coral ‘diseases’ are a
ring due to WBD, bleaching, or scars from snail predation. The disease spreads quickly result of opportunistic, nonspe-
to neighboring colonies, implicating an infectious agent. Using a combination of cultural cific bacteria that exploit the
and molecular techniques, the causative agent of WPX was identified as Serratia marces- compromised health of the coral
cens. Koch’s postulates were fulfilled in experimental infection in 2002 (although a very after exposure to environmental
high inoculum was required, raising some doubts about the ecological significance of the stressors” (Lesser et al., 2007).
However, Work et al. (2008) argue
conclusions). Since this bacterium is commonly associated with the human gut, it appeared
that current knowledge is insuf-
that the infection originated from sewage pollution and this was proved by demonstrating
ficient either to support or refute
that a specific human strain from wastewater acted as a coral pathogen. Non-host corals this claim. They set out a model
and reef snails can serve as reservoir of the pathogen. Other human enteric bacteria can for improved investigation, based
be isolated from the mucus of corals in the Keys; it is likely that the emergence of WPX on standard biomedical and vet-
is linked to expansion of marinas and coastal dwellings in the region and the widespread erinary principles. A decade later,
use of septic tanks, which release sewage runoff to the sea. However, ongoing monitoring this requirement is still unfulfilled.
of diseased corals suggests that other pathogens besides S. marcescens also contribute to Analysis of 492 papers investigat-
WPX signs. ing coral disease showed that only
one used microscopic pathology
(Work and Meteyer, 2014). Sweet
Protistan parasites may cause tissue and Bythell (2017) describe seven
necrosis and skeletal erosion steps combining “traditional C19th
approaches with C21st technologi-
A condition known as Brown Band Disease has been observed in necrotic tissue of Acropora cal advancements” that researchers
spp. and other corals on the GBR. Colonies show a characteristic brown band which is fol- should follow to prove causation.
lowed by bare skeleton as the band progress across the coral branch. Under the microscope, a Burge et al. (2016) provide several
mass of ciliates can be seen actively moving over the surface of the coral at more than 1 mm case studies showing successful
per day, devouring the tissue. Large numbers of ingested zooxanthellae can be seen inside combination of traditional and
modern diagnostic methods for
the ciliate cells, giving rise to the brown color. These are thought to remain photosyntheti-
diseases of corals and other marine
cally active and may provide nutrients to the ciliate, or perhaps more importantly provide
organisms.
oxygen that may offset hypoxia that arises from the intense metabolism of the ciliate mass.
300 Chapter 11

Figure 11.6 Elkhorn coral (Acropora

palmata) infected with white pox
disease. Credit: NOAA Fisheries, SE
Regional Office.

The ciliate is a previously undescribed member of the class Oligohymenophorea, subclass

Scutcociliata.

Another ciliate, Halofolliculina corallasia, has been associated with a condition called
Skeletal Eroding Band—a moving dark band observed on a wide range of branching and
massive corals on Indo-Pacific reefs. In the skeleton behind the moving front, the ciliates are
packed into an indistinguishable dark mass due to the black housing (loricae) of the ciliate.
It kills the tissue and damages the skeleton. Protistan parasites are also frequently found in
aquarium corals. A characteristic “brown jelly” is often observed to move across the surface
of the coral. A ciliated protist, possibly Helicostoma nonatum, has been found in large num-
bers actively feeding on coral tissue.

Viruses have a pivotal role in coral health

Given the ubiquitous role of viruses in diseases of other animals and plants, it is surprising
that their role in diseases of corals and other benthic invertebrates was overlooked for many
years, with only a small number of investigators studying the topic until recently. Early studies
mostly relied on the use of transmission electron microscopy to demonstrate virus-like par-
ticles (VLPs) in corals, some of which are believed to infect the coral animal tissue and some
the zooxanthellae. This led to the hypothesis that the zooxanthellae may contain latent viruses
that are triggered into a lytic infection by environmental changes associated with bleaching,
such as elevated temperature, UV irradiation, or pollution. Other microscopy studies indicate
that a diverse range of viruses infect coral tissues and they can also be observed in the surface
mucus layer, but identifying viruses based on capsid size and shape is unreliable. More useful
evidence comes from recent application of metagenomic and metatranscriptomic techniques,
which have revealed about 60 virus families associated with corals, of which 9–12 families
form a core coral virome. Tailed phages with a head ~ 60 nm (dsDNA, Caudovirales) in the
families Siphoviridae, Podoviridae, and Myoviridae are highly abundant—these infect the
bacterial members of the coral holobiont. Nucleocytoplasmic large DNA viruses (especially
Phycodnaviridae, Mimiviridae, Poxviridae, Iridoviridae, and Ascoviridae) and herpes-like
viruses occur in all or most coral samples investigated – these infect eukaryotic cells, either
the coral host or the zooxanthellae. Other common viral groups encountered include ssDNA
Circoviridae and ssRNA Retroviridae. Several metagenomic studies comparing the virome
of healthy, bleached, and diseased corals have now revealed shifts in the relative abundance
of different virus groups. Temperature and nutrient stress can increase the abundance of
particular types, such as herpes-like viruses. For example, Circoviridae are most abundant in
corals showing signs of WPD and Phycodnaviruses increase in bleached corals, supporting
the latent virus hypothesis for bleaching. Small ssRNA viruses related to HcRNAV (p.213)
may also infect the zooxanthellae. Further evidence comes from examination of whole tran-
scriptomes of cultures of Symbiodiniaceae species, which reveals upregulation of virus-like
gene expression following stress experiments. At present, there is no clear evidence of viral
etiology for any specific coral disease, but there is growing evidence for a connection between
 Microbial Diseases of Marine Organisms 301

the activity of viruses and the health of reefs; not just through overt bleaching, tissue loss and
disease, but also through modifying the flux of nutrients and biogeochemical processes in the
reef water column, through a “viral vortex” of reef decline driven by anthropogenic drivers
such as eutrophication and climate change, as shown in Figure 11.7.

Sponge disease is a poorly investigated

global phenomenon
Sponges are highly important members of tropical, temperate, polar and deep-water marine
ecosystems. As discussed in Chapter 10, sponges are host to a highly diverse microbial popu-
lation comprising heterotrophic bacteria, cyanobacteria, archaea, protists, and fungi, the cells
of which can comprise almost half of the tissue volume in some species. This makes it espe-
cially difficult to determine the involvement of pathogenic microbes in disease. Although
sponge diseases have been reported worldwide for more than 100 years, outbreaks have
increased dramatically in recent years. Most diseases appear as white or colored spots and
patches on the sponge surface, followed by necrosis and damage to the internal structures due
to bacterial digestion of internal fibers—the sponge body crumbles as a result of water pres-
sure. One of the best-documented major epizootics occurred in the Caribbean in 1938, result- Figure 11.7 Overview of the virus-
ing in loss of 70–95% of sponges in some parts. The commercial collection of sponges in the mediated vortex of coral reef decline.
Mediterranean and Ligurian Seas suffered massive losses in the 1980s and in 2008–2009 and Coral reef viruses are envisaged as
some aquaculture systems for sponge cultivation have also experienced disease outbreaks. having a substantial effect on eco-
The microbiological investigation of sponge disease remains limited and the etiological system function because nutrients
agents and environmental factors are poorly identified. A novel strain of Pseudoalteromonas that are released through viral lysis
agarivorans has been identified as a pathogen of the GBR sponge Rhopaloeides odorabile. are readily transferred between the
Disease in deep-sea sponges has also been recorded and there are concerns that this could reef benthic (BC) and water column
impact the functioning of benthic systems in Norwegian fjords. As in corals, it seems likely (WC) compartments, as if through
that sponge disease is initiated by imbalance in the holobiont microbiome, followed by oppor- a “revolving door” Stressful condi-
tunistic or polymicrobial infections. Systematic surveys of the health status of sponges in reef tions can trigger excessive microbial
ecosystems, accompanied by detailed microbiological investigation, are being conducted to growth and viral production in coral
complement our growing knowledge of coral diseases. colonies resulting in the release of
coral cellular contents (e.g. mucus
exudates) into the WC. Some of
these materials move from the BC to
the WC in which they can stimulate
primary and ultimately secondary
production. Increased abundances
of heterotrophic bacteria can lead
to O2 drawdown in the WC adja-
cent to corals, which contributes
to hypoxic conditions and colony
mortality. Concurrently, viral lysis
of autotrophic and heterotrophic
microbial blooms in the WC results
in additional C and N flux, which can
be transferred to the BC and cause
further shifts in microbial com-
munities and viral production, and
increased coral disease and mortality.
The nutrients that are subsequently
released then feedback on the
ecosystem (for example, to increase
primary and secondary production
in the WC) and further degrade
reefs. Stony coral die-off on reefs can
ultimately result in a shift in domi-
nant cover on reefs from scleractin-
ian corals to macroalgae, sponges or
soft corals (dashed arrow). Reprinted
from Thurber et al. (2017) with per-
mission of Springer Nature.
302 Chapter 11

BOX 11.2 RESEARCH FOCUS

Caribbean corals under threat from a new disease syndrome

Stony Coral Tissue Loss Disease (SCTLD). A major new coral fastigiata. Intensive efforts are underway to collect tissue samples
disease is currently spreading throughout the Caribbean, result- from different coral species and geographical areas for histopatho-
ing in extremely rapid tissue loss with up to 100% mortality in logical, microscopic, and molecular analysis. Sentinel sites have
the most susceptible species affected. The syndrome is character- also been established to monitor rates of transmission and spatial
ized by rapid tissue loss that progresses as a band from the base progression of the disease (Sharp and Maxwell, 2018).
of the colony to the margins or as multiple coalescing blotches
(Figure 11.8A). Small colonies often die within a few weeks to What is causing the disease? Many scientists investigating the dis-
months, while infections can persist on larger colonies for several ease suspect that waterborne transmission and bacterial infection is
seasons. The rapidity of disease progression and the range of coral responsible but identifying the causative agent(s) has so far been
species affected distinguishes it from other previously character- unsuccessful. As with other marine diseases, it is very difficult to
ized syndromes like WPD and WPX. SCTLD was first identified determine the role of primary and opportunist pathogens and the
in 2014 at the northern end of the Florida Keys near a site with effects of disturbance of the normal microbiome (dysbiosis) The sit-
increased sedimentation due to dredging (Miller et al., 2016; Precht uation is made even more complex by the large number of different
et al., 2016) and by 2019 has spread throughout the entire Florida coral species affected—indeed, we cannot be sure that all species
reef tract, the Mexican Caribbean, Turks and Caicos, Dominican are suffering from the same disease, due to the lack of diagnos-
Republic, and Belize. A survey by Walton et al. (2018) found losses tic tools. The first detailed analysis of the microbiomes of affected
of over 30% of coral colony density and 60% of live tissue area corals was undertaken by Meyer et al. (2019) who sequenced 16S
in the Southeast Florida reef tract and concluded that ecosystem rRNA PCR amplicons from tissue and mucus samples taken from
function in some areas has been altered to a point where recovery diseased and apparently healthy tissue from multiple samples of
is greatly challenged. the coral species Montastraea cavernosa, Orbicella faveolata,
Diploria labyrinthiformis, and Dichocoenia stokesii. As in other
Infection has been described in 22 coral species. Typically, the microbiome studies of coral diseases (Huggett and Apprill, 2019),
maze coral Meandrina meandrites, the star coral Dichocoenia there was evidence of a stochastic shift in microbial diversity pres-
stokesii, and the pillar coral Dendrogyra cylindrus are the first to ent in diseased and apparently healthy tissue, resulting in unique
show signs of infection at a new site. This is followed quickly by altered microbiomes. This complexity was compounded by dif-
infection of other highly susceptible species, including the brain cor- ferences observed in samples taken at different times of the year.
als Colpophyllia natans, Diploria labyrinthiformis, Pseudodiploria Meyer et al. were unable to identify any single potential pathogen,
strigose, and P. clivosa, and the smooth flower coral Eusmilia but five unique ASVs (see p.282) were enriched in the SCTLD

Figure 11.8 A. Rapid progression of SCTLD across a colony of brain coral, Pseudodiploria strigosa. B. Removal of an unaf-
fected star coral, Dichocoenia stokesii, for the Reef Rescue project. C. Long-term maintenance of corals at Florida Aquarium’s
Center for Conservation. D. Experimental “firebreak” procedure to prevent disease progression in Orbicella faveolata (see
text). Credits: A–C. Florida Fish and Wildlife Conservation Commission; D. Brian Walker, Nova Southeastern University.
 Microbial Diseases of Marine Organisms 303

BOX 11.2 RESEARCH FOCUS

lesions of three of the corals. These were classified as belonging skeleton, which is filled with chlorinated epoxy resin to halt disease
to the orders Flavobacteriales, Clostridiales, Rhodobacterales, progression (analogous to a firewall) as illustrated in Figure 11.8D.
Alteromonadales, and Vibrionales. Some of these ASVs were exact The possibility that viruses or protistan parasites are also involved
matches for sequences isolated from other disease syndromes like cannot yet be excluded.
WPD and BBD. Other groups of opportunistic colonizers were also
enriched in diseased corals—these are presumed to be saprophytic, The Florida Reef Rescue Project. The growing severity of the
feeding on the dead tissue Although this study is unable to reveal epizootic has prompted a multi-agency initiative with the aim of
specific pathogens responsible for the disease, it lays the founda- collecting thousands of specimens of unaffected corals of differ-
tion for further studies, including isolation and culture of pathogens ent species ahead of the disease front (Figure 11.8B, 11.8C). These
and better investigation of control methods such as antimicrobial are shipped to land-based holding facilities at the University of
treatments, probiotics, or phage therapy (see p.404). Some pre- Miami and Nova Southeastern University, where samples are taken
liminary experimental treatments in mesocosms and in the field for genotyping to aid individual tracking of each coral. Samples
have involved application of antibiotic pastes, with some success are being distributed to over a dozen aquaria around the USA to
in halting the progress of disease lesions (FKNMS, 2018). Another serve as long-term breeding stock for possible future reef restora-
approach has been the creation of a trench chiseled into the coral tion (FFWCC, 2019).

Mass mortalities of echinoderms have caused

major shifts in reef and coastal ecology
The long-spined black sea urchin Diadema antillarum occurs throughout the Caribbean and
in tropical regions of the Atlantic Ocean. It is omnivorous, grazing on live coral, seagrasses,
and its preferred food source, benthic turf algae. In 1983, a mass mortality event began near
Panama, and quickly spread east and west at up to 3000 km per year, eventually affecting 3.5
million km2 of the entire Caribbean. Mortality was over 90% in all locations and the popula-
tion was completely wiped out in many areas. The deposition of dust from North Africa, fol-
lowing severe drought, has been suspected as a possible trigger for the outbreak (see p.298).
Recovery has been extremely slow, and only small populations at greatly reduced densities
survive. Although a host-specific waterborne infectious agent was suspected, the causative
agent was never discovered. The urchin played a key role in maintaining the balance of the
reefs by grazing on turf algae and massive increases in algal cover occurred within weeks
of the epizootic. Freed from this top-down control, the Caribbean reefs underwent a major
ecological phase shift from a coral-dominated to an algal-dominated system from which
they have never recovered—with the situation compounded by subsequent coral diseases
discussed above, pollution, and overfishing.

In the summer of 2013, diseased asteroids (sea stars) began washing up onto the shores of
the Olympic Peninsula and British Columbia on the Pacific north-west coast. The animals
showed loss of body turgor, missing limbs, and tissue decay, quickly leading to a disintegrated
mush littering the seabed (Figure 11.9). The condition became known as “sea star wasting
disease” (SSWD) and intensive efforts were mobilized to elucidate the cause and investigate
the ecological effects of the epizootic. Over 20 species of asteroids have been affected, with
the ochre star (Pisaster ochraeus) and sunflower star (Pycnopodia heliamthoides) being par-
ticularly hard hit. Within a year, surveys showed that populations of P. ochraeus were drasti-
cally reduced along the entire Pacific coast from Alaska to southern California, with declines
of over 75% in all areas and up to 99% in southern California. Sustained high seawater tem-
peratures in 2014 and 2015 are thought to have exacerbated the disease’s impact. The outlook
for recovery of sea star populations is uncertain. P. ochraeus is a keystone predator and it
is likely that its reduction or disappearance in some regions will result in major ecological
changes to subtidal and intertidal systems.

The causative agent(s) of SSWD are still unclear, despite many attempts to identify viruses,
bacteria, or fungi that might be responsible. The best lead has come from the finding that
0.22 µm-filtered extracts of infected tissue reproduced disease signs in experimental infec-
tion. A ssDNA virus in the Parvoviridae family (sea star associated densovirus, SSaDV) was
identified and associated with the disease in some species. However, the virus has been found
304 Chapter 11

Figure 11.9 The purple ochre sea

star Pisaster ochraeus disintegrat-
ing as it dies from sea star wasting
syndrome. Credit: Elizabeth Cerny-
Chipman, Oregon State University
via Wikimedia Commons, CC BY 2.0.

to be present in some asymptomatic asteroids and in specimens preserved long before SSWD
was observed. Definitive proof of etiology—as we have seen with so many other marine dis-
eases—is inconclusive. Indeed, detailed analysis shows densoviruses are only consistently
associated at a population level with one species, P. helanthioides, and disease signs differ
subtly between different species. This, the term SSWD may wrongly imply common etiology
of a single condition, and there may be a complex repertoire of interactions between environ-
mental conditions and different pathogenic agents.

DISEASES OF MOLLUSCS
Bacteria are a major cause of disease in molluscs
Numerous Vibrio spp. have been described as part of the normal microbiota and as disease
agents in bivalve and gastropod molluscs. Vibrios are the major cause of disease in intensive
hatcheries, where mortalities of larvae and juveniles can reach 100%. Until the introduction
of genomic methods, classification and identification of this group has been very confused,
but several species have been particularly implicated, namely V. alginolyticus, V. anguilla-
rum, V. splendidus, V. aesturianus, V. neptunius, V. tubiashii, and V. coralliilyticus. These
organisms are all found as members of the normal microbiota of seawater and in biofilms on
marine surfaces. Growth is encouraged by accumulation of organic matter, so careful moni-
toring of water quality and temperature is essential to prevent disease outbreaks in hatcher-
ies. Larvae of different bivalve species seem quite variable in their sensitivity to vibrios and
there are also marked differences in the virulence of bacterial isolates. Extracellular toxins
(hemolysins, proteases, and ciliostatic factors) have been identified.

Disease resulting from vibrio infections is less of a problem in adult molluscs, but impor-
tant exceptions in mariculture include regular epizootics in cultured Portuguese oyster
(Crassostrea gigas) caused by V. splendidus and related species, and high mortalities of the
European abalone (Haliotis tuberculata) caused by V. harveyi, which also infects cultured
pearl oysters (Pinctada maxima) in north-western Australia. V. tapetis causes brown ring
disease in cultured Manila clams (Ruditapes philippinarum); the pathogen attaches to the
clam tissue, causing abnormal thickening and a characteristic brown ring along the edge of
the shell. Mass mortalities due to V. tapetis have caused severe economic losses on the entire
Atlantic coast of Europe since the 1980s. The persistence of the bacteria in molluscan tissues
depends partly on their resistance to the bactericidal activity of the hemolymph, a natural
defense mechanism. The effect of environmental factors (especially temperature) and stress
on the neuroendocrine systems of the host are very important in determining whether these
interactions are benign or lead to disease.

Juvenile oyster disease (JOD) results in seasonal mortalities of hatchery-produced juvenile

oysters (Crassostrea virginica) raised in the north-eastern United States. The disease first
 Microbial Diseases of Marine Organisms 305

appeared in the late 1980s, and losses exceeding 90% of total production have since occurred
FULFILLING KOCH’S
in the states of Maine, New York, and Massachusetts. There are several similarities to brown
ring disease, indicating a bacterial etiology. Growth rate is reduced, and the shell becomes i POSTULATES FOR AN
OYSTER DISEASE
fragile and uneven, with proteinaceous deposits (conchiolin) on the inner shell surfaces, fol-
lowed by sudden high mortality (Figure 11.10a). Infected animals are heavily colonized by a The “gold standard” evidence for
previously undescribed species of the Roseobacter group of the Alphaproteobacteria, named proving disease etiology is fulfil-
as Roseovarius crassostreae. Attachment of the bacteria to the oyster tissue via polar fim- ment of Koch’s Postulates (KP).
briae is thought to be a key factor in pathogenesis (Figure 11.10b). The disease is now known The following conditions must be
as Roseovarius oyster disease (ROD) to emphasize its specific etiology. met: (1) the microbe suspected of
causing the specific disease must
be associated with all cases; (2) it
Gliding bacteria of the Cytophaga–Flavobacterium–Bacteroides (CFB) group can infect the
must be grown in pure culture; (3)
hinge-ligament of the bivalve shell, leading to liquefaction via the production of extracellular the disease must be reproduced
enzymes, and interference with respiration and feeding. CFB appear to be weak opportunist by inoculating healthy individuals
pathogens, and infection is precipitated by poor nutrition of the animals, rising water tempera- with the pure culture; and (4) the
tures, or other environmental stresses. Nocardia crassostreae causes high mortalities of cultivated specific microbe must be re-iso-
C. gigas and Ostrea edulis oysters in Japan and on the west coast of Canada and the United States. lated from experimentally infected
hosts. The investigation of juvenile
Intracellular rickettsia-like and chlamydia-like pathogens are widespread and have been oyster disease (JOD) by Maloy
reported in at least 25 species of bivalves. These infections frequently produce little evidence et al. (2007) is a rare example of a
of tissue damage, and mortality in adult animals is usually low, except in conditions of envi- rigorous, model approach to prove
KP for a marine disease. Maloy
ronmental stress such as sudden temperature change. The larvae are usually very susceptible,
et al. showed that Roseovarius
leading to problems in aquaculture hatcheries of scallops (Argopecten irradians) and other crassostreae is consistently associ-
bivalves. The morphology of the bacteria within the cells of the digestive gland and gills is ated with diseased animals (>90%
very similar to that of known rickettsias and chlamydia, but detailed identification and taxo- of the total bacterial community)
nomic studies are limited because these are obligate intracellular pathogens that can only be but is never found in healthy oys-
grown in suitable cell cultures. Francisella halioticida has been recognized as the cause of ters. They developed a laboratory
mass mortalities in some cases. An uncultured rickettsia-like organism, “Ca. Xenohaliotis model to reproduce the disease;
californiensis,” has been identified as the cause of a lethal disease (withering syndrome) exposure to R. crassostreae caused
of the black abalone Haliotis cracherodii at high seawater temperatures. This large edible JOD-like mortality including the
gastropod was once highly abundant on the west coast of North America and was an impor- characteristic pathology. The tim-
ing of bacterial colonization paral-
tant food source harvested for ~10,000 years. Withering syndrome was first reported on the
leled the development of disease
Californian coast in the 1990s and spread rapidly. Stocks were already heavily depleted by
signs in a natural setting. The bac-
overfishing, and the mollusc is now critically endangered. teria were detected simultaneously
to the first microscopic disease
signs in actively growing oysters,
Several protistan diseases affect providing strong evidence that
culture of oysters and mussels they are the primary pathogen.

The most widespread and destructive protistan disease, described since the 1970s, is bon-
amiasis caused by Bonamia ostrae, B. exitiosa, B. perspora, and B. roughleyi, affecting a
wide range of oyster species along the entire Atlantic coast of Europe, Canada, USA, New
Zealand, and SE Australia. The disease is characterized by yellowing of tissue, lesions on
gill and mantle, and breakdown of connective tissue, with mortalities up to 90%. Aber dis-
ease in Ostrea edulis and Mytilus galloprovincialis culture on the European Atlantic and
Mediterranean coasts is caused by Marteilia refringens, leading to a pale digestive gland,
emaciation, and tissue necrosis. A closely related species, M. sydneyii has recently emerged
as a seasonal infection of rock oysters farmed on the east coast of Australia; QX disease leads
to necrosis of the digestive glands, loss of condition, and reabsorption of the gonad. Starvation

Figure 11.10 Roseovarius oyster

disease (ROD). (a) Shell of infected
oyster showing deposition of con-
chiolin on the inner surface of both
valves. (b) TEM of negatively stained
cell of Roseovarius crassostreae show-
ing one bacterium with a flagellum
and a partical cell with a tuft of polar
fimbriae. Credit: Katherine Boettcher,
University of Maine.
306 Chapter 11

and death of the oyster follow when energy reserves are depleted. A polychaete worm has
THE NUCLEAR
i OPTION—BACTERIAL
PARASITES OF DEEP-
been identified as an intermediate host. Dermo disease in the eastern oyster Crassostrea vir-
ginica is caused by Perkinsus marinus. This infects the hemocytes of oysters, causing severe
emaciation, loss of condition, and high mortality rates, depending on temperature and salin-
SEA MUSSELS
ity. It is naturally prevalent in the Gulf of Mexico, south-eastern United States, and Brazil.
Although symbiotic interactions Culture of C. virginica on the Pacific coast of Mexico and California has led to introduction
between bacteria and molluscs of the disease there. Waterborne spores of the parasite are transmitted over large distances.
found at vents and seeps have
been studied intensively (see
Chapter 10), little is known about Virus infections are a major problem in oyster culture
parasitic interactions. Virus-like and
rickettsia-like inclusions have been France is one of the leading producers of cultivated oysters and has been particularly sus-
observed in TEM of Bathymodiolus ceptible to disease caused by viruses, as well as the vibrio infections previously described.
mussels and may be linked to mor- A major epizootic of gill necrosis erupted in France in the 1960s in Portuguese oysters
tality (Ward et al., 2004; Mills et (Crassostrea angulata), with tissues displaying the presence of large inclusion bodies con-
al., 2005). Using 16S rRNA analysis, taining icosahedral iridovirus particles in the tissues. This disease spread rapidly along the
Zielinski et al. (2009) described a
Atlantic coast of Europe, virtually destroying the European oyster fishery. Pacific oysters
parasitic bacterium, which they
named “Ca. Endonucleobacter
(Crassostrea gigas) were introduced to replace the lost stock, but these also suffered high
bathymodioli.” FISH probes mortalities. A serious outbreak of another viral disease killed over eight billion oysters in
showed that the bacterium has France in 2008. This was shown to be due to the ostreid herpesvirus (OsHV1), which had
a very atypical lifestyle, invading previously been associated with epizootics in C. gigas in Japan, the USA, and Australia
the nuclei of intercalary gill cells since the development of high-density culture methods in the 1940s. It is thought that the
between the bacteriocytes that 2008 outbreak followed a mild, wet winter with high nutrient levels, which encouraged the
contain the chemosynthetic symbi- young oysters to invest more energy in the development of sex organs, weakening their natu-
onts. Infection of a nucleus begins ral defense mechanisms against viral attack. Highly virulent variants of OsHV-1 have now
with a single rod-shaped bacte- spread globally and are causing major losses on the US west coast and in Australia, especially
rium, which grows to an aseptate
Tasmania. Susceptibility to the disease has a genetic basis and it may be possible to breed
filament of up to 20 µm in length,
and then divides repeatedly until
more resistant varieties of C. gigas.
the nucleus becomes greatly swol-
len with up to 8 × 104 bacteria. The Another group of viruses known as the marine birnaviruses (MABVs) are widely distributed
nuclear membrane then bursts, as disease agents in marine invertebrates (as well as in fish; see below). In Japan, infection
destroying the host cell. The bac- by MABVs causes considerable losses in the culture of pearl oysters (Pinctada fucata). PCR
teriocytes never become infected, screening shows that 60% of bivalves and 35% of gastropods from a range of wild species in
suggesting that the symbiotic Japanese waters are positive for MABVs and that zooplankton may act as a vector.
bacteria in these cells protect their
host nuclei against the parasite. Papova-like viruses have been described in the gonads of several bivalve species, espe-
cially oysters, leading to hypertrophy of gametocytes. Viruses commonly cause secondary
infections and may be associated with infection by parasites. Many other VLPs have been
observed in marine bivalves and are often associated with disease signs, although proof of
etiology and study of pathogenesis is usually lacking, except in cultured species. In most
cases, identification of VLPs in electron micrographs has been on the basis of size, morphol-
ogy, and intracellular location; molecular characterization and propagation are needed.

Changes in temperature or salinity, human handling, and increasing stocking density are
particularly important factors contributing to susceptibility to bacterial, protistan, and viral
diseases of molluscs. There is currently considerable research on the immune defense mecha-
nisms of molluscs in the hope that genetic improvement of disease resistance will lead to
better disease control.

DISEASES OF CRUSTACEANS
Bacteria cause epizootics with high
mortalities in crustaceans
Various Vibrio species are again the major causes of devastating losses in hatcheries and
grow-out stages of tropical shrimp and prawn culture, the most important of these being V.
harveyi. Because of its bioluminescence, a large V. harveyi outbreak in prawns can result
in a spectacular greenish light in infected ponds, leading to the name “luminous vibriosis.”
In hatcheries, vibrios attach to the feeding appendages and oral cavity of the larvae, which
become weakened and swim erratically. The virulence of different strains of V. harveyi
 Microbial Diseases of Marine Organisms 307

varies greatly. Some have a minimum lethal dose (100% mortality) of only 100 CFU mL−1,
whereas other strains are nonvirulent even at 106 CFU mL ‒1. Virulent strains produce two
lethal protein exotoxins, which probably act in the larval intestinal tract on the gut epithelial
cells by facilitating passage across the gut and colonization of other tissues. The closely
related V. nigripulchritudo and V. penaeicida are major disease agents of prawn and shrimp
culture in New Caledonia, Japan, and southwest India and V. owensii has been identified
associated with diseased cultured crustaceans in Australia. There is some evidence to sug-
gest that acquisition of a plasmid may be responsible for the emergence of highly virulent V
nigripulchritudo, which produce an exceptionally toxic protein, nigritoxin. This is lethal for
crustaceans and also insects and may find application as a novel insecticide.

In lobsters and crabs, the exoskeleton is covered by a thin protective lipoprotein layer, and
damage to this layer enables various bacteria (especially vibrios) to attach to the chitin exo-
skeleton. It is possible that biofilm growth of lipolytic bacteria may facilitate penetration
resulting in “bacterial shell disease,” in which pits in the exoskeleton are produced by bac-
terial chitinase activity. These can erode to form deep lesions and the bacteria may pen-
etrate the tissues. The disease has a major impact on some populations of the American
lobster (Homarus americanus), with consequent serious effects on the lobster fishery of the
north-eastern coast of the United States and Canada. Another very damaging pathogen of
lobsters is gaffkemia, caused by the Gram-positive Aerococcus viridans var. homari (previ-
ously known as Gaffkya homari). The disease is usually found in holding tanks or ponds
and is of considerable concern to lobster fishers because of the high value of their catch. It
is highly contagious and is probably acquired from a reservoir in wild animals, which show
an infection rate of 5–10%. Schemes to restock depleted fisheries have been badly affected
by this disease, and there is considerable evidence that disease spread is linked to the move-
ment of animals caught from infected wild stocks and transported over large distances. The
bacterium gains entry to the lobster hemolymph via abrasions in the shell and multiplies very
rapidly at higher water temperatures (>10°C). This explains differences in severity of the
disease observed during summer and winter impoundment. Virulence seems to be strongly
connected with a capsule that prevents agglutination in the hemolymph. Antibiotics such as
oxytetracycline are sometimes used as a preventive measure in holding ponds, but there is
concern about antibiotic residues in treated lobsters; a withdrawal period of at least 30 days
should be observed.

The intracellular rickettsias and mycoplasmas can cause epizootics with high mortalities in
crustaceans such as crabs, lobsters, and penaeid prawns, owing usually to infection of the
hepatopancreas. Such epizootics can have a marked effect on marine ecology, for example,
the outbreaks caused by currently unidentified rickettsias affecting Dungeness crabs (Cancer
or Metacarcinus spp.) in Europe and the USA. Diagnosis of disease is difficult, being based
largely on histopathology, but the recent introduction of molecular diagnostic techniques is
leading to improved understanding of their ecology.

Expansion of crustacean aquaculture

is threatened by viral diseases
A number of crustacean viral diseases are important in both wild and cultured populations of
decapod crustaceans such as lobsters, crabs, prawns, and shrimp. Intensive and semi-inten-
sive aquaculture of prawn and shrimp now accounts for about half of the total value of aqua-
culture production (>USD 250 billion). The global demand for prawns and shrimps seems
insatiable, and much of the rapid expansion of intensive culture development has occurred in
Asia (especially China, Thailand, Vietnam, and Indonesia) and Central America (especially
Ecuador). Initial lack of government regulation led to severe problems of habitat destruction
and eutrophication of coastal waters. With a better understanding of these problems and
improved regulation of location and operation of prawn and shrimp farms, infectious diseases
are now considered one of the major limiting factors of further development of the industry,
alongside understanding the nutritional needs of animals and improving sustainable supply
of feeds. For most crustacean species, culture still depends on the use of larvae produced in
hatcheries, mainly from sexually mature females caught in the wild. Thus, genetic selection
of disease-resistant animals is not possible, although it is likely that closure of the life cycle
308 Chapter 11

for some species will be achieved soon. This will decrease the pressure on natural stocks and
enable selection of disease-resistant brood stock. Some crustacean species are cannibalistic
or are indiscriminate feeders that will consume feces and molted exoskeletons, leading to
rapid disease transmission through ponds and holding tanks. Until recently, knowledge of the
ecology of pathogens and application of diagnostic methods has been poor. The development
of gene probes and Q-PCR techniques is assisting in control methods such as selection of
larvae and restrictions on movement of stocks unless certified as disease-free. Viruses also
remain infectious in frozen seafood products, and molecular methods are used for screening
imports and exports to prevent disease transmission. However, many of these techniques are
expensive and require highly trained personnel, and it is necessary to develop cheaper and
easier biosecurity methods for use by rural farmers in developing countries.

The expansion of intensive culture of shrimps and prawns, in both volume and geographic
distribution, has resulted in the number of well-characterized viruses increasing from six in
1988 to nearly twenty today. One of the most devastating diseases in Asia has been white spot
syndrome of penaeid prawns. This was first recognized in eastern Asia and has spread world-
wide, with current losses estimated at over USD 1 billion. High mortalities (80–100%) result
within 2‒3 days for juveniles and 7‒10 days for adults. The agent (WSSV) is a large, envel-
oped, rod-shaped, double-stranded DNA virus, recently named as Whispovirus and assigned
as the sole member of the group Nimaviridae. Analysis of the genome sequence of the virus
indicates that it differs from all known viruses, although some genes are weakly homologous
to herpesvirus genes.

Whispovirus has a broad host range among crustaceans and infects many tissues, multiplying
in the nucleus of the target cell. It is not known how the virus enters the shrimp and spreads
within the body, but recent studies suggest that a viral envelope protein may interact with a
host chitin-binding protein. There have been many studies on the immune responses of the
host to infection. Whispovirus is a lytic virus and causes destruction of host tissues. Analysis
of crustacean genomes is revealing the evolutionary history of the virus, showing that exten-
sive acquisition of host–virus interaction-related genes was a key event in the origin of the
extreme virulence of Whispovirus. Disease transmission has been controlled in some regions
by use of integrated ELISA- and PCR-based screening of larvae to ensure that they are free of
the disease (specific pathogen-free, SPF) before transfer to ponds; however, there is consider-
able sequence variation, which may limit the effectiveness of screening techniques. Animals
acquire the infection from the water and show white spots on the inner surface of the shell
(Figure 11.11). Interestingly, the white spots appear to be caused by a chitinase, and this
enzyme may have been transferred to the viral genome by phage conversion.

Crustaceans may also be infected by parvoviruses, including infectious hypodermal and

hematopoietic necrosis virus (IHHNV), hepatopancreatic parvovirus (HPV), and spawner
mortality virus (SMV). SMV emerged in the 1990s in Queensland, Australia, and has been
responsible for mortalities of 25‒50% in black tiger prawns (Penaeus monodon). The virus
originates in wild brood stock, as 25% of female spawners carry SMV. Again, integrated
PCR and ELISA tests are being developed for high-throughput screening in an attempt to
produce SPF stocks. Sequence analysis shows that there is considerable homology between
SMV and some insect viruses. The four shrimp parvoviruses fall into two different clades
that group with different insect parvoviruses.

Figure 11.11 Head of penaeid

prawn infected with Whispovirus,
showing characteristic white spots
on the exoskeleton. Credit: Donald
Lightner, University of Arizona.
 Microbial Diseases of Marine Organisms 309

Taura syndrome virus (TSV) is a small, non-enveloped, icosahedral, ssRNA virus belonging
DO VIRUSES INFECT
to the family Dicistroviridae; it has caused heavy mortalities in central America and the USA
since the 1990s and several variants are known. Other viruses causing significant diseases in ? CRUSTACEAN
ZOOPLANKTON?
shrimp and other crustaceans include baculovirus penaei virus (BPV), monodon baculovirus
(MBV), and midgut gland necrosis virus (MGNV). Animals acquire these viral infections As the most abundant members
from water, although it has been suggested that seabirds could act as a vector of some shrimp of the zooplankton, the small crus-
viruses by transmitting infectious virions in their feces. Larvae of aquatic insects may also taceans known as copepods have
be reservoirs of infection. great significance in global marine
ecology and nutrient cycling.
They are the main food source
Despite the importance of wild crustaceans such as crabs and lobsters in natural popula-
for larger zooplankton, many fish,
tions and fisheries, knowledge of the importance of viruses in these species is very limited.
whales, and seabirds but there is
Baculoviruses, a parvo-like virus, and reoviruses have been reported in the hepatopancreas also evidence of non-predatory
of Carcinus spp., and early reports of viruses resembling herpesviruses, picornaviruses, mortality (Elliott and Tang, 2011).
rhabdoviruses, and bunyaviruses in other crabs have also largely not been pursued since their Could their population density be
original descriptions in the 1970s. Application of microscopic and molecular methods is controlled by viral infection, as
needed to demonstrate the importance of viruses in the ecology of wild decapod crustaceans. occurs with many phytoplankton
(see Chapter 7)? Drake and Dobbs
(2005) investigated this in a model
Parasitic dinoflagellates also cause disease in crustaceans system by exposing Arcartia tonsa
copepods to natural concentrations
Dinoflagellates in the genus Hematodinium and related genera can be very significant patho- of viruses in seawater. They con-
gens of commercially important crustaceans in many parts of the world, with devastating cluded that exposure to viruses had
effects on fisheries for crabs and lobsters in North America and langoustines in Europe. no effect on copepod fecundity,
Infection causes pathological alterations to the organs, tissues, and hemolymph of the animal. or survival of larvae or adults. The
Juveniles and females often show high prevalence of infection, resulting in sudden crashes authors suggested that the effect
in populations. At lower levels of infection, biochemical changes in the tissues result in the of viruses on higher trophic levels
production of bitter-tasting compounds, making the crabmeat unmarketable. may be much lower than that on
smaller hosts such as protists and
that the effects of viral infection in
crustaceans may be below the lim-
DISEASES OF FISH its of detection (although they did
not determine whether infection
Microbial diseases of fish cause major losses actually occurred). Further studies
in aquaculture and natural populations are needed, as Dunlap et al. (2013)
found molecular and microscopic
The first scientific description of microbial diseases in fish can be traced to the description evidence of abundant and actively
of “red pest” in eels in Italy by Canestrini in 1893. This disease, which we now know as replicating novel ssDNA circo-like
vibriosis caused by the bacterium V. anguillarum, led to mass mortalities in migrating eels viruses in the tissues of two cope-
during the eighteenth and nineteenth centuries. Such large-scale fish kills in the wild occur pod species.
occasionally, particularly in estuaries and on coral reefs, and may be caused by a wide range
of bacterial, viral, or protistan infections, or by harmful algal bloom (HAB) toxins. Viral
hemorrhagic septicemia (VHS) virus (genus Novirhabdovirus) is thought to be responsible
for large periodic fluctuations in the populations of wild shoaling fish in the North Sea, and
major epizootics in Australian pilchards have occurred because of a herpesvirus. Apart from
these mass mortality effects, it is hard to estimate the normal impact of disease on fish popu-
lations in the wild, since sick fish do not last long in the natural environment and are quickly
removed by predators. One of the few pieces of evidence that infection plays a significant
role in controlling natural fish populations comes from observations made in the 1970s that
hatchery-reared salmon immunized against vibriosis before transfer to the ocean had a 50%
greater survival than nonimmunized fish, as shown by the return rate of tagged fish. Fish
such as salmon and eels may be particularly susceptible to acute infections owing to the pro-
nounced physiological changes that occur during their migration from fresh- to saltwater, or
vice versa. Some pathogens are found in a high proportion of wild fish; these tend to be those
that cause slow-developing chronic diseases and may affect populations by impairing growth
and reproduction rather than sudden acute mortality.

The development of intensive mariculture in the 1970s led to a rapid growth in the science of
fish pathology. Early attempts to farm salmonid fish in intensive offshore pens and cages were
frustrated by large-scale mortality and heavy economic losses. This experience has recurred
with many different fish species in all parts of the world and at times has threatened the sur-
vival of the industry in some countries. The impact of disease should come as no surprise,
since it is common in all forms of intensive culture in which single species of animals are
310 Chapter 11

reared at high-population densities. Economic factors demand highly intensive systems, and
? WHAT ARE
EPIZOOTICS?
this can lead to stress of the cultured fish, which then succumb to disease transmitted rapidly
through the dense populations.
The term “epizootic” is used to
describe a situation where cases of Today, marine fish culture is a worldwide industry, with production approaching that of
disease within an animal popula- extensive fish capture. Atlantic salmon (Salmo salar) in Norway, Chile, Scotland, Canada,
tion rise sharply or suddenly above and Ireland is the most economically important species, with fish such as gilthead sea bream
the normal levels, which are (Sparus aurata) and sea bass (Dicentrarchus labrax) dominating culture in the Mediterranean
referred to as “enzootic.” It is a Sea. Japan has a long history of intensive mariculture, dominated by yellowtail (Seriola quin-
subjective term and is not quantita-
queradiata), ayu (Plecoglossus altivelis), flounder (Paralichthys olivaceus), and sea bream
tive. A small number of cases of an
unexpected rare disease affect-
(Pagrus major). A wide range of species are cultured in China and southeast Asia, often in
ing a small number of fish farms small-scale operations for local consumption. In recent years, there has been expansion in
might be described as an epizootic, the farming or ranching of high-value fish such as: Atlantic cod (Gadus morhua), mainly in
while the term would also be used Norway; turbot (Scophthalamus maximus), mainly in Spain and Portugal; or tuna (Thunnus
to describe cyclical increases in thynnus, T. orientalis, and T. maccoyi), in the Mediterranean, Japan, and Australia. While
disease causing mass mortalities there has been criticism of the sustainability and environmental impact of these ventures—
across a large geographic region. including disease problems—the benefits of closed life-cycle aquaculture of such commer-
These terms are the equivalent of cially important fish species relieves the pressure on natural populations, whose stocks are in
the familiar terms “epidemic” and rapid decline. In view of the need to feed the world’s growing population and recognition of
“endemic” used when describing
the environmental impacts of land-based agriculture (especially meat production), expansion
human diseases. The origin and
development of epizootics depends
of open-ocean aquaculture is likely in the coming years, but careful regulation of environ-
on particular circumstances affect- mental impacts will be essential.
ing the host and the pathogen,
with host population density and Microbial infections of fish cause a variety of disease signs
environmental factors often play-
ing a critical role. The ecological The need to implement effective control measures following a disease outbreak puts pressure
impact of epizootics depends on on investigators to determine the causes of mortality quickly. Experience, careful observation
the general status of the popula- of the stock by the fish farmer, and good record keeping play a large part in identifying dis-
tion before the disease outbreak eases. Microbial infection is usually characterized by a rising level of mortality accompanied
and the resilience of the ecosystem
by characteristic disease signs. These are very varied but may be broken down into three
in recovering from mortalities.
broad categories: (1) systemic bacteremia or viremia, in which there is rapid growth of the
pathogen, often with few external disease signs other than hemorrhaging; (2) skin, muscle,
and gill lesions; and (3) chronic proliferative lesions. Some highly lethal systemic infections
such as those caused by Aeromonas, Piscirickettsia, and infectious salmon anemia virus
(ISAV) may cause high mortality rates even when no external signs are present. In contrast,
other agents causing diseases with relatively low mortality rates—such as Tenacibaculum
maritimum, Moritella viscosa, and Lymphocystis disease virus—can result in external
lesions such as ulcers, necrosis, and tumor-like growths that make fish unmarketable.

Postmortem changes are very rapid as a result of overgrowth by the normal microbiota of
fish, so it is important to examine fish showing signs of infection before they succumb com-
pletely. External examination will often reveal the presence of gill and tissue erosion, eye
damage, hemorrhages, abscesses, ulcers, or a distended abdomen. Internal inspection may
reveal organ damage and fluids in the body cavity. To the experienced eye, these signs will
often indicate a particular disease agent, but the diagnosis must be confirmed by identifica-
tion of the pathogen.

For bacteria, traditional culture methods are still widely used, involving plating tissue sam-
ples onto various selective media and performing biochemical tests using diagnostic keys.
Not all bacterial pathogens are amenable to this approach, as some grow very slowly in
culture (notably Renibacterium and some mycobacteria). An alternative approach is the use
of methods such as ELISA or fluorescent antibody techniques (FAT) to detect bacterial anti-
gens in the blood or tissues, or to detect a high titer of host antibodies against the pathogen.
The diagnosis of viral infections is more difficult and time-consuming because it relies on
propagation of the virus in a suitable cell culture. FAT and ELISA are therefore the main
methods used for rapid identification of viral infection. There has been some success with
“dipstick”-type kits for rapid diagnosis, based on modifications of the ELISA technique.
Monoclonal antibodies have been produced against specific pathogens, allowing rapid iden-
tification of disease; they can also be used in fish health programs for screening brood stock
for previous exposure to pathogens. For many bacterial and viral pathogens, there are now
 Microbial Diseases of Marine Organisms 311

accurate molecular diagnostic tests based on PCR amplification and gene probes. Genomic
fingerprinting methods are often used for strain characterization, which is especially impor-
tant in studies of epizootics.

Fish-pathogenic bacteria possess a

range of virulence mechanisms
All bacterial infections involve a number of stages, with different bacterial products impor-
tant as virulence factors. The possession and expression of genes encoding these factors
varies greatly among the broad range of bacteria in various taxonomic groups that have been
associated with disease in marine fish. The initial infection and colonization of the host often
depends on the pathogen sensing the presence of exudates and excretion products from the
host. Chemotaxis using flagellar or gliding motility may result in association with specific
surfaces, such as the gut, skin, or gills. Bacteria often produce pili or some other surface
structures that enable firm attachment to epithelial surfaces.

The most critical stage for bacteria during colonization is the ability to overcome the innate
host defense mechanisms and adaptive immune system. Fish mucus provides a protective
inhibitory function and infection of undamaged skin is rare. Bacteria that penetrate the tis-
sues are subject to the antibacterial collectins and complement proteins found in serum,
and resistance to these is often due to modifications of the cell envelope. Virulent bacte-
ria may also possess mechanisms to resist phagocytic cells of the host, usually by posses-
sion of hydrophobic surface layers or by producing enzymes that cause lysis of phagocytic
cells. Some bacteria can survive within phagocytic cells of the host; these possess a range
of mechanisms to overcome the hostile low pH and antibacterial compounds found in the
intracellular environment.

As noted on p.237, bacteria require iron for the activity of essential cellular functions.
Vertebrate animals possess highly efficient systems for the transport and sequestration of
iron. The serum protein transferrin binds iron with extreme avidity, reducing the concen-
tration of free iron in tissues to 10 ‒18 M, about 108 times lower than the concentration that
bacteria require for growth. During infection, an even more efficient iron-binding protein
(lactoferrin) is released and sequestered iron is removed to storage in the liver and other
organs. Thus, successful pathogens must compete with the host’s iron-binding system to
obtain sufficient iron for growth, and many pathogens achieve this via the production of
siderophores and an iron-uptake system.

The final stage in disease is the damage to host tissues and body systems. Often this occurs
as a result of the host response “over-reacting” to the presence of the pathogen, producing
damaging cytokines and inflammatory products. Many pathogens produce toxins that can
cause death by affecting major organs or extracellular enzymes that cause destruction of
cells or tissues.

Vibrios are responsible for some of the

main infections of marine fish
As seen in the sections on invertebrate diseases, members of the family Vibrionaceae are
again prominent as pathogens. As noted above, V. anguillarum (also classified as Listonella
anguillarum) was the first species to be identified as a fish pathogen causing vibriosis in eels
(from which it derives its specific name), producing ulcers, external and internal hemorrhages,
and anemia. This “classical” vibriosis causes heavy losses in eel culture, especially in Japan,
and in a very wide range of marine species; it has caused particular problems in the culture of
salmon in North America, sea bass and sea bream in the Mediterranean (Figure 11.12), and
yellowtail and ayu in Japan. In all species, the disease is characterized by a very rapid gen-
eralized septicemia usually accompanied by external hemorrhages. A number of serotypes
of the pathogen exist, distinguished by differences in the O-antigen (lipopolysaccharide of
the outer membrane), but only the O1 and O2 serotypes are widespread. There are some
genetic differences between strains isolated from different geographic regions and fish spe-
cies; such knowledge is important in the development of vaccines for particular applications
312 Chapter 11

Figure 11.12 Vibrio abguillarum

infection of European seabass
Dicentrarchus labrax showing
characteristic hemorrhagic lesions.
Reprinted from Kalatzis et al. (2018),
CC BY 4.0.

(see Chapter 14). Vibriosis was one of the first fish diseases for which vaccines were devel-
oped and they have generally been very effective.

One of the major factors in virulence, which explains the extremely rapid growth in vivo, is
the possession of a 65-Mb plasmid (pJM1) that confers ability to acquire iron in the tissues
of the host. Proof of the key role of the plasmid in virulence of V. anguillarum was obtained
by inducing loss of the plasmid by curing and reintroduction by conjugation and by directed
mutagenesis of key genes. Mutation of any of the genes can result in a reduction of virulence
by as much as 105 times. In experimental infections in which iron uptake by transferrin is
swamped by injection of excess iron, possession of the plasmid does not confer an advantage.
The plasmid pJM1 contains genes for synthesis of the siderophore anguibactin, an unusual
hydroxamate–catechol compound, and outer-membrane transport proteins and regulators of
transcription involved in iron uptake (Figure 11.13). Expression of the multiple components
occurs only under low-iron conditions, owing to negative control at the transcriptional level
by the chromosomally encoded Fur protein and at least three plasmid-encoded regulators,
the best studied of which is AngR. Outer-membrane proteins only expressed under iron-
restricted conditions are known as IROMPs. Recently, additional siderophores and iron-
uptake mechanisms have been found.

Figure 11.13 Iron uptake via the

siderophore anguibactin in Vibrio
anguillarum. Anguibactin (Ang) is
secreted to the external environ-
ment, where it competes with the
host protein transferrin (T) for
circulating iron. Chelated iron (Fe) is
reabsorbed into the cell via an outer-
membrane receptor and membrane
transport proteins (TonB, ExbB,
ExbD) before transport across the cell
membrane via the ATP-dependent
fat system. Biosynthesis of anguibac-
tin, the outer-membrane receptor
and transport proteins are subject to
complex regulation at the transcrip-
tional level (not shown). Credit:
based on a drawing by Jorge Crosa.
 Microbial Diseases of Marine Organisms 313

Quorum sensing (QS; p.102) via acyl homoserine lactone (AHL) signaling molecules is also
an important factor in the growth and survival of V. anguillarum in its host and in the aquatic
environment. There are three separate pathways involved in gene regulation of biofilm for-
mation and production of protease, pigment, and exopolysaccharide. The hierarchical QS
system consists of regulatory elements homologous to those found in both Aliivibrio fischeri
(the LuxRI homologs VanRI,) and V. harveyi (the LuxMN homologs, VanMN). It is possible
that some of the AHLs have effects on eukaryotic host cells, including an immunomodula-
tory function.

Other virulence factors are also involved in V. anguillarum pathogenicity, including a metal-
loprotease and a powerful hemolysin causing lysis of erythrocytes. This may contribute to
the acquisition of iron by release of hemoglobin and probably accounts for the pale gills and
anemia typical of infection. The structure of the LPS confers resistance to the bactericidal
effects of the complement system present in normal serum. Also, the flagellum is essential for
motility across the fish integument as directed mutations in the flaA gene result in reduced
virulence. LPS antigens on the sheath of the flagellum (genes virA and virB) are also essential
for virulence, probably by contributing to biofilm formation in vivo.

Vibrio ordalii was originally thought to be a biovar of V. anguillarum but was designated as
a separate species based on a number of biochemical and genomic differences. It has caused
major losses in the cage culture of salmon in Pacific coastal waters off Oregon, Washington
State, and British Columbia. Infections with V. ordalii tend to be localized in muscle tissue
rather than the generalized infections seen with V. anguillarum. Virulence has not been so
well studied as that of V. anguillarum but complement resistance and a leucocytolytic toxin
have been described.

Aliivibrio salmonicida (formerly Vibrio salmonicida) causes cold-water vibriosis (Hitra

disease) in salmon and Atlantic cod farmed in northern regions such as Canada, Norway,
and Scotland during the winter, when water temperatures drop below 8°C. Disease signs
are broadly similar to those of V. anguillarum, but the two bacteria are serologically and
genetically distinct. The pathogen is excreted in the feces of infected fish and seems to have
good powers of survival in marine sediments, thus causing reinfection even if farm sites
are “fallowed” for a season. Epidemiological studies suggest that there is interchange of the
bacterium between populations of cod and salmon. Various strains can be distinguished by
different plasmid profiles, but there does not seem to be a close association between plasmid
possession and virulence. The bacterium produces various extracellular enzymes, a hydroxa-
mate siderophore, and a fur-regulated iron-uptake system, all of which are implicated in
virulence. Interestingly, significant amounts of the siderophore and the IROMPs required for
transport are only expressed at temperatures below 10°C, which may explain the increased
incidence of disease in the winter. This fact has been important in vaccine manufacture,
since temperature-regulated surface proteins are important antigens. Another species that
has caused salmon infections in cold northern waters was originally designated as V. viscosus
but was reclassified as Moritella viscosa. This causes a condition known as winter ulcer dis-
ease, in which ulcerous lesions progress from the skin to the underlying muscle. Mortality
rates are relatively low, but the appearance of the fish lowers its economic value considerably.

V. vulnificus causes infection of eels in Japan. The fish isolate is closely related to clinical
isolates associated with human disease (p.337) but has been designated a separate taxon
(biogroup 2) based on phenotypic, cultural, and serological properties. Iron acquisition and
production of a capsule, hemolysin, protease, and other toxins have all been implicated in
virulence.

Pasteurellosis affects warm-water marine fish

The expansion in the 1990s of sea bass and sea bream aquaculture in the Mediterranean and
yellowtail culture in Japan was accompanied by outbreaks of a new disease that has caused
heavy losses. The pathogen responsible was originally identified as Pasteurella piscicida, and
the disease is still known as pasteurellosis, although the causative agent has been reclassified
as Photobacterium damselae subsp. piscicida. When water temperatures exceed 20°C, acute
mortalities up to 70% can occur, especially in the larval and juvenile stages. A more chronic
314 Chapter 11

condition occurs in older fish. Interest in virulence mechanisms has focused largely on extra-
GENOMIC EVIDENCE
i SHOWS HOW FISH
PATHOGENS ADAPT
cellular proteases, adhesive mechanisms, and the presence of a polysaccharide capsule. Iron
concentration seems to be important in the regulation of expression of superoxide dismutase
and catalase, which protect bacteria against reactive oxygen species (p.108) and may be impor-
TO SPECIFIC HOSTS
tant in intracellular survival. An iron-uptake system is associated with virulence, but as well as
The genome sequence of a strain uptake by siderophores, the bacterium seems able to acquire iron by direct interaction between
of Aeromonas salmonicida subsp. hemin molecules and outer-membrane proteins. Introduction of rapid PCR-based diagnostic
salmonicida revealed that acquisi- methods and vaccines are proving effective in controlling the disease.
tion of mobile genetic elements,
genome rearrangements, and
gene loss appear responsible for Aeromonas salmonicida has a broad geographic
adaptation of the bacterium to
its specific host, salmonid fish range affecting fish in fresh and marine waters
(Reith et al., 2008). Many genome Aeromonas salmonicida was first described as a pathogen of trout in Europe in 1890 but is
rearrangements have occurred, now known to have a broad host and geographic range. The taxonomy of Aer. salmonicida has
affecting regulation of gene expres-
been the subject of much debate over the years and it is now generally recognized that acute
sion. Analysis of further strains has
revealed evolution of a diversity of
“typical” furunculosis in salmonids is caused by the subspecies salmonicida. In mariculture,
constrained mesophilic and psy- this usually presents as a severe septicemia with acute mortalities. At the peak of the furun-
chrophilic lifestyles (Vincent et al., culosis outbreaks in Scotland and Norway in the 1980s, total industry losses neared 50% of
2016). The Aer. salmonicida genome stock. Externally, the fish show darkening and hemorrhages around the fins and mouth and
contains numerous pseudogenes— internally there is extensive hemorrhaging and destruction of the organs. Other subspecies
defunct relatives of known genes (achromogenes, masoucida, pectinolytica, and smithia) can be distinguished by differences
that no longer encode proteins. in standard biochemical tests, pigment production, and molecular techniques such as gene
The genome sequence of Aliivibrio probes and DNA–DNA hybridization. These subspecies are generally associated with der-
salmonicida also reveals evidence mal ulcerations in other marine species such as turbot (Scophthalmus maximus) and halibut
of extensive gene rearrangements
(Hippoglossus hippoglossus). The name furunculosis derives from a boil-like necrotic lesion
through incorporation of insertion
sequences and the degeneration of
seen in a chronic form of the infection in older or more resistant fish (Figure 11.14A). Most
many genes (Hjerde et al., 2008). strains can be easily isolated on laboratory media, although some are fastidious and have a
Of particular interest is the loss specific requirement for heme.
of genes involved in the utiliza-
tion of chitin, which are highly The virulence mechanisms of Aeromonas salmonicida are extremely complex, and this
important in enabling vibrios to pathogen is an excellent example of multifactorial virulence that has attracted the atten-
colonize surfaces in the marine tion of numerous investigators for over half a century. There are many different interacting
environment (p.332). Gene loss or components of the factors responsible for entry, colonization, growth in vivo, and tissue
decay is t­ ypical of recently evolved damage. Genes for many of these virulence factors have been cloned and their properties
genomes, often reflecting adapta- studied in detail, with full genome sequences also now available. Virulent and avirulent
tion to a specific host (Pallen and
strains are usually distinguished by the presence or absence of a regularly structured sur-
Wren, 2007).
face S-layer (p.69) known here as the “A-protein,” the hydrophobic nature of which confers
auto-agglutinating properties in culture (Figure 11.14B). Isolates possessing the 49-kDa
Figure 11.14 Infection of fish by
Aeromonas salmonicida. A. Large
boil-like lesion (furuncle) on the
surface of infected salmon. B. The
swollen skin lesion is filled with pink
fluid containing blood and necrotic
tissue. C. Growth of A+ (left) and
A ‒ (right) cultures in broth: A+ cul-
tures autoagglutinate owing to the
hydrophobic S-layer (A-protein). D.
Transmission electron micrograph
(TEM, negatively stained) of an A+
strain showing regular structure of
the surface A-protein. E. TEM section
of a virulent A+ strain showing the
additional layer (arrowed) external to
the outer membrane. Note outer-
membrane vesicles (V). Credits: A–B.
Reprinted from Dallaire-Dufresne et
al. (2014) with permission of Elsevier.
C–E. Nigel Parker and Colin Munn,
University of Plymouth.
 Microbial Diseases of Marine Organisms 315

A-protein (A+) can also be distinguished from A‒ strains by their growth on agar media
containing Coomassie Blue or Congo Red dyes, which the A-protein absorbs. Electron
microscopy shows the A-protein to be present as a structured array of tetrahedral subunits
(Figure 11.14C) external to the typical Gram-negative outer membrane (Figure 11.14D). It is
linked to the cell surface via the O polysaccharide sidechain of LPS. The main function of
the A-protein is as a protective layer, which contributes to the bacterium’s resistance to the
bactericidal effects of complement in the serum of the host fish. Because of its hydrophobic
nature, it also plays a role in adhesion to host tissue and survival within macrophages. Aer.
salmonicida also produces a range of toxins, and many studies have been carried out on the
activity of purified components, although it is now clear that these have synergistic inter-
actions. The enzyme glycerophospholipid:cholesterol acyltransferase (GCAT) and a serine
protease are particularly implicated as key virulence factors. GCAT forms a complex with
LPS, which is hemolytic, leukocytolytic, cytotoxic, and lethal. Serine protease expression is
regulated by an AHL-mediated QS mechanism, and this enzyme (in synergy with GCAT)
is responsible for hemolysis. As in V. anguillarum, growth in vivo is dependent on the pro-
duction of a siderophore (2,3 diphenol catechol) and the uptake of sequestered iron via
IROMPs. A type III secretion system (T3SS, see p.339) also has a major role in pathogenesis
by delivering a range of effector molecules that switch off the host’s ability to recognize
infection and prevent onset of disease. Elucidation of the various components in virulence
and their immunological properties was a critical step in the formulation of modern vac-
cines, which have been largely successful in control of the disease in salmon mariculture
since the mid-1990s.

The ecology of Aer. salmonicida has been the subject of much controversy, with some inves-
tigators suggesting that it is an obligate fish pathogen and others suggesting that it survives in
the environment. Historical evidence suggests that it spread throughout Europe more than a
century ago and spread into the wild salmon population, where it caused major losses in the
1930s. It is very likely that some wild populations are latently infected with Aer. salmonicida.
The bacterium may enter the VBNC state (p.95), during which the cells undergo various mor-
phological changes. One certainty is that the development of mariculture in enclosed bodies
of water such as lochs in Scotland and fjords in Norway or Chile has led to a shift in the nor-
mal microbiota. Whereas Aer. salmonicida seems to have been primarily a freshwater organ-
ism, it has now adapted to life in seawater, and recent isolates may have a sodium requirement
lacking in earlier strains. Farmed fish can transmit the disease to wild fish around sea cages
and these can spread the pathogen to other sites. Wrasse introduced to net pens to remove sea
lice parasites from salmon can become infected and thus constitute a reservoir of infection.
Atlantic salmon have been shown to harbor latent Aer. salmonicida infections when they
return from the ocean to spawn, so spread of the disease may be contributing to decline of
wild populations.

Marine flexibacteriosis is caused by a weakly

virulent opportunist pathogen
Members of the CFB group (p.305), many of which are pigmented and show gliding motil-
ity, are responsible for infections in a wide range of fish. Mostly, these bacteria are rather
weak opportunist pathogens that colonize damaged tissue in fish weakened by stress, espe-
cially due to increased water temperature and nutritional deficiency. Most diseases caused
by this group occur in fresh water, but Tenacibacter maritimus (previously classified as
Flexibacter marinus) causes a disease called marine flexibacteriosis, sometimes referred
to as “marine columnaris” or “black patch necrosis,” which is widely distributed in numer-
ous species of wild and farmed fish in Europe, Japan, North America, and Australia. It
is characterized by excess mucus production, damage to the gills, tissue necrosis around
the mouth and fins, skin lesions, and eventual death. It is most severe in juvenile fish at
temperatures above 15°C. The clinical signs and the observation of long Gram-negative
gliding bacteria in microscopic preparations are very distinctive. The bacteria are not easy
to grow in the laboratory, as they require special low-nutrient media. There appear to be
several O-antigen serotypes, which may be related to specific hosts. The disease responds
to antibiotics administered by bath and some vaccines have been developed, with variable
degrees of success.
316 Chapter 11

Piscirickettsia and Francisella are intracellular

proteobacteria infecting salmon and cod
Piscirickettsia salmonis was identified in 1989 as the etiological agent of a septicemic disease
that caused severe economic losses in farmed salmon in Chile. Here, it causes predictable
annual epizootics, but it has also caused sporadic outbreaks of disease in Norway, Iceland,
and Canada. Infected fish stop feeding, become lethargic, and may show small white lesions
or ulcers on the skin. The disease progresses rapidly, and death often occurs with few or no
external symptoms. Internally, the most distinctive sign is off-white to yellow nodules in the
liver. The disease seems to be transmitted both horizontally by blood-sucking ectoparasites
and vertically via the eggs, making the testing and elimination of infected brood stock with
ELISA and PCR assays an important element of disease control. Piscirickettsia was origi-
nally considered to be an obligate intracellular parasite and was formerly propagated only in
suitable cell lines, but it can now be grown in broth or agar cultures using appropriate media.
A range of serological tests has been developed, and genetic heterogeneity among strains
can be studied by analysis of the 23S rRNA operon. Antibiotic treatment is ineffective, since
there are very few agents capable of attacking intracellular bacteria without causing damage
to host cells. A recombinant vaccine (p.402) based on an outer-membrane lipoprotein, OspA,
shows promise for disease control.

Epizootics caused by Francisella have emerged as a particular problem in culture of Atlantic

cod in Norway. This is a granulomatous inflammatory disease, resulting in extensive internal
lesions, with nodules in the spleen, heart, kidney, and liver. The Francisella isolated from
cod has recently been characterized as closely related to Francisella philomiragia, which is
widely distributed in aquatic environments and may cause disease in humans with impaired
immunity. Other Francisella spp. also cause infections in a range of other fish.

Intracellular Gram-positive bacteria

cause chronic infections of fish
Renibacterium salmoninarum causes bacterial kidney disease (BKD) and is widely dis-
tributed in both wild and cultured salmon in many countries including Canada, the USA,
Chile, Japan, Scotland, and Iceland. The expansion of salmon culture through international
movement of eggs has assisted the spread of BKD and it causes significant losses in both
Pacific and Atlantic salmon. BKD pathology is characterized by chronic, systemic tissue
infiltration, causing granular lesions in the internal organs, especially the kidney. External
signs include darkening of the skin, distended abdomen, exophthalmia, and skin ulcers.
Significant changes in blood parameters are consistent with damage to the hematopoietic and
lymphopoietic tissues of the kidney, liver, and spleen. The pathological signs are the result
of the interactions between the host’s cellular immune response and the pathogen’s virulence
mechanisms. Tissue destruction forms a focus of necrosis, owing to release of hydrolytic and
catabolic enzymes and liberation of lytic agents from the bacteria.

Our understanding of the mechanisms of virulence of R. salmoninarum is hampered by the

fact that the bacterium takes several weeks to grow on culture plates and does not form dis-
crete colonies. Reproducible infection is also difficult to achieve in aquarium experiments.
One important virulence factor is a 57-kDa surface protein, whose hydrophobic properties
facilitate attachment to host cells. The key feature of R. salmoninarum is its ability to enter,
survive, and multiply within host phagocyte cells. Binding of the complement component C3
to the bacterial surface enhances internalization of the bacterium because phagocytic cells
possess a receptor for C3. Why does R. salmoninarum encourage uptake by cells that nor-
mally kill invading pathogens? The answer seems to lie in the pathogen’s ability to survive (at
least in part) the intracellular killing mechanisms of the macrophage and to replicate (albeit
slowly) within the cells. As well as being resistant to reactive oxygen species, R. salmonina-
rum lyses the phagosome membrane in order to escape its strongly antibacterial environment
(Figure 11.15). In the past, BKD has been difficult to diagnose, but disease management and
control are helped by serological techniques (ELISA and FAT), together with recently devel-
oped gene probes and techniques for accurate differentiation of clinical isolates (based on
PCR amplification of length polymorphisms in the tRNA intergenic spacer regions). These
 Microbial Diseases of Marine Organisms 317

Figure 11.15 Intracellular growth

of Renibacterium salmoninarum.
Transmission electron micrograph
showing lysis of the phagosome
membrane (arrows), prior to entrance
of the bacterium into the cytoplasm.
Image courtesy of S. K. Gutenberger
and J. R. Duimstra, Oregon State
University.

methods are used for certification of brood stock and eggs as disease-free and for implement-
ing quarantine procedures to contain disease outbreaks. There are no effective antibiotic
treatments and there have, therefore, been many attempts to develop a vaccine. These are
thwarted by the slow growth of the pathogen, and recent work has focused on the use of
recombinant DNA technology to produce fusion proteins and DNA vaccines (see p.402), as
well as the use of a live preparation of a non-pathogenic Arthrobacter sp., which has some
antigenic similarity to R. salmoninarum. Genes for several virulence factors, including the
p57 protein and two hemolysins, have been cloned and expressed in Escherichia coli and the
nature of the immune response investigated.

Species of the genera Mycobacterium and Nocardia also cause chronic, persistent infec-
tions with long periods of intracellular survival. Many different species of mycobacteria
occur; these are widely distributed in seawater and sediments and colonize many species
of fish. Mycobacterium marinum and Mycobacterium fortuitum are the best-known agents
of disease. Disease develops slowly and usually affects mature fish; for this reason, myco-
bacteriosis is a particular problem in marine aquaria, but it has also emerged as a serious
problem in the culture of sea bass and turbot. Most species of fish develop few exter-
nal disease signs other than emaciation, although histopathological investigation reveals
extensive granulomatous lesions with caseous necrotic centers. For this reason, the disease
is often called fish tuberculosis, since there is some resemblance to the basic pathologi-
cal mechanisms seen in the human condition caused by Mycobacterium tuberculosis. The
delayed hypersensitivity reactions and involvement of cell-mediated immunity certainly
show some parallels in fish and humans. Nocardia spp. cause similar chronic granuloma-
tous conditions to mycobacteria, but the organisms can be distinguished in the laboratory.
Again, because of their intracellular nature, there are few antimicrobial agents effective
against these pathogens. Valuable aquarium fish are sometimes treated with isoniazid and
rifampicin, but this is unwise given the danger of encouraging resistance to these drugs,
valuable for the treatment of human tuberculosis. It should also be noted that M. marinum
(and possibly other species) can cause a zoonotic infection in humans known as fish tank
or aquarist’s granuloma.

Some Gram-positive cocci affect the

central nervous system of fish
The Gram-positive bacteria Lactococcus garvieae (formerly Enterococcus seriolicida) and
Streptococcus iniae have been especially problematic as the cause of disease, often termed
“streptococcosis” in both fresh and marine warm-water culture, especially in Japan. Several
related species have also been described. They cause central nervous damage characterized
by abnormal swimming behavior, often with exophthalmia and meningitis. Extracellular
cytolytic toxins and an antiphagocytic capsule have been implicated in pathogenic-
ity. Streptococcus iniae has caused epizootics in barramundi (Lates calcarifer) culture in
318 Chapter 11

Australia and has also been implicated in extensive fish kills in the Caribbean Sea. It can also
cause zoonotic infection by infection of wounds in workers handling infected fish, leading to
severe cellulolytic infection.

Viruses cause numerous diseases of marine fish

Viruses have long been suspected as causative agents of disease in fish, but it was not until
the successful development of fish cell culture methods in the 1960s that progress was made
in diagnosing viral fish diseases. As with bacterial diseases, intensive aquaculture has exac-
erbated viral disease problems in fish. The mechanisms by which viruses produce disease in
their fish hosts are less well defined than those of bacteria. Viruses must possess a mechanism
to enter a susceptible host cell and grow within it; this is usually very specific for a particular
type of cell and is determined initially by interactions between surface molecules on the
virion and receptors on the host cell. Viruses possess many mechanisms to resist innate host
defense mechanisms such as enzymes and antiviral substances in fish mucus. Following local
replication at the point of entry, viruses are spread to adjacent cells or carried to other target
tissues via the blood, lymph, or nervous systems. The outcome of infection depends on the
balance between viral replication and its control by innate defense, interferons, antibodies,
and cell-mediated immunity. Host damage is caused by various mechanisms, including kill-
ing infected cells because of production of toxic factors during viral replication, cytokines,
inflammatory modulators, and necrosis. It is often the intensity of the host response to viral
infection that produces the most damage.

Infectious salmon anemia (ISA) is one of the

most serious diseases in salmon culture
ISA emerged as a new disease in farmed Atlantic salmon in Norway in the 1980s and has
been a continuing problem since that time. Subsequent outbreaks causing major economic
losses occurred in Canada, Scotland, the Faroe Islands, the USA, and Chile and has brought
the industry to the brink of collapse on several occasions. There is significant variation in the
severity of ISA, with mortalities in sea pens depending on the virus strain and susceptibil-
ity of fish stock. It is likely that the Salmon isavirus (ISAV) originated in wild salmonids in
lakes of Norway, and that transmission probably evolved in the freshwater phase of trout. The
development of salmon farming is thought to have changed the balance between ISAV and
wild fish, leading to the rapid evolution of a highly pathogenic type. Wild fish may serve as
reservoirs of infection. The virus infects the erythrocytes, leading to their lysis. ISA-infected
fish are lethargic, with anemia (shown by very pale gills), exophthalmia, and hemorrhages
in the eye chamber, skin, and muscle. ISAV is an enveloped negative-strand RNA virus of
the Orthomyxoviridae family, which includes influenza virus. Influenza virus is known for
its property of antigenic variation by mutation (“drift”) and recombination (“shift”) events in
the segmented genome, leading to the emergence of new strains. This may also be occurring
in ISAV, as isolates from Norway and North America now seem to be genetically distinct.
Inactivated virus vaccines have been used in Chile and Norway and recombinant DNA vac-
cines have been developed.

Viral hemorrhagic septicemia (VHS) virus

infects many species of wild fish
VHS virus (VHSV) is a rhabdovirus most commonly known as a causative agent of disease
in the freshwater stage of trout and salmon culture, but which can also cause disease in
marine net pens. VHSV has now been isolated from nearly 50 species of marine fish and
is responsible for severe epizootics in wild populations, especially of shoaling species such
as herring (Clupea harengus), mackerel (Scomber spp.), and sprat (Sprattus sprattus). As
well as the impact of overfishing, fluctuations in natural populations due to viral infection
may be driving some commercially important species close to extinction. Detailed phylo-
genetic studies based on nucleoprotein gene sequences of marine VHSV show that there
are distinct populations of the virus and that aquaculture isolates probably originated in
wild marine fish.
 Microbial Diseases of Marine Organisms 319

Lymphocystis virus causes chronic skin infection of fish

Infection with lymphocystis virus (a member of the iridovirus group) results in hypertrophy
of skin cells, causing characteristic papilloma-like lesions, usually on the skin, fins, and tail.
The virus seems to have worldwide distribution, causing particular problems in cultured and
wild fish in the Mediterranean Sea and in many species of tropical coral reef fish. It is also
one of the most common infections of long-lived aquarium fish. Its role in the ecology of
natural populations is unclear; although the disease is not usually lethal, it does affect growth
rates and probably makes infected fish more susceptible to predation. A large number of other
iridoviruses have been reported from more than 140 different species of fish worldwide; there
is considerable heterogeneity in genome sequence in this group.

Birnaviruses appear to be widespread

in marine fish and invertebrates
DID IMPORT OF
Several isolates of birnavirus have been identified in various wild fish in waters around
the British Isles, Japan, and the northwest Atlantic. These were originally considered to be
? TUNA FEED LEAD
TO THE WORLD’S
closely related to the Infectious pancreatic necrosis virus (IPNV), which is best known as BIGGEST FISH KILL?
a cause of heavy mortalities in small fry in freshwater aquaculture hatcheries. Phylogenetic
Two major epizootics occurred in
analysis of the marine birnaviruses from wild fish, based on amino acid sequence of the poly-
the 1990s in the waters of South
protein gene, indicates that they form a group distinct from IPNV. Little is known about entry
and Western Australia, resulting in
of these viruses and replication in the host, and their impact on wild populations is unclear. the biggest ever single-cause fish
kills. Mass mortalities of pilchards
(Sardinops sagax) occurred, spread-
Viral nervous necrosis (VNN) is an ing at over 30 km a day and cover-
emerging disease with major impact ing 6000 km of coastline.
There was a reduction in the
VNN is characterized by erratic behavior in infected fish, due to spongiform lesions in the spawning biomass of more than
brain (encephalopathy). The causative agent, Betanodavirus (a member of the Nodaviridae), 75% (Ward et al., 2001), result-
has been reported in cultured marine fish worldwide and has emerged since the 1990s to ing in severe ecological conse-
have a major impact on the culture of several marine species. PCR analysis shows that wild quences for other marine life and
and cultured fish are subclinically infected with betanodaviruses and constitute a reservoir seabird predators, as well as major
of infection. The spread of these infections seems to be a direct consequence of the intensity economic losses to an important
of aquaculture and the movement of stock both within aquaculture areas and between distant commercial fishery. The causative
geographic regions. agent was identified as a previ-
ously unknown herpesvirus. These
epizootics coincided with the
Protists cause disease in fish via infections, development of tuna ranching in
South Australia; tuna were kept in
toxins, and direct physical effects large offshore sea cages and fed
A large number of protists can infect wild, farmed, and aquarium fish. Often these are free- with frozen pilchards imported
from California, Peru, Chile, and
living or benign parasites and the environmental and nutritional conditions that promote
Japan to fatten them for the sushi
disease are largely unknown. For example, Paramoeba perurans causes sporadic, severe and sashimi market. Although
outbreaks in salmon culture in Tasmania. Diplomonad flagellates are commonly found in the conclusive proof is hard to obtain, it
gut of fish, in which they seem to have little effect, but under some circumstances they can seems likely that inadvertent intro-
cause systemic infection with high mortalities. Myxosporeans such as Kudoa sp. commonly duction of the virus with feed from
cause muscle infections, causing cysts and softening of the tissue that impairs marketability a different geographic region may
of the fish. Loma salmonae is an obligate intracellular microsporidean parasite infecting the have resulted in a disease outbreak
gills of many economically important fish, including both wild and cultured salmon and cod. because the Australian pilchard
population had not previously
Excessive growth of phytoplankton leading to HABs in coastal waters can be responsible for mor- encountered this variant of the virus
talities in fish. Some of these blooms are caused by toxin-producing species that can also affect and therefore had no immunity.
Alternative explanations might be
marine mammals (see below) or humans (see Chapter 12). Toxic genera such as Alexandrium,
transmission by ballast water or by
Gymnodinium, Karenia, and Pseudo-nitzschia have all been responsible for mass mortality in seabirds. Since 1998, there have
many parts of the world, both in wild and farmed fish. The increased use of coastal waters for been no more mass mortalities and
fish and shellfish farming has undoubtedly contributed to stimulation of blooms by input of the virus may now be endemic in
excess nutrients from the fish excreta and waste food. Caged fish cannot escape the effects. Australia; because of the develop-
ment of host “herd” immunity, it
Nontoxic algae can also kill fish. For example, the diatom Chaetoceros convolutus produces may now cause only low levels of
long barbs, which clog fish gill tissue causing excess mucus production, leading to death from mortality or exist as a latent infec-
reduction in oxygen exchange. Incidents involving the loss of more than 250,000 farmed tion (Whittington et al., 2008).
320 Chapter 11

salmon at a time have occurred. Blooms of the flagellate Heterosigma carterae also cause
heavy mortalities in Pacific coast salmon farms. Nontoxic microalgae can also have indirect
effects on fish, as large blooms reduce light penetration and decrease the growth of sea-
grass beds, which are often important nursery grounds for the young stages of commercially
important fish. Also, clumping, sinking, and decay of phytoplankton can generate anoxic
conditions.

DISEASES OF MAMMALS
Dinoflagellate and diatom toxins affect marine mammals
Marine mammals are susceptible to a variety of infectious disease conditions, as well as the
effects of toxic HABs, as shown in Figure 11.16. The significance of HAB toxins has been
extensively investigated because of their effects on human health, and the origin of blooms
and the nature of the toxins is discussed in Chapter 13. Most HAB toxins have effects on the
nervous system. Ingestion of HAB toxins by humans results in symptoms such as nausea,
vomiting, diarrhea, temperature reversal effects, and paralysis, so it is reasonable to conclude
that similar effects will also occur in marine mammals. If so, this would clearly affect feed-
ing, buoyancy, heat conservation, breathing, and swimming. Neuropathological symptoms
and mass mortality events in marine mammals are increasingly linked to toxins from HABs.
One explanation for some mass strandings or beaching of social mammals such as whales
and dolphins is that they become disoriented because of accumulation of toxins after feeding
on contaminated fish or shellfish (Figure 11.17). In some studies, high levels of saxitoxins
have been found in tissues of stranded killer whales (Orcinus orca) known to have been
feeding on mussels in areas with a toxic Alexandrium bloom. Although the levels of toxin
recovered from most tissues at postmortem often do not seem high enough to account for
symptoms, it is likely that the toxins become concentrated in the brain. Brevetoxin shows a
very high affinity for nerve tissue from the manatee (Trichechus manatus), which are often
killed off the Florida coast during blooms of Karenia brevis, both by ingestion and inhala-
tion. Saxitoxins also bind strongly to nerve tissue from several species of whale. Since the
blood flow diverts to the brain during diving, this could deliver toxins absorbed from the gut
to the brain, where they could accumulate in high enough concentrations to cause disorien-
tation and other neurological effects, leading to stranding. Large-scale mass mortalities of
baleen whales on the coast of Chile have been attributed to HAB toxins through association
with El Niño events. However, it is important to recognize that stranding appears to have
multiple environmental, physiological, and behavioral causes.

There is strong evidence for the role of the toxin domoic acid (DA), produced by the diatom
Pseudo-nitzschia spp., in outbreaks of neuropathological illness in marine mammals and
birds. The toxin was first associated with amnesic shellfish poisoning in humans (see p.347)

Figure 11.16 Microbes as agents of

diseases in marine mammals. Various
factors influence the susceptibil-
ity of the host and transmission of
diseases.
 Microbial Diseases of Marine Organisms 321

and subsequently linked to periodic mass mortalities of sea lions, pelicans, and cormorants
HOW DID A
along the coast of northern California. Animals are poisoned by feeding on planktivorous
fish and shellfish (such as anchovies, sardines, mussels, and clams) that have ingested large ? MICROBIAL TOXIN
INSPIRE A CLASSIC
amounts of toxic Pseudo-nitzschia, which bloom as a result of climatic conditions and
HOLLYWOOD MOVIE?
upwelling of nutrients. A pulse of domoic acid enters the food chain and the affected animals
assimilate the toxin, which accumulates in the brain. Pregnant sea lion females are regularly Occasional mass mortality and
exposed to DA in the diet, and this is especially harmful to development of the fetus brain, erratic behavior of seabirds has
leading to abortion and neonatal death. Recent studies have also found that krill—the princi- been reported along the coast of
pal component of the diet of squid, baleen whales, and seabirds—can accumulate DA, which northern California for many years.
One such incident is thought to
is then passed to higher trophic levels of the food chain.
have been the inspiration—with
some imaginative additions!—for
the classic 1963 Alfred Hitchcock
Virus diseases cause mass mortalities thriller movie The Birds, in which
in cetaceans and pinnipeds residents of a small town are ter-
rified by the erratic behavior of
In 1988, there were mass mortalities of more than 18000 common or harbor seals (Phoca
birds. While on vacation in 1961,
vitulina) in Northern Europe, with over 60% population loss in some areas. At about the same Hitchcock heard of an incident in
time, an epizootic with similar mortality rates occurred in bottlenose dolphins (Tursiops the town of Capitola, California, in
truncatus) on the east coast of the USA. In 1992, there were mass mortalities of porpoises which seabirds crashed into cars
(Phocoena phocoena) in the Irish Sea and of striped dolphins (Stenella coeruleoalba) in the and houses and died in large num-
Mediterranean Sea, which reduced the population to 30% of its previous level. These disease bers. This inspired the idea for the
outbreaks attracted considerable public concern; under pressure from environmentalist lobby film, adapted from a story set in
groups, several government sponsored research projects were set up to investigate the prob- Cornwall, England, by Daphne du
lem. As a result of this work and subsequent studies, we now know that virus diseases are a Maurier. In 1991, another incident
significant cause of mortality in cetaceans (whales, dolphins, and porpoises) and pinnipeds in Monterey Bay provided the first
evidence to suggest that the birds
(seals and sea lions). In the wild, it is difficult to study diseases of marine mammals except
had eaten fish containing the toxin
when mass mortalities occur, when animals are stranded, or if they become caught in fishing domoic acid, originating from a
nets. Networks of marine conservation volunteers have helped in the collection of data, and diatom bloom. Extensive investiga-
it is now common practice to test stranded cetaceans and pinnipeds for viral diseases using tion involving scientists from many
immunological and molecular-based diagnostic tests, so that a worldwide picture of their disciplines of another incident
importance is beginning to emerge. Obtaining blood or tissue samples from live animals is on this coast in 1998 provided
difficult and postmortem deterioration happens very quickly, so data are often sparse. Some conclusive evidence of domoic acid
knowledge about these diseases has also come from the study of captive animals in zoos and poisoning in birds and sea mam-
marine parks; the study of diseases of marine mammals has become a specialized branch of mals and explained why this area is
veterinary medicine. a hotspot for such events (Scholin
et al., 2000).

Viruses from nine different families have been

linked to diseases of marine mammals
The most important viruses infecting cetaceans and pinnipeds are the morbilliviruses; these
are RNA viruses of the paramyxovirus group causing several mammalian diseases, of which
the best-known examples are measles in humans and distemper in dogs. The first clues to

Figure 11.17 Mass stranding of

false killer whales in W. Australia,
possibly caused by a toxic algal
bloom. Image credit: Bahnfrend,
CC BY 3.0 via Wikimedia Commons.
322 Chapter 11

the identity of the virus infecting seals in the 1988 epizootic came when serum was found

i
DOGS WITH A NOSE to neutralize canine distemper virus (CDV). At first, it was thought that the disease might
FOR WHALE POOP be linked to an outbreak of distemper in sled dogs in Greenland. However, using sequencing
Doucette et al. (2006) investigated of the viral capsid proteins, the seal virus (now called phocine distemper virus, PDV) was
whether the transfer of marine shown to be a new species, although it shares some antigenic cross-reactions with CDV. One
algal toxins through the food chain possible explanation for the sudden epizootic is transfer from another species (probably the
could explain poor health and harp seal, Phoca groenlandica), which migrates from a different geographic region in which
impaired reproduction of highly the disease is enzootic. Serological studies have shown that PDV-like viruses are present in
endangered North Atlantic right
several species of marine mammals. By contrast, the virus responsible for outbreaks in seals
whales (Eubalaena glacialis) in the
in the Caspian Sea and Lake Baikal (P. caspica and P. siberica respectively) seems to be
Bay of Fundy. This required maneu-
vering a small boat into the vicinity identical to CDV, and it almost certainly did come from dogs. Infection causes respiratory,
of feeding whales, spotting a gastrointestinal, and neurological symptoms, often with secondary bacterial infections such
whale in the act of defecation, and as pneumonia.
scooping up the sample (scat) in
a net. Doucette used this method The morbilliviruses isolated from diseased porpoises, dolphins, and several species of whale
to show high levels of saxitoxins appear to be closely related antigenically and genetically; they are now recognized as dif-
in the feces of the whales and in ferent strains of cetacean morbillivirus (CeMV). Phylogenetic studies show that CeMV is
the whales’ primary food source close to the ancestor of the morbillivirus group, so these viruses probably have a terrestrial
(the copepod Calanus finmarchius), origin and may have infected cetaceans for the several millions of years since they have
associated with a bloom of toxic
populated the oceans. Infected animals show pneumonia-like symptoms and disturbance of
dinoflagellates (Alexandrium spp.)
Although right whale scat is bright
diving, swimming, and navigation ability. In the enzootic state, the virus probably has long-
orange and floats for about 30 term effects on population dynamics, causing mortalities mainly in young animals in which
min, sample collection at sea no immunity exists. Morbilliviruses typically lead to either rapid death or recovery with
is a very hit-and-miss affair! A life-long immunity, with no persistent carrier state. They therefore require large populations
novel solution was pioneered by to sustain themselves through the input of new susceptible hosts. As with seal distemper,
researchers at the New England epizootics in cetaceans probably occur as a result of cross-infection from different species
Aquarium, who trained dogs to or animals from other geographic regions once a sufficiently large population of susceptible
detect the scent of fresh scat; the individuals has built up to allow spread (this is similar to the decline in “herd immunity” that
dogs point to the direction of the results in periodic measles epidemics in humans). The nutritional status of the host popula-
source up to 1500 m away so that
tion and environmental factors are also important in determining the onset of epizootics.
the helmsman can steer the boat
After the 1988 outbreak of phocine distemper in northern Europe, the harbor seal population
toward it (Rolland et al., 2007).
slowly recovered. In 2002, a new epizootic of phocine distemper in harbor seals emerged,
resulting in death of over 30000 seals, owing to the buildup of a threshold number of nonim-
mune animals (Figure 11.18).

Poxviruses have been implicated as the cause of tattoo-like skin lesions in several species of
small cetaceans, although they do not usually cause significant mortalities. Papillomaviruses
and herpesviruses cause genital warts (proliferation of the squamous epithelium) in porpoises,

Figure 11.18 Transmission of

morbilliviruses in marine mammals.
Major epizootics occur in the seal
and dolphin host species shown.
Black and gray silhouettes of host
species indicate status as a possible
reservoir or dead-end host, respec-
tively. Phocine distemper virus (PDV)
circulates mainly in seals with occa-
sional spillover to sea otters and wal-
ruses. Canine distemper virus (CDV)
circulates in terrestrial carnivores,
causing occasional outbreaks in seals.
Cetacean morbillivirus (CeMV) circu-
lates mainly among cetaceans, with
occasional spillover to pinnipeds. It
is unknown if whale species act as
disease carriers or spillover hosts.
Redrawn from Jo et al. (2018)with
permission of Elsevier.
 Microbial Diseases of Marine Organisms 323

dolphins, and whales. As with their human equivalent, they are sexually transmitted and
may affect population dynamics by disturbing reproduction and social behavior, since sexual
“play” is an important part of cetacean social group interactions and genital warts can some-
times be large enough to interfere with copulation.

Various types of caliciviruses occur in a wide range of marine mammals, having been first
described as San Miguel sea lion virus. From time to time, influenza A virus has been associ-
ated with mortalities in cetaceans and pinnipeds. As occurs in humans, animals infected with
influenza become very weak and often succumb to secondary bacterial infections. The iso-
lates are closely related to avian influenza in birds and are highly virulent. Seabirds are often
associated with marine mammals during feeding at the surface and this favors transmission.

Several species of bacteria and fungi

infect marine mammals
One of the most important globally distributed diseases of cetaceans and pinnipeds is brucel-
losis, caused by the bacterium Brucella, a highly contagious intracellular pathogen that also
occurs in cattle, sheep, goats, and pigs. It infects the reproductive tract, especially the pla-
centa and amniotic sac, leading to abortion. Frequent abortions have been observed to occur
DOES POLLUTION
?
in closely monitored dolphin pods, and the bacterium could therefore have a significant effect
on fertility and population dynamics. Several strains have been isolated from many different MAKE MARINE
species, and serological studies show that up to 30% of small marine mammals surveyed MAMMALS MORE
have evidence of exposure. Phylogenetic analysis of the marine isolates of Brucella show that PRONE TO DISEASE?
these may constitute two distinct new species named Brucella ceti and Brucella pinnipedialis Environmental pollutants such as
with cetaceans and seals, respectively, as the preferred hosts. mercury, polychlorinated biphenyls
(PCBs) and organochlorine pesti-
Leptospirosis, caused by several species of Leptospira, occurs in many populations of seals cides accumulate in lipid-rich tis-
and sea lions. Populations of Californian sea lions suffer a cyclical pattern of infection every sues, especially blubber and milk,
4–5 years; the mode of transmission is unclear, but probably occurs when the animals haul with a variety of deleterious effects
out on land. As in the human equivalent—Weil’s disease, transmitted by rats—the main on marine mammals, especially
reproductive failure. These com-
symptom is renal failure.
pounds also impair the function of
the immune system, so they may
Tuberculosis is a chronic multiorgan disease caused by Mycobacterium tuberculosis and M. increase susceptibility of marine
bovis, while other mycobacteria such as M. marinum and M. fortuitum can cause lesions mammals to bacterial or viral dis-
in the skin and lungs. In seals, sea lions, and cetaceans, there have been several cases in eases. An experiment in which dif-
zoos and marine parks. Other bacterial pathogens known to cause sudden mortality in cap- ferent groups of seals were fed on
tive cetaceans include Burkholderia pseudomallei, Bordetella bronchiseptica, and species of fish caught in either highly polluted
Erysipelothrix, Streptococcus, Salmonella, and Pasteurella. The significance of these patho- or less polluted areas showed that
gens in wild animals is not known. pollutants impaired the immune
system, resulting in reduction in
the natural killer cells that are criti-
In addition to primary pathogens, marine mammals can succumb to a host of opportunistic
cal in defense against virus (Ross
skin and respiratory infections caused by bacteria and fungi, such as pneumonia caused by
et al., 1996; Ross, 2002). Mortality
Aspergillus spp., which is a particular problem in stressed captive animals. Direct acquisition of porpoises and dolphins due
of microbes of human origin, especially opportunist bacterial infections of wounds, may also to infectious diseases was linked
occur when mammals swim in sewage-contaminated waters. with elevated levels of PCBs and
heavy metals (Jepson et al., 1999;
Bennett et al., 2001). Despite
DISEASES OF SEA TURTLES restrictions on the use of PCBs,
biomagnification of these toxins is
Sea turtles are affected by a virus still occurring and continues to be
promoting growth of tumors linked with infectious diseases and
other effects (Jepson et al., 2015).
Some novel herpesviruses are associated with infection of sea turtles (Figure 11.19). The Because of the practical and ethical
chelonid fibropapilloma-associated herpesvirus has emerged as a major cause of debilitat- problems of conducting controlled
ing tumors of the skin and internal organs in green (Chelonia mydas), loggerhead (Caretta experiments, it is impossible to
caretta), and olive ridley (Lepidochelys olivacea) turtles. The large tumors interfere with prove conclusively that pollution is
feeding, behavior, and reproduction and are frequently observed in stranded animals, so there directly responsible for increased
infections of marine mammals, but
is concern that this disease may cause additional problems for the survival of these endan-
the circumstantial evidence is very
gered species. The disease has been reported in Florida, Hawaii, and Australia and appears strong.
to have increased dramatically since its first description in 1938. Genomic analysis shows
324 Chapter 11

Figure 11.19 Hawaiian green sea

turtle (Chelonia mydas) with a large
tumor caused by the fibropapilloma-
associated turtle herpesvirus. Credit:
George Balazs, NOAA.

that viruses belong to the alphaherpesvirus group, but those isolated from different species
of turtle from different geographic regions are genetically distinct. The increase in disease
seems to be due to unknown environmental factors affecting host susceptibility, rather than
emergence of a more virulent form of the virus. A leech, Ozobranchus sp., acts as a mechani-
cal vector for the virus. Other herpesviruses and papillomaviruses have also been implicated
as the cause of disease in turtles reared in mariculture.

DISEASES OF SEAGRASSES AND SEAWEEDS

Heterokont protists cause ecologically
important mortality of seagrasses
Diseases also affect the productivity of seagrasses such as eelgrass (Zostera marina) and
turtlegrass (Thalassia testudinum) in shallow coastal habitats. Seagrass meadows are major
habitats for marine life such as juvenile fish, turtles, dugongs, and manatees, and also act
as feeding grounds for ducks and geese. Seagrass beds are under intense pressure due to
destruction accompanying coastal development, boat activity, pollution, and the impacts of
climate change, resulting in considerable ecological impact through loss of habitats and eco-
nomic effects through impairment of recruitment to fishery stocks. Since the 1920s, there
have been many reports of “wasting disease” affecting large areas of seagrass meadows in
Europe and North America, resulting in the total loss of Z. marina from some areas. The
causative agent was isolated in the 1980s and originally identified as a fungus or slime mold.
Subsequent phylogenetic studies using rRNA gene sequencing revealed the primary patho-
gen to be a new species, Labyrinthula zosterae, a member of a monophyletic clade within
the heterokonta protist group. Members of this group have an unusual morphology; the cells
secrete an ectoplasmic membrane leading to a slimy net-like structure of actin-rich filaments,
through which the nuclei are transported. The pathogen moves rapidly through the plant
tissue through enzymatic degradation of cell walls, leading to blackening and eventually
complete destruction of the tissue (Figure 11.20). The photosynthetic ability of tissue fur-
ther away from the focus of infection is also severely inhibited, so the ecological impact of
reduced productivity in eelgrass showing even minor disease signs could be significant. Other
pathogens include the fungus-like oomycete Phytopthora gemini, which kills dormant seeds
and developing seedlings. Disease outbreaks are predicted to rise due to climate change and
pollution, but there is currently limited knowledge of the factors affecting these diseases.

Bacteria, fungi, and viruses infect marine macroalgae

Marine macroalgae (seaweeds) have a major function in forming habitat structure of tidal
and subtidal waters and temperate reefs, providing food and shelter to a wide range of inver-
tebrates and fish. Descriptions of seaweed diseases in natural habitats are surprisingly rare.
In part, this is due to this being a neglected area of research, but it also seems that many
seaweeds possess very effective chemical defense mechanisms against microbial coloniza-
tion and infection. Indeed, many of these antimicrobial properties have been investigated
for their biotechnological potential (see Chapter 14). Intensive aquaculture of red, brown,
 Microbial Diseases of Marine Organisms 325

Figure 11.20 A. Eelgrass (Zostera

marina) meadow in Peconic Estuary,
Long Island, NY, showing character-
istic black lesions indicating infec-
tion with Labryrinthula zosterae. B.
Closeup of infected leaf. C. Light
micrograph of Labyrinthula sp. on the
surface of agar. Cells move through
the outer matrix membrane of the
trackway (arrows). D. Transmission
electron micrograph of a single
Labyrinthula sp. cell within infected
tissue of the seagrass Thalassia testu-
dinum. L, lipid vesicle; M, mitochon-
dria; N, nucleus; OM, outer matrix
membrane; V, vacuole. Credits:
A. Kimberly Petersen Manzo and
Chris Pickerell, Cornell Cooperative
Extension Eelgrass Program; B–D.
Tim Sherman, University of South
Alabama.

and green algae as a source of foods, nutritional supplements, pharmaceuticals, and prod-
ucts such as carageenins and agar is a major industry in many countries, especially China,
Japan, Indonesia, Korea, the Philippines, and Chile and some disease outbreaks have been
described. The red alga Porphyra, harvested as the Japanese food crop nori, is susceptible to
a wasting disease resulting in coalescing lesions caused by the oomycete Pythium porphy-
rae. Ascomycete fungi cause infections of Laminaria, Sargassum, and other kelps. Bacterial
infections include “ice-ice disease” of the red seaweed Kappachus alvarezii, which is one of
the most important cultured species for the production of carageenins. As we have seen in
many other situations, Vibrio spp. are the most likely causative agents, colonizing the algal
fronds if they are stressed by low salinity, high temperature, or poor light. Recent research
on the microbiomes of kelps suggests that disease increases might be expected with ocean
warming and acidification, due to shifts in the microbiome.

Reports of disease in natural ecosystems include those affecting the coralline algae associated
with reefs, originally observed in the 1990s in the Cook Islands and now reported throughout
the Indo-Pacific. Bacteria belonging to the genera Planococcus, Bacillus, and Pseudomonas
have been identified in association with coralline lethal orange disease (CLOD), an infection
that affects encrusting coralline algae. CLOD is identified by the appearance of bright orange
coloration spreading to dead areas, although details of the pathogenic mechanisms are scant.
Another disease is associated with a black fungal infection. As with other diseases, increased
sea-surface temperatures seem to encourage the spread of infection. Other diseases lethal to
coralline algae, with currently unknown etiology, have been reported from the Caribbean.
Because coralline algae produce very hard deposits of calcium carbonate, they play an
important role in cementing the structure of reefs and protecting them from wave damage.
Disturbance of the microbiome also affects the settlement and metamorphosis of coral larvae
and other invertebrates, so the emergence of these diseases is a cause for concern.

More than 50 species in most of the algal orders have been shown to contain VLPs, and about
20 algal viruses have been isolated and characterized. Most studies have been concerned with
the role of viruses in infection of planktonic algae, especially prymnesiophytes, diatoms,
and dinoflagellates (see Chapter 7) and there are surprisingly few reports of viruses in sea-
weeds. VLPs have been observed in the reproductive organs of several species of filamentous
326 Chapter 11

brown algae in different geographic areas. Production of infectious virions is variable and
SHIFTING
i PARADIGMS—FROM
“PATHOGEN” TO
host viability may not always be affected. The Ectocarpus siliculosus virus (EsV-1) has been
sequenced, and study of the factors involved in viral replication is yielding valuable informa-
tion about reproductive development in these algae, because virion release is synchronized
“PATHOBIOME”
with the release of spores or gametes, enabling interaction of viruses with their susceptible
Propelled by advances in high- host cells. EsV-1 is designated as the founding member of the phaeoviruses, a group of the
throughput sequencings and Phycodnaviridae. Host-specific phaeoviruses have been identified in several species of brown
bioinformatic analysis, our growing algae including kelps (Laminarinales) and are widely distributed in all temperate coastal
recognition of multicellular organ- waters. Their importance as diseases agents in the ecology and productivity of these algae is
isms and their associated microbes
still unclear.
as a holobiont and knowledge of
the importance of the complex
interactions of the microbiome is
changing our view of the nature
of disease. Casadevall and Pirofski
(2014) argue that “research on
Conclusions
infectious diseases continues to
The ecological and economic impacts of many marine diseases caused by bacteria, protists,
be dominated by reductionist
approaches … what is needed and viruses are increasing. Climate change, pollution, and other anthropogenic factors are
is the simultaneous analysis of undoubtedly influencing the incidence of disease in marine ecosystems. This chapter has
microbial and host variables using provided many examples of evolution of new strains, changes in virulence, and transfer of
new analytical tools.” We need to pathogens between hosts and ecosystems—marine to freshwater, freshwater to marine, and
shift our focus onto the multipar- terrestrial to marine. These shifts are often precipitated by movement of organisms between
tite interactions between hosts and geographic areas, which may occur naturally through migration and inter-mingling of host
microbes. Pitlik and Koren (2017) species. However, we have also seen examples of major epizootics and the development of
hypothesize “that probably all new reservoirs of infection emerging from the spread of intensive aquaculture and the inter-
diseases of holobionts, acute or
national movement of fish, shellfish, feedstuffs, and seafood products. Adverse environmen-
chronic, infectious or non-infec-
tal factors may compromise the immune status or general physiology of the host and affect
tious, and regional or systemic, are
characterized by a perturbation of the virulence of pathogens. In many of the examples described, we have seen that it has
the healthy microbiome into a dis- proved impossible to reliably assign causality to a single pathogen and how difficult it is to
eased pathobiome.” We now have unravel the complexity of host–environment–microbe interactions. In many cases, trying to
strong evidence that many com- distinguish organisms as “primary” or “opportunist” pathogens is no longer useful. Although
plex disease of humans—including interest in diseases of marine organisms has increased greatly in the past few years, the num-
non-infectious conditions—are ber of scientists directly involved in the detailed study of etiology and pathology of microbial
due to this imbalance in microbial infections of marine life remains very small and lags well behind those involved with study
communities and a huge number of disease in humans or in other natural ecosystems. Since climate change and other anthro-
of variables shape its composition, pological impacts are causing increased incidence and emergence of new diseases in some
This concept of dysbiosis can help
groups, we need much better baseline knowledge before we can implement strategies to pre-
to explain why we have found
it so difficult to assign causes of
vent or limit the impact of diseases in the marine environment. This can only be achieved by
disease in corals and other marine increasing the number of researchers studying marine diseases, especially scientists with an
organisms. integrated approach to the ecology of marine diseases, who employ the most effective tech-
nologies available to study organisms within a holobiont perspective. Finally, it is important
to consider that the legacy of Koch, Pasteur, and the other founders of pathogenic microbiol-
ogy in the 1880s may have unwittingly conditioned us to always seek a single microorganism
as the causative agent of a specific disease. The reality is that many infectious diseases are
multifactorial and depend on shifts in the normal microbial community and complex changes
in the host, influenced by multiple environmental factors.

Kasai, H., Nishikawa, S., & Watanabe, K. (2018) Viral diseases affecting
References and further reading aquaculture. In: Seafood Safety and Quality, Bari, Md.L. & Yamakzaki,
General articles K. (eds.), CRC Press, pp. 112–128.

Baskin, Y. (2006) Sea sickness: the upsurge in marine diseases. Bioscience Lafferty, K.D. (2009) The ecology of climate change and infectious dis-
56: 464–469. eases. Ecology 90: 888–900.

Burge, C.A., Friedman, C.S., Getchell, R. et al. (2016) Complementary Lafferty, K.D. (2017) Marine infectious disease ecology. Annu. Rev. Ecol.
approaches to diagnosing marine diseases: a union of the modern and the Evol. Syst. 48: 473–496.
classic. Philos. Trans. R. Soc. B Biol. Sci. 371: 20150207. Lafferty, K.D., Porter, J.W., & Ford, S.E. (2004) Are diseases increasing in
Burge, C.A., Mark Eakin, C., Friedman, C.S. et al. (2014) Climate change the ocean? Ann. Rev. Ecol. Evol. System. 35: 31–54.
influences on marine infectious diseases: implications for management McCallum, H.I., Kuris, A., Harvell, C.D. et al. (2004) Does terrestrial
and society. Ann. Rev. Mar. Sci. 6: 249–277. epidemiology apply to marine systems? Trends Ecol. Evol. 19: 585–591.
Casadevall, A. & Pirofski, L. (2014) Microbiology: ditch the term patho- Munn, C.B. (2006) Viruses as pathogens of marine organisms—from bac-
gen. Nature 516: 165–166. teria to whales. J. Mar. Biol. Assoc. U.K. 86: 1–15.
 Microbial Diseases of Marine Organisms 327

Pitlik, S.D. & Koren, O. (2017) How holobionts get sick—toward a unify- Hallegraeff, G., Coman, F., Davies, C. et al. (2014) Australian dust storm
ing scheme of disease. Microbiome 5: 64. associated with extensive Aspergillus sydowii fungal “bloom” in coastal
waters. Appl. Environ. Microbiol. 80: 3315–3320.
Starr, M., Lair, S., Michaud, S. et al. (2017) Multispecies mass mortal-
ity of marine fauna linked to a toxic dinoflagellate bloom. PLoS One 12: Harvell, C.D., Montecino-Latorre, D., Caldwell, J.M. et al. (2019) Disease
e0176299. epidemic and a marine heat wave are associated with the continental-scale
collapse of a pivotal predator (Pycnopodia helianthoides). Sci. Adv. 5:
Takemura, A.F., Chien, D.M., & Polz, M.F. (2014) Associations and
eaau7042.
dynamics of Vibrionaceae in the environment, from the genus to the pop-
ulation level. Front. Microbiol. 5: 38. Hewson, I., Bistolas, K.S.I., Quijano Cardé et al. (2018) Investigating the
complex association between viral ecology, environment, and Northeast
Diseases of corals, sponges and echinoderms Pacific sea star wasting. Front. Mar. Sci. 5: 77.

Austin, B., Austin, D., Sutherland, R. et al. (2005) Pathogenicity of vibrios Hewson, I., Button, J.B., Gudenkauf, B.M. et al. (2014) Densovirus associ-
to rainbow trout (Oncorhynchus mykiss, Walbaum) and Artemia nauplii. ated with sea-star wasting disease and mass mortality. Proc. Natl. Acad.
Environ. Microbiol. 7: 1488–1495. Sci. USA 111: 17278–17283.

Ben-Haim, Y. & Rosenberg, E. (2002) A novel Vibrio sp. pathogen of the Hoff, M. (2007) What’s behind the spread of white syndrome in Great
coral Pocillopora damicornis. Mar. Biol. 141: 47–55. Barrier Reef corals? PLoS Biol. 5: e164.

Ben-Haim, Y., Zicherman-Keren, M., & Rosenberg, E. (2003) Huggett, M.J. & Apprill, A. (2019) Coral microbiome database: integra-
Temperature-regulated bleaching and lysis of the coral Pocillopora tion of sequences reveals high diversity and relatedness of coral‐associ-
damicornis by the novel pathogen Vibrio coralliilyticus. Appl. Environ. ated microbes. Environ. Microbiol. Rep. 11: 372–385.
Microbiol. 69: 4236–4242. Kimes, N.E., Grim, C.J., Johnson, W.R. et al. (2012) Temperature regula-
Bourne, D.G., Boyett, H. V, Henderson, M.E. et al. (2008) Identification of tion of virulence factors in the pathogen Vibrio coralliilyticus. ISME J. 6:
a ciliate (Oligohymenophorea: Scuticociliatia) associated with brown band 835–846.
disease on corals of the Great Barrier Reef. Appl. Environ. Microbiol. 74: Lawrence, S.A., Floge, S.A., Davy, J.E. et al. (2017) Exploratory analysis
883–888. of Symbiodinium transcriptomes reveals potential latent infection by large
Cárdenas, A., Neave, M.J., Haroon, M.F. et al. (2018) Excess labile carbon dsDNA viruses. Environ. Microbiol. 19: 3909–3919.
promotes the expression of virulence factors in coral reef bacterioplank- Lesser, M.P., Bythell, J.C., Gates, R.D. et al. (2007) Are infectious dis-
ton. ISME J. 12: 59–76. eases really killing corals? Alternative interpretations of the experimental
Corinaldesi, C., Marcellini, F., Nepote, E. et al. (2018) Impact of inor- and ecological data. J. Exp. Mar. Bio. Ecol. 346: 36–44.
ganic UV filters contained in sunscreen products on tropical stony corals Lohr, J., Munn, C.B., & Wilson, W.H. (2007) Characterization of a latent
(Acropora spp.). Sci. Total Environ. 637–638: 1279–1285. virus-like infection of symbiotic zooxanthellae. Appl. Environ. Microbiol.
Danovaro, R., Bongiorni, L., Corinaldesi, C. et al. (2008) Sunscreens 73: 2976–2981.
cause coral bleaching by promoting viral infections. Environ. Health Luter, H.M. (2017) Sponge disease and climate change. In: Climate
Perspect. 116: 441–447. Change, Ocean Acidification and Sponges, Carballo, J.L. & Bell, J.J.
Davy, S.K., Burchett, S.G., Dale, A.L. et al. (2006) Viruses: agents of coral (eds.), Springer, Cham, pp. 411–428.
disease? Dis. Aquat. Organ. 69: 101–110. Luter, H.M., Bannister, R.J., Whalan, S. et al. (2017) Microbiome anal-
Downs, C.A., Kramarsky-Winter, E., Segal, R. et al. (2016) ysis of a disease affecting the deep-sea sponge Geodia barretti. FEMS
Toxicopathological effects of the sunscreen UV filter, oxybenzone (ben- Microbiol. Ecol. 93: fix074.
zophenone-3), on coral planulae and cultured primary cells and its envi- Mera, H. & Bourne, D.G. (2018) Disentangling causation: complex roles
ronmental contamination in Hawaii and the US Virgin Islands. Arch. of coral-associated microorganisms in disease. Environ. Microbiol. 20:
Environ. Contam. Toxicol. 70: 265–288. 431–449.
FFWC (2019) Florida Fish and Wildlife Conservation Commission. Stony Meron, D., Efrony, R., Johnson, W.R., et al. (2009) Role of flagella in
Coral Tissue Loss Disease. Online at: https​://my​fwc.c​om/re​searc​h /hab​ virulence of the coral pathogen Vibrio coralliilyticus. Appl. Environ.
itat/​coral​/dise​ase/ (accessed 4 July 2019). Microbiol. 75: 5704–5707.
FKNMS (2018) Florida Keys National Marine Sanctuary. Case Definition: Meyer, J.L., Castellanos-Gell, J., Aeby, G.S. et al. (2019) Microbial com-
Stony Coral Tissue Loss Disease (SCTLD). Online at: https​://nm​sflor​idake​ munity shifts associated with the ongoing stony coral tissue loss disease
ys.bl​ob.co​r e.wi​ndows​.net/​f lori​d akey​s -pro​d /med​ia/do​cs/20​18100​2 -sto​ outbreak on the Florida Reef Tract. bioRxiv, 626408.
ny-co​ral-t​issue​-loss​-dise​ase-c​ase-d​efini​tion.​pdf (accessed 4 July 2019).
Miller, M.W., Karazsia, J., Groves, C.E. et al. (2016) Detecting sedimen-
Garren, M., Son, K., Raina, J.-B. et al. (2014) A bacterial pathogen uses tation impacts to coral reefs resulting from dredging the Port of Miami,
dimethylsulfoniopropionate as a cue to target heat-stressed corals. ISME Florida USA. PeerJ 4: e2711.
J. 8: 999–1007.
Mills, E., Shechtman, K., Loya, Y., & Rosenbergg, I. (2013) Bacteria
Garren, M., Son, K., Tout, J. et al. (2016) Temperature-induced behavioral appear to play important roles in both causing and preventing the bleach-
switches in a bacterial coral pathogen. ISME J. 10: 1363–1372. ing of the coral Oculina patagonica. Mar. Eco. Prog. Ser. 489: 155–162.
Gavish, A.R., Shapiro, O.H., Kramarsky-Winter, E., & Vardi, A. (2018) Miner, C.M., Burnaford, J.L., Ambrose, R.F. et al. (2018) Large-scale
Microscale tracking of coral disease reveals timeline of infection and het- impacts of sea star wasting disease (SSWD) on intertidal sea stars and
erogeneity of polyp fate. bioRxiv, 302778. implications for recovery. PLoS One 13: e0192870.
Gibbin, E., Gavish, A., Domart-Coulon, I. et al. (2018) Using NanoSIMS Montalvo-Proaño, J., Buerger, P., Weynberg, K.D., & van Oppen, M.J.H.
coupled with microfluidics to visualize the early stages of coral infection (2017) A PCR-Based assay targeting the major capsid protein gene of
by Vibrio coralliilyticus. BMC Microbiol. 18: 39. a dinorna-like ssRNA virus that infects coral photosymbionts. Front.
Gibbin, E., Gavish, A., Krueger, T. et al. (2019) Vibrio coralliilyticus infec- Microbiol. 8: 1665.
tion triggers a behavioural response and perturbs nutritional exchange and Munn, C.B. (2015) The role of vibrios in diseases of corals. Microbiol.
tissue integrity in a symbiotic coral. ISME J. 13: 989–1003. Spectr. 3: doi: 10.1128/microbiolspec.VE-0006-2014.
328 Chapter 11

Mydlarz, L.D., Holthouse, S.F., Peters, E.C., & Harvell, C.D. (2008) Stacy, B.A., Wellehan, J.F.X., Foley, A.M. et al. (2008) Two herpesviruses
Cellular responses in sea fan corals: granular amoebocytes react to patho- associated with disease in wild Atlantic loggerhead sea turtles (Caretta
gen and climate stressors. PLoS One 3: e1811. caretta). Vet. Microbiol. 126: 63–73.
Nissimov, J., Rosenberg, E., & Munn, C.B. (2009) Antimicrobial prop- Sussman, M., Mieog, J.C., Doyle, J., et al. (2009) Vibrio zinc-metallopro-
erties of resident coral mucus bacteria of Oculina patagonica. FEMS tease causes photoinactivation of coral endosymbionts and coral tissue
Microbiol. Lett. 292: 210–215. lesions. PLoS One 4: e4511.
Page, C. & Willis, B. Epidemiology of skeletal eroding band on the Great Sussman, M., Willis, B.L., Victor, S., & Bourne, D.G. (2008) Coral patho-
Barrier Reef and the role of injury in the initiation of this widespread coral gens identified for White Syndrome (WS) epizootics in the Indo-Pacific.
disease. Coral Reefs 27: 257–272. PLoS One 3: e2393.
Pascelli, C., Laffy, P.W., Kupresanin, M. et al. (2018) Morphological char- Sutherland, K.P., Berry, B., Park, A. et al. (2016) Shifting white pox aeti-
acterization of virus-like particles in coral reef sponges. PeerJ 6: e5625. ologies affecting Acropora palmata in the Florida Keys. Phil. Trans. R.
Soc. B 371: 20150205.
Patterson, K.L., Porter, J.W., Ritchie, K.E. et al. (2002) The etiology of
white pox, a lethal disease of the Caribbean elkhorn coral, Acropora pal- Sutherland, K.P., Shaban, S., Joyner, J.L. et al. (2011) Human pathogen
mata. Proc. Natl. Acad. Sci. USA. 99: 8725–8730. shown to cause disease in the threatened eklhorn coral Acropora palmata.
PLoS One 6: e23468.
Pita, L., Rix, L., Slaby, B.M. et al. (2018) The sponge holobiont in a chang-
ing ocean: from microbes to ecosystems. Microbiome 6: 46. Sweet, M. & Bythell, J. (2012) Ciliate and bacterial communities associ-
ated with White Syndrome and Brown Band Disease in reef-building cor-
Pollock, F.J., Wada, N., Torda, G. et al. (2017) White syndrome-affected
als. Environ. Microbiol. 14: 2184–2199.
corals have a distinct microbiome at disease lesion fronts. Appl. Environ.
Microbiol. 83: e02799-16. Sweet, M. & Bythell, J. (2017) The role of viruses in coral health and dis-
ease. J. Invertebr. Pathol. 147: 136–144.
Precht, W.F., Gintert, B.E., Robbart, M.L. et al. (2016) Unprecedented
disease-related coral mortality in Southeastern Florida. Sci. Rep. 6: 31374. Sweet, M., Bulling, M., & Cerrano, C. (2015) A novel sponge disease
caused by a consortium of micro-organisms. Coral Reefs 34: 871–883.
Reshef, L., Koren, O., Loya, Y. et al. (2006) The Coral probiotic hypoth-
esis. Environ. Microbiol. 8: 2068–2073. Sweet, M.J. & Bulling, M.T. (2017) On the importance of the microbiome
and pathobiome in coral health and disease. Front. Mar. Sci. 4: 9.
Reshef, L., Ron, E., & Rosenberg, E. (2008) Genome analysis of the coral
bleaching pathogen Vibrio shiloi. Arch. Microbiol. 190: 185–194. Sweet, M.J., Jones, R., & Bythell, J.C. (2011) Coral diseases in aquaria and
in nature. J. Mar. Biol. Assoc. UK 92: 791–801.
Richardson, L.L. (1998) Coral diseases: what is really known? Trends
Ecol. Evol. 13: 438–443. Thurber, R.V., Payet, J.P., Thurber, A.R., & Correa, A.M.S. (2017) Virus–
host interactions and their roles in coral reef health and disease. Nat. Rev.
Rosenberg, E., Kushmaro, A., Kramarsky-Winter, E. et al. (2009) The role
Microbiol. 15: 205–216.
of microorganisms in coral bleaching. ISME J. 3: 139–146.
Tout, J., Siboni, N., Messer, L.F. et al. (2015) Increased seawater tempera-
Rypien, K. & Baker, D. (2009) Isotopic labeling and antifungal resistance
ture increases the abundance and alters the structure of natural Vibrio
as tracers of gut passage of the sea fan pathogen Aspergillus sydowii. Dis.
populations associated with the coral Pocillopora damicornis. Front.
Aquat. Organ. 86: 1–7.
Microbiol. 6: 432.
Rypien, K.L. (2008) African dust is an unlikely source of Aspergillus
Ushijima, B., Videau, P., Burger, A.H. et al. (2014) Vibrio coralliilyticus
sydowii, the causative agent of sea fan disease. Mar. Ecol. Ser. 367: 125–131.
strain OCN008 is an etiological agent of acute Montipora white syndrome.
Rypien, K.L., Andras, J.P., & Harvell, C.D. (2008) Globally panmictic Appl. Environ. Microbiol. 80: 2102–2109.
population structure in the opportunistic fungal pathogen Aspergillus
van de Water, J.A.J.M., Allemand, D., & Ferrier-Pagès, C. (2018) Host-
sydowii. Mol. Ecol. 17: 4068–4078.
microbe interactions in octocoral holobionts - recent advances and per-
Santos, E. de O., Alves, N., Dias, G.M. et al. (2011) Genomic and pro- spectives. Microbiome 6: 64.
teomic analyses of the coral pathogen Vibrio coralliilyticus reveal a
Walton, C., Hayes, N.K., & Gilliam, D.S. (2018) Impacts of a regional,
diverse virulence repertoire. ISME J. 5: 1471–1483.
multi-year, multi-species coral disease outbreak in Southeast Florida.
Sato, Y., Civiello, M., Bell, S.C. et al. (2016) Integrated approach to Front. Mar. Sci. 5: 323.
understanding the onset and pathogenesis of black band disease in corals.
Ward, J.R. & Lafferty, K.D. (2004) The elusive baseline of marine disease:
Environ. Microbiol. 18: 752–765.
are diseases in ocean ecosystems increasing? PLoS Biol. 2: e120.
Sato, Y., Ling, E.Y.S., Turaev, D. et al. (2017) Unraveling the microbial
Webster, N.S. Sponge disease: a global threat? Environ. Microbiol. 9:
processes of black band disease in corals through integrated genomics.
1363–1375.
Sci. Rep. 7: 40455.
Wegley Kelly, L., Haas, A.F., & Nelson, C.E. (2018) Ecosystem microbiology
Shapiro, O.H., Kramarsky-Winter, E., Gavish, A.R. et al. (2016) A coral-
of coral reefs: linking genomic, metabolomic, and biogeochemical dynamics
on-a-chip microfluidic platform enabling live-imaging microscopy of reef-
from animal symbioses to reefscape processes. mSystems 3: e00162-17.
building corals. Nat. Commun. 7: 10860.
Weynberg, K.D. (2018) Ecosystems: from open waters to coral reefs.
Sharp, W. & Maxwell, K. (2018) Investigating the ongoing coral disease
In: Malmstrom, C. (ed), Environmental Virology and Virus Ecology.
outbreak in the florida keys: collecting corals to diagnose the etiological
Academic Press, pp. 2–38.
agent(s) and establishing sentinel sites to monitor transmission rates and
the spatial progression of the disease. Report, Florida Fish & Wildlife Wilson, B., Muirhead, A., Bazanella, M. et al. (2013) An improved detec-
Research Institute. Online at https​://fl​orida​dep.g​ov/si​tes/d​efaul​t /fil​es/FW​ tion and quantification method for the coral pathogen Vibrio coralliilyti-
C-Sen​tinel​-Site​-Repo​r t-Fi​nal.p​df (accessed 4 July 2019). cus. PLoS One 8: e81800.
Soffer, N., Brandt, M.E., Correa, A.M. et al. (2014) Potential role of Wilson, W.H. (2011) Coral viruses. In: Hurst, C.J. (ed), Studies in Viral
viruses in white plague coral disease. ISME J. 8: 271–283. Ecology: Animal Host Systems. Wiley-Blackwell, pp. 143–151.
Soler-Hurtado, M.M., Sandoval-Sierra, J.V., Machordom, A., & Diéguez- Wood, E. (2018) Impacts of sunscreens on corals. Report by the
Uribeondo, J. (2016) Aspergillus sydowii and other potential fungal pathogens International Coral Reef Initiative (ICRI). https​://cr​m.gov​.mp/w​p-con​tent/​
in gorgonian octocorals of the Ecuadorian Pacific. PLoS One 11: e0165992. uploa​ds/cr​m /ICR​I_Sun​scree​n_0.p​df (accessed 4 March 2019).
 Microbial Diseases of Marine Organisms 329

Work, T. & Meteyer, C. (2014) To understand coral disease, look at coral Kawato, S., Shitara, A., Wang, Y. et al. (2018) Crustacean genome explora-
cells. Ecohealth. 11: 610–618. tion reveals the evolutionary origin of white spot syndrome virus. J. Virol.
Work, T.M., Richardson, L.L., Reynolds, T.L., & Willis, B.L. (2008) 93: e01144-18.
Biomedical and veterinary science can increase our understanding of Labreuche, Y., Chenivesse, S., Jeudy, A. et al. (2017) Nigritoxin is a bacte-
coral disease. J. Exp. Mar. Bio. Ecol. 362: 63–70. rial toxin for crustaceans and insects. Nat. Commun. 8: 1248.

Diseases of molluscs Munro, J., Oakey, J., Bromage, E., & Owens, L. (2003) Experimental bac-
teriophage-mediated virulence in strains of Vibrio harveyi. Dis. Aquat.
Arzul, I., Corbeil, S., Morga, B., & Renault, T. (2017) Viruses infecting Organ. 54: 187–194.
marine molluscs. J. Invertebr. Pathol. 147: 118–135.
Oakey, H.J., Cullen, B.R., & Owens, L. (2002) The complete nucleotide
Bruto, M., Labreuche, Y., James, A. et al. (2018) Ancestral gene acquisi- sequence of the Vibrio harveyi bacteriophage VHML. J. Appl. Microbiol.
tion as the key to virulence potential in environmental Vibrio populations. 93: 1089–1098.
ISME J. 12: 2954–2966.
Wang, W. (2011) Bacterial diseases of crabs: a review. J. Invertebr. Pathol.
de Lorgeril, J., Escoubas, J.-M., Loubiere, V. et al. (2018) Inefficient 106: 18–26.
immune response is associated with microbial permissiveness in juvenile
oysters affected by mass mortalities on field. Fish Shellfish Immunol. 77: Diseases of fish
156–163.
Austin, B. & Newaj-Fyzul, A. eds. (2017) Diagnosis and Control of
Genard, B., Miner, P., Nicolas, J.-L. et al. (2013) Integrative study of physi- Diseases of Fish and Shellfish. John Wiley & Sons.
ological changes associated with bacterial infection in Pacific oyster lar-
vae. PLoS One 8: e64534. Austin, B., Pride, A.C., & Rhodie, G.A. (2003) Association of a bacterio-
phage with virulence in Vibrio harveyi. J. Fish Dis. 26: 55–58.
Maloy, A.P., Ford, S.E., Karney, R.C., & Boettcher, K.J. (2007)
Roseovarius crassostreae, the etiological agent of Juvenile Oyster Disease Austin, B. & Zhang, X.-H. (2006) Vibrio harveyi: a significant pathogen of
(now to be known as Roseovarius Oyster Disease) in Crassostrea virgi- marine vertebrates and invertebrates. Lett. Appl. Microbiol. 43: 119–124.
nica. Aquaculture 269: 71–83. Beaz-Hidalgo, R. & Figueras, M.J. (2013) Aeromonas spp. whole genomes
Meyer, G.R., Lowe, G.J., Gilmore, S.R., & Bower, S.M. (2017) Disease and virulence factors implicated in fish disease. J. Fish Dis. 36: 371–388.
and mortality among Yesso scallops Patinopecten yessoensis putatively Cárdenas, C., Ojeda, N., Labra, Á., & Marshall, S.H. (2019) Molecular
caused by infection with Francisella halioticida. Dis. Aquat. Organ. features associated with the adaptive evolution of Infectious Salmon
125:79–84. Anemia Virus (ISAV) in Chile. Infect. Genet. Evol. 68: 203–211.
Mills, A.M., Ward, M.E., Heyl, T.P., & Van Dover, C.L. (2005) Parasitism as Dallaire-Dufresne, S., Tanaka, K.H., Trudel, M.V. et al. (2014) Virulence,
a potential contributor to massive clam mortality at the Blake Ridge Diapir genomic features, and plasticity of Aeromonas salmonicida subsp. sal-
methane-hydrate seep. J. Mar. Biol. Assoc. United Kingdom 85: 1489–1497. monicida, the causative agent of fish furunculosis. Vet. Microbiol. 169:
Romalde, J.L., Dieguez, A.L., Lasa, A., & Balboa, S. (2014) New Vibrio spe- 1–7.
cies associated to molluscan microbiota: a review. Front. Microbiol. 4: 413. Escobar, L.E., Escobar-Dodero, J., & Phelps, N.B.D. (2018) Infectious dis-
Ward, M.E., Van Dover, C.L., & Shields, J.D. (2004) Parasitism in species ease in fish: global risk of viral hemorrhagic septicemia virus. Rev. Fish
of Bathymodiolus (Bivalvia: Mytilidae) mussels from deep-sea seep and Biol. Fish. 28: 637–655.
hydrothermal vents. Dis. Aquat. Organ. 62: 1–16. Guanhua, Y., Wang, C., Wang, X. et al. (2018) Complete genome sequence
Zielinski, F.U., Pernthaler, A., Duperron, S. et al. (2009) Widespread of the marine fish pathogen Vibrio anguillarum and genome-wide trans-
occurrence of an intranuclear bacterial parasite in vent and seep bathymo- poson mutagenesis analysis of genes essential for in vivo infection.
diolin mussels. Environ. Microbiol. 11: 1150–1167. Microbiol. Res. 216: 97–107.
Hjerde, E., Lorentzen, M.S., Holden, M.T.G. et al. (2008) The genome
Diseases of crustaceans sequence of the fish pathogen Aliivibrio salmonicida strain LFI1238
Bateman, K.S. & Stentiford, G.D. (2017) A taxonomic review of viruses shows extensive evidence of gene decay. 9:616.
infecting crustaceans with an emphasis on wild hosts. J. Invertebr. Pathol. Ina-Salwany, M.Y., Al-Saari, N., Mohamad, A. et al. (2019) Vibriosis in
147: 86–110. fish: a review on disease development and prevention. J. Aquat. Anim.
Cano-Gõmez, A., Goulden, E.F., Owens, L., & Høj, L. (2010) Vibrio Health. 31: 3–22 .
owensii sp. nov., isolated from cultured crustaceans in Australia. FEMS Kalatzis, P., Castillo, D., Katharios, P. et al. (2018) Bacteriophage inter-
Microbiol. Lett. 302: 175–181. actions with marine pathogenic vibrios: implications for phage therapy.
Drake, L.A. & Dobbs, F.C. (2005) Do viruses affect fecundity and sur- Antibiotics 7: 15.
vival of the copepod Acartia tonsa Dana? J. Plankton Res. 27: 167–174. Kashulin, A., Seredkina, N., & Sørum, H. (2017) Cold-water vibriosis.
Dunlap, D.S., Ng, T.F.F., Rosario, K. et al. (2013) Molecular and micro- The current status of knowledge. J. Fish Dis. 40: 119–126.
scopic evidence of viruses in marine copepods. Proc. Natl. Acad. Sci. U. Krkošek, M. (2017) Population biology of infectious diseases shared by
S. A. 110: 1375–1380. wild and farmed fish. Can. J. Fish. Aquat. Sci. 74: 620–628.
Elliott, D. & Tang, K. (2011) Influence of carcass abundance on estimates Pallen, M.J. & Wren, B.W. (2007) Bacterial pathogenomics. Nature 449:
of mortality and assessment of population dynamics in Acartia tonsa. 835–842.
Mar. Ecol. Prog. Ser. 427: 1–12.
Reith, M.E., Singh, R.K., Curtis, B. et al. (2008) The genome of Aeromonas
Flegel, T.W. (2009) Current status of viral diseases in Asian shrimp aqua- salmonicida subsp. salmonicida A449: insights into the evolution of a fish
culture. Isr. J. Aquac. 61: 229–239. pathogen. BMC Genomics 9: 427.
Flegel, T.W. (2019) A future vision for disease control in shrimp aquacul- Rønneseth, A., Castillo, D., D’Alvise, P. et al. (2017) Comparative assess-
ture. J. World Aquac. Soc. doi.org/10.1111/jwas.12589 ment of Vibrio virulence in marine fish larvae. J. Fish Dis. 40: 1373–1385.
Flegel, T.W., Pasharawipas, T., Owens, L., & Oakey, H.J. (2005) Evidence Vanden Bergh, P. & Frey, J. (2013) Aeromonas salmonicida subsp.
for phage-induced virulence in the shrimp pathogen Vibrio harveyi. 329– salmonicida in the light of its type-three secretion system. Microb.
338. Dis. Asian Aqiuac. V: 329–327. Biotechnol. 7: 381–400.
330 Chapter 11

Vincent, A.T., Trudel, M.V., Freschi, L. et al. (2016) Increasing genomic Rolland, R.M., Hamilton, P.K., Kraus, S.D. et al. (2007) Faecal sampling
diversity and evidence of constrained lifestyle evolution due to insertion using detection dogs to study reproduction and health in North Atlantic
sequences in Aeromonas salmonicida. BMC Genomics 17: 44. right whales (Euhalaena glacialis). J. Cetac. Res. Man. 8: 121–125.
Wallace, I.S., McKay, P., & Murray, A.G. (2017) A historical review of Ross, P.S. (2002) The role of immunotoxic environmental contaminants
the key bacterial and viral pathogens of Scottish wild fish. J. Fish Dis. 40: in facilitating the emergence of infectious diseases in marine mammals.
1741–1756. Hum. Ecol. Risk Assess. 8: 277–292.
Ward, T., Hoedt, F., McLeay, L. et al. (2001) Effects of the 1995 and 1998 Ross, P.S., De Swart, R.L., Timmerman, H.H. et al. (1996) Suppression of
mass mortality events on the spawning biomass of sardine, Sardinops natural killer cell activity in harbour seals (Phoca vitulina) fed Baltic Sea
sagax, in South Australian waters. ICES J. Mar. Sci. 58: 865–875. herring. Aquat. Toxicol. 34: 71–84.
Whittington, R.J., Crockford, M., Jordan, D., & Jones, B. (2008) Scholin, C.A., Gulland, F., Doucette, G.J. et al. (2000) Mortality of sea
Herpesvirus that caused epizootic mortality in 1995 and 1998 in pil- lions along the central California coast linked to a toxic diatom bloom.
chard, Sardinops sagax neopilchardus (Steindachner), in Australia is now Nature 403: 80–84.
endemic. J. Fish Dis. 31: 97–105. Smith, J., Connell, P., Evans, R.H. et al. (2018) A decade and a half of
Pseudo-nitzschia spp. and domoic acid along the coast of southern
Disease of marine mammals California. Harmful Algae 79: 87–104.
Bennett, P.M., Jepson, P.D., Law, R.J. et al. (2001) Exposure to heavy met- Whatmore, A.M., Dawson, C., Muchowski, J. et al. (2017) Characterisation
als and infectious disease mortality in harbour porpoises from England of North American Brucella isolates from marine mammals. PLoS One
and Wales. Environ. Pollut. 112: 33–40. 12: e0184758.
Doucette, G.J., Cembella, A.D., Martin, J.L. et al. (2006) Paralytic shell-
fish poisoning (PSP) toxins in North Atlantic right whales Eubalaena gla- Diseases of turtles
cialis and their zooplankton prey in the Bay of Fundy, Canada. Mar. Ecol. Chaloupka, M., Balazs, G.H., & Work, T.M. (2009) Rise and fall over 26
Prog. Ser. 306: 303–313. years of a marine epizootic in Hawaiian green sea turtles. J. Wildl. Dis.
Doucette, G.J., Medlin, L.K., McCarron, P., & Hess, P. (2018) Detection 45: 1138–1142.
and Surveillance of Harmful Algal Bloom Species and Toxins. In: Harmful Greenblatt, R.J., Work, T.M., Balazs, G.H. et al. (2004) The Ozobranchus
Algal Blooms. John Wiley & Sons, Ltd, Chichester, UK, pp. 39–114. leech is a candidate mechanical vector for the fibropapilloma-associated
Gulland, F.M.D., Dierauf, L.A., & Whitman, K.L. (2018) CRC Handbook turtle herpesvirus found latently infecting skin tumors on Hawaiian green
of Marine Mammal Medicine, 3rd edition. CRC Press. turtles (Chelonia mydas). Virology 321: 101–110.
Gulland, F.M.D. & Hall, A.J. (2007) Is marine mammal health deteriorat- Greenblatt, R.J., Work, T.M., Dutton, P. et al. (2005) Geographic variation
ing? Trends in the global reporting of marine mammal disease. 4: 135–150. in marine turtle fibropapillomatosis. J. Zoo Wildl. Med. 36: 527–530.
Häussermann, V., Gutstein, C.S., Beddington, M., Cassis, D., Olavarria,
C., Dale, A.C. et al. (2017) Largest baleen whale mass mortality during Diseases of seagrasses and algae
strong El Niño event is likely related to harmful toxic algal bloom. PeerJ Hughes, R.G., Potouroglou, M., Ziauddin, Z., & Nicholls, J.C. (2018)
5: e3123. Seagrass wasting disease: nitrate enrichment and exposure to a herbicide
Jepson, P.D., Bennett, P.M., Allchin, C.R. et al. (1999) Investigating poten- (Diuron) increases susceptibility of Zostera marina to infection. Mar.
tial associations between chronic exposure to polychlorinated biphenyls Poll. Bull. 134: 94–98.
and infectious disease mortality in harbour porpoises from England and Lachnit, T., Thomas, T., & Steinberg, P. (2016) Expanding our understand-
Wales. Sci. Total Environ. 244: 339–348. ing of the seaweed holobiont: RNA viruses of the red alga Delisea pul-
Jepson, P.D., Deaville, R., Barber, J.L. et al. (2015) PCB pollution contin- chra. Front. Microbiol. 6: 1489.
ues to impact populations of orcas and other dolphins in European waters. McKeown, D., Schroeder, J., Stevens, K. et al. (2018) Phaeoviral infections
Sci. Rep. 6: 18573. are present in Macrocystis, Ecklonia and Undaria (Laminariales) and are
Jo, W.K., Osterhaus, A.D., & Ludlow, M. (2018) Transmission of morbil- influenced by wave exposure in Ectocarpales. Viruses 10: 410.
liviruses within and among marine mammal species. Curr. Opin. Virol. Qiu, Z., Coleman, M.A., Provost, E. et al. (2019) Future climate change is
28: 133–141. predicted to affect the microbiome and condition of habitat-forming kelp.
Lefebvre, K.A., Bargu, S., Kieckhefer, T., & Silver, M.W. (2002) From Proc. Roy. Soc. B 286: 20181887.
sanddabs to blue whales: the pervasiveness of domoic acid. Toxicon 40: Sullivan, B.K., Trevathan-Tackett, S.M., Neuhauser, S., & Govers, L.L.
971–977. (2018) Host-pathogen dynamics of seagrass diseases under future global
Nymo, I.H., Tryland, M., & Godfroid, J. (2011) A review of Brucella infec- change. Mar. Poll. Bull. 134: 75–88.
tion in marine mammals, with special emphasis on Brucella pinnipedialis Zozaya-Vald, E., Roth-Schulze, A.J., Egan, S., & Thomas, T. (2017)
in the hooded seal (Cystophora cristata). Vet. Res. 42: 93. Microbial community function in the bleaching disease of the marine
Pyenson, N.D., Gutstein, C.S., Parham, J.F. et al. (2014) Repeated macroalgae Delisea pulchra. Environ. Microbiol. 19: 3012–3024.
mass strandings of Miocene marine mammals from Atacama Region
of Chile point to sudden death at sea. Proc. R. Soc. B Biol. Sci. 281:
20133316–20133316.
Chapter 12
Marine Microbes
as Agents of
Human Disease

The overwhelming majority of marine microbes that are naturally present in the sea (indige-
nous or autochthonous organisms) are nonpathogenic for humans. However, there are several
important diseases caused by infection with bacteria naturally associated with marine ani-
mals as part of their normal microbiota. Humans can become infected when marine bacteria
colonize the intestinal tract through ingestion of contaminated seawater or seafood; or they
may enter the body via damaged skin or wounds. Such diseases are not normally transmit-
ted from person-to-person, although cholera provides a very important exception. Humans
may also become ill following exposure to toxins produced by bacteria, dinoflagellates, and
diatoms, usually via ingestion of fish or shellfish that have accumulated the toxins in their
tissues. A small number of infectious diseases of fish and marine mammals (zoonoses) can be
transmitted to humans. Infectious diseases are also caused by viruses and bacteria introduced
into the sea via sewage pollution; these are discussed in Chapter 13.

Key Concepts
• Several species of Vibrio whose normal habitat is seawater, plankton, and sediments
in coastal and marine environments are major human pathogens that are becoming
increasingly important as a result of climate change.
• Interactions with phages, extensive acquisition and exchange of genetic information,
and complex regulatory mechanisms are responsible for the population structure and
evolution of Vibrio spp. in the aquatic environment and their virulence in human
infections.
• Several syndromes of “seafood poisoning” occur following consumption of fish or
shellfish containing certain marine bacteria, their toxins, or their metabolic products.
• Zoonotic infections can be acquired by contact with fish or marine mammals.
• Toxins produced by cyanobacteria and microalgae (dinoflagellates and diatoms) may
cause illness after ingestion of fish or shellfish that have accumulated the toxins by
feeding.
• The incidence, distribution, and effects of toxic harmful algal blooms are increasing
due to climate change and other anthropogenic factors; detection and management of
outbreaks is a major activity of public health and fisheries authorities.
332 Chapter 12

MICROBIAL INFECTIONS
Pathogenic vibrios are common in marine
and estuarine environments
Most members of the genus Vibrio can tolerate a range of salinities and they are therefore
widely distributed in marine and estuarine environments throughout the world, especially in
warmer waters above 17°C. They are associated with plankton and the surfaces of marine
animals and occur in high densities in filter-feeding shellfish. In particular, vibrios have
numerous adaptations for colonization and utilization of chitin, which means that they are
particularly associated with the exoskeleton of crustaceans, such as copepods. As discussed
in Chapter 11, several species are important pathogens of marine fish and invertebrates, but
there are also about 12 species that have been associated with human diseases. Of these, the
most important are Vibrio cholerae, V. vulnificus, and V. parahaemolyticus, discussed in
detail in this chapter. Other species are occasionally associated with diarrhea or wound infec-
tions, usually following consumption or handling of seafood.

Vibrio cholerae is an autochthonous aquatic bacterium

The disease cholera (caused by V. cholerae) was originally confined to parts of India until
the early nineteenth century, when six major pandemics of the classical form of the disease
spread to Europe and North America. Throughout the twentieth century, cholera continued
CHOLERA REMAINS to spread to become a major cause of mortality throughout the world, aided by increased
i A MAJOR WORLD
HEALTH PROBLEM
intercontinental transport, wars, and natural disasters. The 1960s saw the emergence of the
El Tor biotype from the Celebes Islands of Indonesia, which has spread to become the sev-
enth global pandemic. The El Tor biotype has reduced virulence compared with the previous
Cholera has been endemic in the
Indian subcontinent for millen-
“classical” V. cholerae strains but is better adapted for transmission. In recent years, another
nia. Seven global pandemics have more virulent El Tor variant has emerged in several parts of Asia and Africa. There are >200
resulted in untold human misery. serotypes of the bacterium, distinguished by the lipopolysaccharide O-antigens on their cell
Today, major epidemics occur surface, although they are biochemically similar. Until 1993, all epidemic cases of cholera
regularly, especially in Africa and were caused by V. cholerae possessing antigen O1 (including the El Tor biotype), but in 1993
Asia during floods, earthquakes, a new serotype, O139, emerged as the cause of epidemics that began in the Bay of Bengal. To
wars, and other disasters. The date, this serotype has been largely confined to Asia and its incidence has recently declined.
disease requires notification to
the World Health Organization Cholera is characterized by profuse diarrhea caused by fluid and electrolyte loss from the
(WHO) under international health
small intestine and is therefore easily spread from person-to-person via contamination of
regulations, but most cases are
water and food. This concept developed from John Snow’s famous deduction in 1854 that a
not reported because countries
fear damage to trade and tourism. particular point source (the Broad Street water pump) was responsible for clusters of disease
It is estimated that there are ~2.9 outbreaks in London. However, the history of cholera through the ages reveals a striking
million cases, with up to 95,000 association with the sea, and epidemiologists have long suspected that there might be an
deaths annually, mostly in young aquatic environmental reservoir of V. cholerae. Research by Rita Colwell and colleagues,
children (WHO, 2017). Attempts to beginning in the 1970s, changed our thinking about the ecology of V. cholerae and we now
develop long-lasting effective vac- realize that the bacterium possesses numerous signaling and regulatory adaptations that
cines for use in endemic areas have equip it for life both as an autochthonous inhabitant of marine and estuarine waters and for
had only partial success. Formalin- colonization of the human host.
killed whole cells combined with
purified B-subunits has proved the
The survival of the pathogen in water is greatly affected by environmental conditions, particu-
most successful and can provide
up to 85% protection, although
larly salinity, temperature, and nutrient concentration. V. cholerae can survive for long periods
this is short-lived. A live attenu- in seawater, but the numbers that can be isolated from water are often very low and insuffi-
ated vaccine—in which the active cient to initiate infection. Comparison of direct epifluorescence and viable counts shows that
A-subunit of the toxin has been a large proportion of the cells enter the “viable but nonculturable” (VBNC) state (see p.95).
deleted by genetic modification— During transition to the VBNC state, the cells initiate an active program of change, resulting
has also been used. Vaccination is in reduced size, alteration of mRNA, and changes to the cell surface. This adaptation allows
regarded as only a complementary bacteria to survive adverse changes in nutrient concentration, salinity, pH, and temperature.
prevention and control measure, V. cholerae associates with a range of phytoplankton and zooplankton, especially copepods,
with improvement of sanitation and with egg masses of chironomid flies. Like other vibrios, V. cholerae possesses chitin-
and clean water of paramount
binding proteins and chitinase, which play key roles in colonization of surfaces and biofilm
importance (WHO, 2017). This
formation in the environment, under the control of a two-component sensor system (see p.103).
remains a major challenge in many
overpopulated areas of the world. Attachment to chitin has been shown to lead to upregulation of numerous genes in chitin
degradation and protein synthesis, whilst genes for motility and chemotaxis are suppressed.
 Marine Microbes as Agents of Human Disease 333

Complex regulatory networks control human

colonization and virulence of V. cholerae i OLD CLOTHES CAN
PREVENT CHOLERA
Major changes in gene expression occur in the transition from the aquatic environment to Knowledge of the close association
the human host, and vice versa. These involve hundreds of genes, in sets (regulons) under of V. cholerae with zooplankton
the control of master regulators that in turn respond to environmental factors via signaling as a key factor in the ecology of
mechanisms and quorum sensing (p.49). Following ingestion, the bacteria encounter dra- cholera has led to a simple but
matic environmental changes, particularly the shock of temperature increase in the body effective method of control. In
and low pH as it encounters the acid of the stomach. There is a complex process of gene Bangladesh, efforts to provide
clean water from wells have led to
regulation as the bacterium responds to chemotactic signals, swims toward the surface of the
disastrous levels of poisoning by
lumen of the small intestine and attaches to the epithelium via pili on the bacterial surface.
arsenic leached from the rocks,
The bacteria then produce a toxin that affects membrane permeability, stimulating loss of and many people still rely on the
electrolytes and water from the small intestine (although it does not lead to permanent dam- collection of untreated surface
age to the cells). Each molecule of the toxin is composed of two different protein subunits. water from ponds and rivers. This
There are five B-subunits that bind to membrane receptors (carbohydrate-containing lipids is an obvious source of cholera and
known as gangliosides) on the surface of the gut cells, and one A-subunit that is inserted into other diseases, especially as fuel
the host cell membrane. The A-subunit is an ADP-ribosylating enzyme that modifies a host to make water safe by boiling is in
regulatory protein. This results in continued activation of the enzyme adenyl cyclase in the very short supply. Therefore, field
gut cells, leading to increased levels of cyclic AMP responsible for excessive fluid and elec- workers have trained local people
to use a simple cloth filter when
trolyte loss by activating a transmembrane regulator chloride channel. Fluids in the gut cells
collecting water. An old sari folded
are replaced from interstitial tissue, plasma, and deeper tissues, and cholera victims usually
eight times provides an effec-
die from extreme dehydration as fluids flow uncontrolled from the body. The dehydration can tive trap for the zooplankton and
be limited by oral rehydration therapy or intravenous drips; in these circumstances, a patient small particles, to which most of
can pass up to 20 liters of fluid a day. However, despite the effectiveness of this simple treat- the V. cholerae are attached. Trials
ment, it is often very difficult to deliver during epidemics in field hospitals in remote areas. conducted by Colwell et al. (2003)
over a number of years have shown
The cholera pathogen possesses an amazingly complex set of interconnected signaling and that this simple method can pre-
regulatory pathways to maximize its dissemination and survival, both in the host and in the vent almost 50% of new cases in
aquatic reservoir. Infection results in the copious secretion of water and salts by cells of the villages where it is used. Follow-up
small intestine, leading to the characteristic “rice-water stool” containing over 107 viable studies after several years showed
that the method was sustainable
bacteria per milliliter. The excreted bacteria are hyperinfectious, with an infectious dose
and still provided protection (Huq
about 100–1000 times lower than laboratory-cultured bacteria. This transient phenotypic
et al., 2010). Although this has little
change drives the rapid person-to-person transfer during a cholera epidemic, but decays when effect on person-to-person transfer
the bacterium enters the aquatic environment. In the late stages of infection, V. cholerae once the disease is established
expresses genes that prepare it for survival outside of the host. These include genes involved in the population, this method
in motility, transport, and regulation of attachment to chitin and its subsequent utilization. reduces the opportunities for the
The ecology of V. cholerae in the aquatic environment and human hosts is illustrated in pathogen to transfer from the
Figure 12.1. aquatic environment into humans.

The ecology of V. cholerae in the natural environment, and the initiation and decline of
cholera epidemics is greatly affected by dynamic interactions between the bacteria and vib-
riophages. Some filamentous vibriophages are lysogenic (p.201) including CTXΦ, which
encodes the cholera toxin, plus a wide range of lytic tailed vibriophages. These regulate the
density and population structure of V. cholerae during plankton blooms due to environmen-
tal factors. In the host, the density of lytic vibriophages increases concomitantly with the
accumulation of large numbers of V. cholerae in the intestinal fluids during late stages of
infection. The continued action of these lytic vibriophages when they return to the aquatic
environment may explain the decline of cholera epidemics.

Mobile genetic elements play a major

role in the biology of Vibrio spp.
There is evidence that the ancestors of the modern Vibrio spp. have undergone major shifts in
their genetic makeup, which explains their transition from autochthonous aquatic bacteria to
pathogens. Genetic information in vibrios is divided between two distinct chromosomes. The
larger chromosome (Chr1) is of similar size in all vibrios (typically ~3.0 Mb) and contains
most of the “housekeeping” genes for growth and viability, as well as the major pathogenic-
ity factors, which are clustered in distinct regions known as pathogenicity islands that have
a different G:C base pair ratio to other parts of the chromosomes. Many of these genes show
334 Chapter 12

Figure 12.1 Schematic diagram of

the ecological interactions of Vibrio
cholerae. (1) In the aquatic reservoir,
V. cholerae strains of diverse genetic
makeup (shown by different col-
ors) are associated with biofilms on
particles, copepods etc. Attachment
to chitin promotes natural transfor-
mation by DNA released via lytic
phages and T6SS action. (2) Under
appropriate environmental condi-
tions, enrichment of pathogenic V.
cholerae (shown as red cells) occurs;
these can infect humans when
contaminated water is ingested.
(3) Bacteria migrate through the
mucus to colonize the epithelium of
the small intestine (4), mediated by
attachment via toxin co-regulated
pili (TCP). (5) Production of cholera
toxin (CT) leads to excessive secre-
tion of body fluids and salts from the
intestine, leading to watery diarrhea evidence of acquisition by horizontal transfer. The small chromosome (Chr2) is more vari-
containing large numbers of hyper- able in size (0.8–2.4 Mb) and gene composition. Although the chromosomes are independent,
infectious V. cholerae (shown by red the replication of Chr2 depends on the replication of a specific gene in Chr1, so that termina-
cells with yellow marking). (6) These tion of replication of both chromosomes and the subsequent cell division are coordinated.
escape from the mucus and propa- The genes on both chromosomes are expressed differently in the aquatic and human environ-
gate widespread epidemic transmis- ments. This genome flexibility is thought to play a major role in the adaptation of vibrios to
sion and seeding of the environment. various niches during their evolution.
Note that changes in the abundance
of lytic phages in the host and envi- In V. cholerae, the ctxA and ctxB genes encoding the two subunits of the toxin are derived
ronment also have a major effect on from integration into the chromosome of the genome of a phage (CTXΦ). Outside of areas
initiation and decline of epidemics contaminated by excreta from infected persons, most environmental isolates do not possess
(not shown, see text).
the ctx genes. Other genes, encoding accessory toxins, are involved in the morphogenesis of
CTXΦ. The second major factor in pathogenicity is the presence of toxin coregulated pili
(TCP) that coat the bacterial surface. The tcp genes encoding these pili are clustered in a
40 kb chromosomal pathogenicity island (VPI-1). This region may also be a phage genome
(VPIΦ) and some authorities argue that the pili act as a receptor for CTXΦ. VPI-1 also
encodes ToxT, which is a key regulatory element for the transcription of both ctxAB and the
tcp operon; thus, the remarkable situation occurs in which a regulator on one mobile element
controls the expression of a toxin on another.

ToxR and ToxS are transmembrane proteins that also regulate pili and toxin production
in response to environmental signals. Intriguingly, homologs of both proteins have been
found to be widely distributed in both pathogenic and nonpathogenic species in the fam-
ily Vibrionaceae, so it appears that they evolved in the distant past as part of the ancestral
genome. Their function is to control the synthesis of outer membrane porins that regulate
transport of ions and small molecules; their relative abundance in the membrane alters in
response to osmotic pressure. In V. cholerae, these signals have evolved to control toxin and
pilus gene expression through a complex signal transduction cascade involving other mem-
brane signaling proteins that are encoded by the CTX and TCP pathogenicity islands. The
evolution of pandemic V. cholerae strains can be explained by these shifts in genetic com-
position. Two islands in the genome of seventh pandemic (El Tor) strains (VSP-1 and VSP-2)
encode genes responsible for altered carbohydrate metabolism and functions thought to be
essential for the characteristic environmental persistence and spread of the El Tor biotype.
Another shift involved the horizontal gene transfer of genes encoding enzymes involved
in synthesis of lipopolysaccharide, which led to the emergence of strain O139. The key to
this extensive genome evolution is the large integron island (123.5 kb) found on the small
chromosome; this is a cluster of genes that captures DNA sequences from a wide variety
of sources, integrating them by site-specific recombination and rearranging them to form
functional genes.
 Marine Microbes as Agents of Human Disease 335

As noted above, attachment to chitin plays a key role in the life cycle of vibrios. Recently,
attachment to chitin has been shown to induce production of a Type VI Secretion System
(T6SS) whose primary function is the delivery of effector molecules that kill other bacte-
ria in the environment (Figure 12.2a). This is thought to provide a competitive advantage
to vibrios—by the liberation of nutrients from other bacteria—growing in biofilm com-
munities on the surface of crustacean exoskeletons or on chitin rich particles in the water.
Cells of closely related V. cholerae are protected from self-destruction by immunity pro-
teins. Furthermore, this process releases DNA from other bacteria. Chitin also induces
natural competence—the ability to acquire exogenous DNA by transformation—in which
a type IV pilus produced on the cell surface can attach to DNA and retract into the cyto-
plasm (Figure 12.2b) before processing prior to integration of selected genes. This hori-
zontal transfer of a large cluster of genes is believed to explain the continuous evolution
of new strains of V. cholerae, such as the sudden emergence of the epidemic O139 Bengal
serotype in 1992, and its rapid spread in India and Bangladesh. Chitin-induced regula-
tory programs and T6SS systems have also been demonstrated in other pathogenic and
symbiotic vibrios—V. vulnificus, V. parahaemolyticus, V. alginolyticus, V. coralliilyticus,
and Aliivibrio fischeri—and undoubtedly play a key role in their survival, population
structure, and genetic adaptation in the natural aquatic environment as well as in their
animal hosts.

Non-O1 and non-O139 serotypes of Vibrio cholerae are

widely distributed in coastal and estuarine waters
There are numerous other strains of V. cholerae that do not possess either the O1 or O139
antigens characteristic of epidemic cholera strains. These non-O1 and non-O139 serotypes
lack the major cholera toxin but they can cause localized mild to moderate gastroenteritis and

Figure 12.2 Processes in chitin-

induced DNA acquisition in Vibrio
spp. (a) Model of the structure and
mechanism of the Type 6 Secretion
System, showing the “aim, load, fire”
mechanism of a contractile sheath-
like structure and expulsion of a
cell-puncturing device. Toxins and
other effector molecules with various
functions are delivered directly into
target cells (other bacteria or animal
host cells). (b) Model of the “catch
and reel” mechanism of natural
transformation, based on real-time
fluorescence video microscopy of
Vibrio cholerae. dsDNA released from
other bacterial cells binds to the tip
of a specialized pilus that is pulled
through the PilQ pore via a molecu-
lar ratchet mechanism involving
the periplasmic protein ComEA.
Abbreviations: IM inner membrane,
OM outer membrane, PG peptido-
glycan. Credits: A. Reprinted from
Cianfanelli et al. (2016) with permis-
sion of Elsevier. B. Reprinted from
Ellison et al. (2018) with permission
of Springer Nature.
336 Chapter 12

BOX 12.1 RESEARCH FOCUS

Are human Vibrio infections a microbial barometer for climate change?

Sea surface temperatures (SST) are increasing. The well-known Although the number of cases remains small, official reporting of
effects of high levels of atmospheric CO2 (together with other gases non-cholera Vibrio infections is not mandatory in most countries,
such as methane and nitrous oxide) in promoting global warming and the real public health impact is probably considerably greater
via the “greenhouse effect” are a major topic of current socio- than realized. Increasing SST and rising sea levels is leading to
political concern. The most immediate effects in our oceans are greater risk of infection from vibrios in estuarine and coastal waters
the rise in sea surface temperature (SST) and ocean acidification in temperate regions, where they thrive in warm (>15˚C) seawater
(OA). Since 1950, the mean SST of the Indian, Atlantic, and Pacific with low to medium salinity (<25 ppt NaCl).
Oceans has increased by 0.7˚C, 0.5˚C and 0.3˚C and respectively
(IPCC, 2018; Figure 12.3). Within this, there are great regional Long-term increase in the relative abundance of vibrios.
variations, with extended summer periods and extreme tempera- Determining whether there have been changes in the numbers of
ture anomalies (“heatwaves”) more frequently observed at high particular microbes over long periods is difficult because of the
latitudes and in enclosed bodies of water. lack of historical records. A study by Vezzulli et al. (2012) devel-
oped a pioneering approach by developing methods to apply modern
Seasonality of Vibrio cholerae epidemics. Rita Colwell and her DNA analysis techniques to archived samples from the Continuous
numerous co-workers conducted many detailed studies of the envi- Plankton Recorder (CPR) survey maintained by the SAHFOS/MBA
ronmental factors affecting the epidemiology of cholera outbreaks Laboratory in Plymouth, UK. This sampler is towed by commercial
(reviewed by Lipp et al. (2002); Almagro-Moreno and Taylor ships and provides the longest running dataset of plankton samples
(2013). Clear seasonal patterns linked to climate are apparent in collected over 60 years. It has a mesh size of 270 µm and was designed
areas where cholera is endemic. In India, retrospective analysis of for collecting zooplankton, which are trapped on a moving band of
epidemics since the nineteenth century shows strong linkage to silk. However, small zooplankton and phytoplankton and associated
monsoon events. In the Bengal region, regular screening of water epibiotic bacteria are also trapped. Vezzulli and colleagues were able
for the abundance of V. cholerae shows strong correlation with to extract DNA from formalin-preserved silk samples that had been
increased SST, mean wave height, and blooms of phyto- and zoo- collected from transects in two areas of the North Sea from 1961
plankton. Baracchini et al. (2017) developed a model that shows to 2005. After, preliminary work to overcome the damaging effects
how rainfall and temperature explain the seasonality of cholera and of long-term formalin storage on DNA integrity, they used Q-PCR
the dynamics of outbreaks in different years, depending on the fluc- (p.47) with general and Vibrio-specific primers to provide estimates
tuations in the aquatic reservoir of the pathogen. of the relative abundance of Vibrio spp. as a proportion of the total
bacterial counts associated with the plankton. The authors found
Other Vibrio spp. do not cause epidemics. However, V. vulnificus, a long-term increase in relative Vibrio abundance, with a statisti-
V. parahaemolyticus, V. alginolyticus, non-O1/O139 V. cholerae, cally significant correlation with the summer SST in samples taken
and other Vibrio spp. can all cause significant disease, either by around the Rhine estuary (but not the Humber estuary, which experi-
wound infection or consumption of contaminated seafood. Baker- ences lower peaks of summer temperature). Using high-throughput
Austin et al. (2017) have dubbed vibrios “the microbial barometer sequencing to compare samples taken in the 1961–1972 and 1998–
of climate change,” because Vibrio infections are increasing and 2004 periods, Vezzulli et al. (2012) also showed that a major shift
being recorded in areas where they have not occurred previously. in bacterial community composition had occurred, with members of

Figure 12.3 Changes in North

Atlantic SST calculated as the differ-
ence between SST averages for the
years 2000–2011 and 1890–1958.
Reprinted from Vezzulli et al.
(2016) with permission of National
Academies of Science.
 Marine Microbes as Agents of Human Disease 337

BOX 12.1 RESEARCH FOCUS

the Vibrionaceae increasing in dominance. This was explained by distribution of annual cases of Vibrio infections is strongly cor-
a “regime shift” in plankton composition, known to have occurred related with peaks in summer SST. Most infections are probably
at this time due to changes in water circulation in the North Sea. acquired through leisure activities such as swimming and water-
Because of the strong association of vibrios with plankton, much of sports. Subsequently, Vezzulli et al. (2016) showed that the rise in
the observed increase observed in this study could be explained by Vibrio infections in Northern Europe matched the increase in abun-
a temperature-related increase in the abundance of plankton, espe- dance of plankton-associated vibrios in the archived CPR samples.
cially chitin-rich copepods. It is not known if the number of free- This was especially notable in the heatwave summer of 2006, when
living vibrios has also increased. In their follow-up study, Vezzulli an unprecedented 60 reported Vibrio infections in the North Sea
et al. (2016) extended their analysis to 133 CPR samples collected at and Baltic Sea areas coincided with one of the highest levels of the
nine locations in the North Atlantic (marked as dots on Figure 12.3). Vibrio abundance index from the CPR analysis. Similar trends of
In most of the areas, there was a strong positive correlation between increasing Vibrio diseases were observed using data from the US
decadal SST values and the relative abundance of vibrios. The Atlantic coast and northwest Spain.
increase was greatest in the last ten years surveyed, during which
there was an abrupt rise in SST. Statistical modeling showed that the A monitoring system to evaluate the risks of Vibrio infections.
main driver of changes in Vibrio abundance is the general increase Models developed using the information from the Baker-Austin
in the Northern Hemisphere Temperature, although the influence et al. (2012) study were used by the European Centre for Disease
of other natural oscillations in chemical and physical processes are Prevention and Control to develop a web-based mapping system
an additional contributory factor. As in their 2012 North Sea study, based on remote sensing data of SST and salinity of coastal waters
Vezzulli and colleagues found evidence from the CPR samples that from sampling stations throughout the world (Semenza et al., 2017).
long-term changes in phyto- and zooplankton abundance and species This provides daily prediction of areas that provide particularly
composition affected the relative abundance of vibrios and hypoth- amenable conditions for the growth of vibrios. In a study of the
esized that “a poleward transport of Vibrio species mediated by zoo- Swedish coast in the Baltic Sea, Semenza et al. found that areas
plankton will occur as a result of global warming.” predicted suitable for growth of vibrios showed a record increase in
reported vibrio infections (mostly of the ear, wounds, or septicemia),
Increased incidence of Vibrio infections. Baker-Austin et al. during the heatwave summer of 2014. They suggest that the mapping
(2012) provided the first empirical evidence of a clear link between tool could be used to issue risk-based public health warnings for rec-
climate change and an unexpected increase in Vibrio infections reational activities and harvesting shellfish at critical times. Baker-
observed in Northern Europe, especially around the Baltic Sea. Austin et al. (2017) review the properties of vibrios that make them
This is one of the largest low-salinity seas, where recent increases ideal candidates to assess the effect of climate change on the spread
in SST have been about seven times the global average, with par- of marine waterborne diseases and discuss the further research and
ticularly high summer temperatures. Statistical methods employed improved surveillance measures needed to provide reliable, quanti-
by Baker-Austin and colleagues showed that the number and fiable data for future risk-assessment models.

watery diarrhea, possibly as a result of the production of accessory toxins that damage the
membranes or disrupt the cytoskeleton of gut cells. These strains are widely distributed as
autochthonous components of coastal and estuarine waters and can survive and multiply in a
wide range of seafoods; infections are most commonly associated with eating raw or under-
cooked seafood, especially oysters. Infections are seasonal and show a peak with increased
water temperatures in late summer and early fall, coinciding with the warmest water tem-
peratures. They are also capable of infecting wounds and causing redness and swelling at
the site of infection. Septicemia can occur in people with reduced immunity or liver disease.

Vibrio vulnificus is a deadly opportunistic pathogen

V. vulnificus is part of the natural microbiota of estuarine and coastal water, sediment, plank-
ton, and shellfish throughout the world, especially tropical or subtropical regions. Human
infection occurs either by the infection of wounds or by primary septicemia following the
consumption of contaminated shellfish. Increased awareness has recently led to recognition
of cases in the Caribbean islands, Japan, and Taiwan, and there is growing evidence of the
spread of infections to new areas, probably linked to climate change. In healthy individuals,
consumption of contaminated raw shellfish usually results in an unpleasant, but not life-
threatening, gastroenteritis. However, in individuals with chronic underlying diseases (espe-
cially liver cirrhosis, immunodeficiency, or diabetes) a septicemic form of the disease may
result, with about 90% of cases requiring hospitalization and a mortality rate of about 50%.
In the USA it has the highest economic impact of all food-related illnesses. The US Gulf
338 Chapter 12

states require mandatory health warnings to be displayed wherever raw oysters are served.
The infective dose is believed to be very low, with estimates ranging from 100 to 300 bacteria
needed to infect susceptible persons. Males are much more susceptible to infection, possibly
owing to hormonal differences. A key feature of this pathogen is its extremely rapid growth in
host tissues, and death can occur as little as 24 h after infection. Infected persons show fever,
chills, shock, and large skin lesions filled with bloody fluid (Figure 12.4A). These can become
necrotic and so extensive that surgical removal of tissue or amputation of limbs is often
necessary. As well as causing food-borne disease, V. vulnificus can also cause a very serious
infection following contamination of wounds. This can occur if existing open wounds are
contaminated by seawater containing the pathogen. Wounds obtained from shucking oyster
shells, fishing hooks, fish spines, and the like are a particularly likely source (Figure 12.4B).

Pathogenicity of V. vulnificus is due to the

interaction of multiple gene products
V. vulnificus is clearly a very aggressive organism able to circumvent many of the body’s
defense mechanisms, especially when these are undermined by host susceptibility factors.
An extracellular capsule is especially important in conferring resistance to phagocytosis and
the bactericidal effect or complement in the serum. Strains of the pathogen that naturally
lack the capsule, or mutants in which the genes encoding the capsule are disrupted, lose their
virulence. A second factor contributing to the rapid growth of the pathogen is its ability to
scavenge iron as a nutrient from the host. V. vulnificus also produces powerful cytolytic and
proteolytic enzymes that are responsible for dissemination from the intestine and the exten-
sive tissue damage. In fatal cases, death is primarily due to the effect of endotoxin (LPS)
which stimulates overproduction of cytokines with consequent host damage. Many attempts
have been made to identify genotypic or phenotypic markers of pathogenicity using DNA
analysis—including recent full genome comparisons—but this is complicated by the exten-
sive horizontal gene transfer prevalent in the vibrios.

Environmental factors affect the

pathogenicity of V. vulnificus
Three biotypes of V. vulnificus have been described: Biotype 1 contains human clinical and
related environmental strains; Biotype 2 is a pathogen of farmed fish; and Biotype 3 appears
to be a hybrid strain causing wound infections associated with handling fish. Two distinct
genotypes of V. vulnificus Biotype 1 have been described, based on allelic variation of a
randomly amplified gene marker vcg (virulence correlated gene). Almost all human clinical
isolates from septicemic cases were thought to belong to the C-type containing this marker,
whilst most environmental strains (E-type) isolated from shellfish and water do not, although
these do cause wound infections. Genome sequencing has not yet provided a satisfactory
explanation for this. Updated phylogenetic classification systems are being developed,

Fig. 12.4 Consequences of Vibrio

vulnificus infection. A. Gangrene
and hemorrhage in a patient with
septicemia following infection by
ingestion. B. Infection of a fish
bone injury from which septicemia
developed. C. Gram-negative curved
bacilli isolated from a blood sample.
Reprinted from Hsueh et al. (2004)
with permission of CDC.
 Marine Microbes as Agents of Human Disease 339

incorporating more detailed consideration of the divergence of strains based on the host and
IRON ACQUISITION
geographic sources of isolation. It is likely that some complex regulation of gene expression
under different conditions—either in water, shellfish, or the human host—will explain this i IS IMPORTANT
IN BACTERIAL
enigma and this may be revealed by transcriptomic studies.
INFECTIONS
V. vulnificus can be detected at low levels in seawater from many parts of the world, but it is All bacteria require iron for the
most common in warm coastal waters of medium salinity, where it can reach densities of up activity of essential cellular func-
to 104 CFU mL−1. Filter-feeding shellfish such as oysters concentrate high levels of the bac- tions. Vertebrates possess highly
terium, often exceeding 105 per gram of tissue. There is a strong seasonality to V. vulnificus efficient systems for transporting
infections, and this correlates with an inability to detect the pathogen in the winter, when the and storing iron. The blood serum
protein transferrin binds iron with
temperature drops below about 15°C. Like V. cholerae and many other pathogens, the bacte-
extreme avidity, reducing the
rium enters a VBNC state as a stress response (in this case, reduced temperature rather than concentration of free iron in tissues
starvation). VBNC V. vulnificus cells have been shown to be virulent and can be resuscitated to 10 −18 M, about 108 times lower
and cultured if protected from oxidative stress (see p.95). than the concentration that bacte-
ria require for growth. In response
to infection, an even more efficient
Vibrio parahaemolyticus is the leading cause iron-binding protein (lactoferrin)
of seafood-associated gastroenteritis is released, and this removes iron
from the blood to storage in the
V. parahaemolyticus occurs in marine and coastal waters throughout the world, coloniz- liver and other organs. To over-
ing the surface of many types of marine animals and plankton. It is the leading cause of come this host defense mechanism,
gastroenteritis associated with the consumption of seafood and is the commonest cause of invading pathogens must obtain
all types of food poisoning in Asia (especially Japan), because of the popularity of raw sea- enough iron for growth by compet-
food. V. parahaemolyticus has recently become more important in the USA, Australasia, and ing with the host’s iron-withdrawal
Europe, largely due to the increasing international trade in seafood products, but possibly system. One of the reasons that
Vibrio vulnificus is able to cause
also due to rising seawater temperatures. After an incubation period of 12–24 h, infection
rapid septicemic infection in people
results in watery diarrhea, abdominal cramps, nausea, vomiting, headache, fever, and chills.
who have underlying liver damage
Most infected persons recover within a few days, but antibiotic treatment may be necessary is because the iron-storage and
for more serious infections. Mild cases are usually not reported, so the true incidence of transport systems are disturbed,
infection is unknown. As with other vibrios, V. parahaemolyticus can also cause infection leading to excess levels of iron in
of wounds—occasionally leading to necrotizing fasciitis or septicemia—after exposure to the serum. This allows the bacteria
contaminated seawater. to multiply so rapidly that they
overwhelm the body’s defenses
The disease is associated with raw or incompletely cooked seafood (particularly shrimp, (Baker-Austin and Oliver, 2018).
crabs, and bivalve mollusks). It is thought that a large infective dose is required for sufficient
numbers to survive the acidity of the stomach—use of antacid medicines may reduce this—
and outbreaks often follow the cross-contamination of seafood that has been kept in a warm
kitchen. The doubling time of the bacterium at 37°C can be as low as 10 min, permitting a
massive expansion of the population within a few hours.

Although high numbers of V. parahaemolyticus can usually be recovered from harvested

fish and shellfish, especially those from estuarine and coastal environments, only a small
proportion of isolates appear to carry factors that make the bacterium pathogenic. Isolates
from environmental samples are usually much less virulent than those isolated from clini-
cal samples, so it is likely that there are either multiple types of the organism, or that it
undergoes genetic changes and selective enrichment in the gut as seen with V. cholerae. The
exact mechanisms by which the bacterium produces disease remain unclear, despite exten-
sive research and the identification of numerous virulence factors, including adhesins, toxins,
iron-sequestering mechanisms, and Type III and IV secretion systems (T3SS, T6SS).

Most strains isolated from clinical samples are “Kanagawa positive”—this is a reaction on
a specialized type of blood agar—because they possess a thermostable direct hemolysin
(TDH) and/or a thermolabile related toxin (TRH). Although defined by their pore-forming
hemolytic activity on blood cells in the laboratory, the natural target of these toxins in vivo
is the intestinal epithelial cells. Binding to epithelial cell membranes in the colon leads to
disturbance of osmotic pressure, fluid loss and cell death. By contrast, most environmental
isolates produce neither TDH nor TRH—even if they contain the tdh or trh genes detected
using gene probes. About 10% of strains isolated from clinical cases are also non-toxigenic.

Further insight into the mechanisms of pathogenicity of V. parahaemolyticus came with the
recognition of genes for two Type III secretion systems (T3SS-1 and T3SS-2) in genome
340 Chapter 12

sequences. Like other vibrios, V. parahaemolyticus has two chromosomes. T3SS-1 is

encoded on the large chromosome and is present in environmental and clinical isolates,
whereas T3SS-2 is encoded on the small chromosome in a pathogenicity island that also
encodes TDH and is present only in clinical isolates. The T3SS can be likened to a molecular
syringe which pathogens use to inject proteins into host cells; this is known to be a major
factor in the pathogenicity of other enteric pathogens such as Shigella, Salmonella, Yersinia,
and enteropathogenic Escherichia coli. The T3SS-1 of V parahaemolyticus injects a number
of proteins that induce autophagy in the host cell (a process of self-digestion of cell contents),
followed by rounding and release of nutrients for the pathogen on cell death, whereas the
T3SS-2 is thought to be responsible for cytotoxicity and enterotoxicity due to disruption of
the cytoskeleton of intestinal cells. Recently, new Type VI Secretion Systems (T6SS) have
been described in enteric pathogens (Figure 12.2A). V. parahaemolyticus possesses two such
systems—T6SS-1 and T6SS-2—encoded on chromosomes 1 and 2 respectively. These have
been shown to contribute to adhesion and cellular disruption of cell cultures, but their role
WORMS AND MOTHS
i
in vivo is unclear.
AS MODELS FOR
PATHOGENICITY There are numerous serotypes of V. parahaemolyticus, characterized by the O (LPS) and K
STUDIES (capsular) antigens. Until the 1990s, outbreaks of V. parahaemolyticus showed no clear asso-
Many advances in the understand- ciation with particular serotypes, but since the mid-1990s, a clonal pandemic strain O3:K6
ing the genetic and biochemical has spread throughout the world and is now the cause of most seafood-associated outbreaks.
mechanisms by which bacteria Further investigation reveals that this serotype designation conceals variations of genetic
produce human diseases come diversity of V. parahaemolyticus that can only be revealed by multi-locus sequencing typing
from the use of cell culture, but (MLST, p.45) or full genome comparison. This shows that there are several genetic regions
proof of in vivo effects depends characteristic of pathogenicity islands, mostly present in the O3:K6 pandemic strains. All
on the use of animal models. For isolates of the O3:K6 pandemic serotype contain the genes for T3SS-2 and TDH. It seems
enteric diseases such as those
that these strains have evolved within a short time frame by acquiring large regions of new
caused by Vibrio spp., this often
DNA—possibly due to phage infection—and this may explain their prevalence and spread
requires administration of patho-
gens directly into the peritoneum as human pathogens.
of mice or the exposed intestine
(ileal loop) of rabbits, to mea- Research to develop reliable methods—combining classical culture methods with modern
sure inflammation or accumula- molecular approaches—for laboratory testing of V. parahaemolyticus and identification of
tion of fluid due to toxins. Such pathogenic strains in the face of this continuing evolution of genetic variants is especially
experiments require expensive important due to the growing global trade in seafood products and the need to adopt interna-
specialized facilities and there is tionally accepted standards to ensure public safety.
pressure to develop more ethically
acceptable alternatives that can
be used to conduct large-scale Microbes associated with fish and marine
screening programs of multiple
bacterial strains. In the case of
mammals can be transmitted to humans
Vibrio parahaemolyticus, it has Diseases of animals that are transmitted to humans are known as zoonoses. Some fish
proved very difficult to unravel the pathogenic bacteria can infect humans, usually via skin abrasions or wounds. Infection by
complexities of pathogenicity. To
Mycobacterium marinarum (“fish tank granuloma”) is an important hazard for aquarium
tackle this problem, researchers
workers and hobby aquarists. It usually causes lesions on the hands but can cause more
have used the nematode worm
Caenorhabditis elegans (Durai et al., serious infections of the bones and joints in people with impaired immunity. Streptococcus
2011) and larvae of the wax moth iniae and Edwardsiella tarda have been responsible for small outbreaks of serious invasive
Galleria mellonella (Wagley et al., infections in workers handling infected warm-water fish from aquaculture facilities. Other
2018) —these are both widely used bacteria such as Vibrio spp., Aeromonas species, Lactococcus garviae, and Erysipelothrix
in different branches of biomedical rhusiopathiae from the skin or intestines of fish can cause range of skin and gastrointestinal
research—to determine the patho- infections, including some chronic conditions.
genic potential of the highly diver-
gent strains isolated from different There is also a definite, albeit small, risk of acquiring diseases from marine mammals. There
courses. Wagley and colleagues have been cases of infection by Mycobacterium sp. (causing respiratory and skin lesions) and
found the wax moth model par-
Brucella infection (causing severe lethargy, fever, and headaches) in aquarium and research
ticularly useful, as the larvae were
susceptible to both toxigenic and
staff working with mammals. Marine conservation organizations highlight the need for cau-
non-toxigenic strains of V. parahae- tion when engaging in the rescue of stranded cetaceans. “Seal finger” is an extremely painful
molyticus, and this is leading to the infection of the hands caused by Staphylococcus or Erysipelothrix, which occurs in seal
development of new genetic mark- hunters and research workers, especially in the Arctic regions of Norway and Canada. Seal
ers for the effective screening of bites can also cause serious infections, caused by a variety of poorly characterized bacteria,
clinical and environmental isolates and are notoriously difficult to treat unless antibiotics are used promptly. The popularity of
of V. parahaemolyticus. tourist activities such as swimming with dolphins, could also lead to a rise in transmission
 Marine Microbes as Agents of Human Disease 341

of zoonoses. Such activities also increase the likelihood of transmission from humans to
marine mammals, although there is limited evidence of this to date. Of the viruses, influenza
poses the most important risk and several instances of direct transmission from persons in
close contact with infected seals have occurred. Evolution of new strains of influenza virus
occurs regularly through antigenic changes, owing to recombination events in the fragmented
genome. It is therefore possible that marine mammals could provide an opportunity for trans-
mission of new variants to humans. Protozoan parasites such as Giardia may also have a
reservoir in marine mammals.

DISEASES CAUSED BY MARINE MICROBIAL TOXINS

Scombroid fish poisoning results from
bacterial enzyme activity
This type of food-borne intoxication is associated with eating fish of the family Scombridae,
which include tuna, mackerel, marlin, and bonito. The tissue of these fish contains high levels
of the amino acid histidine. If there is a delay or breakdown at any stage in the refrigeration
process between catching the fish and consumption, bacteria from the normal microbiota
can multiply and convert histidine to histamine (Figure 12.5). Bacteria isolated from fish
associated with scombroid fish poisoning include Raoultella planticola, Morganella morga-
nii, Hafnia alvei, and Photobacterium phosphoreum. The histamine is heat stable and will
withstand normal processing such as canning. The spoilage may not be enough to alter the
taste or smell of the fish, but levels of histamine above 1 mg g−1 can be enough to induce a
rapid allergic-type response. Reddening of the face and neck, shortness of breath, and, in
severe cases, respiratory failure can result within a few minutes of eating contaminated fish.
Treatment with antihistamine drugs is beneficial and the person usually recovers fully. The
incidence of scombroid poisoning is increasing because of the rising popularity of sushi and
the import of fresh tuna and swordfish by airfreight over long distances, with opportunities
for breaks in the cold chain. Some outbreaks involving groups of affected people may neces-
sitate intervention by health authorities and destruction of suspected fish products.

Botulism is a rare lethal intoxication from seafood

Botulism is one of the most feared food-borne diseases. It is caused by Clostridium botuli-
num, a Gram-positive bacterium that forms endospores that survive high temperatures. The
bacterium is a strict anaerobe and under appropriate conditions it produces a powerful neu-
rotoxin in a range of different foodstuffs. Botulinum toxin is the most lethal toxin known;
if ingested, tiny amounts of the toxin—estimated to be only about 200 ng—cause a severe
flaccid paralysis. The toxin is a protease that cleaves proteins involved in the docking of
neurotransmitter vesicles, blocking release of the neurotransmitters and therefore prevent-
ing muscle contraction. C. botulinum is normally incapable of growth within the adult gut,
so disease is dependent on ingestion of preformed toxin, with symptoms usually occurring
12–36 h after ingestion. Unlike other fast-acting neurotoxins, the delay is caused by absorp-
tion of the toxin and transport via the central nervous system. (Note that the bacterium can
grow in the gut of babies, and infant botulism is typically associated with spore-contaminated
honey.) C. botulinum is classified into types A to G based on the serological properties of the
toxin produced. Type E is most commonly associated with seafood and can be isolated from
marine sediments and the intestinal contents of fish and crustaceans. Fortunately, food-borne

Histidine
decarboxylase
Histidine Histamine
(non-toxic) (toxic)

Figure 12.5 Enzymatic conversion of histidine to histamine.

342 Chapter 12

botulism—especially associated with seafoods—is very rare because of strict regulations

RUMMAGING
i IN THE TRASH
REVEALS SOURCE OF
concerning commercial canning and other methods of preservation. Most salting, drying,
and smoking methods of fish preservation are safe, but some cases have occurred following
consumption of smoked fish in modified atmosphere packaging (p.360). Since smoking kills
BOTULISM OUTBREAK
most, but not all, of the spores, such products should always be refrigerated to inhibit growth
Commercial canning, the most of any survivors and sold with restricted shelf life. Most cases occur from home-prepared
common method of preserving ethnic foods. Botulism is a persistent serious health problem in indigenous Arctic communi-
fish, is designed to ensure destruc- ties in northern Canada and Alaska, where some traditional foods such as fish, whale, or seal
tion of Clostridium botulinum meat are wrapped, buried in the permafrost and left to rot. Wound botulism can occur in
spores. The “12D process” of cook-
workers handling fish, causing local paralysis.
ing at high temperature ensures
a 1012-fold reduction in spore
numbers—or expressed another
way, only a one in a trillion chance
Fugu poisoning is caused by
that a can will contain a spore— a neurotoxin of bacterial origin
offering a very high safety margin.
This intoxication is caused by the ingestion of tetrodotoxin (TTX; Figure 12.6a), which is
However, one famous outbreak
of botulism in 1978 occurred in
found in the intestines, liver, and gonads of pufferfish, especially Takifugu rubripes and
Birmingham, England, involving related species. The toxin is probably synthesized by environmental bacteria and accumu-
one of the world’s best-known fish- lates in the organs of the fish. TTX is one of the most active neurotoxins known, and acts
processing companies. In this case, by blocking the flow of sodium ions in the nerves. Within a few minutes or hours of eating
four people were infected, and fish contaminated with the toxin, victims feel tingling sensations in the mouth and a sense
two died, following consumption of lightness, followed quickly by the onset of total paralysis. The person remains conscious
of canned Alaskan salmon. Careful but totally immobilized until the moment of death, which occurs in up to 50% of cases. In
investigation by public health offi- Japan, fugu is a prized delicacy for which diners in specialist restaurants (Figure 12.6b) are
cials resulted in the discarded can prepared to pay large amounts for the thrill of eating tiny portions of this risky food—usually
being recovered from a rubbish
served as elegant plates of translucent sashimi—with the added frisson of possibly ingest-
bin. It was found to have a minute
ing a miniscule dose of the toxin in order to give a “buzz.” Fishing for T. rubipres and the
hole in the seal. Despite being
properly heated, the investigators sale of fugu are now tightly regulated in Japan and fugu chefs must be specially licensed, so
concluded that a small amount of most of the 50 or so cases each year occur among fishermen and amateur cooks who prepare
contaminating material from fish the dish. There are a few licensed fugu restaurants in the USA, but it is banned in Europe.
viscera was sucked into the can as Fish farmers in some parts of Japan have developed systems for production of nonpoisonous
it cooled on the production line, fugu by rearing fish on special diets and centralized preparation facilities have improved
and this contained enough viable safety. Consumption of fugu has a long history and it is deeply embedded in Japanese culture,
spores to germinate and produce a with associated economic significance. The development of safe fugu has met with problems
lethal dose of toxin. The economic
consequences for the company
and the Alaskan salmon industry
were very serious because of the
costs of product recall and adverse
publicity.

OH

O O
HO OH

OH
HO
HN N OH
HN H

(a)

Figure 12.6 (a) Representative

structure of tetrodotoxin (there are
numerous analogues of the basic
structure). (b) A typical fugu restau-
rant in Tokyo, where live puffer fish
are displayed in tanks to be prepared
by licensed chefs. Credit: Richard (b)
Tucker, CC-BY-SA 2.0.
 Marine Microbes as Agents of Human Disease 343

of consumer acceptance and controversy about whether rules prohibiting the sale of fugu
liver—“kimo,” the most expensive and prized delicacy—should be relaxed.

TTX is widespread amongst marine animals

Besides pufferfishes, a wide variety of animals including triggerfishes, sunfishes, xanthid
crabs, seastars, molluscs, and worms are known to contain TTX, in which its main function
seems to be in defense against predators. Some marine animals also use TTX as a mecha-
nism of attack; among the best-known examples are the blue-ringed octopus (Hapalochlaena
maculosa) of northern Australia, which uses TTX in its venom to paralyze its prey of fish
and crustaceans, and the tropical flatworm Planocaeta multicentaculata, which paralyzes
its prey of cowrie mollusks. A few species of terrestrial salamanders, newts, and toads
also produce the toxin. Animals that contain TTX appear to have single-point mutations in
the amino acid sequence of a sodium channel protein, which makes them resistant to the
toxin’s effects. Because of the wide distribution of TTX in phylogenetically diverse groups it
seems unlikely that the toxin is synthesized by the animals themselves—it is a very complex
molecule with no known critical metabolic role—so the most likely explanation is that it is
acquired through the food web. This is supported by the fact that farmed pufferfish that have
been fed special diets are non-toxic. Various isolates of Vibrio, Pseudomonas, Alteromonas,
Shewanella, Flavobacterium, and other bacteria have been isolated from the skin, intestinal
tract, gonads, salivary glands, and elsewhere of TTX-containing animals. These bacterial
isolates have also been reported to produce the toxin in culture, although there is much con-
troversy about these results due to variations in growth conditions and detection methods.
Some investigators consider the TTX-producing bacteria to be symbionts, but the functions
of TTX for these bacteria and the evolution of this possibly symbiotic relationship remain
unknown.

There is some concern that populations of TTX-containing fish have spread to other regions,
where the dangers are less well known. For example, populations of the toadfish Lagocephalus
scleratus have increased in the Eastern Mediterranean and there have been cases of TTX poi-
soning from this source in Turkey. Recently, there have been some concerns for public health
following the detection of low levels of TTX in marine bivalves and gastropods in European
waters. Although there are very few known cases of TTX poisoning associated with this
source, studies are in progress to assess potential risks.

Some dinoflagellates and diatoms produce harmful toxins

The growth of marine plankton is affected by a wide range of factors that influence their
spatial and temporal distribution. Seasonal periodic increases in plankton density (blooms)
obviously have great ecological importance in ocean food webs. In addition, some planktonic
microbes produce toxins that affect human health, and many of these can form exceptional
blooms under certain conditions. These are frequently referred to as “red tides,” although this
colloquial term is something of a misnomer since not all toxic blooms are red in color—they
may not even reach high enough densities to discolor the water. A more generally accepted
term is “harmful algal blooms” (HABs), although this term is also not entirely accurate
because health effects are not always associated with a distinct “bloom.” Among the many
thousands of species of phytoplankton, only about 150 are known to produce toxins, and only
a few of these actually cause problems for human health. Examples of some of the main toxic
dinoflagellates and diatoms are illustrated in Figure 12.7. The main health hazard for humans
comes from eating fish or shellfish that have accumulated the toxin in their tissues through
feeding in water containing high levels of toxin-producing plankton. Some toxins may also
produce disease symptoms as a result of direct contact with contaminated water or inhala-
tion of aerosols. Representative structures of the major algal toxins are shown in Figure 12.8.
All of these phycotoxins are non-proteinaceous substances that remain active after cooking.
This, together with the usually very rapid onset and neurological symptoms, is an important
distinction of HAB intoxications from most bacterial and viral infections associated with
consumption of shellfish. Most of the toxins cause paralysis by binding to the voltage-gated
sodium channel of nerve cells, inhibiting influx of Na+ ions and therefore blocking the gen-
eration of action potentials.
344 Chapter 12

Figure 12.7 Examples of toxin-

producing diatoms and dinoflagel-
lates. Upper: light micrographs of
A. Dinophysis acuti and B. Pseudo-
nitzschia pseudodelicatissima. Lower:
scanning electron micrographs
(SEM) of C. Karenia brevis and D.
Gambierdiscus toxicus. Credits: A.
D. Cassis, T. Ivanochko, J. Shiller, B.
Moore-Maley, J. Kim, S. Huang, A.
Sheikh, G. Oka. Phytopedia: The
Phytoplankton Encyclopedia Project.
Published at www.eoas.ubc.ca/
research/phytoplankton/B. Susanne
Busch, IOW. C, D. Paula Scott,
Florida Fish and Wildlife Commission.

TOXIC HABS ARE

i NOT NEW, BUT
THEY ARE GROWING
IN IMPORTANCE
Sudden changes in the appear-
ance of coastal waters have been
described for many centuries. For
example, it is often surmised that Paralytic shellfish poisoning is caused by
the reference in the Bible to the saxitoxins produced by dinoflagellates
“first plague … and the river [Nile]
turned to blood” is a description Paralytic shellfish poisoning (PSP) has a worldwide distribution and is the best-known human
of such a bloom, thought to have illness associated with microalgae. PSP is caused by a group of over 50 closely related water-
occurred about 1290 BCE). Ancient soluble toxins known as saxitoxins (STXs) and gonyautoxins. The main organisms respon-
cultures such as Pacific Islanders sible for production of PSP toxins are Alexandrium spp. and Gymnodinium catenatum, which
and coastal tribes of North have been described for many years in temperate waters of North America, Northern Europe,
America had scouts who “read Scandinavia, and southeast Asia. In the past few decades these taxa have been increasingly
the sea” for changes in color, in reported from countries in the southern hemisphere, including Australia, New Zealand, and
order to warn communities about
Chile. In tropical waters, PSP is largely caused by Pyrodinium bahamense. PSP results from
the dangers of harvesting fish
eating shellfish, especially clams, oysters, mussels, and certain species of crabs, in which
and shellfish. From the fifteenth
century onwards, seafaring explor- toxins produced by the dinoflagellates have built up through filter feeding. Crabs and lobsters
ers recorded outbreaks of fish and can also accumulate the toxin by feeding on contaminated bivalve mollusks.
shellfish poisoning that affected
them and their crews during their PSP has a dramatic onset, often within minutes of consuming contaminated shellfish. Tingling
journeys. Fossil records show that around the lips is quickly followed by numbness of the face and neck, nausea, headache, and
blooms of toxic microalgae have difficulty in speech. In severe cases, muscular and respiratory paralysis can occur, and mortal-
occurred for millennia. Despite ity can be more than 10% unless there is rapid access to medical services. Treatment consists
their long history, HABs have of stomach pumping and administration of charcoal to absorb the toxins. In severe cases, arti-
assumed growing importance since ficial respiration may be required, but there are usually no long-lasting effects.
the late twentieth century because
they are occurring more frequently
and in regions where they have not
Many countries operate PSP-management programs, which include monitoring seawater for
previously occurred. This is due to the density of toxin-producing dinoflagellates and the assay of shellfish samples for saxitoxins.
stimulation of blooms by nutrient Careful analysis has shown that there is no straightforward relationship between the presence
runoff, climate change, increased of toxin-producing dinoflagellates and the level of particular toxins in shellfish tissue. Shellfish
aquaculture in coastal waters, and may modify the toxins or excrete them differentially according to climatic and physiological
transport of algae in ships’ ballast conditions, making management of outbreaks more difficult. Commercial collection of shell-
water. Besides affecting human fish is prohibited when saxitoxin levels in the tissue exceed a certain threshold concentration
health, HABs damage other marine (typically, 800 μg kg−1). The traditional internationally accepted assay method involves intra-
life (see Chapter 11) and have peritoneal injection of an extract of shellfish meat into mice. The toxin is quantified as “PSP
significant economic and social
equivalents” by the time taken for the mice to die, but this assay does not give any informa-
impacts on fisheries, aquaculture,
tion about the precise amounts of individual toxin components or their breakdown products.
and tourism.
There have been many attempts to develop alternative methods that do not depend on the use
 Marine Microbes as Agents of Human Disease 345

Figure 12.8 Representative struc-

tures of major dinoflagellate and
diatom structures. Note that there is
considerable variation in the struc-
ture of brevetoxins, ciguatoxins, and
saxitoxins.

of animals and that give more reliable and informative results. This is complicated by the fact
that the profile of different toxin analogues and their toxicity can vary greatly between differ-
ent regions. High-pressure liquid chromatography fluorescence detection (HPLC-FLD) and
liquid chromatography/mass spectrometry (LSMS) are now widely used as official methods
by regulatory authorities. A receptor-binding assay based on the ability of native toxin in a
sample to compete with radioactively-labeled saxitoxin for binding sites in rat cell membrane
preparations also provides quantitative results. Commercially available immunological rapid
“dipstick” test kits produced by Scotia™ and Neogene™ have been approved for regulatory
use and allow shellfish producers to conduct simple qualitative tests on a boat or dockside
before harvesting. A sample of shellfish tissue is ground and applied to a lateral flow strip,
which provides a positive result (negative if below the 800 μg kg−1 saxitoxin equivalent level)
within a few minutes. Quantitative results can be obtained using enzyme-linked immunosor-
bent assay (ELISA) methods including Abraxis™ and Europroxima™ kits. This screening is
very cost-effective for use by shellfish producers to determine whether full testing by regula-
tory authorities is needed. The closure of commercial fisheries can have devastating economic
and social effects, affecting the livelihood of whole communities for many years, so valida-
tion of methods appropriate to a particular fishery and area is essential. Authorities will also
impose bans on collection of shellfish for personal consumption, and notices on beaches are a
common sight in affected areas (Figure 12.9).
346 Chapter 12

Figure 12.9 A typical beach notice

prohibiting the collection of shell-
fish due to contamination by PSP
or other toxins. Credit: Washington
State Department of Health.

Brevetoxin causes illness via ingestion

or inhalation during red tides
Neurotoxic shellfish poisoning (NSP) is caused by the toxin brevetoxin, produced by the
dinoflagellate Karenia brevis (Figure 12.7C) and a few other species. The lipophilic toxin
binds to voltage-gated sodium channels in nerve cells, leading to disruption of normal neu-
rological function. Typical signs of intoxication include dilated pupils, paresthesia (abnormal
sensations on the skin, such as tingling or burning), a reversal of hot–cold temperature sen-
sation, vertigo, muscle pains, diarrhea, and nausea. Blooms of K. brevis are highly distinc-
tive—these are the archetypal red tides—and are usually seasonal, starting in late summer
and lasting for many months. They have been known in the Gulf of Mexico, especially the
Florida coast, for several centuries; however, in recent years blooms have occurred more
frequently and have been of longer duration. Since the late 1980s, blooms have also occurred
on the eastern US coast and in New Zealand, probably involving other Karenia spp. Like
PSP, NSP is caused by consumption of shellfish that have accumulated the toxin, although
symptoms are usually milder. However, an additional threat comes from inhalation of the
toxin. During extensive blooms, lysis of K. brevis cells and disruption by surf action leads to
aerosols containing the toxin—these can be carried by wind up to 100 km inland—causing
shortness of breath and eye irritation. In Florida, beach closures are necessary during red
tides and health advisory notices are issued warning asthmatics and other susceptible people
to stay indoors. The toxin also kills marine mammals, birds, and turtles (see p.320) and the
blooms also cause massive fish kills, leading to further problems as coastal waters become
anoxic. Thus, red tides have a major impact on fisheries and tourism, and government agen-
cies are involved in close monitoring using satellite imagery and other techniques, so that
precautionary management procedures can be put in place at the first signs of a bloom.

Dinophysiotoxins and azaspiracid toxins from

shellfish result in gastrointestinal symptoms
Diarrhetic shellfish poisoning (DSP) is caused by eating shellfish (usually bivalve mollusks)
that contain the toxin okadaic acid and other dinophysiotoxins produced by dinoflagellates of
 Marine Microbes as Agents of Human Disease 347

the genera Dinophysis (Figure 12.7A) and Prorocentrum. Okadaic acid is an inhibitor of pro-
BENEFICIAL
tein phosphatases. Ingestion causes the intestinal cells to become highly permeable to water,
leading to profuse diarrhea and vomiting which usually lasts for a day or two without serious i APPLICATIONS
OF MARINE
effects. The symptoms resemble those caused by infection by enteric viruses from shellfish
ALGAL TOXINS
grown in sewage-polluted waters (p.362), but a growing body of evidence since the 1980s has
shown that DSP can result from consuming shellfish from high-quality waters that are free of Tetrodotoxin (TTX) and saxitoxins
sewage contamination. Many outbreaks have been described in Japan and in European coun- both cause paralysis by inhibiting
tries, particularly on the Atlantic coast of Spain, France, and the British Isles. A European transmission of the nerve impulse
Union Directive requires the monitoring of shellfish for the toxin—by LCMS and immunoas- by interfering with the pas-
sage of sodium ions through the
say kits like those used for STX—and this has resulted in closure of many traditional shellfish
membrane of nerve cells. Indeed,
harvesting and aquaculture operations in areas where toxins have been detected. Unlike the
studying the interaction of these
other shellfish poisonings, a DSP risk can occur even in the absence of a bloom; Dinophysis toxins with membrane proteins
densities as low as 200 cells mL−1 can lead to accumulation of unacceptable levels of toxin was vital for our understanding
in shellfish. Okadaic acid has been shown to be carcinogenic in rats, and long-term exposure of how sodium channels work.
has been suggested as a possible cause of cancer in the digestive tract, although there seems Ion channels are composed of
to be insufficient epidemiological evidence to raise human health concerns. proteins in the membrane which
form a pore, enabling the passage
In the 1990s, outbreaks of DSP-like disease associated with mussels harvested from Irish of specific ions from the aqueous
waters was linked to the consumption of mussels (Mytilus edulis). DSP toxins could not be external environment, through
identified, and a new group of related toxins called the azaspiracids (AZAs) was shown to be the hydrophobic membrane to the
interior of the cell. These toxins
responsible. Since then, several outbreaks of AZA poisoning have been reported in several
interact with the external surface of
countries. The symptoms are severe diarrhea and vomiting, nausea, and headache, which
the channel and prevent the flux of
may persist for several days. In a mouse bioassay, AZAs cause death from serious tissue and ions. Besides their use as biological
organ damage, with neurotoxic effects at high doses. However, their mode of action seems research probes, these molecules
quite different from the other shellfish toxins. Development of improved chromatographic also have medical applications as
assays and immunoassays means that routine monitoring for AZAs is now being introduced, anesthetics and therapeutic agents
with the result that they are much more widespread than previously thought. The toxins seem (Assunção et al., 2017). Modified
to be associated particularly with M. edulis and there are no obvious links with bloom events. TTX and saxitoxins have been
Rather, it seems as if the toxin accumulates gradually in the tissue of the mussel as a result used for prolonged anesthesia and
of long-term exposure to low levels of the dinoflagellate Azadinium spinosum. The toxin can relief of severe pain, prevention of
brain damage after strokes, and as
persist in the shellfish tissue for many months.
a treatment for heroin addiction.
Sadly, their intense toxicity has
meant that they have also been
Amnesic shellfish poisoning is caused by toxic diatoms developed for use in biological
Amnesic shellfish poisoning (ASP) was discovered in 1987, following an unusual outbreak warfare, terrorism, and espionage.
of seafood poisoning in Prince Edward Island, Canada, in which more than 100 people were
affected and three died. The symptoms included rapid onset of vomiting, disorientation, and
dizziness that did not match those of other known diseases. As is usual in such large out-
breaks, health professionals interviewed those affected and were surprised that many could
not answer the usual questions about what they had eaten before feeling ill; they showed
other signs of amnesia, which persisted for many months. This suggested that a new type
of neurotoxin was involved, and this was identified as domoic acid produced by the diatom
Pseudo-nitzschia multiseries. We now know that several other species of Pseudo-nitzschia
(e.g. P. pseudodelicatissima, Figure 12.7B) can produce domoic acid and these have been
isolated from many parts of the world. Human cases of ASP are rare, but they are almost
certainly underreported, and in areas where Pseudo-nitzschia blooms occur, they may be
associated with eating planktivorous fish such as anchovies. As discussed on p.320, domoic
acid from Pseudo-nitzschia blooms also causes disease in marine birds and mammals and is
increasingly recognized in the marine food chain.

Ciguatera fish poisoning has a major impact

on the health of tropical islanders
In terms of public health, ciguatera fish poisoning (CFP) is undoubtedly the most important
of the diseases caused by marine toxins, especially for inhabitants of tropical islands who
rely on fishing. It is also a well-known hazard among sailors and travelers in the tropics
(35°N–35°S) and was first documented by explorers of the Caribbean and Pacific in the 15th
and 16th centuries. In 1774, Captain James Cook gave a detailed account in his ship’s log of
poisoning that affected him and his entire crew after eating fish in the New Hebrides, which
348 Chapter 12

was almost certainly a description of CFP. The causative agent of CFP was not discovered
DID CIGUATERA
? CAUSE THE
MIGRATION OF
until 1977, when the toxin-producing dinoflagellate Gambierdiscus toxicus (Figure 12.7D)
was isolated in the Gambier Islands in French Polynesia. Related species occur worldwide,
and isolates vary greatly in the amount and types of toxin produced. Cyanobacteria have also
PACIFIC ISLANDERS?
recently been implicated in some outbreaks. The fat-soluble toxins become concentrated by
In the years between 1000 and biomagnification as they pass up the food chain, and disease is almost always associated with
1450, multiple waves of migration eating large predatory fish, as illustrated in Figure 12.10. More than 400 species of fish have
occurred between the islands of been implicated in CFP, with moray eel and barracuda being the most common. CFP is char-
Polynesia and distant lands, espe- acterized by a distinctive sequence of diarrhea, vomiting, abdominal pain, and neurological
cially New Zealand. The reasons
effects within a few hours of eating contaminated fish. Unusual symptoms include numbness
why islanders undertook such
and weakness in the extremities, aching teeth, and reversal of temperature sensation (cold
perilous voyages into the unknown
across vast expanses of ocean things feel hot and hot things feel cold). In severe cases, this can progress rapidly to low
have always been something of a blood pressure, coma, and death. However, the type and severity of symptoms depends on the
mystery. By studying past climate particular “cocktail” of toxins consumed, and this varies with species of fish and geographic
conditions and examination of origin. There seem to be important differences in structure between the main CFP toxins
archaeological artifacts, a native implicated in Pacific and Caribbean cases. Doses that cause severe symptoms in one person
of the Cook Islands has suggested may be harmless in another. It is also likely that other ancillary toxins are involved. In some
that the migrations may have been people, neurological problems can persist for many years, and a repeat attack can be triggered
due to increased chronic incidence by eating any fish (whether or not it contains toxin) or by drinking alcohol. The toxin has also
of CFP affecting the population, been shown to cross the placenta to affect the fetus, and there are even reports of transmission
who were heavily dependent on
to another person via sexual intercourse. In countries where the disease is endemic, there are
ciguatera-susceptible large carnivo-
rous fish at the time (Rongo et al.,
a range of local remedies, but these are of unproven value. In severe cases, it is necessary to
2009). Reliance on fishing in the administer intravenous mannitol, which reverses the effect on sodium transport.
islands declined after this time and
the remaining population seem The incidence of CFP can be quite localized but unpredictable. Local fishermen will often
to have moved to smaller, safer believe that fish from a particular island or reef will be hazardous, whilst others nearby
fish species. However, the highest are not. Many travelers have learnt—to their cost—that such claims are often not reliable.
incidence of CFP still occurs today Gambierdiscus adheres to macroalgae on the surface of dead coral and on the seabed, but
in the Cook Islands, and modern little is known about the factors promoting colonization of the macroalgae by the dinoflagel-
outbreaks have led to a decline in lates and its subsequent proliferation and toxicity. Damage to the reef by dynamite fishing,
fishing and reliance on processed
foods rather than fish, with a new
wave of mass emigration to New
Zealand and Australia in the 1990s.

Figure 12.10 Schematic diagram

showing the accumulation of cigua-
toxin through the food chain on
coral reefs.
 Marine Microbes as Agents of Human Disease 349

diving, boat anchors, military action, or construction work promotes the initial colonization
by the macroalgae. Increased frequency of bleaching and diseases of corals may also be
leading to higher occurrence of CFP. However, this is only part of the story, as the incidence
of toxic fish is very dependent on environmental factors such as rainfall, pollution, and nutri-
ent runoff from the land. Also, it is likely that there are genetic differences among different
strains of Gambierdiscus and factors that affect dinoflagellate growth may have a different
effect on toxin production. Recently, reports of ciguatera-like illness associated with dino-
flagellates in the genus Ostreopsis have been reported in Italy and Puerto Rico. The complex
ecological, toxicological, and physiological factors seen with CFP illustrate dramatically how
careful we must be when looking for a simple “cause and effect” in the etiology of a disease.

Bacteria influence the production of HAB toxins

Early in the study of the origin of PSP, microbiologists observed bacteria-like structures
in electron micrographs of dinoflagellate cells and asked if these could be symbionts and
whether they are involved in toxin production. Besides the intrinsic interest in the possible
role of symbiosis in providing their eukaryotic host with a possible defensive mechanism,
answering these questions is of importance in the monitoring and mitigation of fish and
shellfish toxicity. Cyanobacterial toxins, especially microcystin, are an important health haz-
ard in freshwater lakes and it is now known that certain cyanobacteria possess gene clus-
ters involved in saxitoxin synthesis, although phylogenetic analysis suggests that these have
evolved separately rather than by gene transfer. Recently, there have been accounts of PSP
linked to cyanobacterial blooms in coastal lagoons.

Dinoflagellates form tight associations with numerous bacteria and many experiments have
been conducted to compare levels of toxin production in dinoflagellate cultures before and
after curing of contaminating bacteria, by treatment with antibiotics or enzymes. Although
there is little evidence for autonomous production of dinoflagellate toxins by associated bac-
teria, they do influence the biosynthesis of the toxins—probably by supplying cofactors or
precursors necessary for toxin production. Bacterial involvement also seems to be impli-
cated in the production of domoic acid by the diatom Pseudo-nitzschia. Axenic cultures have
greatly reduced levels of domoic acid, whereas levels are increased when certain species of
bacteria isolated from the original culture are reintroduced.

Although the precise role of bacteria in synthesis of HAB toxins remains uncertain, there
seems little doubt that the presence of bacteria has a considerable influence on the levels and
nature of toxins released by the algae. An additional complication is that some of the bacteria
naturally associated with dinoflagellates and diatoms can cause lysis of the cells, leading to
sudden release of high levels of toxin, whilst other bacteria may metabolize algal products.
The importance of such algicidal bacteria as a possible method of control of HABs is dis-
cussed in Chapter 14.

Dinoflagellate and diatom toxins may be

antipredator defense mechanisms
The toxins produced by dinoflagellates and diatoms are usually complex secondary metabo-
lites, and diversion of energy to their synthesis implies some ecological benefit. The extreme
sensitivity of humans and other animals to these toxins and the fact that most act on similar
mechanisms of nerve action is coincidental. It is usually presumed that toxins have evolved to
deter predation by zooplankton. If this is correct, once physical and chemical conditions are
favorable for the initial bloom of a toxic species, the production of toxins will deter predation
and prolong the maintenance of high densities of the species. It seems that many zooplankton
species, such as copepods, can discriminate dinoflagellates with low toxin content and feed
selectively, and some flagellates and ciliates appear to be killed by dinoflagellate toxins.
However, toxins do not give universal protection against grazing by all species of predator,
and it may be that the toxins offer a selective advantage to organisms that produce them under
low-nutrient conditions, by directing predation pressure onto competitors. It is notable that
the production of toxin, and its cellular concentration, is very dependent on the supply of
nutrients (especially phosphorus). This has led to the suggestion that the toxins may act as a
350 Chapter 12

reserve for storage when the nutrient supply is unbalanced. As discussed in Chapter 6, many
dinoflagellates are mixotrophic, with the capacity for both phototrophic and heterotrophic
nutrition, and the production of such secondary metabolites may be associated with major
switches in metabolic pathways.

Complex factors affect the incidence of

HABs and toxin-associated diseases
The expansion of HABs and associated diseases is due to complex interactions between a
variety of climatic, environmental, and biological factors. Blooms occur due to particular
combinations of physical conditions (temperature, sunlight, water stratification, circulation)
and chemical conditions (levels of oxygen and specific nutrients). Species dispersal due to
large-scale water movements is important, and unusual currents and storms may account for
appearance of atypical blooms. For example, in some recent years, the K. brevis red tide off
Florida has moved much further north because of formation of unusual circulation patterns
in the Gulf Stream. The occurrence of some HABs has been closely correlated with the El
Niño Southern Oscillation phenomenon. The increased incidence of HABs is often cited as
evidence of disturbance to our oceans and atmosphere due to increased carbon dioxide levels.

One explanation for the increased geographic distribution of some HABs is transport of
the causative organisms via ship movements. Modern large vessels pick up many thousands
of liters of water as ballast, which can be transported to completely different geographic
regions. Several studies have shown that the cysts of toxic dinoflagellates and other harmful
“alien” organisms can be transported in this way. For example, the introduction of PSP into
southern Australia in the 1980s is suspected to be due to transport from Japan and Korea. As
a consequence, regulations of the International Maritime Organization require ships to have
programs for management of ballast water—such as offshore exchange of ballast water, dis-
infection of the water before it is discharged, or transfer to a shore-based disinfection facility.

Nutrient enrichment of coastal waters is often linked to the increased incidence of HABs.
Upwelling of nutrients into cold oligotrophic waters is responsible for many natural blooms,
such as those that occur regularly off the California coast. Such natural phenomena may now be
overlaid with eutrophication from anthropogenic sources of excess nutrients, and some studies
have provided clear evidence for this. Although there are few long-term studies and it is difficult
to compare the results from different areas, it seems likely that a significant increase in plankton
growth can result from nutrient input—from sewage and runoff of agricultural fertilizers—into
coastal waters and estuaries with limited exchange with the open ocean. For example, regular
blooms of non-toxic microalgae such as Phaeocystis and Emiliania have become a regular
occurrence in the North Sea and English Channel (see Figure 6.8A). Could such eutrophication
also lead to an increase in toxic species? One widely held idea is that nutrient inputs from sew-
age or agricultural land runoff alter the ratio of particular nutrients, as well as the total loading,
and that this may change the balance of the different plankton groups. Sewage is rich in nitro-
gen and phosphorus but has a low silicon content. Since diatoms specifically require silicon,
this could favor the selective growth of dinoflagellates. The emergence of dinoflagellate blooms
as a problem in the North Carolina and Maryland estuaries was linked to increased nutrient
levels due to an expansion in poultry and pig farming on the surrounding land. The dinoflagel-
late Pfiesteria piscicida was implicated as the cause of fish mortalities and was also linked to
neuropsychological effects in humans (although this was not proven, and the true nature of the
problems remains something of a mystery). The expansion of mariculture, such as high-density
salmon or shrimp culture, has been directly implicated in the increased frequency of HABs due
to nutrient enrichment from uneaten food and excreta. For example, the increased incidence of
Alexandrium blooms and high levels of PSP toxins in shellfish in Scotland during the 1990s
was due to poorly sited salmon pens in lochs, bays, and inlets with limited water exchange.

Another obvious explanation for the apparent increase in HABs and associated diseases is
increased awareness of toxic species and more extensive and effective monitoring, which is
exacerbated by the increased use of coastal waters for aquaculture of shellfish and finfish.
These activities act as sensitive indicators of potential problems and undoubtedly account for
at least part of the apparent spread of PSP and DSP. Mass mortalities in fish farms can occur
due to HABs, although these do not always involve toxic species.
 Marine Microbes as Agents of Human Disease 351

Increasing incidence of HABs—both toxic and non-toxic—also pose considerable threats to

seawater desalination facilities. These use a process of reverse osmosis which is effective at
removing most toxins but can be overwhelmed if toxin concentrations are very high. Blooms ? IS IT POSSIBLE TO
DEVELOP REAL-TIME
HAB FORECASTS?
also clog filters or cause biofouling of osmosis membranes.
The ability to provide early
warning of the development of
Coastal waters must be regularly monitored specific HABs would be of great
to assess the development of HABs value to managers of fisheries and
conservation programs, and for
Regular surveys by microscopic examination of the dynamics of phytoplankton populations residents and tourists in coastal
within an area may give advance warning of an increase in particular species, which may zones. Computer modeling based
sometimes precede a toxic bloom. Such surveys are time-consuming, and it is often difficult on environmental factors, such as
to distinguish toxic species or strains using morphological criteria. Some improvements can water temperature and salinity,
be achieved using flow cytometry or epifluorescence microscopy after labeling the target river flow, winds, and tides, and
biological factors such as abun-
species with specific fluorescent antibodies. Unfortunately, background fluorescence is often
dance and behavior of toxic dino-
a problem. Microphotography and automated image analysis of algal cells that are identified
flagellates have been developed by
by artificial intelligence systems are also being used to identify particular types of algae. NOAA state authorities in the USA
Remote sensing can be employed by equipping satellites with spectral scanners that detect to provide weekly prediction of
pigments in surface waters, and it is becoming possible to identify specific types of algae Karenia and Pseudo-nitzschia HABs
based on their spectral signature. When coupled with physical measurements such as sea in California and Florida. In the
surface temperatures and current flows, satellite images are especially useful in tracking the near future, it is likely that auto-
development and movement of blooms. As genetic sequence data for toxic dinoflagellates matic sensing devices using gene
and diatoms become available, gene probes are being increasingly used. A common method probe technology could be placed
is to amplify microalgal DNA encoding 18S rRNA using eukaryote-specific primers. The on offshore buoys, and information
could be beamed to satellites and
resulting PCR products can be cloned and sequenced, leading to a specific oligonucleotide
integrated with improved remote
probe that will hybridize with DNA of the organism in water samples to provide quantitative
sensing signal-detection systems
estimates of density of particular HAB species using microarray or qPCR technology. linked to powerful computer
models leading to reliable real-time
Surveys can be used to limit the economic and health impacts of HABs, but whether we can forecasts.
use such information to control their development or spread is questionable, since HABs
often occur over huge areas. The effects of weather, ocean currents, and the many physi-
cal and biological factors that determine the development and eventual demise of a bloom
are unpredictable. Chemical treatments, such as adding agents that promote agglutination
and sinking of microalgal cells, can be applied locally to control blooms, but the ecological
impact of such intervention on a large scale needs careful assessment. Flocculating clays are
used widely in Japan and China to control blooms around fish farms. Algicidal bacteria and
viruses capable of initiating lysis of dinoflagellates and diatoms have been isolated from sea-
water and investigated as possible biological control agents. Also, some microalgae have been
shown to contain lysogenic viruses, which can be induced into the lytic cycle by particular
conditions. Some success in initiating the collapse of bloom populations has been achieved
in microcosm experiments, but much more research and evaluation of the ecological and
“scale-up” issues is needed before biological control becomes a practical proposition. Such
biotechnological approaches to the control of HABs are considered further in Chapter 14.

Conclusions
The most important human diseases associated with autochthonous marine bacteria are
infections caused by bacteria in the genus Vibrio. These are adapted to a life in the aquatic
environment but possess complex regulatory systems coordinating the expression of viru-
lence factors when they transfer to the human host. Vibrio spp. show evidence of extensive
gene transfer that explains the emergence of new pathogenic variants. Numerous intoxica-
tions are caused by dinoflagellates and diatoms. Factors that affect plankton blooms and the
production of toxins harmful to humans are numerous, complex, and poorly understood. A
combination of natural and anthropogenic effects is undoubtedly responsible for a worldwide
increase in HABs, a wider range of toxic species and the emergence of “new” human dis-
eases. Climate change is increasing the threat of disease in the human population. Bacterial
pathogens show increased abundance in seawater at higher temperatures and people are
more likely to become infected following coastal flooding storms or hurricanes, whilst the
increased intensity of HABs and their spread to new regions can be linked to climate change.
352 Chapter 12

References and further reading

Diseases caused by Vibrio spp. IPCC (2018) Special Report: Global Warming of 1.5°C above pre-indus-
trial levels. www.ipcc.ch/sr15/. Accessed 10 Jan 2019.
Almagro-Moreno, S. & Taylor, R.K. (2013) Cholera: Environmental res-
ervoirs and impact on disease transmission. Microbiol. Spectrum 1: 1–19. Le Roux, F. & Blokesch, M. (2018) Eco-evolutionary dynamics linked to
horizontal gene transfer in vibrios. Annu. Rev. Microbiol. 72: 89–110.
Baker-Austin, C. & Oliver, J.D. (2018) Vibrio vulnificus: New insights into
a deadly opportunistic pathogen. Environ. Microbiol. 20: 423–430. Lekshmi, N., Joseph, I., Ramamurthy, T., & Thomas, S. (2018) Changing
facades of Vibrio cholerae: An enigma in the epidemiology of cholera.
Baker-Austin, C., Trinanes, J., Gonzalez-Escalona, N., & Martinez- Indian J. Med. Res. 147: 133–141.
Urtaza, J. (2017) Non-cholera vibrios: The microbial barometer of climate
change. Trends Microbiol. 25: 76–84. Lipp, E.K., Huq, A., & Colwell, R.R. (2002) Effects of global climate
on infectious disease: The cholera model. Clin. Microbiol. Rev. 15:
Baker-Austin, C., Trinanes, J.A., Taylor, N.G.H. et al. (2012) Emerging 757–770.
Vibrio risk at high latitudes in response to ocean warming. Nat. Clim.
Change 3: 73–77. Nelson, E.J., Harris, J.B. Morrisa Jr, J.G. et al. (2009) Cholera transmis-
sion: The host, pathogen and bacteriophage dynamic. Nat. Rev. Microbiol.
Baracchini, T., King, A.A., Bouma, M.J. et al. (2017) Seasonality in chol- 7: 693–702.
era dynamics: A rainfall-driven model explains the wide range of patterns
in endemic areas. Adv. Water Resour. 108: 357–366. Okada, K., Iida, T., Kita-Tsukamoto, K., & Honda, T. (2005) Vibrios com-
monly possess two chromosomes. J. Bacteriol. 187: 752–757.
Borgeaud, S., Metzger, L.C., Scrignari, T., & Blokesch, M. (2015) The
type VI secretion system of Vibrio cholerae fosters horizontal gene trans- Phillips, K.E. & Satchell, K.J.F. (2017) Vibrio vulnificus: From oyster
fer. Science 347: 63–67. colonist to human pathogen. PLOS Pathog. 13: e1006053.

Chourashi, R., Das, S., Dhar, D. et al. (2018) Chitin-induced T6SS in Vibrio Roig, F.J., González-Candelas, F., Sanjuán, E. et al. (2018) Phylogeny of
cholerae is dependent on ChiS activation. Microbiology 164: 751–763. Vibrio vulnificus from the analysis of the core-genome: Implications for
intra-species taxonomy. Front. Microbiol. 8: 2613.
Church, S.R., Lux, T., Baker-Austin, C. et al. (2016) Vibrio vulnificus
type 6 secretion system 1 contains anti-bacterial properties. PLoS One Sakib, S.N., Reddi, G., & Almagro-Moreno, S. (2018) Environmental role
11: e0165500. of pathogenic traits in Vibrio cholerae. J. Bacteriol. 200: e00795-17.

Cianfanelli, F.R., Monlezun, L., & Coulthurst, S.J. (2016) Aim, load, Semenza, J.C., Trinanes, J., Lohr, W. et al. (2017) Environmental suitabil-
fire: The type VI secretion system, a bacterial nanoweapon. Trends ity of Vibrio infections in a warming climate: An early warning system.
Microbiol.24: 51–62. Environ. Health Perspect. 125: 107004.

Colwell, R.R., Huq, A., Islam, M.S. et al. (2003) Reduction of cholera Shapiro, B.J., Levade, I., Kovacikova, G. et al. (2016) Origins of pan-
in Bangladeshi villages by simple filtration. Proc. Natl. Acad. Sci. 100: demic Vibrio cholerae from environmental gene pools. Nat. Microbiol.
1051–1055. 2: 16240.

Durai, S., Pandian, S.K., & Balamurugan, K. (2011) Changes in Vezzulli, L., Brettar, I., Pezzati, E. et al. (2012) Long-term effects of ocean
Caenorhabditis elegans exposed to Vibrio parahaemolyticus. J. warming on the prokaryotic community: Evidence from the vibrios. ISME
Microbiol. Biotechnol. 21: 1026–1035. J. 6: 21–30.

Ellison, C.K., Dalia, T.N., Ceballos, A.V. et al. (2018) Retraction of DNA- Vezzulli, L., Grande, C., Reid, P.C. et al. (2016) Climate influence on
bound type IV competence pili initiates DNA uptake during natural trans- Vibrio and associated human diseases during the past half-century in the
formation in Vibrio cholerae. Nat. Microbiol. 3: 773–780. coastal North Atlantic. Proc. Natl. Acad. Sci. USA 113: E5062–71.

Faruque, S.M. & Mekanolos, J.M. (2012) Phage-bacterial interactions in Wagley, S., Borne, R., Harrison, J. et al. (2018) Galleria mellonella as
the evolution of Vibrio cholerae. Virulence 3: 556–565. an infection model to investigate virulence of Vibrio parahaemolyticus.
Virulence 9: 197–207.
Ghenem, L., Elhadi, N., Alzahrani, F., & Nishibuchi, M. (2017) Vibrio
parahaemolyticus: A review on distribution, pathogenesis, virulence WHO (2017) Cholera vaccines: WHO position paper – August 2017. Wkly.
determinants and epidemiology. Saudi J. Med. Med. Sci.93: 93–103. Epidemiol. Rec. 92: 477–500.

Hartnell, R.E., Stockley, L., Keay, W. et al. (2019) A pan-European ring Zoonoses from marine animals
trial to validate an International Standard for detection of Vibrio cholerae,
Vibrio parahaemolyticus and Vibrio vulnificus in seafoods. Int. J. Food Gauthier, D.T. (2014) Bacterial zoonoses of fishes: A review and appraisal
Microbiol. 288: 58–65. of evidence for linkages between fish and human infections. Vet. Rec. 203:
27–35.
Heng, S.-P., Letchumanan, V., Deng, C.-Y. et al. (2017) Vibrio vulnificus:
An environmental and clinical burden. Front. Microbiol. 8: 997. Haenen, O.L., Evans, J.J., & Berthe, F. (2013) Bacterial infections from
aquatic species: Potential for and prevention of contact zoonoses. Rev. Sci.
Hsueh, P.-R., Lin, C.-Y., Tang, H.-J. et al. (2004) Vibrio vulnificus in
Tech. (Int. Off. Epizoot.) 32: 497–507.
Taiwan. Emerg. Infect. Dis. 10: 1363–1368.
Hunt, T.D., Ziccardi, M.H., Gulland, F.M. et al. (2008) Health risks for
Huq, A., Yunus, M., Sohel, S.S. et al. (2010) Simple sari cloth filtration
marine mammal workers. Dis. Aquat. Org. 81: 81–92.
of water is sustainable and continues to protect villagers from cholera in
Matlab, Bangladesh. MBio 1: e00034-10.
 Marine Microbes as Agents of Human Disease 353

Waltzek, T.B., Cortés‐Hinojosa, G., Wellehan Jr, J.F.X., & Gray, G.C. Jal, S. & Khora, S.S. (2015) An overview on the origin and production of
(2012) Marine mammal zoonoses: A review of disease manifestations. tetrodotoxin, a potent neurotoxin. J. Appl. Microbiol. 119: 907–916.
Zoonoses Pub. Health 59: 521–535. Magarlamov, T.Y., Melnikova, D.I., & Chernyshev, A.V. (2017)
Tetrodotoxin-producing bacteria: Detection, distribution and migration of
Bacterial and algal toxins the toxin in aquatic systems. Toxins 9: 166.
Assunção, J., Guedes, A.C., & Malcata, F.X. (2017) Biotechnological and McLauchlin, J., Little, C.L., Grant, K.A., & Mithani, V. (2006)
pharmacological applications of biotoxins and other bioactive molecules Scombrotoxic fish poisoning. J. Pub. Health 28: 61–62.
from dinoflagellates. Mar. Drugs 15:.393.
Rongo, T., Bush, M., & van Woesik, R. (2009) Did ciguatera prompt the late
Dorantes-Aranda, J.J., Campbell, K., Bradbury, A. et al. (2017) Holocene Polynesian voyages of discovery? J. Biogeogr. 36: 1423–1432.
Comparative performance of four immunological test kits for the detec-
tion of Paralytic Shellfish Toxins in Tasmanian shellfish. Toxicon 125: Ecology and Detection of Harmful Algal Blooms
110–119.
Gilbert, P.M., Berdalet, E., Burford, M.A. et al. (2018) Global Ecology
EFSA Panel on Contaminants in the Food Chain, Knutsen, H.K., and Oceanography of Harmful Algal Blooms. Springer Nature.
Alexander, J., Barregard, L. et al. (2017) Risks for public health related to
the presence of tetrodotoxin (TTX) and TTX analogues in marine bivalves McPartlin, D.A., Loftus, J.H., Crawley, A.S., Silke, J. et al. (2017)
and gastropods. EFSA J. 15: e4752. Biosensors for the monitoring of harmful algal blooms. Curr. Opin.
Biotechnol. 45: 164–169.
Feng, C., Teuber, S., & Gershwin, M.E. (2016) Histamine (scombroid)
fish poisoning: A comprehensive review. Clin. Rev. Allergy Immunol. 50: Ottesen, E.A. (2016) Probing the living ocean with ecogenomic sensors.
64–69. Curr. Opin. Microbiol. 31: 132–139.
Chapter 13
Microbial Aspects of
Marine Biofouling,
Biodeterioration,
and Pollution

The first section of this chapter discusses the economically important detrimental effects
of marine microbes through biofouling of marine surfaces and structures, the biodeteriora-
tion of materials, and the spoilage of seafood. The second section considers the microbial
diseases arising from the pollution of the sea by sewage and wastewater, together with meth-
ods of monitoring water and shellfish using microbial indicators and new approaches for
detection of pathogens. In the final section, the role of marine microbes in biodegradation
and bioremediation of oil and other chemical pollutants is discussed. The chapter concludes
with discussion of the roles that microbes play in two areas of growing concern, namely the
mobilization of toxic mercury and other pollutants into marine food webs, and plastic pollu-
tion of the oceans.

Key Concepts
• Microbial colonization often provides the first stage in biofouling of surfaces in the
sea, leading to economic losses through interference with the efficient operation of
ships and damage to structures such as piers and aquaculture facilities.
• Corrosion and deterioration of metal and wooden structures is initiated by microbes.
• Microbial activity results in harmful spoilage of fish and shellfish; it can also be used
to produce seafood products with altered desirable properties.
• Pathogenic microbes introduced to the sea via sewage and wastewater constitute a
health hazard through recreational use of coastal environments and contamination of
shellfish.
• Bacterial indicators are used for monitoring coastal waters for fecal pollution but may
be unreliable predictors of health risks, which are mainly associated with viruses.
• Many microbes degrade oil naturally and this may be enhanced using bioremediation.
• Microbial activity in sediments and pelagic aggregates leads to production of toxic
mercury compounds that can contaminate fish.
• Interactions of microbes with polluting microplastics may affect their transport and
persistence, with potentially major implications for ocean processes such as carbon
cycling.
356 Chapter 13

BIOFOULING AND BIODETERIORATION

Microbial biofilms initiate the process of biofouling
As previously discussed in Chapter 3, the surfaces of inanimate objects and living organ-
isms in the sea are colonized by mixed microbial communities that form biofilms showing
complex physical structures and chemical interactions. As shown in Figure 13.1, the process
of biofouling usually begins with the formation of a molecular conditioning film formed by
the deposition of organic matter on the surface within a few minutes of immersing a material
into seawater. The conditioning film is composed of many compounds including amino acids,
proteins, lipids, nucleic acids, and polysaccharides. This is followed over the next few hours
by the attachment of pioneer motile bacteria as primary colonizers, followed within a few
days by a diverse association of bacteria and benthic diatoms. This leads to a slimy biofilm
up to 500 μm thick and consolidated by the production of sticky exopolymers. Compounds
produced in the microbial biofilm act as cues for the settlement of other microbes, planktonic
algal spores, and invertebrate larvae. Colonization by some macrofouling organisms such as
barnacles can occur very rapidly.

The complex dense community that develops leads to the biofouling of all types of marine
surfaces, including coastal plants, macroalgae, animals, piers and jetties, oil-drilling rigs,
boat hulls, fishing gear, aquaculture cages, engineering materials, concrete and metal struc-
tures. (Figure 13.2). Other effects include blockage of pipes and filters, reduced efficiency of
heating and cooling plants, and interference with the efficient operation of boats and ships.

Sailing enthusiasts know the inconvenience and cost—in lost weekends—of scraping the
bottom of yachts, while the economic effects of biofouling of larger ships are immense, lead-
ing to extreme increases in fuel usage due to frictional drag. Formation of a microbial biofilm
alone can cause a 1–2% increase in drag. Subsequent colonization by macroalgae increases
this to about 10% and, if unchecked, extensive colonization by hard-shelled invertebrates
such as barnacles, tubeworms, bryozoans, or mussels can lead to 30–40% drag. This costs
shipping companies and navies throughout the world over $200 billion every year because of
losses in fuel efficiency and the cost of antifouling measures. Increased fuel usage also leads
to increased emissions of CO2 and sulfurous pollutants. In addition, ships and other floating
debris can transport non-indigenous species around the world.

Various methods for the prevention of macrofouling of ships’ hulls, such as the use of copper
sheeting, have been in use for hundreds of years. Self-polishing copolymer paints contain-
ing copper or tin compounds have been used extensively in the second half of the twentieth
century. Tributyl tin (TBT) is particularly effective at controlling the problem, but its use has
been banned since 2008 by the International Maritime Organization, following the recogni-
tion that it has serious environmental consequences, with particular effects on the ecology
and reproductive behavior of marine invertebrates and the immunity of marine mammals
(see p.323). There is, therefore, an active search for effective “environment-friendly” alterna-
tives. Ideally, products isolated from marine organisms should have low toxicity, be effective
at low concentrations, and break down quickly if released into the environment. Hundreds of

Figure 13.1 Schematic represen-

tation of stages in the biofouling
process, showing the critical role of
microbial activities. Soft biofouling
organisms include algae, bryozoans,
tunicates, anemones, soft corals, and
sponges. Hard biofouling organ-
isms include barnacles, mussels, and
tubeworms. Modified from Martín-
Rodríguez et al., 2015, CC BY 4.0.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 357

Figure 13.2 Examples of biofoul-

ing. (a) Polythene mesh immersed
in North Atlantic water shows
extensive colonization by hydroids
(background), whereas copper alloy
mesh (foreground) shows no biofoul-
ing. (b) Biofouling on a cast-iron
statue in Antony Gormley’s “Another
Place” installation on Crosby Beach,
Merseyside, England—shown four
years after erection in the intertidal
zone. Credits: A. Environment,
B. Rept0n1x; CC BY-SA 3.0 via
Wikimedia Commons.

molecules with antifouling properties have been investigated, with the richest sources being
sessile marine organisms—especially sponges, corals, seaweeds, and ascidians—as these
often possess mechanisms to avoid overgrowth by other organisms. As shown in Figure 13.1,
there are various critical points at which the process of biofouling may be inhibited. From
a microbiological perspective, our interest is in the prevention of the initial formation of a
biofilm. One example of a successful development in this field is the discovery of bromi-
nated furanone compounds produced by the seaweed Delisea pulchra, which prevent biofilm
formation by interfering with quorum sensing. A range of other such “quorum quenching”
compounds are now being investigated,

The surfaces of some species of algae are colonized by bacteria that seem to interfere with
cues for settlement of invertebrate larvae or spores of other algae. Bacteria belonging to the BACTERIA CAN
genera Pseudoalteromonas and Phaeobacter, isolated from bacterial biofilms on the surface
of algae, produce a range of inhibitory compounds that have potential applications in anti-
i PRODUCE DEADLY
GAS IN OIL
fouling treatments. A number of coatings for plastic surfaces have also been developed by PRODUCTION
mixing extracts of marine isolates of Bacillus, Pseudomonas, and Streptomyces with water-
The activity of sulfate-reducing
based resins. These “living paints” have shown promise in experimental trials, but problems
bacteria (SRB) in oil and gas
with formulation and delivery of active compounds means that laboratory assays are often reserves can lead to high levels
not confirmed in field trials. of sulfide that cause “souring”
of the oil that cause problems in
Biofouling is also a major problem in desalination plants, in which seawater is treated by refining. More importantly, there
reverse osmosis. Microbial biofilms on filters decreases the flow of water and efficiency of have been numerous examples of
salt extraction. They also cause reduced efficiency of heat exchangers in coastal powerplants. leaks of toxic levels of H2S gas on
offshore drilling platforms that have
It is important to note that while the negative aspects of these processes are considered here resulted in fatalities to oil-rig work-
because of their deleterious consequences, microbial biofilms also play a very important, ers. H2S is toxic at very low levels.
Trace amounts of the gas are easily
positive role. Microbial products may stimulate the settlement and metamorphosis of algae
detected by its characteristic “rot-
and invertebrate larvae, which is very important in marine ecology (e.g. in reef formation)
ten eggs” smell, but it paralyzes the
and aquaculture (e.g. in settlement of mussels in suspended rope culture). olfactory nerve at levels above 30
ppm—so a worker may smell the
gas, investigate its source, and then
Microbes induce corrosion of metals, be unable to smell higher concen-
alloys, and composite materials trations. At low levels, H2S causes
headaches, blurred vision, and con-
A range of aerobic and anaerobic bacteria are responsible for microbially induced corro- junctivitis. It can be instantly fatal at
sion (MIC), causing pitting and fracture stresses of metal structures, vessels, equipment, and high levels, and even a few minutes
instruments in the marine environment. Corrosion occurs due to attached bacteria in bio- exposure to 500–1000 ppm
films at the interface of the metal and seawater. Anaerobic sulfate-reducing bacteria (SRB) can cause long-term illness.
have long been known to be the main cause of marine corrosion, but we now recognize the Modern rigs employ a variety of
importance of complex multi-species mixed biofilms containing microbes that carry out a sensors to detect escaping gas and
range of biochemical and electrochemical processes synergistically. Where surfaces are open workers receive special training in
to oxygenated seawater, SRB can occupy the lower levels of a biofilm because oxygen is dealing with leaks, but in some oil-
well “blowouts” drilling companies
depleted by aerobic microbes in the upper layers. Many other species of bacteria and fungi
have been forced to set fire to rigs
can produce organic acids by fermentation that corrode steel, manganee and zinc alloys. to burn off the escaping H2S.
MIC is a particular problem for the offshore oil and gas industry because it causes corrosion
358 Chapter 13

of carbon steel pipelines, drilling platforms and equipment and also leads to the “souring”
SHIPWORM USES
i BACTERIAL ENZYMES
TO FEED ON WOOD
of crude oil by the production of H2S, with significant health and safety implications. Other
bacteria can attack corrosion-resistant manganese or iron oxide films on the surface of iron
and steel. Thermophilic sulfate-reducing archaea also occur in marine hydrothermal systems
Teredinibacter turnerae is the only and oil reservoirs.
member of the endosymbiotic
consortium of microbes in the ship- Control of biofilms is particularly difficult, because they are protected from the effects of
worm Bankia spp. that has been many chemical treatments and repeated high concentrations of biocides—such as glutar-
cultured. The genome sequence aldehyde, quaternary ammonium, and phosphonium compounds—are needed. Corrosion
reveals that the bacterium contains
of steel structures can be limited using cathodic protection, but this can lead to structural
genes for a range of carbohydrate-
weakness unless the electrical potential is carefully applied and monitored; formation of
degrading enzymes that specialize
in the breakdown of wood as well excess hydrogen can cause metal fatigue. Routine monitoring by measuring levels of cultur-
as genes for nitrogen fixation (Yang able acid-producing bacteria together with the application of biocides to fluids injected into
et al., 2009). Stable isotope analysis the well can help to control the problem. Testing for SRB is more problematic because of the
shows that almost all of the carbon need for anaerobic culture, and development of molecular-based test kits will prove useful
and up to half of the nitrogen in to the industry. Problems with the use of cathodic protection and environmental concerns
the shipworm is derived from the about large-scale use of biocides at sea favor a biotechnological solution to the control of
symbiont metabolism (Charles corrosive biofilms. Nitrate is sometimes added to injection fluids pumped into oil reservoirs
et al., 2018). O’Connor et al. (2014) during extraction, because nitrate-reducing bacteria may compete with SRB. However, this
made the surprising discovery that
process needs careful monitoring because if nitrate enters the pipelines, nitrate reduction can
the digestive bacteria live in bac-
be coupled with iron oxidation, leading to corrosion.
teriocytes in the gills, rather than
the gut. Wood-degrading enzymes
are secreted and transported—by A special type of MIC occurs in the case of wrought iron or steel structures in deep water.
mechanisms that are currently Formations of rust aggregate into bioconcretious structures similar to a stalactite or ici-
unknown—to the caecum, a cle. They were first observed and called “rusticles” during exploration of the wreck of the
part of the gut which is free of a Titanic 76 years after it sank in 1912 to a depth of 3.8 km off the coast of Newfoundland
resident microbiota. O’Connor and (Figure 13.3). Rusticles consist of iron oxides, carbonates, and hydroxides colonized by a
colleagues conclude that separa- complex syntrophic community of bacteria and fungi that corrode the metal to produce the
tion of the symbionts from the site iron compounds. This includes a novel species named Halomonas titanicae. The meter-
of wood digestion eliminates com- length fragile structures are permeated by water channels that allow water to flow through
petition with the host for soluble
the structure and are expected to result in complete destruction of the wreck by 2030. Such
nutrients and enables the host
formations have subsequently been described on other deep-water wrecks and oil platform
to have greater control over the
process. Identification of the small mooring chains.
number of key bacterial enzymes
transported to the gut opens
up the possibility of developing Microbes cause biodeterioration of timber
efficient biotechnological digestion and marine wooden structures
of lignocellulose—a major goal for
the production of biofuels. Microbial colonization and decomposition of timber transported or stored at sea, or of wooden
structures—such as wharves, jetties, piers, and boats—causes an immense amount of dam-
age costing many billions of dollars. The main cause of damage is penetration by wood-
boring invertebrates, of which the most important are the shipworms, so called because of the

Figure 13.3 Microbially induced

rusticles on the steel hull of the
Titanic. The rusticles pass through
a cycle of growth, maturation and
then fall away, over a probable 5–10-
year cycle. Credit: Lori Johnston, RMS
Titanic Expedition 2003, NOAA-OE.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 359

damage caused to wooden boats throughout history. These are not worms, but small bivalves
SHIPWRECKED
with long stretched bodies—sometimes up to 60 cm long and ~ 1 cm wide—in the family
Teredinidae, which contains over 70 species. They tunnel into wooden structures using their i SHIPS WRECKED
serrated shell valves and secrete a calcium carbonate rich coating that protects the delicate The preservation of submerged
tissue from crushing (Figure 13.4). As previously mentioned in Chapter 11, shipworms and archaeological timbers is a special-
other marine invertebrates that digest wood rely on an obligate symbiotic association with ized process. Waterlogged wood
wood-degrading microbes. In one species of shipworm, Bankia setacea, the gammaproteo- is protected when buried in anoxic
bacterium Teredinibacter turnerae has been identified as a symbiont of a specialized region sediments, but fungi and bacteria
decay it rapidly once it is brought
of the host gill. Wood is also degraded by small isopod crustaceans known as gribbles, such
to the surface. The Swedish navy
as Limnoria lignorum and related species that break down the surface of wood with cut-
ship Vasa, which sank in Stockholm
ting and grinding mouthparts. (Although notorious for their destructive role, shipworms and harbor on its maiden voyage in
gribbles have an important ecological role due to the decay of fallen wood in coastal areas 1628, was discovered 333 years
such as mangroves.) later in remarkably well-preserved
condition because the hypoxic,
Most types of wood used throughout history are susceptible to damage by wood borers, low salinity conditions in the
although shipbuilders have traditionally used the denser heartwood as it has some resistance. Baltic Sea prevented colonization
A few kinds of tropical hardwoods have high resistance to wood borers, but their use is by shipworms and degradative
unsustainable. The use of various preservative wood treatments delays deterioration, but microbes. When the vessel was
these are expensive and cannot be used on new timber during transport. Treatments such as raised, impregnation with polyeth-
ylene glycol was used to fill voids
creosote and chromated copper arsenate cause considerable environmental damage and their
in the wood structure. Despite
use is now tightly regulated in many countries. Furfurylation is an environmentally-friendly this treatment, it was realized after
process in which wood is modified by impregnation with furfuryl alcohol (C5H6O2)—made ~20 years that unexpected decay
from plant waste such as sugar cane or corn cobs—combined with heat and pressure treat- was occurring after all. During its
ment. This improves mechanical strength and resistance to biodeterioration. Identification of long period in anoxic conditions
the symbionts and their mode of transmission provide a potential weak link in the process of on the seabed—favored by sewage
biodeterioration, as it may be possible to identify compounds that inhibit bacterial coloniza- dumping in the harbor—sulfate
tion or metabolism. reduction by bacteria led to accu-
mulation of > 5 tonnes of sulfur in
the timbers. Iron nails in the struc-
Microbial growth and metabolism cause ture catalyzed an oxidative reaction
forming sulfuric acid that is now
spoilage of seafood products hydrolyzing cellulose in the timber,
Fish and shellfish deteriorate very rapidly as a result of microbial activities, resulting in the reducing its stability (Sandström
rapid production of discoloration, slime, and unpleasant odors and flavors. The composition of et al., 2002). The same problem
microbial communities in freshly caught seafood varies considerably according to the animal has occurred with other recovered
shipwrecks such as the Mary Rose
species and the methods of fishing, handling, processing, and storage. Natural transformation
and Batavia.
of compounds within the fish tissue begins soon after death owing to autolytic processes,
making catabolites available for decomposition by members of the bacteria that are normal
inhabitants of the gut and skin of fish. Specific spoilage organisms include Pseudomonas,
Shewanella, Alteromonas, Moraxella, Acinetobacter, Cytophaga, and Flavobacteriumpolio.
Psychrophilic strains of Pseudomonas and Shewanella become the dominant spoilage organ-
isms when fish is chilled on ice. Seafood is rich in poly-unsaturated fatty acids, which are
prone to lipid oxidation leading to rapid loss of flavor and nutritional quality. Specific spoil-
age organisms are most often associated with the production of ammonia, amines, ketones,
organic acids, and sulfur compounds that characterize “off” seafood. Trimethylamine is the
most important of these metabolites, produced by reduction of trimethylamine oxide, natu-
rally present in the tissue of many fish.

Figure 13.4 (a) Driftwood show-

ing extensive boring damage by
shipworm. (b) Teredo shipworm
extracted from mangrove wood,
Brazil. The body is ~0.5 m long.
Credits: A. Michael C. Rygel. B.
Depleswsk. CC-BY-SA-3.0 via
Wikimedia Commons.
360 Chapter 13

Bacteria grow rapidly to high levels (over 108 CFU per gram) in the nutrient-rich environment
of fish tissue, which contains high concentrations of readily utilizable substrates such as free
amino acids. Metabolic consortia develop in the mixed microbial community; for example,
lactic acid bacteria (e.g. Lactobacillus and Carnobacterium) degrade the amino acid arginine
to ornithine, which is further degraded to putrescine by enterobacteria. The production of
siderophores is important for the acquisition of iron, since fish tissue contains limiting con-
centrations (see p.312). Lactic acid bacteria characteristically produce bacteriocins, which are
highly specific membrane-active peptide antibiotics that inhibit certain other bacteria.

Most fish spoilage organisms are easily cultured and identified using standard microbiologi-
cal methods, although there has been some recent use of molecular methods for characteriza-
tion and early detection of spoilage organisms (e.g. gene probes for Shewanella putrefaciens)
and chemical assays for total volatile basic nitrogen after extraction with trichloroacetic or
perchloric acid. The growth kinetics of specific spoilage organisms can be determined by
absorbance measurements in culture and used to construct mathematical models to predict
rates of spoilage and shelf-life of fish products.

Processing, packaging, and inhibitors of

spoilage are used to extend shelf-life
The food industry has devised various methods of extending shelf-life in “value-added”
products by inhibiting or delaying spoilage, but many processes used with other foods are
unsuitable because the texture and flavor of fish products is easily destroyed by processing,
leading to reduced consumer acceptance. Many processes like salting, pickling, and smok-
ing are based on traditional methods, while recent innovations include modified atmosphere
packaging (Figure 13.5) in which products are packaged in a gas mixture containing low
levels of oxygen and high levels of carbon dioxide and nitrogen. The composition of the gas
mixture and the permeability of the plastic film used to wrap the product is optimized for
specific food types. This can shift the ecology of the contaminating microbiota; for example,
while CO2 packaging of fresh fish in ice is highly effective, it can suppress the growth of
respiratory bacteria so that fermentative Photobacterium, lactic acid bacteria, and enterobac-
teria become dominant. Spore-forming bacteria (Bacillus and Clostridium) can survive mild
heat treatment (pasteurization) of vacuum-packed products. As well as spoilage, some fish
products are the source of human pathogens. Fish- and shellfish-associated bacterial, viral,
and toxic diseases were considered in Chapter 12, but other pathogens such as Salmonella
spp., Staphylococcus aureus, and Listeria monocytogenes may be introduced from human
sources during handling and processing. Listeria is of particular concern in lightly preserved
ready-to-eat products such as cold-smoked fish and shellfish because of its ability to grow
over a very wide temperature range and salinity. Control of these pathogens is an important
requirement for commercial processors and modified atmosphere packaging is widely used
to limit health risks.

Besides improvements in packaging methods, antioxidants and chemical inhibitors of

microbial growth may be employed. Antibiotics such as tetracyclines were once added to
ice on board fishing vessels and in markets to prevent spoilage of fresh fish. This practice
is now prohibited because of concerns over antibiotic residues and resistance. However,
bacteriocins from lactic acid bacteria are generally regarded as safe and are permitted in

Figure 13.5 Fish preservation,

old and new. (a) Fishwives packing
herring into barrels for salting, circa
1930. (b) A factory in Vietnam for
processing fish reared by aquaculture,
showing preparation under high
levels of hygiene. Credits: (a) North
Shields Library Services; (b) AccuDB.
com.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 361

foods (e.g. nisin is widely used in dairy products). With fish, addition of purified bacte-

?
riocins has produced some promising results, as has the addition of non-spoilage lactic HOW FRESH IS
acid bacteria as competitors of pathogens or spoilage organisms. It has been shown that THAT FISH?
quorum-sensing signaling molecules (see p.102) are produced during development of the Wise consumers who purchase
microbial spoilage community, and interference with the quorum-sensing mechanism fresh fish know how to tell if fish
might offer a potential new method of control. Some natural preservatives such as plant is fresh and good to eat, based on
extracts, essential oils, and bioactive peptides may be effective because of this quorum appearance and smell. However,
quenching effect. it is more difficult to assess
prepacked fish bought from a
supermarket, and “best before”
Some seafood products are made by deliberate and “use by” dates can be very
unreliable, leading to excessive
manipulation of microbial activities wastage for retailers as they neces-
The spoilage activities of microbes in fish are generally regarded as detrimental but use sarily err on the side of quality and
safety. One novel solution to this
of fermented fish sauces have a long history, for example, the garum of the ancient Roman
problem is the development of
Empire. In some parts of the world, there are numerous ethnic food products that depend “smart packaging” incorporating
on microbial activities for preservation and flavors. Most of these are encountered in Asian real-time sensors. For fish, the most
countries, especially Indonesia, Thailand, the Philippines, and Japan. Many processes are promising approach is to detect
conducted according to traditional recipes, but the microbiology of some has been inves- the production of total volatile
tigated during commercialization of production for export, and pure starter cultures may basic nitrogen (ammonia and vola-
sometimes now be used. Perhaps the best-known product is nam-pla (fish sauce), which is tile amines, TVB-N) produced by
now widely used in the West due to the popularity of Thai cuisine. Nam-pla is made by microbial activity during spoilage.
fermentation of fish hydrolyzed by a high-salt concentration (15–20%), which encourages For example, pH-sensitive dyes
growth of the extreme halophile Halobacterium salinarum, leading to characteristic flavors encapsulated within a polymer and
covered by a gas-permeable film
and aromas. This appears to be the only food product that relies on activities of a member of
can change color in response to
the Archaea. Other high-salt products include som-fak, burong-isda, and jeikal, traditional accumulation of TVB-N (Pacquit
Korean foods made from fermented shrimp. Other fermented fish products such as plaa-som et al., 2006; Wells et al., 2019).
use lower salt concentrations; in these, a microbial flora dominated by lactic acid bacteria Other approaches include elec-
and yeasts leads to a characteristic aroma. There must be enough salt present (2–8%) and a tronic sensors and indicator labels
final pH of less than 4.5 to inhibit the growth of pathogens; nevertheless, contamination by that are activated during packag-
Staphylococcus can be a problem. Garlic is a major ingredient of some recipes; not only does ing; these measure temperature
this serve as a carbohydrate source, but it also has antibacterial activity (some constituents and time elapsed to indicate if the
of garlic inhibit quorum sensing). Ika-shiokara is made from squid and fish guts pickled in product is still fresh. Such sensors
2–30% salt in a process that depends on growth of the yeast Rhodotorula; if this sounds will have an important future in the
retail market if their use proves to
appealing, you will find it as a delicacy in Hokkaido, Japan. Nordic countries also have
be a reliable indicator of freshness
several traditional fermented fish dishes, of which the most famous is hákarl from Iceland.
in various fish products under dif-
This is made from Greenland shark, traditionally buried on the seashore for several weeks ferent conditions.
or months, during which urea is converted to ammonia by Moraxella and Acinetobacter
spp. before drying. Skate can also be processed in the same way. These foods were once an
essential winter staple source of protein and energy for Icelanders but have recently gained
a reputation as an item to be sampled as a dare by curious tourists, accompanied by shots of
Brennivín liquor.

MARINE POLLUTION BY SEWAGE AND WASTEWATER

Coastal pollution by wastewater is
a source of human disease
A large proportion of the world’s population lives near the coast. Many of the largest urban
settlements have grown up around river estuaries and natural harbors because of their impor-
tance for trade. As these towns and cities developed, it was an easy option to dispose of
untreated sewage directly into the rivers and sea; the adage “the solution for pollution is dilu-
tion” was applied. In many developed countries, awareness of the problems arising from dis-
posal in close proximity to the population led to long pipelines off the coast, but the grounds
for this were usually esthetic rather than health-related. In many countries, vast quantities of
untreated sewage are still disposed directly to sea. Even in highly developed countries, where
most medium-sized communities will have sewage treatment works, human waste containing
potential pathogens still finds its way into coastal waters from isolated dwellings, houseboats,
and marinas. Mixture of untreated sewage with storm-water runoff is a regular occurrence,
even in large cities with well-developed sewage systems.
362 Chapter 13

Since the mid-twentieth century, increased wealth and leisure time has led to greater use
REDUCING DANGERS
i FROM SEWAGE
POLLUTION
of the sea for recreational use, and a trip to the seaside has become an important feature
of many peoples’ lives. In the 1950s, awareness began to be focused on the potential haz-
ards of swimming in sewage-polluted water. Public awareness of the problem became
Every day, each human being acute because of the growing popularity of seawater bathing and sports such as surfing,
produces between 100 and 500 sail boarding, and diving. As discussed in Box 13.1, there are few well-documented large
g of feces and 1–1.5 L of urine, outbreaks of serious disease associated with recreational use of marine waters, but there
which are disposed via sewers, are many epidemiological studies showing health risks for swimmers in waters polluted by
together with water used for flush- wastewater. In most adults, these diseases are troublesome but not usually life-threatening;
ing and washing. Sewage treat-
however, young children, old people, and those with an impaired immune system are at risk
ment involves three stages, which
successively reduce the load of
of acquiring more serious infections. Coastal waters used for recreation contain a mixture of
biological and chemical contami- pathogenic, opportunistic and non-pathogenic microorganisms derived from sewage efflu-
nants. Primary treatment consists ents, bathers themselves, seabirds, and runoff from agricultural land contaminated by waste
of screening and separating solid from livestock, as well as autochthonous marine bacterial pathogens such as the vibrios
material, before a second stage in discussed in Chapter 12. Most of these pathogens are transmitted via the fecal–oral route
which microbial processes occur in and cause disease when swimmers unwittingly ingest seawater. Infection via the ears, eyes,
biofilms or flocs in trickling filters, nose, and upper respiratory tract and via open wounds may also occur. Aerosols may be a
activated sludge systems, or other significant route of infection for surfers, causing mild respiratory illness. The number of
types of bioreactors. Before dis- organisms required to initiate infection will depend on the specific pathogen, the conditions
charge, sewage is usually subject to
of exposure, and the susceptibility and immune status of the host. Viruses and parasitic pro-
tertiary treatment such as filtration
tozoa may require only a few viable units (perhaps 100 or less) to initiate infection, whereas
through sand or activated char-
coal, or coagulation with iron or most bacteria require large doses of thousands or millions. The types and numbers of vari-
aluminum salts. Tertiary treatment ous pathogens in sewage vary significantly according to the distribution of disease in the
results in approximately 107–fold population from which sewage is derived, as well as geographic and climatic factors. As a
reductions in fecal bacteria and up general rule, the more serious infections transmitted via the fecal–oral route—diseases such
to 104 –fold reductions in the num- as cholera, hepatitis, typhoid, and poliomyelitis—only become significant for swimming-
ber of viruses (Wang et al., 2018). associated illness if there is an epidemic in the local population. In this case, direct person-
Disinfection via ultraviolet light, to-person transmission is likely to be far more important; however, recall the special case of
chlorination, or ozone reduces cholera transmission discussed in Chapter 12.
pathogen levels further still, but
these are expensive options requir-
Although bacteria are used as indicators of sewage pollution in seawater, as discussed below,
ing complex engineering works.
Raw or partially treated sewage is
they are much less important than viruses as a source of infection. Pathogenic enteric bac-
still discharged in huge quantities teria include Salmonella, Shigella, and pathogenic strains of Escherichia coli. Dermatitis
into coastal waters in many parts of (“swimmers’ itch”) and infection of cuts and grazes can be caused by bacteria such as
the world. Staphylococcus, Pseudomonas, and Aeromonas and the yeast Candida (as well as autochtho-
nous Vibrio spp.). These organisms can also cause ear and eye infections.

Many of the pathogens in sewage effluent or land runoff also become associated with inter-
tidal sediments, which provide a long-term reservoir and health risk. Viruses are protected
by adsorption to sediment particles and bacteria may persist for long periods in a biofilm
matrix or enter the VBNC state (see p.95). Levels of pathogens in the water column are cor-
related with resuspension of sediment particles, which increases water turbidity and prolongs
survival.

Human viral pathogens occur in sewage-polluted seawater

More than 100 human enteric viruses belonging to three main families are transmitted via
the fecal–oral route. The Adenoviridae are nonenveloped, icosahedral, double-stranded
DNA viruses most commonly associated with infections of the upper respiratory tract, but
there are several adenovirus types that cause gastroenteritis, leading to diarrhea and vomit-
ing, primarily in children. The family Caliciviridae includes caliciviruses, astroviruses, and
noroviruses. These are single-stranded RNA viruses, with small round or hexagonal virions.
Norovirus, previously known as the Norwalk agent, is the most significant enteric virus in
developed countries, probably causing about half of all cases of gastroenteritis. Millions of
cases of infection by noroviruses occur each year through contamination of drinking water
and foods by infected persons; the disease is a major problem as a cause of local epidemics in
hospitals, hotels, holiday camps, and cruise ships. The infectious dose is thought to be very
low (perhaps 1–10 virions), as stomach acid does not inactivate the virus and immunity to the
virus is often incomplete —noroviruses also mutate rapidly leading to antigenic variation—
so they are a major cause of disease associated with swimming or consumption of seafood.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 363

The disease causes severe nausea, vomiting, and diarrhea, often accompanied by painful
abdominal cramps, headaches, and chills.

The Picornaviridae are single-stranded RNA viruses with small icosahedral virions. This
group includes polio virus (Enterovirus C), which infects the central nervous system, caus-
ing paralysis and death. Infections were at their peak in the 1940s and 1950s and prompted
the development of effective vaccines and a World Health Organization global eradication
program; there are now only a few thousand cases a year in a few regions of Africa and the
Indian subcontinent. Coxsackieviruses also belong to this group and are probably the most
abundant type in sewage; besides gastroenteritis, they are associated with meningitis and can
sometimes cause paralysis.

Infection with hepatitis A virus is endemic in many low and middle-income countries and
over 90% of children may be infected and excreting the virus. It is most commonly acquired
by travelers to such countries, although there have been occasional outbreaks in Europe and
North America associated with consumption of shellfish and other foods. Infection damages
the function of the liver but is not usually serious.

Finally, the Reoviridae family includes the rotaviruses, which are the most serious causes
of infant mortality in Africa and large parts of Asia. Even in high-income countries, a large
proportion of young children are infected, meaning that sewage contains high numbers of
these viruses.

Fecal indicator organisms (FIOs) are

used to assess public health risks
The concept of an indicator organism stems from the idea that pathogens may be present in
such low numbers in water that their detection is difficult, time-consuming, or expensive. An
indicator is an easily cultivable organism present in sewage, whose presence indicates the
possibility that pathogens may be present. Ideally, the indicator should be present in greater
numbers and survive longer in the environment than the pathogens, providing a built-in safety
margin. Early microbiologists recognized that E. coli and other coliform bacteria are present
in large numbers (over 108 per gram) in the gut of warm-blooded animals and their feces—
they are also very easy to culture—so they became the standard indicator for fecal pollution
of water. The coliform bacteria are facultative anaerobic Gram-negative bacilli characterized
by sensitivity to bile salts (or similar detergent-like substances) and the ability to ferment lac-
tose at 35°C. The definition of the coliform group is operational rather than taxonomic, with
80 species in 19 genera; the best known of these are Escherichia, Citrobacter, Enterobacter,
Erwinia, Hafnia, Klebsiella, Serratia, and Yersinia. The fecal coliforms—they may also
be termed “thermotolerant coliforms,” depending on the methodology used—are a sub-
set of the coliforms originally distinguished by their ability to grow and ferment lactose at
44°C; the main member of this group is E. coli. Growth on selective media and fermentative
properties are used in the traditional methods of identification and enumeration in water
(Figures 13.6, 13.9). A defining feature of the coliforms is the enzyme β-galactosidase, which
can be detected using o-nitrophenol-β-galactopyranoside (ONPG) as a substrate, resulting in
a yellow color. E. coli can be distinguished by the production of β-glucuronidase, which can
cleave either methylumbelliferyl-β-glucuronide (resulting in fluorescence under UV light)
or proprietary chromogenic compounds (resulting in distinctive colored colonies on agar
plates). These reactions form the basis of several commercial testing methods, which have
gained widespread acceptance because they produce results more rapidly.

Coliforms and E. coli are unreliable

FIOs for seawater monitoring
It is now generally recognized that, while appropriate for testing drinking water, coliform
and E. coli counts are poor indicators of marine water quality. The total coliforms can be
derived from a wide variety of sources, including plants and soil, so they often reflect runoff
from the land as well as fecal pollution. It is also very difficult to distinguish whether E. coli
in coastal waters originates from human or animal fecal contamination. Most significant of
364 Chapter 13

Figure 13.6 Outline of membrane

filtration method for enumeration
of E. coli and intestinal enterococci
in marine water samples (based on
the USEPA method). Appropriate vol-
umes of seawater are filtered through
0.45 µm membrane filters, which
trap bacteria on the surface. Filters
are rolled onto selective chromo-
genic agar. After incubation for 24 h,
E. coli at 44˚C produces red-magenta
colonies on mTec agar (contains
indoxyl-β- d -glucoside). Enterococci
at 41˚C produce blue colonies
with a halo on MEI agar (contains
indolyl-β- d -glucuronide).

all, the survival of indicator coliforms and E. coli in the environment does not seem to reflect
COULD BUILDING the survival of viral pathogens. Thus, they fail to meet one of the major requirements of a reli-
? SANDCASTLES
LEAD TO DEADLY
able indicator organism. Many studies have been carried out on the survival of bacteria and
viruses in water. The effects of temperature, sunlight, pH, water turbidity, salinity, and pres-
INFECTION? ence of organic matter are highly complex interacting variables. Of these, the bactericidal
effects of ultraviolet irradiation are the most important. Bacterial indicators have different
We usually think of E. coli as a
survival characteristics in marine and fresh waters, while human viruses are inactivated at
harmless member of the gut
similar rates in both (Figure 13.7). Although many different studies have produced conflict-
microbiota used, but several strains
are pathogenic—of which the most ing results, depending on the environmental conditions, a general conclusion is that levels of
serious is serotype O157:H7. This total coliforms are virtually useless as an indicator of health risks associated with marine
causes severe kidney damage and waters, while levels of fecal coliforms (including E. coli) do not necessarily indicate good or
can be fatal, especially in children. poor water quality and are unreliable as predictors of health risks (see Box 13.1). In the USA,
Cattle are the main reservoir of E. coli is used as an indicator in inland recreational waters but is no longer used in marine
E. coli O157:H7 and large numbers waters for regulatory purposes, although it is still in use in Europe and other countries.
of the bacteria may be excreted
in their feces. Serious epidemics
in humans are usually associated Enterococci are more reliable FIOs for seawater monitoring
with consumption of meat or salad
crops contaminated by infected As early as 1980, it was proposed that the bacterial group known as the enterococci might
animals, but there have also been be a better indicator of human pathogens for monitoring environmental water quality. These
links between infection and envi- bacteria are consistently associated with the intestines of humans and warm-blooded animals
ronmental sources (Muniesa et al., and, as shown in Figure 13.7, generally have longer survival times in water than coliforms
2006; Bintsis, 2018). In southwest and E. coli. It was previously thought that different species were more consistently associ-
England, the death of a child in ated with humans or animals, and a high ratio of fecal coliforms to fecal enterococci has
1999 and several other cases of been interpreted as indicative of primarily human fecal contamination, whereas a low ratio
O157:H7 infections linked to play-
was thought to indicate animal sources. These conclusions are now thought to be unreli-
ing on the beach since that date
are probably due to pollution from
able because of variable survival times of different species in seawater. Previously known as
streams draining from land where fecal streptococci, reclassification led to the recognition of a subgroup including the species
cattle have grazed. This can be a Enterococcus avium, E. faecalis, E. faecium, E. durans, and E. gallinarum. These are differ-
problem on apparently pristine, entiated from other streptococci by their ability to grow in 6.5% sodium chloride, at pH 9.6,
rural beaches far from human and in a wide temperature range of 10–45°C; they are also resistant to 60°C for 30 min. Since
habitation. Williams et al. (2007) the most common types found in the environment fulfil these criteria, the terms “fecal strep-
and others have shown that E. coli tococci,” “intestinal enterococci,” and “Enterococcus” are used interchangeably. Although
O157:H7 can survive in beach sand the number of enterococci excreted in feces is lower than E. coli, they are still much higher
and soil for various time periods, than the numbers of detectable pathogens. Methods of enumeration follow the same prin-
depending on the conditions.
ciples as those for E. coli as shown in Figure 13.6 and a variety of specialized media incor-
Several other studies have also
porating chromogenic substrates, antibiotics, and tests for specific biochemical reactions can
linked digging in beach sand with
enteric illness (Heaney et al. 2009). be used. Some semi-automated alternatives to membrane filtration have the advantage of pro-
ducing same-day results. For example, the Enterolert® test kit quantifies enterococci by the
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 365

Figure 13.7 Simplified guide to

typical survival times of fecal indica-
tor organisms and enteric viruses in
temperate coastal waters. Conditions
that affect microbial survival in water
are complex and interdependent.
Extreme conditions would be typi-
fied by clear water, bright sunlight,
high temperatures; protected condi-
tions would include turbid water,
cloud cover, low temperatures.

most probable number (MPN) method using growth in media with a fluorogenic substrate.
The XplOrer64™ system measures growth of enterococci in a selective medium via changes
in impedance. Under some conditions, autochthonous marine vibrios and aeromonads can
interfere with the reactions and lead to false positive results. Also, runoff from farmland
after heavy rainfall can significantly distort the counts of enterococci, and some studies have
attempted to distinguish bacteria of human and animal origin.

Molecular-based methods permit quicker analysis of

indicator organisms and microbial source tracking
The conventional culture-based methods used to assess microbial quality of water are widely
used because they require relatively simple laboratory equipment, but they suffer from the
disadvantage that it usually takes 18–96 h to obtain results. While regular water monitoring
gives information about the long-term trends at particular sites, it does not provide informa-
tion to the public about the state of the water on a particular day or time. Knowing that the
levels of indicator organisms were unacceptably high a couple of days ago and that the beach
users should have been advised not to swim does not inspire public confidence and does not
adequately protect public health. Conversely, some authorities close beaches for several days
after heavy rains as a precaution because past experience has shown that heavy rains may be
associated with increased levels of indicator organisms; the reality may be that any “spike”
in the levels of pathogens may have been and gone more quickly. This can be of considerable
importance to the economy of coastal communities relying on beach trading as a source of
income. For these reasons, there have been many efforts to develop more rapid methods of
testing that will produce near real-time results in 2–4 h. This would allow management deci-
sions to be made the same day and also permit better tracking of the source of contamination
to its origin.

A number of qPCR-based studies have been carried out using primers targeting sequences
of 16S rRNA genes or functional genes of indicator bacteria. DNA is extracted from water
and amplified by multiplex PCR using a range of primers. PCR based methods suffer from
the requirement for trained laboratory personnel in centralized laboratories, which could
limit their usefulness if samples have to be transported long distances, although some tri-
als of devices to carry out PCR assays in the field have shown some promise. Microarray
techniques (see p.59) have the advantage that they can simultaneously assay for hundreds of
different gene sequences. Beads can be coated with probes designed to hybridize with these
amplicons and are also labeled with spectrally specific fluorescence markers so that quanti-
fication of specific targets can be measured in a fluorimeter. Generally, DNA-based methods
366 Chapter 13

detect genetic sequences from both dead and viable bacteria, which may overestimate con-
centrations when compared with culture-based methods. One promising alternative involves
the immunomagnetic separation of E. coli or enterococci, using magnetic beads coated with
antibodies directed against these bacteria. The beads are mixed with a water sample and
selectively capture the bacteria; the beads are then collected from the sample using a magnet.
The concentration of bacteria present is assessed by removing them from the beads and mea-
suring the amount of ATP present by measuring bioluminescence in a luminometer, using a
luciferin-luciferinase assay (see p.377). Because of considerable variations in composition of
water samples and the possible low number of target FIOs, standardization of new methods
and improvements in reliability and reproducibility of results are required before universal
acceptance for regulatory purposes.

One of the main applications of new methods has been for microbial source tracking. The
aim here is to identify populations of bacteria that are specific to particular animals, because
recreational waters are often contaminated by nonpoint sources of fecal pollution of ani-
mal origin through runoff of farm waste via streams, or from wildlife including seabirds.
Microbial source tracking usually involves culturing isolates of indicator bacteria from a
wide range of sources and building up a library of profiling data, using phenotypic charac-
teristics (such as antibiotic sensitivity or carbohydrate utilization) or genetic fingerprinting
methods (see Table 2.2). These methods are very labor-intensive and time-consuming but
reveal important information about temporal and geographic changes in communities associ-
ated with different animals.

Various alternative indicator species

have been investigated
SPICING UP
i
The limitations of the usual FIOs—coliforms, E. coli, and (to a lesser degree) enterococci—
POLLUTION CHECKS as reliable predictors of health risks in marine waters has led to investigations of alterna-
Metagenomic analyses of viral tive indicators. Clostridium perfringens has been used because it forms resistant endospores
sequences in the human gut has that survive for long periods in the environment. This is particularly valuable in monitoring
revealed a surprising result that long-term effects of sewage disposal, for example, in sediments and offshore sewage sludge
may lead to new pollution indica- dumps. The order Bacteroidales is a group of Gram-negative anaerobic bacteria including
tors. Rosario et al. (2009) found Bacteroides, Prevotella, and Porphyromonas. Bacteroides fragilis is one of the most abun-
that the Pepper mild mottle virus dant organisms in human feces and early studies showed that it dies rapidly on transfer to
(PMMoV, genus Tobamovirus) seawater, making it a good indicator of recent human pollution. Despite some promising
is abundant in human feces, in research results, the requirement for anaerobic cultivation makes these methods unsuitable
treated wastewater and in coastal
for many laboratories undertaking routine monitoring. However, detection and quantification
waters near discharge sites. Most
animal feces do not contain
of Bacteroides and Prevotella by qPCR has proved very useful in tracking sources of pol-
PMMoV. Numerous studies have lution. A variety of primer sets have been developed that distinguish between strains from
confirmed its strong potential as humans and those of other animals such as cattle, horses, pigs, and dogs.
an indicator of human-specific
pollution in environmental waters Methanogenic archaea have also been investigated as potential indicator species. Methanogens
(Symonds et al., 2018). Why such are widely distributed in marine sediments and activated sludge in sewage plants, but most
high numbers of this plant virus members of the genus Methanobrevibacter seem to be restricted to the mouth and intestines
are found in the human gut is of animals and the species M. smithii appears to be a human-specific species, present in
unknown, although PMMoV RNA human feces at densities between 107 and 1010 per gram. PCR assays targeting sequences of
sequences are present in many
the nifH gene specific to M. smithii have been developed.
spicy food products containing
peppers. Colson et al. (2010) pro-
vided preliminary evidence of asso- Phages are especially useful as indicator organisms, using the rationale that these viruses
ciation with immune responses, could be expected to have similar survival characteristics to viruses pathogenic for humans.
abdominal pain and fever. They Direct testing for enteric viral pathogens in seawater is technically very difficult and unreli-
suggested that this might be the able whereas phages are easily measured using plaque assays (plating concentrated water
first example of a plant virus infect- samples on lawns of susceptible host bacteria; see Figure 7.2) or with group-specific oligo-
ing humans. However, no causal nucleotide probes. The most promising phages are the F+ (“male-specific”) RNA coliphages
relationship has been demon- that infect E. coli expressing sex pili on their surface. They are morphologically similar
strated (Balique et al., 2015) and to several groups of enteric viruses and are present in very high numbers in sewage. Some
epidemiological studies would be success has been achieved in distinguishing animal and human sources and in establishing
needed to rule out the effects of
that they may have survival times similar to the human viruses when exposed to UV light
reactions to the spicy foods rather
than viral infection.
under various conditions. Epidemiological studies have shown correlation between numbers
of coliphages and gastrointestinal illness. Metagenomic analysis of viral sequences from the
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 367

human gut has revealed a previously unidentified phage (crAssphage), which probably infects
Bacteroides. It is highly abundant in human fecal waste and qPCR assays to detect its con-
centration in water are being developed for use in pollution monitoring.

Countries have different quality

standards for bathing waters
As discussed in Box 13.1, there have been many studies attempting to prove a link between
the levels of FIOs and adverse health risks from recreational use of marine waters. The meth-
odological or statistical basis of such studies has been the subject of much controversy. The
difficulties of devising regulatory systems and quality measures that reliably reflect risk have
led to variations in the microbiological monitoring programs and standards adopted by envi-
ronmental agencies in different countries. In 2003, the World Health Organization (WHO)
published guideline values for the microbiological quality of recreational waters. These used
a statistical approach in which 95% of the counts must lie below a threshold value. Using this
system, the estimated additional risk of bathing (compared with a control group) in waters
with less than 40 colony-forming units (CFU) of enterococci per 100 mL is less than 1%
(grade A water). Between 41 and 200 CFU per 100 mL (grade B water), this rises to a 1–5%,
and between 201 and 500 CFU per 100 mL (grade C water), the increased risk is 5–10%. In
grade D water, containing more than 500 enterococci per 100 ml, there is a >10% increased
risk of contracting gastrointestinal illness. The risk of acquiring acute febrile respiratory
infections is also significant in grade C and D waters. It should be noted that these risk factors
apply to healthy adults, as children are usually excluded from this type of epidemiological
study for ethical reasons. The risk of infection is probably much higher for lower age groups.
It should also be noted that the data on which the standards are based are mainly obtained
from Northern Europe and North America, so may not be applicable in all regions.

Despite these WHO guidelines, the nature of the indicators, the frequency of sampling, the
methods of quantification, and the threshold values for compliance with standards all vary
between countries. Because there can be considerable day-to-day variation in the numbers of
indicator organisms detected, many systems are based on percentage compliance levels, such
as the requirement in the European Union that 90 or 95% of measurements taken during the
sampling period meet the prescribed standard. Other authorities such as the US Environment
Protection Agency (EPA) use geometric mean values of data to even out unusually high or
low results. The geometric mean is statistically more stable—although it can overlook the
occasional high values at the top of the statistical distribution—whereas the 90/95% com-
pliance system has the advantage of reflecting the occasional high levels of concern and is
marginally easier for the public to interpret.

In Europe, the first definition of microbiological standards was published in 1976 as the
European Community Bathing Water Directive. This defined testing protocols based on total
and fecal coliform counts. The European legislation was used as the basis for improvements—
at great expense—to coastal sewage disposal systems which were designed to comply with
the standards for coliform and E. coli levels. Interpretation of epidemiological studies led to
the realization that, although the motive of these schemes was laudable, the scientific basis for
evaluating their effectiveness was questionable, since an increased risk of illness seemed to be
associated with water-quality indicators well within the European limits. Increasing recogni-
tion that enterococci are a more reliable indicator than fecal coliforms led to the inclusion of
a guide level for enterococci in amendments to the European legislation in 1998. After a long
period of discussion and argument, revised microbiological standards were adopted in 2006,
which shifted the emphasis from treatment and disposal of sewage to expanding the “Blue
Flag” classification system for beaches and improvements in public information. On the basis
of monitoring designated bathing beaches about 20 times from May to September, water qual-
ity is assessed based on samples from the previous four years using threshold values per 100
mL for E. coli (EC) and intestinal enterococci (IE). The categories are: Excellent (<100 IE,
<250 EC); Good (<200 IE, <500 EC); Sufficient (185 IE, 500 EC); or Poor (>185 IE, >500 IE).
The Excellent and Good standards are based on results from the 95th percentile, whereas the
Sufficient standard—with an apparently lower limit than Good—is based on the 90th percen-
tile. This has caused confusion and controversy. Health authorities will have to make decisions
368 Chapter 13

BOX 13.1 RESEARCH FOCUS

Is it safe to swim? The health effects of recreational exposure to seawater

Correlation of FIO levels with bathers’ perceptions of illness. Controlled surveys. One of the major limitations of many studies
The first attempts to investigate this question were carried out in is that they rely on perception of illness by those taking part, who
the 1950s in freshwater bathing sites. These showed that there was a are usually asked to respond to a questionnaire distributed among
significant increase in gastrointestinal, ear, nose, and throat symp- beachgoers. It is very difficult to distinguish the true effect of swim-
toms in swimmers exposed to waters with high coliform counts. ming from other effects; for example, illness might result from food
This was used by the US authorities to set the first standards for consumed during a trip to the beach. In the 1990s, growing public
recreational waters. The epidemiological basis for defining these concern and pressure from groups such as Surfers against Sewage
standards was questioned by many authors, most notably Cabelli and the Marine Conservation Society led the UK Government to
et al. (1982), who undertook more detailed studies at US beaches. sponsor new epidemiological research designed to overcome some
Rigorous attempts to ensure standardization of microbiological of the criticisms of earlier studies. Carefully controlled beach
methods, reporting of symptoms by participants, and the use of surveys were conducted, in which volunteers were given a medi-
suitable control groups were included. Cabelli’s group found that cal examination before randomized groups undertook supervised
levels of fecal coliforms were not a reliable indicator of health risks, swims with water testing at frequent intervals. One week later,
but that there was a statistically significant correlation between participants received a second medical examination plus testing
the levels of enterococci and an increased incidence of gastroin- of throat and ear swabs and fecal specimens. Randomization and
testinal symptoms associated with swimming. As a result, the US detailed analysis of questionnaires allowed other risk factors such
Environment Protection Agency (USEPA) dropped the use of fecal as food intake to be considered. The UK study (Fleisher et al., 1998;
coliforms as indicators in marine waters and set a new standard Kay et al., 2001) confirmed that there was close correlation between
based on a geometric mean of 35 enterococci per 100 ml in five levels of enterococci in bathing water and a significant increase in
samples over a 30-day period. Many similar surveys have been risk of GI disease when the number exceeded 30–40 per 100 ml. A
undertaken, and meta-analyses of the data from numerous epide- convincing dose-response curve was observed. In a new systematic
miological studies have shown a consistent correlation between meta-analysis of 40 studies, Leonard et al. (2018) employed rig-
increased risk of gastrointestinal (GI) symptoms—especially in orous selection criteria to determine the increased risk of becom-
children —and the counts of FIOs, especially enterococci (Pruss, ing ill after seawater bathing and whether the level of increased
1998; Wade et al., 2003; Arnold et al., 2016). Some studies have risk depends on the nature of exposure. The results—based on
reported increased rates of skin symptoms among swimmers, but over 120,000 people in the eligible studies—showed that bathing
controlled studies have produced limited evidence of links with increased the odds of earache by 77% and the odds of GI symptoms
fecal contamination. by 29%, compared with not bathing. All the data in this analysis
were obtained from high-income countries (US, UK, Australia,
New Zealand, Denmark, and Norway) with a high level of pollu-
tion control, but the results show that further improvements would
be beneficial.

Does surfing carry an increased risk? Participants in water

sports such as body boarding, surfing, and scuba diving may have
prolonged exposure and increased risks of ingestion of seawater or
inhalation of aerosols. Despite many anecdotal reports of illnesses
linked to water sports, there have been very few detailed studies of
health risks from microbial infections. In a survey of surfers who
spent a mean of 77 days a year surfing in Oregon, USA, Stone et al.
(2008) concluded that surfers had a mean exposure of 11–86 CFU
enterococci per day, with an epidemiological model predicting that
they have a 23% probability of exceeding the exposure equivalent
to the USEPA maximum acceptable gastrointestinal illness rates.
Excellent Water sports activities may also take place outside of the official
Good “bathing season,” meaning that the authorities do not undertake
checks of water quality. For example, in southwest England the best
Sufficient surfing conditions often occur in the winter, when potential patho-
Poor gens may survive for longer or be resuspended from sediments by
wave action following storms (Bradley and Hancock, 2003). In San
Figure 13.8 Example of beach signage showing EU Diego, California, a prospective cohort study of 654 adult surfers
Bathing Water Directive information. Upper sign indicates over two winters showed that surfing following stormy weather car-
that bathing is not advised (or prohibited); lower sign indi- ried a three-fold increased incidence of earache/infections and a
cates status of the classification of water quality at the beach five-fold increase in infected open wounds (Arnold et al., 2017).
according to the monitoring survey.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 369

BOX 13.1 RESEARCH FOCUS

Despite the high incidence of head immersion and ingestion of reasons—the UK-based survey asked participants to collect their
water in surfers, their overall GI rate was similar to that previously own rectal swabs, which were returned to the laboratory for isola-
reported for swimmers. Arnold and colleagues found that levels tion of E. coli resistant to cefotaxime (a third-generation cephalo-
of FIOs were only a reliable indicator of risk within three days of sporin, 3GC) and meropenem (a carbapenem). Resistant isolates
heavy rain, which they attributed to urban runoff from storm drains. were then screened for the mobile resistance genes using Q-PCR.
Other studies have also noted this lack of correlation between ill- Leonard and colleagues found 3GC-resistant E. coli in 15% of
ness and FIOs when there is no well-defined point source of human coastal sites. They isolated nearly 115000 E. coli, of which 140
fecal contamination (Fewtrell and Kay, 2015). (0.12%) were 3GC-resistant and 83 (0.7%) possessed blaCTX-M-15.
Based on the distribution of the resistant E. coli and using data
Exposure to antibiotic-resistant bacteria. Resistance to antibi- on the number of water sports events, Leonard et al. (2018) con-
otics has emerged as major threat to global health, and threatens cluded that surfers are at a particularly high risk of being exposed.
to return us to the pre-antibiotic era, when many common infec- Further analysis showed that—despite the low overall prevalence
tions will become untreatable and some surgical interventions and of colonization—surfers are four times as likely to be colonized by
other medical procedures will become too dangerous to undertake blaCTX-M-15-containing E. coli as non-surfers. Several of the surfers
(O’Neill, 2016). Antibiotic-resistant bacteria (ARB) are now wide- in the survey were shown to be carrying 3GC-resistant pathogenic
spread as a result of widespread use in medicine, agriculture, and strains of E. coli; although they were asymptomatic, it is possible
aquaculture, which has encouraged the rapid evolution and spread that future health impairment could lead to proliferation of the
of genes conferring antibiotic resistance. Resistance to β-lactam bacteria and disease. Despite the relatively small frequency of
antibiotics (the widely used penicillin and cephalosporin fami- these ARB and the need for additional objective epidemiological
lies) is particularly concerning. Extended-spectrum β-lactamase information about the effects of harboring them, the study never-
(ESBL) enzymes encoded on mobile genetic elements are now theless provides worrying evidence of the spread of ESBL genes
prevalent in bacteria of the family Enterobacteriaceae and other in the environment.
pathogens associated with community and hospital-acquired
infections, which are consequently very difficult to treat. There Concluding remarks. It is important that the general health ben-
is increasing recognition of the transmission of ARB to humans efits—physical outdoor activity and sense of well-being—of swim-
from natural environments. Wastewater effluent has been shown to ming, surfing, and other coastal pursuits are balanced against the
contribute to the dissemination of multidrug-resistant ESBL bacte- mostly small additional risks of acute illnesses, which are usually
ria carrying the blaCTX-M-15 gene (Amos et al., 2014). To determine mild and self-limiting (although risks for children are higher).
whether there was a link between recreational use of coastal waters However, continuing improvements to reduce pollution are essen-
and acquisition of ARB, Leonard et al. (2018) devised an innova- tial and better provision of information is needed to enable the pub-
tive epidemiological survey. A group of 150 frequent surfers was lic to fully understand the meaning of water-quality standards so
recruited, together with 150 matched controls with very little expo- that they can evaluate risks and make informed choices according
sure to seawater. Dubbed the “Beach Bum Survey”—for obvious to local conditions.

on what constitutes an “acceptable risk,” but public information on the interpretation of the
water-quality results posted at many seaside beaches (Figure 13.8) is woefully inadequate to
enable people to make an informed personal choice about the risks of bathing or undertaking
water sports. In view of the importance of international tourism, public confidence in monitor-
ing procedures is essential; and the current diversity of approaches is confusing. Surprisingly,
the current European level of 100 enterococci per 100 mL (indicating “Excellent” water qual-
ity) is considerably greater than the level predicted by epidemiological studies as carrying a
significant increased probability of illness and is much higher than that in many other devel-
oped countries. For example, the USA and Canada use a threshold value of 35 enterococci per
100 mL (based on a geometric mean of five samples over a 30-day period). In Australia and
New Zealand, the standards for primary contact activities such as swimming and surfing are
150 fecal coliforms and 35 enterococci per 100 mL (with maximum permissible levels of 600
and 100 respectively); the authorities also define less stringent standards for secondary contact
activities such as boating and fishing. It is surprising that the rank of “Excellent” water under
European standards is only the equivalent of grade B in the WHO guidelines.

Shellfish from sewage-polluted waters

can cause human infection
Molluscan bivalve shellfish such as oysters, mussels, clams, and cockles concentrate patho-
gens from their environment by filter feeding. For example, a single oyster filters about 5
liters of water per day and each square meter of a dense bed of mussels filters more than
370 Chapter 13

105 liters per day. Thus, even if there are low concentrations of pathogenic microbes in
KEEPING UP THE
i PRESSURE FOR
SAFER SEAFOOD
the water, there is the potential for very large numbers to accumulate in the bodies of the
animals. The rate of accumulation of microbes and their subsequent survival within the
tissue depends on the species involved and numerous environmental factors, especially
Pathogenic bacteria and viruses temperature.
in shellfish are killed by thorough
cooking. However, this alters Shellfish are often subjected to a process of disinfection known as depuration before they
the desired taste and texture are sent to market for human consumption. Depuration involves transferring the animals
of shellfish, especially premium for 24–48 h to free-flowing or recirculating clean seawater, which is disinfected by UV
products such as oysters. Very
light, chlorination, or ozone treatment. The natural filtering activity of the bivalve mol-
high hydrostatic pressure (typically
lusks results in elimination of their intestinal contents. The processes of depuration were
275–400 MPa) can be used instead
for the processing of sensitive developed in the early twentieth century to eliminate the risk of typhoid fever caused by
foodstuffs and has proved effective the bacterium Salmonella typhi; at the time, this caused serious epidemics associated with
in reducing bacterial viability and eating shellfish grown in polluted water. Depuration is effective in removing S. typhi and
prolonging shelf-life of shellfish other fecal bacteria including FIOs from shellfish. However, it is much less effective in
(Murchie et al., 2005). This process the removal of sewage-associated viruses, naturally occurring bacterial pathogens such as
is now widely used in commercial V. parahaemolyticus and V. vulnificus (see p.338), dinoflagellate, and diatom toxins (see
processing of oysters and some p.343), or chemical pollutants. Hepatitis A acquired from shellfish is of particular concern;
other shellfish; it also has the although rarely fatal, it can cause a long, severe debilitating illness and some large out-
advantage of making oysters easier
breaks involving hundreds of cases—one outbreak in China affected 291,000 people—have
to open and “plumping up” the
occurred, for example, following the serving of contaminated shellfish at banquets or recep-
volume of the meat. Importantly,
high-pressure treatment has been tions. Norovirus is the most common cause of seafood-associated gastroenteritis but usually
shown to inactivate V. vulnificus, only comes to the attention of health authorities when outbreaks involving large groups of
V. parahaemolyticus, and V. people occur. It is likely that thousands of undocumented cases of seafood-associated gas-
cholerae, and pathogenic viruses troenteritis occur each year, since most victims do not visit their doctor and, in any case,
such as caliciviruses and hepatitis accurate diagnosis is difficult. As occurs with illnesses caused by the autochthonous marine
A virus (Calci et al. 2005). Some bacterial pathogens V. parahaemolyticus and V. vulnificus, the problem is exacerbated by
tests show that it may also be the fact that shellfish are often eaten raw or only partially cooked. Many of the enteric
effective in inactivation of norovi- viruses are resistant to moderate heating, low pH, freezing, and drying, so normal food-
rus (Imamura et al., 2017; Gyawali processing methods have little effect.
et al., 2019). Unfortunately, there
is little evidence that this method
leads to the inactivation of algal Microbiological standards are used for
toxins, another significant health
hazard associated with shellfish. classification of shellfish production areas
Shellfish growing areas are classified to assess their suitability for production for
human consumption. The standards used in Europe (shown in Table 13.1) are based on
end-product analysis of replicate samples of shellfish flesh by a most probable number

Figure 13.9 Outline of the most

probable number (MPN) method for
determining E. coli in shellfish. Media
and conditions can vary; the version
here is based on the EU Reference
Laboratory protocol. Abbreviations:
MMGB minerals modified gluta-
mate broth; TBGA/TBX tryptone
bile glucuronide agar (contains
5‐bromo‐4‐chloro‐3‐indolyl‐β‐D
glucuronide to detect
β‐glucuronidase activity). Calculation
of MPN is based on the statistical
probability that some tubes will be
inoculated with a single organism
from the dilutions used.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 371

Table 13.1 Classification of shellfish (bivalve molluscs) microbiological quality

in the European Union according to Shellfish Hygiene Directive 2013

Category E. coli (MPN) per 100 g Acceptability for human

consumption

Maximum level Overall

in 90% of maximum
samples level

A 230 700 May be harvested directly

B 4600 46000 Must be depurated in approved

unit, relayed in an approved
category A area, or heat treated

C 46000 − Must be relayed for at least two

months to meet category A or B
requirements, or be heat treated

(MPN) method to quantify E. coli as an indicator of fecal pollution (Figure 13.9). In

other countries, monitoring of FIOs in surface waters is used to classify shellfish pro-
duction areas into one of five classes based on levels of total coliforms or fecal coli-
forms in surface waters. In the USA, approval for marketing of shellfish without prior
treatment requires geometric mean levels in the water of E. coli ≤14 MPN per 100 mL
(90th percentile ≤43). Shellfish from restricted areas with E. coli levels of ≤88 MPN
per 100 mL (90th percentile ≤260) require depuration or relaying. Other areas may have
periodic restrictions or be prohibited. Again, the use of E. coli as an indicator is a very
unreliable predictor of health risk—no classification methods employ enterococci—and
the standards have no epidemiological basis. Many outbreaks of viral illness—some-
times large and severe—have been associated with shellfish that appear to be of high
standard as judged by the levels of bacterial FIOs. Shellfish for human consumption
must also be tested for the presence of microbial toxins, for which there are separate
standards (see p.344).

Direct testing for pathogens in shellfish is

possible with molecular methods
The above discussion suggests that the indicator concept has major shortcomings when
applied to marine waters and shellfish. Would it not be better to undertake direct testing
for the pathogens themselves? The problem here is that most viruses, the pathogens that
cause most concern, are difficult or impossible to culture. They are often also present at
extremely low levels, yet nevertheless present a health hazard due to their low infectious
dose. The usual method is to inoculate cell cultures of suitable human cells (such as entero-
cytes) with water samples (after concentration) and monitor for cell lysis or other cyto-
pathic effects. Special safety precautions must be taken in laboratories using human cell
culture, and the effects, if any, may take several weeks to appear. Attention has therefore
turned to molecular biological techniques, in which the most promising approach has been
detection of viral RNA sequences by the qRT-PCR method (p.47). Difficulties with the
technique include selection of suitable primers that are representative of virus strains cir-
culating in a particular region, and interference with the PCR amplification by inhibitors,
which can be a particular problem in shellfish tissue. Careful optimization of techniques is
required for extraction from the digestive tissue. Suitable primers that target the RNA virus
group under study are added to samples of water or shellfish tissue extracts in the presence
of the enzyme reverse transcriptase to make a cDNA copy, which is then amplified by Taq
polymerase, usually using a nested PCR (p.47). The methods are still too time-consuming
and expensive to be used in routine monitoring, but they are invaluable in confirming the
sources of disease outbreaks and special studies such as environmental impact assessments
of new sewage disposal schemes and classification of shellfish harvesting areas. Future
advances will depend on linking infectious virus levels with epidemiological data on the
incidence of disease.
372 Chapter 13

OIL AND OTHER CHEMICAL POLLUTION

Oil pollution of the marine environment
is a major problem
The world economy is totally dependent on petroleum products as a source of energy and
raw materials for a vast range of products. Despite concerns about the impact of fossil fuels
on climate change and the move towards sustainable sources, about 4 billion metric tons of
crude oil are extracted annually and global demand is rising by about 1% every year. About
1.2 million metric tons of oil enters the oceans every year, of which nearly half comes from
natural seepage from oil deposits under the seabed. As shown in Figure 13.10, pollution
by oil tankers and spills from offshore oil rigs accounts for only a small proportion of oil
entering the ocean, although it attracts the greatest public concern and has major impacts
because of the sudden introduction of large amounts into the environment. Tankers take
on water for ballast and, when this is discharged, considerable local contamination of the
sea can occur if their oil separators are not functioning correctly. Occasionally, tankers
collide or run aground, releasing large quantities of oil into the sea. Fortunately, in recent
years, tighter regulations on ballast and greater use of double-hulled tankers has reduced
the amount of oil spilt in this way, but major incidents still occur. The immediate damaging
effect of this on marine wildlife and the economic impact on fisheries and tourism means
that environmental agencies are under intense pressure to alleviate the problem quickly
and effectively. Apart from immediate effects, there are concerns about toxic residues and
long-term disruption of ecosystems. There are various strategies for dealing with spilled
oil, including trapping with booms and skimming the surface to remove the oil, using
absorbent materials to soak up oil, applying dispersant chemicals, or setting fire to oil
slicks. The efficacy of these processes is very dependent on the location of the spill and
weather conditions, and the use of dispersants or burning can have highly damaging effects
on marine ecology.

Microbes naturally degrade oil in the sea

Fortunately, naturally occurring microbes break down most of the components of petro-
leum, which is a complex mixture of thousands of compounds, predominantly hydro-
carbons. Oil is a natural product formed from algae buried in the sediments of the sea
and lakes under high pressure and temperature over millions of years. Large quantities
of crude oil have therefore been seeping naturally during this time, and many different
microbes in more than 175 bacterial and archaeal genera have evolved efficient mecha-
nisms to degrade the different hydrocarbon components—these have been isolated from
seawater, shores, and sediments all over the world. A similar number of yeasts and fila-
mentous fungi have also been described. Most hydrocarbon-degrading isolates are het-
erotrophs belonging to the Gammaproteobacteria, of which the most important are in
the genera Alcanivorax, Cycloclasticus, Fundibacter, and Oleispira. Cyanobacteria (e.g.
Merismopedia, Microcoleus, and Phormidium) and the alga Ochromonas have also been
linked with hydrocarbon degradation. Some photoautotrophic species appear to accumu-
late hydrocarbons within vesicles, but not to degrade them and it is possible that they form
syntrophic consortia with heterotrophic bacteria, leading to breakdown. Hydrocarbons
are degraded by aerobic microbes in all parts of the water column and the upper surface
of sediments, while anaerobes are active in deeper sediments and hydrocarbon seeps. In

3%

12% Extraction,offshore rigs, blowouts

Transport, tanker spills,
tank washing

Figure 13.10 Sources of oil 47% Landrunoff, atmosphere,

38% marine engines
inputs to the sea (based on data
from Transportation Research Natural seeps
Board, National Research Council,
Washington D.C., 2003.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 373

anoxic environments, SRB and archaeal methanogens can degrade hydrocarbons as a sole
source of carbon and energy. The breakdown of oil depends on the activities of a consor-
tium of microorganisms, each responsible for transformation of a specific fraction. These
degraders are always present, but normally constitute only a small proportion (<1%) of the
microbiota of seawater, except in the location of natural seepage. However, following a
pollution incident they multiply rapidly to 10% or more of the population. Immobilization
of mixed microbial communities on biofilms may be particularly important in efficient bio-
degradation. Box 13.2 describes how the use of dispersants at the wellhead enabled rapid
degradation of oil in the Deepwater Horizon oil well blowout in 2010, one of the worst ever
pollution incidents.

Physical and biological processes affect the fate of oil spills

When oil is released into the seas as a result of a spillage, it floats on the surface and the
low-molecular-weight fractions evaporate quickly. Some components are water-soluble and
photochemical oxidation occurs through the action of sunlight. The fate of an oil slick is very
dependent on wave action and weather conditions. Sometimes, droplets of oil will become
emulsified in the water and disperse quickly; in other cases, a water-in-oil emulsion will
form to produce a thick viscous “chocolate mousse,” which takes a very long time to disperse
and gives rise to the familiar lumps of beach tar. Different crude oils vary, but usually over
70% of the hydrocarbons are biodegradable. Only the asphaltenes and resin components are
recalcitrant to breakdown.

The biochemical processes involved in the biodegradation of oil have been studied exten-
sively. Aerobic processes are responsible for the efficient biodegradation of oil. The initial
step involves incorporation of molecular oxygen into n-alkanes by oxygenases, resulting in
primary alcohols, which are then further oxidized to aldehydes and fatty acids. The fatty
acids are then metabolized to acetyl-coenzyme A, which enters primary metabolism via the
tricarboxylic acid (TCA) cycle. In this way, the n-alkanes are completely converted to car-
bon dioxide and water. There are many different routes for the degradation of the aromatic
hydrocarbons, the key step being the cleavage of the aromatic ring; oxygen is again essential
for this process. Polyaromatic hydrocarbons (PAHs) such as naphthalene, phenanthrene, and
pyrene contain four or more aromatic rings and are only degraded slowly by a few types of
bacteria. Consequently, PAHs that are not degraded in the water column accumulate in sedi-
ments and can persist for very long periods. Many harbors and estuaries that are used exten-
sively by shipping have sediments that are chronically polluted with PAHs. In such anoxic
sediments, anaerobic degradation occurs with the oxidative processes linked to the reduc-
tion of nitrate or sulfate, or the production of methane. Degradation probably begins with
carboxylation reactions, followed by ring reduction and cleavage. The activity of burrowing
benthic animals can enhance aerobic PAH degradation by microbes, owing to the mixing
of sediments and introduction of oxygen through the elaborate ventilation systems that they
construct; isolation of bacteria from such burrows may prove a rich source of new species.

Bioremediation of oil spills may be enhanced

by emulsifiers and nutrients
Biodegradation proceeds most quickly when the oil is emulsified into small droplets and
hydrocarbon uptake by microorganisms is stimulated by the production of biosurfactants
(surface-active agents containing both hydrophilic and hydrophobic regions which reduce
surface tension). Many microbes have hydrophobic surfaces and adhere to small droplets of
oil, and many also produce extracellular compounds that disperse the oil. There has been
great interest in developing natural biosurfactants as an alternative to chemical dispersants.
For example, the bacterium Acinetobacter calcoaceticus is particularly effective in biodegra-
dation of oil because it both adheres to hydrocarbons and produces an extracellular glycolipid
biosurfactant called emulsan. Emulsan-deficient mutants grow very poorly on hydrocarbons.
It may be possible to select naturally occurring strains or use genetic engineering to produce
large amounts of biosurfactants, although optimizing the industrial-scale production of these
compounds may be difficult. Emulsan is also used to reduce viscosity to aid in the extraction
of crude oil.
374 Chapter 13

BOX 13.2 RESEARCH FOCUS

How microbes cleaned up deep-sea oil after the

Deepwater Horizon (DWH) disaster

Response to the DWH blowout. In April 2010, the DWH platform and would be highly flammable if it came to the surface. The injec-
was drilling an exploratory well for oil at a depth of 1500 m, 322 tion of dispersant successfully forced some of the oil to rise to the
miles off the coast in the Gulf of Mexico. A blowout led to a meth- surface at a distance from the wellhead as a result of creating small
ane explosion and fire that killed 11 workers. The fire raged for 36 droplets that were less buoyant and moved more easily by ambient
h, before the platform sank, leaving the wellhead seriously dam- currents. Over 40% of the oil formed a plume—more than 200 m
aged and gushing huge quantities of crude oil into the deep sea. A high and 2 km wide—of highly dispersed microdroplets of hydro-
massive operation involving hundreds of ships, aircraft, and shore- carbons and succinate compounds derived from the dispersants.
based operations was put in place to remove surface oil or trap it in The plume extended in the deep water over 35 km from the source
containment booms in an attempt to protect the thousands of miles (Atlas and Hazen, 2011) (Figure 13.11)
of beaches, estuaries, and wetlands on the Gulf Coast. Throughout
the summer of 2010, nearly five million barrels of oil (7.8 × 105 m3) Effect on deep-sea microbial communities. From a microbio-
of crude oil flowed unchecked into the sea, until a major engineer- logical perspective, one of the most interesting consequences of the
ing operation sealed the well with a concrete cap in mid-July. By spill was the fate of the massive plumes of oil droplets released
the first week of August, oil could no longer be detected in the deep- in the deep water. This is a cold (2–5˚C), high-pressure environ-
water column or on the ocean surface, and in September, the US ment with very low levels of carbon, so it was not known how
Coast Guard declared the well dead. the microbial community would respond to such a sudden influx
of hydrocarbons. An urgently coordinated study of the plume by
Use of chemical dispersants. Because of the scale of the release, a team from the Berkeley National Laboratory found evidence of
more than 4 × 106 L of dispersant (Corexit 9500) were used to treat rapid degradation of the oil (Hazen et al., 2010). The analysis was
the oil as it escaped from the wellhead. Although it was reasoned initially performed using the PhyloChip 16SrRNA microarray that
that dispersing the oil might facilitate its breakdown by microbes, was developed as a rapid method for the profiling of microbial
the main reason for using it at the well head was for safety reasons populations in environmental samples. They found a significant
as the mixture of light crude oil and natural gas was very volatile stimulation of psychrophilic bacteria related to three families in

Figure 13.11 Schematic representation of the Deepwater Horizon oil spill. Not to scale (the well head well is 1500 m below
the surface). Reprinted with permission from Atlas and Hazen (2011). Copyright 2011 American Chemical Society.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 375

BOX 13.2 RESEARCH FOCUS

the order Oceanospirillales, dominated by a novel uncultured spe- insights into microbial succession patterns following the addition of
cies closely related to Oleispira antarctica and Oceaniserpentilla oil, our analysis revealed that the dominant hydrocarbon-degrading
haliotis. In later samples, the community composition shifted bacteria in the simulation was not in the environment, and the one
and was dominated by species of Colwellia and Cycloclasticus in the environment was not in the simulation.”
(Redmond and Valentine, 2012). Analysis of functional microbial
genes in the cloud of dispersed oil correlated with changes in the Effects of dispersants on degradation. The use of such large quan-
concentration of various components of the oil and rapid biodegra- tities of chemical dispersants was highly controversial, because of
dation rates were demonstrated in laboratory studies, with the half- their known toxicity and the fact that they had never been used in
life of alkanes ranging from 1.2 to 6.1 days (Hazen et al., 2010). such deep water. Laboratory simulations to study the effects of the
The authors concluded that the degradation proceeded so rapidly dispersant have also generated some controversy about their role
because the oil contained a high component of readily degradable in biodegradation. Experiments conducted by Kleindienst et al.
volatile components, and because the oil was dispersed into small (2015) found that the dispersants caused a major shift in microbial
droplets, providing a large surface area amenable to attack by the community composition through selection of dispersant-degrading
bacteria. However, there did seem to be a constraint on the rate Colwellia, which also bloomed during the DWH event. However,
of degradation, possibly because of low levels of iron needed as a when oil was added to deep water samples in the absence of dis-
co-factor for degradative enzymes. This was reassuring in view of persants, hydrocarbon-degrading Marinobacter was selected.
the fears that extensive oxygen depletion due to heterotrophic activ- Kleindienst and colleagues concluded that dispersants can suppress
ity would create a massive anoxic “dead zone.” In fact, the oxygen rates of hydrocarbon degradation. However, Prince et al. (2016)
saturation inside the plume was only a few percent lower than out- argued that the laboratory methods used did not truly simulate the
side the plume. Nowhere in the world has a higher concentration action of dispersants in the sea, where turbulence generates small
of natural seeps and offshore drilling than the Gulf of Mexico, so droplets that diffuse apart and remain entrained in the water col-
it is perhaps not surprising that the resident microbes were able to umn. Kleindienst et al. (2016) rebutted these criticisms and main-
respond so rapidly to the influx of oil—resulting in a very different tained their argument that the efficacy of dispersants to reduce oil
outcome to other catastrophic events such as the Exxon Valdez oil spill impacts is unclear, as undegraded chemically dispersed oil
spill from a tanker on the Alaska shoreline (Atlas and Hazen, 2011). was shown to be present on Gulf coastlines several years after the
To obtain more information about the long-term fate of the oil, Hu spill.
et al. (2017) collected deep-sea water samples from the area in 2014
and conducted a laboratory simulation of the plume and recon- Long-term impacts. Although almost all the suspended oil disap-
structed population genomes of the successive blooms of diverse peared within a few months, deposits of undispersed oil have been
bacteria observed, obtaining nearly complete genomes of the major transferred to the seabed and coastal sediments, while up to 30% of
hydrocarbon-degrading species. Hu and colleagues concluded that oil remains unaccounted for. Also, the dispersants used are known
the rapidly degraded oil components were consumed by bacteria to be toxic to marine life and, as they had never been used in the deep
with substrate specialization; these included a new species belong- sea previously, their long-term effects are not known. Whatever
ing to the Oceanospirillaceae, which they named “Ca. Bermanella happens 1500 m down, oil that has contaminated the surface water
macondoprimitus.” However, Delmont and Eren (2017) found no and the shores, creeks, mangroves, and salt marshes of the Gulf
evidence that the 16S rRNA gene of the newly proposed species Coast will behave very differently (reviewed by Kimes et al., 2014;
was present in the original community genome analysis of the King et al., 2015) . The legacy of the DWH disaster will undoubt-
plume and commented: “Although the simulation offers valuable edly affect the ecology and economy of the Gulf region for many
decades to come and some sensitive areas may never fully recover.

Bioremediation is defined as a biological process to enhance the rate or extent of naturally

occurring biodegradation of pollutants, although in a broader sense it can be used for delib-
erate use of any biological process that reverses environmental damage. The hydrocarbons
in oil provide a carbon source for bacteria, but oil is deficient in other nutrients (especially
nitrogen and phosphorus) and the supply of these is the main factor limiting the rate of degra-
dation. Therefore, the most successful approach to bioremediation is the addition of inorganic
or organic nutrients as fertilizers to accelerate natural processes. The process of seeding
oil spills with exogenous microorganisms shown to have high degradative activity in the
laboratory (bioaugmentation) has been less successful because they are rapidly outcompeted
by the enrichment of naturally occurring microbes. Furthermore, the idea—popular in the
1980s—that genetic modification could be used to create a “superbug” suitable for use in the
natural marine environment and capable of digesting all the different components of oil has
proved to be misguided.

Laboratory studies provide little information about how well bioremediation treatments will
work in the field. Some mesocosm and controlled release experiments in the field have been
done, but these obviously must be limited in size and scope, as few authorities are willing
376 Chapter 13

to allow the deliberate pollution of coastal waters. Therefore, most of our knowledge about
LESSONS LEARNED
i FROM THE EXXON
VALDEZ OIL SPILL
the efficacy of bioremediation comes from studies of opportunity following large-scale spills
from tanker accidents, in which investigators have little control over the prevailing condi-
tions. Of such incidents, the cleanup after the Exxon Valdez spill in Alaska has been the most
The oil tanker Exxon Valdez ran extensive study of bioremediation. Deliberate experimental contamination of shorelines in
aground in Prince William Sound, Norway and Canada had shown previously that, even under Arctic conditions, oil applied
Alaska in March 1989, releasing to shorelines would be degraded naturally within a few years, but that addition of agricul-
11 million gallons of crude oil that tural fertilizers like ammonium phosphate, ammonium nitrate, or urea increased the initial
contaminated 500 km of coastline rate of biodegradation up to ten times. Bioremediation has also had some success in other
in a pristine wilderness environ-
major oil spills, such as the Prestige tanker spill that affected the coast of Galicia in Spain
ment. Under the control of the US
Environment Protection Agency, a
in 2002 and in oil reserves deliberately released in Kuwait during the 1991 Gulf War. The
range of approaches to bioreme- best approach to fertilization seems to be to use oleophilic compounds, which stick to the oil
diation were tested, and the les- and/or release nutrients slowly. One such slow-release fertilizer is a proprietary compound
sons learned have informed many called Inipol™ EAP22; this is a microemulsion of urea in brine, encapsulated in an external
other bioremediation applications. phase of oleic acid and lauryl phosphate, co-solubilized by butoxy-ethanol. Bioremediation
Nitrogen fertilizer was used in of sediment contamination is more problematic. In terrestrial situations, biodegradation is
large amounts and its effective- enhanced by tilling of contaminated soil, which introduces oxygen, but this is obviously
ness was found to depend on the impractical in marine situations. It may be possible to introduce oxygen to marine sediments
nature of the substrate to which by aeration pumps, chemical oxidants, or alternate electron acceptors, but these require care-
the oil was bound. Enhancement
ful evaluation of ecosystem effects. Bioremediation in sensitive habitats like mangroves and
of breakdown on pebbles, gravel,
salt marshes is particularly difficult and the ecological impact is especially severe.
and large sand particles was much
better than on fine sand particles.
Bioremediation of exposed surfaces Bioremediation of petroleum products is applied extensively to clean up contaminated soil
in Prince William Sound was suc- (e.g. to reclaim land polluted by spillage from oil tanks), and many commercial products
cessful—microbial activity was developed for this purpose have been marketed for use in harbors and marinas. However,
enhanced, and oil biodegradation the scientific rigor with which these have been tested for marine bioremediation is ques-
was stimulated two- to five-fold. tionable. A major problem in testing the effectiveness of bioremediation is monitoring the
However, large amounts of oil extent of degradation. Because breakdown proceeds in a progressive fashion, disappearance
remain in subsurface sediments of compounds such as the alkanes and small aromatic compounds can be measured easily but
trapped under boulders and monitoring removal of the more recalcitrant compounds is more problematic. One internal
pebbles, and it is unlikely that
standard method is to measure the disappearance of biodegradable components in compari-
further bioremediation treatments
would enhance degradation of
son with the concentration of hopanes, which are highly recalcitrant to breakdown.
the polyaromatic hydrocarbons,
because few nutrients permeate In summary, bioremediation by the application of oleophilic or slow-release fertilizers is now
down into the subsurface lay- generally accepted to have proved its worth as one component in the response to an oil spill.
ers. Although most of the surface However, before it is used, careful attention must be paid to the nature of the substratum
oil has gone, the ecological and and the degree of penetration of oil into the sediments. Addition of exogenous organisms
economic legacy of the oil spill has not been successful, but further development of surfactant-producing strains and their
remains. formulation into products that allow them to compete with the stimulation of indigenous
microorganisms may hold promise for the future. More research on changes in microbial
community composition in response to introduction of oil containing different mixtures of
hydrocarbons may also yield valuable information about the best ways to enhance natural
processes of degradation.

Microbes can detoxify heavy metals

from contaminated sediments
Pollution of the marine environment by heavy metals is a major threat to all forms of life
due to their toxicity. The elements usually considered to be most important as environmental
pollutants are chromium, arsenic, cadmium, tin, cobalt, nickel, antimony, lead, and mercury.
These elements cause a range of biological effects, mainly through their interference with
enzyme function. They can be carcinogenic, cause damage to the nervous and reproductive
systems and are generally toxic at low levels to many aquatic organisms. Because they do not
degrade, they accumulate and persist in the environment. The major sources in the marine
environment are effluent from mining activity, discharge of industrial waste, atmospheric
input from power stations, or terrestrial runoff from contaminated soils or landfill sites. Some
marine bacteria and microalgae can gain resistance to high concentrations of metals in con-
taminated environments and are effective at immobilizing metals into a non-bioavailable
form via biosorption or bioaccumulation.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 377

Microbial systems can be used for ecotoxicological testing

The Microtox® system is an established biosensor for the rapid toxicity testing of water, sedi-
ment, and soil samples. It depends on inhibition of bioluminescence of Aliivibrio fischeri (p.103),
which is supplied as a standardized freeze-dried culture. Light emission is measured using a
photometer and is very sensitive to the presence of toxic chemicals at sublethal concentrations.
Portable systems for field testing of chemical spills are now available. Bioassays based on the
inhibitory effects on bioluminescence of dinoflagellates such as Lingulodinium polyedra or
Pyrocystis lunula have also been developed. These methods are many times more sensitive and
much easier to carry out than conventional ecotoxicological bioassays using fish or amphipods.

The study of natural marine microbial communities is an important aspect of monitoring the
effects of pollution, temperature shifts, and other environmental disturbance. The DGGE
technique for community analysis has been used widely in such studies, before its replace-
ment by high-throughput sequencing (see p.49). Pollutant-degrading organisms can also be
genetically modified to link the expression of degradative genes to reporter systems such as
lux genes from A. fischeri or GFP genes from jellyfish (see p.35). These reporter systems have
wide applications in cell biology as well as environmental investigations.

Microbial adsorption and metabolism

affect accumulation of mercury
An indirect consequence of the activity of bacteria in soil and water is the transformation
by microbial methylation reactions of trace elements such as mercury, arsenic, cadmium,
and lead. These methylated elements are mobilized and can enter the food chain, causing
deleterious health effects. The accumulation of mercury in marine fish is a particular health
concern. Although it is a naturally occurring element released from volcanic eruptions and
soil runoff, most mercury now enters the environment from anthropogenic sources, such
as mining, coal-fired power stations, waste incinerators, and industrial processes. Mercury
is also present in some pesticides and is an important component of discarded electronic
equipment and batteries—erosion of coastal landfill sites due to rising sea levels and storms
is a growing problem. It is estimated that up to 4.5 × 103 tonnes of mercury are deposited in
the oceans each year, and levels are rising because of the increased emissions from rapidly
developing industrialized countries in Asia.

Most mercury enters the sea in the form of the Hg2+ ion and adsorbs readily to particles in
which it can be transformed by several types of anaerobic bacteria, especially sulfate- and
iron-reducing bacteria and methanogens, to methylmercury (CH3Hg+ or MeHg) in a reac-
tion mediated by the gene pair hgcAB. A second methyl group can also be attached to give
dimethylmercury [(CH3)2Hg or Me2Hg]. Hg methylation mainly occurs in anoxic sediments
and oxygen minimum zones (see p.235), but it also occurs in the mesopelagic zones of the
open ocean, at great distances from industrial discharges. This explains why Pacific tuna—
one of the major sources of mercury intake by humans—accumulate such high levels of
MeHg even though they are living very distant from sources of mercury. It is likely that Hg
from industrial sources is transported over long distances by atmospheric deposition and
ocean currents and that dying diatoms and other algae scavenge Hg from the water column
by adsorption to their cells as they sink into deeper waters. Anerobic bacteria containing
the hgcAB genes convert the mercury during breakdown of organic matter within oxygen-
depleted niches within marine snow aggregates. Thus, the mesopelagic habitat is a major
entry point for mercury into marine food webs. The tissue levels of MeHg in different fish
species depends on the depth at which they feed. In areas with high productivity and an effi-
cient biological pump, MeHg adsorbed to diatom cells is rapidly transported to the seabed
and accumulates in siliceous sediments (diatom oozes).

Microbial cycling is important in the distribution

of persistent organic pollutants
Persistent organic pollutants include the organochlorine insecticides (e.g. DDT, aldrin, and
chlordane), industrial chemicals (e.g. polychlorinated biphenyls, PCBs) and by-products (e.g.
378 Chapter 13

dioxins and furans). They reach the sea via terrestrial runoff and atmospheric deposition. In

i THE HAZARDS OF
MERCURY IN FISH
numerous countries the manufacture and use of many of these chemicals has been prohibited,
but they are highly persistent in sediments and dump sites and produced during incineration
Mercury is a potent neurotoxin; it of waste. PCBs occur in many electronic products, and the disposal of unwanted equipment is
causes liver, kidney, and cardiovas- a significant source of these chemicals. These chemicals are very resistant to photochemical,
cular damage in adult humans and biological, and chemical degradation. Most persistent compounds are halogenated and highly
severely affects development of the soluble in lipids and they accumulate particularly in fatty tissues. Their semi-volatile nature
fetal nervous system if consumed allows them to vaporize or to be adsorbed onto atmospheric particles and they are therefore
by pregnant women. Because
transported over great distances. For example, animals and humans living in remote polar
MeHg is lipid soluble, it is concen-
regions have high levels of organic pollutants in their tissues despite these compounds not
trated in the food chain, especially
in fish. One of the world’s most being used there in any significant amounts. Extensive evidence links these compounds to
serious environmental disasters reproductive failure, impairment of the immune system, deformities, and other malfunctions
occurred in Japan in the 1970s, in a wide range of marine life, and they are highly toxic to humans through ingestion of fish
when thousands of people who and other routes.
had eaten fish from the heavily pol-
luted Minamata Bay were seriously Microbial cycling of organic pollutants acquired by plankton in the upper parts of the ocean
affected by MeHg poisoning. In plays an important role in their distribution through ocean food webs. The bacterioplankton
this case, hundreds of deaths and presents a large surface area for the adsorption of organic pollutants, and microbial loop pro-
long-term health effects occurred. cesses release compounds during settlement of plankton debris and organic particles through
Because of its sequential concentra-
the water column, but some particles with adsorbed pollutants will be buried in sediments.
tion at each step of the food chain,
high levels of MeHg occur in top
Disturbance of sediments by tides, currents, dredging, and the activity of benthic animals can
predators such as tuna, shark, and release large quantities of the chemicals back into the water. PCBs are highly recalcitrant to
marine mammals. Very high levels degradation and there is an active search for microbes capable of breaking down the chlorine
of mercury have been found in the bonds, thus offering a potential use in bioremediation. Many aerobic bacteria can degrade
tissues of Eskimo and Inuit com- the biphenyl ring in PCBs, but not the heavily chlorinated congeners. Anaerobic degradation
munities who eat large amounts of PCBs is known to occur, but isolation and identification of the organisms responsible has
of fish, seal, and whale meat. The been elusive; different bacteria with distinct dehalogenases and congener specificities occur.
general health advice that fish is Selective enrichment of PCB-contaminated sediments, accompanied by high throughput
a good source of nutrition with sequencing, can be used to identify new species of PCB degraders and the genes responsible
significant health benefits has to
for reductive dechlorination. Syntrophic consortia of members of the Dehalococcoidia class
be balanced against the possibil-
and sulfur-oxidizing Epsilonproteobacteria appear to be important. Bioelectrochemical sys-
ity of consuming unsafe levels of
MeHg. Health authorities in many tems (microbial fuel cells) are a promising technology for bioremediation of contaminated
countries advise pregnant women, sediments—solid electrodes serve as an electron acceptor or donor to facilitate microbial
nursing mothers, and young chil- oxidation or reduction.
dren to completely avoid fish with
high levels of MeHg (e.g. sword-
fish, marlin, or shark) and to limit
their intake of fish with moderate
Plastic pollution of the oceans is a major global problem
levels (e.g. tuna) to no more than Different types of plastic have a variety of uses, replacing traditional materials and leading to
twice a week. many novel applications (Table 13.2). Every year, society produces over 350 million metric
tons of plastic, and it is estimated that about 8 million MT of end-of-life plastics—equivalent
to a garbage truckload every minute—finds its way into the marine environment. Because
most plastics are resistant to degradation they are accumulating at an alarming rate and are
currently the cause of great public concern, prompting government and international action
to curb the use of single-use plastics, promote recycling and new approaches to reduce the
current linear use of plastics, and move towards a more “circular economy.” Some plastic
polymers are now made using plant biomass (e.g. wood, starch, or cellulose), vegetable oils,
or bacterial production. Manufacture of these bioplastics uses a renewable rather than fossil
carbon source but changing the carbon source does not necessarily make the polymer any
more degradable. A few biobased polymers, including polylactic acid, polyhydroxyalkono-
ates, and poly-3-hydroxy butyrate are fully biodegradable, but biologically derived PE or
PA are not. Although they have some environmental benefits and some types biodegrade
under controlled composting conditions, many so called biodegradable plastics (such as PE
shopping bags incorporating corn starch) do not truly degrade; instead, they break up slowly
and release many small particles of plastic. Beaches and coastlines throughout the world are
littered with plastic debris and it is estimated that more than 105 tonnes of plastic waste circu-
lates in the North Pacific and Atlantic gyres. Accumulations also occur in the South Atlantic
and Pacific, as well as the Indian Ocean gyres, but these have not been so well sampled.
Plastics can be found in waters from all over the world—from the surface to the deepest
parts of the sea floor. Plastic waste is harmful to marine mammals, birds, and fishes. Ropes,
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 379

Table 13.2 Examples of the most commonly used plastics

Plastic Abbreviation Examples of uses

Polyethylene terephthalate PET Polyester fibers, packaging, bottles, clothing

Low-density polyethylene LDPE Shopping and food bags, agricultural mulches

High-density polyethylene HDPE Robust packaging containers

Polyvinyl chloride PVC Pipes, insulation, construction, clothing

Polypropylene PP Packaging, manufactured goods, textiles

Polystyrene PS Foam packaging, laboratory equipment

Polylactic acid PLA Fully biodegradable containers, 3D-printed parts

Polycarbonate PC Glasshouses, protective shields

Acrylic PMMA Transparent sheeting, optical devices

Polyamide (nylon) PA Clothing, tires, car parts, ropes, nets

Polyurethane PU Foam insulation and furnishings, moldings

Acrylonitrile butadiene styrene ABS Machined, molded, and 3D-printed parts; toys

discarded fishing nets, and plastic bags cause entanglement and suffocation, while syringes,
cigarette lighters, straws, nurdles (pellets used in plastic manufacture), and other debris is
mistaken for food by seabirds, fish, and turtles. But the readily visible pieces of plastic are
just part of the problem. Most of the plastic that has entered the ocean cannot be accounted
for. Some sinks to the bottom and will be buried, but much of it becomes smaller and smaller
in size—eventually reaching nanometer size—due to fragmentation, abrasion, and weather-
ing. Microplastics also enter the ocean in a primary form as beads used in cosmetic products
and fibers from clothing. The importance of microbial interactions with these polluting par-
ticles is discussed in Box 13.3.

Conclusions
This chapter has shown that microbes have some serious deleterious effects due to coloni-
zation of marine surfaces, biodeterioration and corrosion of structures and materials, and
spoilage of seafood products. These activities cause significant economic losses to maritime
industries. Some control methods have been discussed here, while other biotechnological
solutions are considered in the next chapter. We have seen that microbes play an important
role in pollution of coastal waters by sewage, both as human pathogens introduced with fecal
waste and also as indicators of pollution for environmental monitoring and evaluating risks
to public health. There are severe limitations with conventional techniques, but considerable
improvements are now being achieved through the use of new methodologies. By contrast,
microbes play a highly beneficial role in the degradation of oil, most of which finds its way
into the ocean from natural seepage. Augmentation or enhancement of these natural degra-
dation processes for the bioremediation of coastal pollution from oil spills have been used
with varying degrees of success. It remains to be seen whether microbes will be effective in
the removal of other industrial pollutants of the modern world, especially persistent organic
pollutants and plastics. One of the most worrying developments discussed is the realization
that ocean microbes can convert mercury to a toxic form as part of the normal food web pro-
cesses. This threatens the safety of pelagic fish as a food source and can only be controlled by
limiting anthropogenic emissions from newly industrialized countries. We also need urgent
solutions to prevent end-of-life plastics entering the ocean and better understanding of the fate
and effects of the massive quantities of microplastics already in our oceans. Microbiologists
will have a major role in investigating and ameliorating this global pollution problem.
380 Chapter 13

BOX 13.3 RESEARCH FOCUS

Microorganisms and microplastics

How big (or small) is the problem? The term microplastics was first used every 3 tonnes of fish by 2025 and that plastics will exceed the weight of
by Thompson et al. (2004) to describe microscopic fragments of plastic. fish by 2050. Understanding the implications of this for ocean ecology and
Since then, many studies have adopted a working definition of micro- the consequences for human health (Wright and Kelly, 2017; Oliveira et
plastics as <5 µm diameter. Recently, an international group of experts al., 2019) is one of the most pressing scientific issues of our time.
in this field have discussed the need for a unified terminology and sys-
tem for categorizing different materials by size, shape, composition, and Colonization of plastics by microbial biofilms. A number of studies have
other parameters (Hartmann et al., 2019), so that researchers are “speak- shown that plastics become rapidly colonized by microbes after introduc-
ing the same language.” They propose that the term microplastics is used tion to seawater. Lobelle and Cunliffe, 2011) showed that LDPE food bags
for items 1 to <1000 µm and nanoplastics for 1 to <1000 nm. One of the suspended in seawater became colonized within a few days by a firmly-
main sources of primary microplastics are minute fibers released during attached bacterial biofilm, which decreased the hydrophobicity and buoy-
the washing of garments made of synthetic fabrics. De Falco et al. (2018) ancy, causing the bags to sink below the surface after three weeks. Zettler
estimated that a typical 5 kg wash load of woven polyester garments et al. (2013) examined PP and PE plastic marine debris from the North
can release up to 6 × 106 microfibers. Other sources include microbeads Atlantic gyre using scanning electron microscopy (SEM) and high-through-
(Napper et al., 2015) used in cosmetic exfoliants (now banned or restricted put tag sequencing (p.54). They showed the presence of mixed biofilms con-
in many countries), air-blasting media, dust from vehicle tires, and road taining diverse bacteria (over 1000 OTUs), diatoms, picoeukaryotes, and
markings. Until quite recently, the fate of most of the microplastics was other protists covering up to 8% of the surface of the plastic—a community
unknown—“Lost at sea – where is all the plastic?” (Thompson et al., they termed the “plastisphere” (Figure 13.12). Sequence analysis showed
2004). We now know that microplastics can remain suspended for vary- that the microbial community was consistently distinct from that of the sur-
ing periods (depending on their density, size, and shape); they are trans- rounding seawater. As discussed earlier in this chapter, such biofilms recruit
ported via ocean currents and are found globally in the water column at a variety of other colonizers. Reisser et al. (2014) also observed bacteria-like
all depths and in tidal, sub-tidal, and deep-sea sediments (Woodall et al., cells, diatoms, coccolithophores, bryozoans, barnacles, and other organ-
2014). Microplastics are an appropriate size to be ingested by plankton- isms colonizing ocean microplastics around Australia. These plastic par-
feeding animals, as demonstrated in a range of species, from zooplankton ticles might therefore serve as a floating “microbial reef” that could disperse
(Cole et al., 2013) to fish (Lusher et al., 2013), leading to concerns that bacterial pathogens (such as Vibrio spp.), harmful algae, and invasive inver-
they could interfere with natural feeding and negatively impact ecosystem tebrates. Significantly, both studies also observed bacterial cells embedded
function. Furthermore, plastics may also transfer toxic chemicals, either in pits and grooves on the surface of the plastic, suggesting that bacterial
those present as a result of manufacture (e.g. phthalates and bisphenol A) biodegradation of the plastics might be occurring. Zettler and colleagues
or by adsorption of PCBs and other hydrophobic organic contaminants noted that some OTUs matched bacteria known to be capable of degrading
from the environment (Teuten et al., 2009), although Bakir et al. (2016) hydrocarbons. Harrison et al. (2014) conducted a mesocosm experiment in
provide evidence that ingestion of microplastic is unlikely to provide a which they showed successional changes in community composition of bac-
major pathway for the transfer of adsorbed chemicals from seawater to teria on LDPE microplastics, detected over a two-week period using SEM
animals via the gut. Microbeads and microfibers in wastewater might also and CARD-FISH (p.38). This study suggested that specific bacteria asso-
become colonized by pathogenic or ARB as they pass through water treat- ciated with hydrocarbon degradation might be selected. Oberbeckmann
ment plants. A report by the World Economic Forum (2016) suggests that, et al. (2016) also used SEM and tag sequencing to study PET bottles sus-
if current trends continue, the oceans could contain 1 tonne of plastic for pended in the North Sea. Bottles developed surface communities that were

Figure 13.12 SEM images showing examples of the rich microbial community on plastic marine debris. A. Pennate diatom with
possible filaments produced by Hyphomonas-like bacteria. B. Filamentous cyanobacteria. C. Stalked suctorian ciliate covered with
ectosymbiotic bacteria (inset), along with diatoms, bacteria, and filamentous cells. D. Microbial cells pitting the surface. All scale
bars are 10 μm. Reprinted with permission from Zettler et al. (2013). Copyright 2013 American Chemical Society.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 381

BOX 13.3 RESEARCH FOCUS

significantly different from those in seawater, but not from glass surfaces, of PE and PP microplastics in sterile artificial seawater led to rapid leach-
suggesting that the PET did not influence the nature of settling biofilm ing of dissolved organic carbon (DOC). This material probably includes
microbes. However, some OTUs were characteristic of the PET-associated truly dissolved substances such as additives used in plastic manufacture
biofilms, notably members of the Sphingobacteriales and Mycococcales and nanoparticles released by fragmentation. After removal of the micro-
and filamentous cyanobacteria in the genus Phormidium, as well as repre- plastics, the water was inoculated with natural bacterial community from
sentatives of other bacterial families known to degrade hydrocarbons—as surface sea water. Bacteria assimilated over half of the leached DOC and
also observed by Zettler et al. (2013) in their North Atlantic gyre study. bacterial growth was stimulated by ~one order of magnitude. Although
acknowledging the complexities of extrapolating the laboratory findings
How significant is microbial degradation of plastics? As noted in to ocean conditions, Romera-Castillo et al. (2018) show calculations that
Box 13.2, many marine bacteria have evolved to degrade the hydrocarbons indicate the potential large impact of microplastics on carbon cycling. In
in crude oil, which has been seeping into the oceans for millions of years. another simulation, Galgani et al. (2018) showed that PS particles increased
However, plastics—mostly derived by polymerization and polycondensa- the production of CDOM (chromophoric or light-absorbing DOM, which
tion of petroleum products—have only been accumulating in the natural is subject to photochemical transformations). This may increase the avail-
environment for the past 50 or so years, and it might be expected that few ability of low molecular weight compounds for microbial growth. Gewert
organisms have evolved to degrade them efficiently. Yoshida et al. (2016) et al. (2018) identified numerous degradation products (mainly dicarboxylic
isolated a distinct consortium of PET-degrading bacteria, fungi and protists acids) from floating plastic fragments exposed to UV light. Galgani and
from sediments at a PET bottle recycling site. Through enrichment cultures, colleagues extrapolated their findings to suggest that stimulation of CDOM
they isolated a novel bacterial species, given the proposed name Ideonella release by microplastics acting as a hotspot for high microbial activity could
sakaiensis. Experimental treatments of PET film with this isolate resulted in alter carbon cycling at the sea surface. This could modify the penetration
complete degradation within six weeks at 30˚C. The bacterium was shown of solar radiation and change the air-sea flux of CO2. Another way in which
to produce two key enzymes that hydrolyze PET and enable it to use it as microplastics might affect ocean processes is via alteration of the rate of
a carbon and energy source. Through genome database searches, Yoshida transport of carbon from the surface layers. Most plastics are buoyant, so
postulated that the enzymes for PET metabolism may have evolved from how can the deep ocean be a major sink for microplastics? The answer
hydrolases that target naturally occurring polymers such as cutin. However, seems to be that, after colonization by biofilms and association with TEPs,
Yang et al. (2016) commented that the results presented by Yoshida and diatom cells, zooplankton feces, and other debris, they form an integral part
colleagues exaggerated the rate of PET degradation because they used a of marine snow and sink through the water column. To simulate this pro-
low-crystallinity PET as substrate and did not demonstrate reduction in cess, Long et al. (2015) developed a flow-through roller tank to mimic the
molecular weight of the polymer. Crystallinity of plastics such as PE, PP, behavior of diatom aggregates as they settle through a layer of microplastic
and PET refers to the aggregation of segments of the long polymer chains in (PS) beads. Incorporation of beads into the aggregates increased the sinking
hard crystal-like domains, embedded in a soft amorphous polymer matrix. rates of the beads from tens to hundreds of meters per day. However, dif-
This has a great bearing on the weathering and degradation of plastics ferent diatom aggregates varied in the amount of plastic incorporated and
(Andrady, 2017). Yang et al. (2014, 2015) had previously demonstrated PE the effects this had on permeability, fragility, and sinking rates. The dense
and PS degradation by bacteria in the guts of insect larvae. Such findings aggregates of diatom species that naturally sink rapidly showed decreased
raise hopes that accelerated biodegradation of plastics might become fea- sinking rates after incorporating the low-density PS microbeads. In a simi-
sible, under carefully controlled conditions, in industrial recycling plants lar roller bottle simulation, Porter et al. (2018) showed that laboratory gener-
through modification of such bacterial isolates or their enzymes, as shown ated marine snow can incorporate microplastics of different shapes, sizes
for PETase by Austin et al. (2018). A better approach might be the artificial and polymer types. Sinking rates of all tested microplastics (PA, PS, PE, PP
selection of natural microbial communities with improved biodegradation fibers, and beads) increased when incorporated into marine snow. Buoyant
potential of recalcitrant polymers. Some success with this approach has polymers became negatively buoyant once incorporated into aggregates and
been achieved by Wright et al. (2019) using chitin as a case study, although sank rapidly. For example, PE microbeads floated on the surface, but sank
careful optimization of incubation times between generations is needed. Of at a relative rate of 659 m per day when incorporated into marine snow. The
course, such methods will be of no use to treat plastic waste already in the authors point out that these relative sinking rates may not be truly repre-
oceans, but the observations of apparent deformation of the surface of plas- sentative of real-world conditions, where a range of factors affects the rate
tics in the studies of Zettler et al. (2013) and Reisser et al. (2014) hint that of downward migration. Nevertheless, it seems that microplastics do have
natural biodegradation does occur in the sea, albeit at a rate too slow to have a major impact on the dynamics of marine snow transport and, hence, on
any meaningful effect on plastic accumulation. Might it be possible that microbially induced POM-DOM transition as the particles settle through
increased abundance and local concentration of microplastics in the envi- the water column. It may be that changes in their surface properties and
ronment could select for more efficient plastic-eating microbes? Might the fluctuations in density might cause some particles to remain suspended.
usual efficient processes of adaptation and evolution through mutation and Recently, it was found that zooplankton copepods ingest PS microbeads
gene transfer then accelerate their spread? Perhaps, but this is likely to be a and excrete them in their fecal pellets, which consequently have reduced
very slow process that can only be revealed by long-term mesocosm studies. density and structural integrity (Cole et al., 2016). This leads to greater
fragmentation and reduction in sinking rates. Since zooplankton fecal pel-
Could microplastics affect carbon cycling and food webs? There are lets—containing large amounts of dense organic material—are responsible
various ways in which large numbers of microplastic particles or fibers for the rapid transport of POM to the deep ocean (see p.230), natural pro-
could affect microbial processes in the food web and carbon pumps. Like cesses of carbon flux could be altered considerably. Many further studies
all particles, they create a micro-environment that differs significantly from are needed to disentangle the complexities of the influence of microplastics
the surrounding water, creating “hotspots” of nutrients and microbial activ- and their interactions with microorganisms on their passage from source to
ity (see Figure 1.11). Zettler et al. (2013) suggested that the hydrophobic sink. In particular, future studies need to recognize that the fate of plastics
nature of plastic surfaces could lead to the surface concentration of micro- in the environment, and the organic pollutants that they absorb, depends
nutrients—an effect first described by Zobell (1943)—that could stimu- on the physical and structural characteristics of the polymers, as well as
late microbial activity in the upper layer of oligotrophic ocean gyres. In their chemical composition (Andrady, 2017). Complex variables determine
laboratory experiments designed to simulate the fate of microplastics in the effects of weathering, fragmentation, microbial colonization, and decay.
the surface ocean, Romera-Castillo et al. (2018) showed that suspension
382 Chapter 13

References and further reading Boehm, A.B., Graham, K.E., & Jennings, W.C. (2018) Can we swim yet?
Systematic review, meta-analysis, and risk assessment of aging sewage in
Biofouling, bideterioration, and surface waters. Environ. Sci. Technol. 52: 9634–9645.
corrosion of materials Bradley, G., & Hancock, C. (2003) Increased risk of non-seasonal and
Bixler, G.D. & Bhushan, B. (2012) Biofouling: Lessons from nature. body immersion recreational marine bathers contacting indicator micro-
Trans. R. Soc. A 370: 2381–2417. organisms of sewage pollution. Mar. Poll. Bull. 46: 791–794.

Charles, F., Sauriau, P.-G., Aubert, F., et al. (2018) Sources partitioning Cabelli, V.J., Dufour, A.P., McCabe, L.J., & Levin, M.A. (1982)
in the diet of the shipworm Bankia carinata (J.E. Gray, 1827): An experi- Swimming-associated gastroenteritis and water quality. Am. J. Epidemiol.
mental study based on stable isotopes. Mar. Environ. Res. 142: 208–213. 115: 606–616.

de Carvalho, C.C.C.R. (2018) Marine biofilms: A successful microbial Colson, P., Richet, H., Desnues, C., et al. (2010) Pepper mild mottle virus,
strategy with economic implications. Front. Mar. Sci. 5: 126. a plant virus associated with specific immune responses, fever, abdominal
pains, and pruritus in humans. PLoS One 5: e10041.
Enning, D. & Garrelfs, J. (2014) Corrosion of iron by sulfate-reducing
bacteria: New views of an old problem. Appl. Environ. Microbiol. 80: Dutilh, B.E., Cassman, N., McNair, K., et al. (2014) A highly abundant
1226–1236. bacteriophage discovered in the unknown sequences of human faecal
metagenomes. Nat. Commun. 5: 4498.
Jia, R., Unsal, T., Xu, D., et al. (2019) Microbiologically influenced cor-
rosion and current mitigation strategies: A state of the art review. Int. Fewtrell, L. & Kay, D. (2015) Recreational water and infection: A review
Biodeterior. Biodegrad. 137: 42–58. of recent findings. Curr. Environ. Heal. Rep. 2: 85–94.

Kip, N. & Van Veen, J.A. (2015) The dual role of microbes in corrosion. Fleisher, J.M., Kay, D., Wyer, M.D., & Godfree, A.F. (1998) Estimates of
ISME J. 9: 542–551. the severity of illnesses associated with bathing in marine recreational
waters contaminated with domestic sewage. Int. J. Epidemiol. 27: 722–726.
Little, B.J., Lee, J.S., & Ray, R.I. (2008) The influence of marine biofilms
on corrosion: A concise review. Electrochim. Acta 54: 2–7. Haramoto, E., Kitajima, M., Hata, A.,et al. (2018) A review on recent prog-
ress in the detection methods and prevalence of human enteric viruses in
Martín-Rodríguez, A.J., Babarro, J.M.F., Lahoz, F., et al. (2015) From water. Water Res. 135: 168–186.
broad-spectrum biocides to quorum sensing disruptors and mussel repel-
lents: Antifouling profile of alkyl triphenylphosphonium salts. PLoS One Hassard, F., Gwyther, C.L., Farkas, K., et al. (2016) Abundance and distri-
10: e0123652. bution of enteric bacteria and viruses in coastal and estuarine sediments—
A review. Front. Microbiol. 7: 1692.
O’Connor, R.M., Fung, J.M., et al. (2014) Gill bacteria enable a novel
digestive strategy in a wood-feeding mollusk. Proc. Natl. Acad. Sci. USA Heaney, C.D., Sams, E., Wing, S., et al. (2009) Contact with beach sand
111: E5096–104. among beachgoers and risk of illness. Am. J. Epidemiol. 170: 164–172.

Pal, S., Qureshi, A., & Purohit, H.J. (2018) Intercepting signalling mecha- Jofre, J. & Blanch, A.R. (2010) Feasibility of methods based on nucleic
nism to control environmental biofouling. 3Biotech 8: 364. acid amplification techniques to fulfil the requirements for microbiologi-
cal analysis of water quality. J. Appl. Microbiol. 109: 1853–1867.
Pandit, S., Sarode, S., Sargunaraj, F., & Chandrasekhar, K. (2018)
Bacterial-mediated biofouling: Fundamentals and control techniques. In: Kay, D., Wyer, M., Fewtrell, L., et al. (2001) Health effects from recre-
Biotechnological Applications of Quorum Sensing Inhibitors. Springer ational water contact.
Singapore, Singapore, pp. 263–284. Leonard, A.F.C., Singer, A., Ukoumunne, O.C., et al. (2018) Is it safe to
Sandström, M., Jalilehvand, F., Persson, I., et al. (2002) Deterioration of go back into the water? A systematic review and meta-analysis of the risk
the seventeenth-century warship Vasa by internal formation of sulphuric of acquiring infections from recreational exposure to seawater. Int. J.
acid. Nature 415: 893. Epidemiol. 47: 572–586.

Vigneron, A., Alsop, E.B., Chambers, B., et al. (2016) Complementary Leonard, A.F.C., Zhang, L., Balfour, A.J., et al. (2018) Exposure to and
microorganisms in highly corrosive biofilms from an offshore oil produc- colonisation by antibiotic-resistant E. coli in UK coastal water users:
tion facility. Appl. Environ. Microbiol. 82: 2545–2554. Environmental surveillance, exposure assessment, and epidemiological
study (Beach Bum Survey). Environ. Int. 114: 326–333.
Vigneron, A., Head, I.M., & Tsesmetzis, N. (2018) Damage to offshore
production facilities by corrosive microbial biofilms. Appl. Microbiol. Mendes Silva, D. & Domingues, L. (2015) On the track for an efficient
Biotechnol. 102: 2525–2533. detection of Escherichia coli in water: A review on PCR-based methods.
Ecotoxicol. Environ. Saf. 113: 400–411.
Yang, J.C., Madupu, R., Durkin, A.S., et al. (2009) The complete genome
of Teredinibacter turnerae T7901: An intracellular endosymbiont of Muniesa, M., Jofre, J., Garcia-Aljaro, C., & Blanch, A.R. (2006)
marine wood-boring bivalves (shipworms). PLoS One 4: e6085. Occurrence of Escherichia coli O157 : H7 and other enterohemorrhagic
Escherichia coli in the environment. Environ. Sci. Technol. 40: 7141–7149.
Sewage and wastewater pollution O’Neill, J. (2016) Tackling drug-resistant infections globally: Final report
and recommendations. The review on antimicrobial resistance. Available
Amos, G.C.A., Hawkey, P.M., Gaze, W.H., & Wellington, E.M. (2014) online: https​://am ​r-rev​iew.o​rg/si​tes/d​efaul​t /fil​es/16​0518_​Final​%20pa​per_
Waste water effluent contributes to the dissemination of CTX-M-15 in the w​ith%2​0cove​r.pdf​ (accessed 2 February 2019).
natural environment. J. Antimicrob. Chemother. 69: 1785–1791.
Oliver, D.M., Hanley, N.D., van Niekerk, M., et al. (2016) Molecular tools
Arnold, B.F., Schiff, K.C., Ercumen, A., et al. (2017) Acute illness among for bathing water assessment in Europe: Balancing social science research
surfers after exposure to seawater in dry- and wet-weather conditions. Am. with a rapidly developing environmental science evidence-base. Ambio
J. Epidemiol. 186: 866–875. 45: 52–62.
Arnold, B.F., Wade, T.J., Benjamin-Chung, J., et al. (2016) Acute gastroen- Pruss, A. (1998) Review of epidemiological studies on health effects from
teritis and recreational water: Highest burden among young US children. exposure to recreational water. Int. J. Epidemiol. 27: 1–9.
Am. J. Public Health 106: 1690–1697.
Quilliam, R.S., Taylor, J., & Oliver, D.M. (2019) The disparity between
Balique, F., Lecoq, H., Raoult, D., et al. (2015) Can plant viruses cross regulatory measurements of E. coli in public bathing waters and the public
the kingdom border and be pathogenic to humans? Viruses 7; 2074–2098. expectation of bathing water quality. J. Environ. Manage. 232: 868–874.
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 383

Rosario, K., Symonds, E.M., Sinigalliano, C., et al. (2009) Pepper mild Tsironi, T.N. & Taoukis, P.S. (2018) Current practice and innovations in
mottle virus as an indicator of fecal pollution. Appl. Environ. Microbiol. fish packaging. J. Aquat. Food Prod. Technol. 27: 1024–1047.
75: 7261–7267. Ventura De Souza, R., Campos, C., Hamilton, L., et al. (2017) A critical
Stachler, E., Kelty, C., Sivaganesan, M., et al. (2017) Quantitative analysis of the international legal framework regulating the microbiologi-
CrAssphage PCR assays for human fecal pollution measurement. Environ. cal classification of bivalve shellfish production areas. Rev. Aquacult. 10:
Sci. Technol. 51: 9146–9154. 1025–1033.
Stone, D.L., Harding, A.K., Hope, B.K., & Slaughter-Mason, S. (2008) Wells, N., Yusufu, D., & Mills, A. (2019) Colourimetric plastic film indi-
Exposure assessment and risk of gastrointestinal illness among surfers. J. cator for the detection of the volatile basic nitrogen compounds associated
Toxicol. Environ. Heal. A-Curr. Issues 71: 1603–1615. with fish spoilage. Talanta 194: 830–836.
Symonds, E.M., Nguyen, K.H., Harwood, V.J., & Breitbart, M. (2018) Pepper Wu, L., Pu, H., & Sun, D.-W. (2019) Novel techniques for evaluating fresh-
mild mottle virus: A plant pathogen with a greater purpose in (waste)water ness quality attributes of fish: A review of recent developments. Trends
treatment development and public health management. Water Res. 144: 1–12. Food Sci. Technol. 83: 259–273.
USEPA Office of Science & Technology (2015) Review of coliphages
as possible indicators of fecal contamination for ambient water quality. Oil pollution
Available online: www.e​pa.go​v/wqc​/revi​ew-co​lipha​ges-p​ossib​le-in​dicat​ Atlas, R. & Bragg, J. (2009) Bioremediation of marine oil spills: When
ors-f​ecal-​conta​m inat​ion-a​mbien​t-wat​er-qu​ality​ (accessed 2 Feb 2019). and when not—The Exxon Valdez experience. Microb. Biotechnol. 2:
Wade, T.J., Pai, N., Eisenberg, J.N.S., & Colford Jr, J.M. (2003) Do US 213–221.
Environmental Protection Agency water quality guidelines for recre- Atlas, R.M. & Hazen, T.C. (2011) Oil biodegradation and bioremediation:
ational waters prevent gastrointestinal illness? A systematic review and A tale of the two worst spills in U.S. history. Environ. Sci. Technol. 45:
meta-analysis. Environ. Health Perspect. 111: 1102–.1109. 6709–6715.
Wang, H., Sikora, P., Rutgersson, C., et al. (2018) Differential removal of Baelum, J., Borglin, S., Chakraborty, R., et al. (2012) Deep-sea bacteria
human pathogenic viruses from sewage by conventional and ozone treat- enriched by oil and dispersant from the Deepwater Horizon spill. Environ.
ments. Int. J. Hyg. Environ. Health 221: 479–488. Microbiol. 14: 2405–2416.
Williams, A.P., Avery, L.M., Killham, K., & Jones, D.L. (2007) Bik, H.M., Halanych, K.M., Sharma, J., & Thomas, W.K. (2012) Dramatic
Persistence, dissipation, and activity of Escherichia coli O157: H7 within shifts in benthic microbial eukaryote communities following the
sand and seawater environments. FEMS Microbiol. Ecol. 60: 24–32. Deepwater Horizon Oil Spill. PLoS ONE 7: e38550.
Delmont, T.O. & Eren, A.M. (2017) Simulations predict microbial
Fish and shellfish spoilage and safety
responses in the environment? This environment disagrees retrospec-
Bellou, M., Kokkinos, P., & Vantarakis, A. (2013) Shellfish-borne viral tively. Proc. Natl. Acad. Sci. USA 114: E8947–E8949.
outbreaks: A systematic review. Food Environ. Virol. 5: 13–23. Hazen, T.C., Dubinsky, E.A., DeSantis, T.Z., et al. (2010) Deep-sea oil
Bintsis, T. (2018) Microbial pollution and food safety. AIMS Microbiol. plume enriches indigenous oil-degrading bacteria. Science 330: 204–208.
4: 377–396. Hazen, T.C., Prince, R.C., & Mahmoudi, N. (2016) Marine oil biodegrada-
Calci, K.R., Meade, G.K., Tezloff, R.C., & Kingsley, D.H. (2005) High- tion. Environ. Sci. Technol. 50: 2121–2129.
pressure inactivation of hepatitis A virus within oysters. Appl. Environ. Hu, P., Dubinsky, E.A., Probst, A.J., et al. (2017) Simulation of Deepwater
Microbiol. 71: 339–343. Horizon oil plume reveals substrate specialization within a complex
Gram, L. & Dalgaard, P. (2002) Fish spoilage bacteria - problems and community of hydrocarbon degraders. Proc. Natl. Acad. Sci. USA 114:
solutions. Curr. Opin. Biotechnol. 13: 262–266. 7432–7437.
Gyawali, P., Fletcher, G.C., McCoubrey, D.-J., & Hewitt, J. (2019) Kimes, N.E., Callaghan, A.V., Suflita, J.M., & Morris, P.J. (2014) Microbial
Norovirus in shellfish: An overview of post-harvest treatments and their transformation of the Deepwater Horizon oil spill—Past, present, and
challenges. Food Control 99: 171–179. future perspectives. Front. Microbiol. 5: 603.
Hassard, F., Sharp, J.H., Taft, H., et al. (2017) Critical review on the pub- King, G.M., Kostka, J.E., Hazen, T.C., & Sobecky, P.A. (2015) Microbial
lic health impact of norovirus contamination in shellfish and the environ- responses to the Deepwater Horizon oil spill: From coastal wetlands to the
ment: A UK perspective. Food Environ. Virol. 9: 123–141. deep sea. Annu. Rev. Mar. Sci. 7: 377–401.
Imamura, S., Kanezashi, H., Goshima, T., et al. (2017) Effect of high-pres- Kleindienst, S., Seidel, M., Ziervogel, K., et al. (2015) Chemical disper-
sure processing on human noroviruses in laboratory-contaminated oysters sants can suppress the activity of natural oil-degrading microorganisms.
by bio-accumulation. Foodborne Pathog. Dis. 14: 518–523. Proc. Natl. Acad. Sci. USA 112: 14900–5.
Murchie, L.W., Cruz-Romero, M., Kerry, J.P., et al. (2005) High pressure Kleindienst, S., Seidel, M., Ziervogel, K., et al. (2016) Reply to Prince et
processing of shellfish: A review of microbiological and other quality al.: Ability of chemical dispersants to reduce oil spill impacts remains
aspects. Innov. Food Sci. Emerg. Technol. 6: 257–270. unclear. Proc. Natl. Acad. Sci. USA 113: E1422–3.
Odeyemi, O.A., Burke, C.M., Bolch, C.C.J., & Stanley, R. (2018) Seafood spoil- Prince, R.C., Coolbaugh, T.S., & Parkerton, T.F. (2016) Oil dispersants do
age microbiota and associated volatile organic compounds at different storage facilitate biodegradation of spilled oil. Proc. Natl. Acad. Sci. USA 113:
temperatures and packaging conditions. Int. J. Food Microbiol. 280: 87–99. E1421.
Olatunde, O.O. & Benjakul, S. (2018) Natural preservatives for extending Redmond, M.C. & Valentine, D.L. (2012) Natural gas and temperature
the shelf-life of seafood: A revisit. Compr. Rev. Food Sci. Food Saf. 17: structured a microbial community response to the Deepwater Horizon oil
1595–1612. spill. Proc. Natl. Acad. Sci. USA 109: 20292–7.
Pacquit, A., Lau, K.T., McLaughlin, H., et al.(2006) Development of a vol- Ron, E.Z. & Rosenberg, E. (2014) Enhanced bioremediation of oil spills in
atile amine sensor for the monitoring of fish spoilage. Talanta 69: 515–20. the sea. Curr. Opin. Biotechnol. 27: 191–194.
Potasman, I., Paz, A., & Odeh, M. (2002) Infectious outbreaks associated Transportation Research Board & National Research Council (2003) Oil
with bivalve shellfish consumption: A worldwide perspective. Clin. Infect. in the Sea III: Inputs, Fates, and Effects. The National Academies Press,
Dis. 35: 921–928. Washington, DC.
384 Chapter 13

Organic pollutants and heavy metals Cole, M., Lindeque, P.K., Fileman, E., et al. (2016) microplastics alter the
properties and sinking rates of zooplankton faecal pellets. Environ. Sci.
Ayangbenro, A.S. & Babalola, O.O. (2017) A new strategy for heavy metal Technol. 50: 3239–3246.
polluted environments: A review of microbial biosorbents. Int. J. Environ.
Res. Public Health 14: 94. De Falco, F., Gullo, M.P., Gentile, G., et al. (2018) Evaluation of micro-
plastic release caused by textile washing processes of synthetic fabrics.
Chakraborty, J. & Das, S. (2016) Molecular perspectives and recent Environ. Pollut. 236: 916–925.
advances in microbial remediation of persistent organic pollutants.
Environ. Sci. Pollut. Res. 23: 16883–16903. Galgani, L., Engel, A., Rossi, C., et al. (2018) Polystyrene microplastics
increase microbial release of marine chromophoric dissolved organic mat-
Chen, F., Li, Z.-L., & Wang, A.-J. (2019) Acceleration of microbial dehalor- ter in microcosm experiments. Sci. Rep. 8: 14635.
espiration with electrical stimulation. In Bioelectrochemistry Stimulated
Environmental Remediation. Springer Singapore, Singapore, pp. 73–92. Gewert, B., Plassmann, M., Sandblom, O., & MacLeod, M. (2018)
Identification of chain scission products released to water by plastic
Cheng, K.Y., Karthikeyan, R., & Wong, J.W.C. (2019) Microbial electro- exposed to ultraviolet light. Env. Sci. Technol. Lett. 5: 272–276.
chemical remediation of organic contaminants: Possibilities and perspec-
tive. In Microbial Electrochemical Technology, pp. 613–640, Elsevier. Geyer, R., Jambeck, J.R., & Law, K.L. (2017) Production, use, and fate of
all plastics ever made. Sci. Adv. 3: e1700782.
Harding, G., Dalziel, J., & Vass, P. (2018) Bioaccumulation of methylmer-
cury within the marine food web of the outer Bay of Fundy, Gulf of Maine. Gregory, M.R. (2009) Environmental implications of plastic debris in
PLoS One 13: e0197220. marine settings—Entanglement, ingestion, smothering, hangers-on,
hitch-hiking and alien invasions. Philos. Trans. R. Soc. Lond. B. Biol. Sci.
Matturro, B., Frascadore, E., & Rossetti, S. (2017) High-throughput 364: 2013–25.
sequencing revealed novel Dehalococcoidia in dechlorinating micro-
bial enrichments from PCB-contaminated marine sediments. FEMS Harrison, J.P., Schratzberger, M., Sapp, M., & Osborn, A.M. (2014) Rapid
Microbiol. Ecol. 93: fix134. bacterial colonization of low-density polyethylene microplastics in coastal
sediment microcosms. BMC Microbiol. 14: 232.
Matturro, B., Ubaldi, C., & Rossetti, S. (2016) Microbiome dynamics of
a polychlorobiphenyl (PCB) historically contaminated marine sediment Hartmann, N.B., Hüffer, T., Thompson R.C., et al. (2019) Are we speaking
under conditions promoting reductive dechlorination. Front. Microbiol. the same language? Recommendations for a definition and categorization
7: 1502. framework for plastic debris. Environ. Sci. Technol. 53: 1039–1047.
Outridge, P.M., Mason, R.P., Wang, F., et al. (2018) Updated global Lobelle, D. & Cunliffe, M. (2011) Early microbial biofilm formation on
and oceanic mercury budgets for the United Nations Global Mercury marine plastic debris. Mar. Pollut. Bull. 62: 197–200.
Assessment 2018. Environ. Sci. Technol. 52: 11466–11477. Long, M., Moriceau, B., Gallinari, M., et al. (2015) Interactions between
Qureshi, A.A., Bulich, A.A., & Isenberg, D.L. (2018) Microtox* toxicity microplastics and phytoplankton aggregates: Impact on their respective
test systems—Where they stand today. In Microscale Testing in Aquatic fates. Mar. Chem. 175: 39–46.
Toxicology, Wells, P.G. & Lee, K. (eds.), CRC Press, pp. 185–199. Lusher, A.L., McHugh, M., & Thompson, R.C. 2013. Occurrence of micro-
Regnell, O. & Watras, C.J. (2019) Microbial mercury methylation in plastics in the gastrointestinal tract of pelagic and demersal fish from the
aquatic environments: A critical review of published field and laboratory English Channel. Mar. Poll. Bull. 67: 94–99.
studies. Environ. Sci. Technol. 53: 4–19. Napper, I.E., Bakir, A., Rowland, S.J., & Thompson, R.C. (2015)
Sanchez-Ferandin, S. (2015) Assessing ecotoxicity in marine environ- Characterisation, quantity and sorptive properties of microplastics
ment using luminescent microalgae: Where are we at? Am. J. Plant Sci. extracted from cosmetics. Mar. Poll. Bull. 99: 178–185.
6: 2502–2509. Oberbeckmann, S., Osborn, A.M., & Duhaime, M.B. (2016) Microbes on a
Senn, D.B., Chesney, E.J., Blum, J.D., et al. (2010) Stable isotope (N, C, bottle: Substrate, season and geography influence community composition
Hg) study of methylmercury sources and trophic transfer in the northern of microbes colonizing marine plastic debris. PLoS One 11: e0159289.
Gulf of Mexico. Environ. Sci. Technol. 44: 1630–1637. Oliveira, M., Almeida, M., & Miguel, I. (2019) A micro(nano)plastic boo-
Sunderland, E.M., Krabbenhoft, D.P., Moreau, J.W., et al. (2009) Mercury merang tale: A never ending story? Trends Anal. Chem. 112: 196–200.
sources, distribution, and bioavailability in the North Pacific Ocean: Porter, A., Lyons, B.P., Galloway, T.S., & Lewis, C. (2018) Role of marine
Insights from data and models. Global Biogeochem. Cycles 23: GB2010. snows in microplastic fate and bioavailability. Environ. Sci. Technol.
Wang, Y. & Tam, N.F.Y. (2019) Microbial remediation of organic pollut- 52:.7111–7119.
ants. In World Seas an Environmental Evaluation, Sheppard, C. (ed.). Reisser, J., Shaw, J., Hallegraeff, G., et al. (2014) Millimeter-sized marine
Acdemic Press, pp. 283–303. plastics: A new pelagic habitat for microorganisms and invertebrates.
Zaferani, S., Pérez-Rodríguez, M., & Biester, H. (2018) Diatom ooze-A PLoS One 9: e100289.
large marine mercury sink. Science 361: 797–800. Romera-Castillo, C., Pinto, M., Langer, T.M., et al. (2018) Dissolved
organic carbon leaching from plastics stimulates microbial activity in the
Pollution by plastics ocean. Nat. Commun. 9: 1430.
Andrady, A.L. (2017) The plastic in microplastics: A review. Mar. Poll. Teuten, E.L., Saquing, J.M., Knappe, D.R.U., et al. (2009) Transport and
Bull. 119: 12–22. release of chemicals from plastics to the environment and to wildlife.
Philos. Trans. R. Soc. B Biol. Sci. 364: 2027–2045.
Austin, H.P., Allen, M.D., Donohoe, B.S., et al. (2018) Characterization
and engineering of a plastic-degrading aromatic polyesterase. Proc. Natl. Thompson, R.C., Olsen, Y., Mitchell, R.P., et al. (2004) Lost at sea: Where
Acad. Sci. 115: E4350–E4357. is all the plastic? Science 304: 838.
Bakir, A., O'Connor, I.A., Rowland, S.J., et al. (2016) Relative importance Woodall, L.C., Sanchez-Vidal, A., Canals, M., et al. (2014) The deep sea is
of microplastics as a pathway for the transfer of hydrophobic organic a major sink for microplastic debris. R. Soc. Open Sci. 1: 140317.
chemicals to marine life. Environ. Pollut. 219: 56–65. World Economic Forum (2016) The New Plastics Economy: Rethinking
Cole, M., Lindeque, P., Fileman, E., et al. (2013) Microplastic ingestion by the future of plastics. Available online: http:​//www​3.wef​orum.​org/d​ocs/W​
zooplankton. Environ. Sci. Technol. 47: 6646–6655. EF_Th​e_New​:Plas​tics_​Econo​my.pd​f (accessed 2 February 2019).
 Microbial Aspects of Marine Biofouling, Biodeterioration, and Pollution 385

Wright, R.J., Gibson, M.I., & Christie-Oleza, J.A. (2019) Understanding Yang, Y., Yang, J., Wu, W.-M., et al. (2015) Biodegradation and miner-
microbial community dynamics to improve optimal microbiome selection. alization of polystyrene by plastic-eating mealworms: Part 2. role of gut
Microbiome 7: 85. microorganisms. Environ. Sci. Technol. 49: 12087–12093.
Wright, S.L. & Kelly, F.J. (2017) Plastic and human health: A micro issue? Yoshida, S., Hiraga, K., Takehana, T., et al. (2016) A bacterium that
Environ. Sci. Technol. 51: 6634–6647. degrades and assimilates poly(ethylene terephthalate). Science 351:
Yang, J., Yang, Y., Wu, W.-M., et al. (2014) Evidence of polyethylene bio- 1196–1199.
degradation by bacterial strains from the guts of plastic-eating waxworms. Zettler, E.R., Mincer, T.J., & Amaral-Zettler, L.A. (2013) Life in the “plas-
Environ. Sci. Technol. 48: 13776–13784. tisphere”: Microbial communities on plastic marine debris. Environ. Sci.
Yang, Y., Yang, J., & Jiang, L. (2016) Comment on “A bacterium that Technol. 47: 7137–7146.
degrades and assimilates poly(ethylene terephthalate)”. Science 353: 759. Zobell, C.E. (1943) The effect of solid surfaces upon bacterial activity. J.
Bacteriol. 46: 39–56.
Chapter 14
Marine Microbial
Biotechnology

Biotechnology is defined broadly as the application of scientific and engineering principles

to provide goods and services through mediation of biological agents. Biotechnological
approaches used for monitoring and remediating pollution are considered in Chapter 13.
The first part of this chapter considers examples of products from marine microorganisms,
including enzymes, pharmaceuticals, antifouling agents, polymers, and biofuels. This is fol-
lowed by discussion of biotechnological processes for the diagnosis and prevention of disease
in marine systems, especially aquaculture. Major economic benefits derive from the indus-
trial and biomedical exploitation of marine microbial processes—their enormous natural
diversity and range of metabolic activities provide great opportunities for future exploitation.
As our appreciation of the great variety of marine microbes grows with the study of different
habitats, so the collection of microbes, their genes, and their products with useful properties
continues to expand.

Key Concepts
• Marine microbes are the source of important enzymes and polymers used in many
branches of industry.
• Microalgae may provide a significant future source of biofuels.
• Many pharmaceuticals and other health products have been obtained from marine
microbes, but their potential has not yet been fully realized commercially.
• Study of bacterial colonization of marine surfaces has led to new approaches to anti-
fouling and prevention of infection through interference with signaling mechanisms.
• Structural components of marine microbes are being exploited in nanotechnology, bio-
electronics, and the development of new materials.
• Marine microbes and their products are increasingly used in vaccines, probiotics, and
immunostimulants for aquaculture of finfish and shellfish, in order to overcome the
problems of disease and resistance to antimicrobial compounds.
• Phages have potential uses in the biological control of bacterial diseases of fish, shell-
fish, and corals.
388 Chapter 14

Enzymes From Marine Microbes Have Many Applications

Among the many beneficial activities of marine microbes shown in Table 14.1, enzymes
feature prominently. Enzymes are widely used by industry and the global market in 2021
is estimated to be about USD 5 billion, growing by 4% a year. Many research institutes
and commercial organizations have developed culture collections of marine bacteria, fungi,
and microalgae, with initial attention being focused on culturable organisms that are easily
collected from near-shore habitats. The full potential of the thousands of marine microbes
already in culture collections has not yet been fully investigated. In recent years, the grow-
ing recognition of the importance of fungi as autochthonous inhabitants of diverse marine
habitats has led to increased focus on their potential for biotechnological exploitation. The
most commonly exploited enzymes are those that degrade polymers, especially proteins and
carbohydrates. Examples of early successes include the production by Vibrio spp. of various
types of extracellular proteases, some of which are tolerant of moderate salt concentrations
and detergents. Another is the extraction of glucanases and other carbohydrate-degrading
enzymes from Bacillus spp. isolated from mud samples.

Table 14.1 Some beneficial effects of activities or products from marine

microbes

Application Examples

Aquaculture Disease diagnostics

Nutritional supplements
Pigments
Probiotics
Vaccines

Cosmetics Liposomes
Polymers
Sunscreens

Environmental protection Bioremediation of pollution

Disease diagnostics
Non-toxic anti-fouling agents
Toxicology bioassays
Waste processing

Food processing Enzymes

Flavors
Preservatives
Texture modifiers

Manufacturing industry Bioelectronics

Polymers
Structural components

Fuels Desulfurization of oil and coal

Oil extraction
Production of biofuels

Nutraceuticals Anti-oxidative compounds

Dietary supplements
Health foods

Pharmaceuticals and biomedical devices Antibacterial, antifungal, and antiviral agents

Anti-tumor and immunosuppressive agents
Biosensors
Drug delivery
Enzymes
Neuroactive agents
Self-cleaning implants

Textiles and papers Enzymes

Surfactants
 Marine Microbial Biotechnology 389

The most successful developments have occurred with the isolation of enzymes from extremo-
philic microorganisms; some of the products that have resulted are shown in Table 14.2.
Enzymes from thermophiles and hyperthermophiles have particular attractions for industrial
processes, which often require high temperatures. Even at milder temperatures, thermophilic
enzymes are beneficial because of their much greater stability. The structural features of
thermophilic enzymes that confer stability and function at high temperatures (sometimes
>100°C) are discussed in Chapter 3.

The use of proteases and lipases as stain removers in detergents is a particularly important
application, since properties that confer high thermostability and activity are often com-
bined with resistance to bleaching chemicals and surfactants used in these products. Some
washing powders incorporate thermophilic cellulases and hemicellulases, which digest loose
fibers and help to prevent “bobbles” on clothes after washing; these enzymes are also impor-
tant in the manufacture of “stone-washed” denim. Enzyme production for biological deter-
gents is a very large market accounting for approximately 30% of the total global production
of enzymes. The first source of these enzymes was soil bacteria of the genus Bacillus, but
enzymes from marine thermophiles have higher temperature optima and superior stability.

Modern food processing uses a wide range of enzymes. Almost all processed foods now rely
on some form of modified starch product for the improvement of texture, control of moisture,
and prolonged shelf life. Amylases hydrolyze α-1,4-glycosidic linkages in starch to produce a

Table 14.2 Some biotechnological applications of extremophilic microbes

Product Applications

Thermophiles and hyperthermophiles

Amylases, pullulanases, lipases, proteases Baking, brewing, food processing

DNA polymerases PCR amplification of DNA

Lipases, pullulanases, proteases Detergents

S-layers Ultrafiltration, electronics, polymers

Xylanases Paper bleaching

Halophiles

Bacteriorhodopsin Bioelectronic devices, optical

switches, photocurrent generators

Compatible solutes Protein, DNA, and cell protectants

Lipids Liposomes (drug delivery, cosmetics)

S-layers Ultrafiltration, electronics, polymers

γ-linoleic acid, β-carotene, cell extracts Health foods, dietary supplements,

food colors, aquaculture feeds

Psychrophiles

Ice nucleating proteins Artificial snow, frozen food processing

Polyunsaturated fatty acids Food additives, dietary supplements

Proteases, lipases, cellulases, amylases Detergents

Alkaliphiles and acidophiles

Acidophilic bacteria Fine papers, waste treatment

Elastases, keratinases Leather processing

Proteases, cellulases, lipases Detergents

Sulfur-oxidizing acidophiles Recovery of metals, sulphur removal

from coal and oil
390 Chapter 14

mixture of glucose, malto-oligosaccharides, and dextrins. All the remaining α-1,4-glycosidic

branches in the products are hydrolyzed by pullulanase. When starch is treated with amylase
and pullulanase simultaneously at high temperatures, it produces higher yields of desired end
products. Pullulanases and other enzymes derived from the hyperthermophilic bacterium
Thermotoga maritima have recently been introduced into food processing. A range of other
carbohydrate-modifying enzymes have been isolated from marine thermophiles.

Agar-degrading enzymes can be isolated from many species of bacteria, including Vibrio,
Pseudomonas, Pseudoalteromonas, Alteromonas, Thalassomonas, Cytophaga, and
Agarivorans. Zones or craters of agar degradation are often seen surrounding colonies of
bacteria isolated from seawater, sediments, algae, and marine invertebrates when grown on
agar plates. Agar, obtained from red seaweeds such as Gracilaria, is added to many foods
such as ice cream, glazes, and processed cheese to improve texture. Industrially, agar-degrad-
ing enzymes (agarases) are used to produce diverse oligosaccharides with a variety of prop-
erties and numerous applications in health foods and cosmetics. Agarases are also widely
used in the molecular biology laboratory as a method of purifying DNA after separation on
agarose gels. The enzymes are also useful for degradation of the cell wall of algae for extrac-
tion of labile substances with biological activities such as unsaturated fatty acids, vitamins,
and carotenoids. As with many enzymes and other proteins isolated from bacteria, it is often
SUNKEN WHALE more convenient for manufacturing to clone the genes responsible and express them in E. coli
i CARCASSES MAY
YIELD NOVEL
or another recombinant host. However, there can be problems with expression and correct
folding of thermostable proteins in mesophilic hosts, and if post-translational modifications
ENZYMES are required for enzyme function, the recombinant enzyme may not function as required.

As described on p.269, when the Enzymes from psychrophilic microbes have many uses and isolates from the deep sea and
carcass of a dead whale reaches the
polar regions have been exploited for use in food-processing applications in which low tem-
seafloor, it passes through sequen-
peratures are required to prevent spoilage, destruction of key ingredients (e.g. vitamins), or
tial stages that support a diverse
community of organisms for many loss of texture. For example, galactosidases from cold-adapted bacteria have been used for
years (Smith et al., 2015). Deep- the removal of lactose from milk to improve digestibility, and xylanases are used in baking
sea whale carcasses thus present a to improve crumb texture in bread. Proteases from psychrophiles can be used for tender-
unique high-pressure, low-temper- izing meat at low temperatures. Significant savings in energy costs result from the use of
ature environment for the selection cold-water laundry detergents that incorporate proteases and lipases active at temperatures
of microbes with potential benefits of 10°C or less.
for industrial applications, but the
difficulties of studying deep-sea
whale falls means that only limited DNA polymerases from hydrothermal vent
investigations have been possible.
Metagenomic analysis of a whale
organisms are widely used in molecular biology
fall at 4240 m in the South Atlantic The discovery of thermostable DNA polymerases used in the polymerase chain reaction
Ocean revealed a number of micro- (PCR) has been one of the most spectacular scientific advances since the discovery of DNA
bial genes with biotechnological itself. The use of the PCR in research, disease diagnostics, and forensic investigations has led
potential (de Freitas et al., 2019).
to a huge market; the market value of PCR enzymes, kits, and equipment was USD 7.4 billion
Many had little sequence identity
in 2017 and is growing by ~6% each year. The original PCR enzyme, Taq polymerase, was
to previously described proteins
and some were recognized as isolated from Thermus aquaticus from a terrestrial hot spring. Although Taq is still the least
having potential uses in biore- expensive and most widely used enzyme for both research and diagnostic uses of the PCR,
mediation of organic pollutants. a number of alternative enzymes are now available that are superior in certain applications,
Often, the skeleton is colonized by and some of these were isolated from marine bacteria (Table 14.3). The choice of enzyme
Osedax worms (Figure 10.9), which depends on the specific activity and sensitivity required with different amounts and lengths of
secrete acid to dissolve calcium template, nucleotide specificity, and various other factors, including cost. Enzymes from the
phosphate in the bones. The root hyperthermophilic vent bacteria have greater thermostability and activity at higher tempera-
of the worm contains symbiotic tures, particularly useful when amplifying GC-rich sequences. They may also have a higher
Oceanospirillales bacteria that can fidelity of replication (due to integral proofreading ability), although careful optimization of
metabolize complex carbon com-
the reaction is required; enzymes such as Vent™ or Pfu are often used in a mixture with Taq.
pounds (Goffredi et al., 2005), but
the specific roles of host cells and
symbionts are not clear (Goffredi
et al., 2007; Miyamoto et al.,
Metagenomics and bioinformatics lead to
2017). Further information about new biotechnological developments
these interactions could lead to
more novel enzymes with interest-
Molecular biology methods have led to exciting new approaches in the search for new prod-
ing properties. ucts from marine microbes. In an analogy with the gold rush, “bioprospectors” can use
oligonucleotide hybridization probes to “pan” for genetic sequences of interest in the hope
 Marine Microbial Biotechnology 391

Table 14.3 Some thermostable DNA polymerases and their sources

DNA Organism Sourcea Half-life Proofreading

Polymerase at 95°C

Taq Thermus aquaticus T (N or R) 40 −

Amplitaq® Thermus aquaticus T (R) 40 −

Vent™ Thermococcus litoralis M (R) 400 +

Deep Vent™ Pyrococcus GB-D M (R) 1380 +

Tth Thermus thermophilus T (R) 20 −

Pfu Pyrococcus furiosus M (N) 120 +

ULTma™ Thermotoga maritima M (R) 50 +

aT = terrestrial hot spring; M = marine hydrothermal vent; (N) = natural;
(R) = recombinant.

of reaping rich rewards. This enables the screening of communities without the need for
culture. For example, comparison of sequence data for genes encoding proteins with a par-
ticular function can allow the identification of consensus sequences. These can be used to
construct complementary oligonucleotide probes, which are then used to search for genes
in DNA amplified from a community of interest. Bioinformatics data-mining tools can be
used to identify gene homologs, and genes can then be cloned and expressed in suitable
hosts. The disadvantage of this method is that it may fail to detect truly novel proteins with
AFTER THE GOLD
sequences unlike those previously described. This limitation is borne out by the results
from metagenomic analyses and genome sequencing of cultured microbes. In both cases, i RUSH—MICROBIAL
NANOPARTICLES
large numbers of open reading frames (ORFs) in the genomes have no match with existing
genes in databases. To overcome this, direct functional screening of metagenomic libraries Nanotechnology—the manipula-
can be employed to identify candidate genes associated with observed enzymatic activities. tion of atoms and molecules for the
PCR primers for amplification of sequences from libraries can be rationally designed using fabrication of products on micro-
sequence features characteristic of genes for particular enzymes and tailored for the pos- scopic scale—is a rapidly expand-
session of properties compatible with specific industrial applications. These approaches are ing field. Biosynthetic Au, Ag, Cd
nanoparticles (typically 10–20 nm
leading to new enzymes such as esterases, lipases, chitinases, amidases, cellulases, alkane
diameter) have many potential
hydrolases, and proteases. Advances in sequencing technology and bioinformatics mean
applications in electronics, new
that we will undoubtedly be able to develop numerous new enzymes carrying out reactions antimicrobial treatments, biological
that are completely unknown to us at present. imaging, and as delivery agents for
DNA vaccination or CRISPR gene-
editing. Manufacture of nanopar-
Polymers from marine bacteria have many applications ticles by top-down size reduction
involves environmentally damaging
Microbial polymers are used in bioremediation, industrial processes, manufacturing indus-
physical and chemical processes,
try, and food processing. The best-exploited compounds are extracellular polysaccharides
so efforts to develop cost-effective
that form the glycocalyx associated with biofilm formation and protection from phagocy- eco-friendly methods are being
tosis. Applications in bioremediation were mentioned in Chapter 13. Other potential appli- developed. This can occur intracel-
cations include underwater surface coatings, bioadhesives, drag-reducing coatings for ship lularly due to reduction of metallic
hulls, dyes, and sunscreens. It has been suggested that the high production (up to 80% cell nanoparticles by enzymes on the
dry weight under appropriate conditions) of poly-hydroxy-β-hydroxyalkanoates by some cytoplasmic membrane or cell wall.
marine bacteria as food reserves could be exploited to produce fully biodegradable plastics. Extracellular synthesis depends on
Oil-derived plastics continue to dominate the market, but bacterially produced plastics are extracellular microbial enzymes
becoming more competitive because of increasing oil prices and the polluting effects of non- such as nitrate reductase—elec-
degradable plastics (Box 13.1). trons are transferred to metal ions,
which precipitate as nanoparticles
(Patil and Kim, 2018). Examples
Microalgae can produce biofuels and edible oils of marine microbes employed for
such synthesis include Rhodococcus
The global economy is dependent on oil for energy and as a source of chemicals. With increas- spp., Marinobacter pelagius, and
ing energy demands from a growing population, limited supplies of fossil fuels, concerns Vibrio alginolyticus (bacteria),
about CO2 emissions and political instability in oil-producing countries, there has been con- Euglena spp. and Nannochloropis
siderable recent interest in using renewable biological materials to produce biofuels. Crops spp. (microalgae), and Aspergillus
sydowii (fungus);(Vala, 2014).
such as corn, sugar, soybeans, sunflowers, or palms can be fermented to alcohol for addition
392 Chapter 14

to gasoline or used to produce biodiesel. However, the costs of production, competition with
food production for use of high-quality land, and the ecological consequences of deforesta-
tion associated with such schemes make the use of land crops unsustainable. Therefore, there
has been much interest in growing microalgae as a source of biodiesel and other biofuels
including hydrogen, methane and ethanol.

Several species of microalgae may be suitable for large-scale culture. Marine examples
include cyanobacteria (e.g. Synechococcus), diatoms (e.g. Phaeodactylum tricornutum and
Chaetoceros muelleri), prymenesiophytes (e.g. Emiliania huxleyi), and chlorophytes (e.g.
Dunaliella tertiolecta and Botryococcus braunii). Careful selection of species is needed
depending on culture conditions and required end-products and the bacteria associated with
the microalgae may influence growth and productivity. Autotrophic microalgae can be grown
in saline or brackish ponds, lagoons or open circulating units with seawater and wastewater,
requiring sunlight and some simple nutrients. Alternatively, they can be grown in photobio-
reactors; these are flow-through transparent glass or plastic columns, panels or tubes exposed
to light (Figure 14.1). Although offering many advantages of better control and prevention of
contamination, photobioreactors require large capital investment. Hybrid systems use photo-
bioreactors to produce large volumes of single-species microalgae, which are then exposed
to nutrient rich open ponds for production of biomass. Growth is very efficient because,
unlike vascular plants, all cells in a microalgal culture are photosynthetic and carbon diox-
ide and nutrients are available directly to the cells. In addition, the concentration of carbon
dioxide in an algal suspension is much higher than that in the atmosphere above a land plant.
Algae therefore have the potential to produce about 30 times more energy per unit area than
arable crops such as soybeans. Microalgae can also be grown heterotrophically, which can be
applied for bioremediation of wastewater or effluent from food processing.

Microalgae can be harvested by flocculation or sedimentation, followed by filtration or cen-

trifugation before drying. At present, downstream processing requires high energy inputs and
accounts for 50–80% of the cost of production. Lysis of the cell wall to extract oils is problematic
due to strengthening polymers. Various methods of lysis have been used, including mechanical
shearing, solvent extraction, thermochemical treatment, microwave, and ultrasound disruption.

The starting points for oil production from microalgae are the lipids and fatty acids found in
their membranes and stored in intracellular compartments as an energy reserve. Algal strains
vary greatly in the types and quantity of lipids produced; selection of appropriate strains for
biodiesel production depends on many factors. Careful control of growth conditions can lead
to the cells containing over 75% of their mass as lipids. These lipids can be converted into a
variety of compounds with similar properties to petroleum products, including diesel.

Figure 14.1 A serpentine photo-

bioreactor Phyco-Flow™ installed at
Cyanotech in Hawaii for culture of
Haematococcus pluvialis for astaxan-
thin production. Image courtesy of
Varicon Aqua, UK.
 Marine Microbial Biotechnology 393

Whilst there have been considerable successes in small-scale production of microalgal bio-
fuels, the costs of production of biodiesel remain much higher than other biofuels such as
soybeans, jatropha, or palm oil to be a viable competitor. Much ingenuity will be needed to
develop this process into a sustainable and cost-effective industry that will deliver the con-
siderable environmental benefits. Supplies of carbon dioxide and sunlight are clearly not a
problem, but organisms require other major nutrients, especially nitrogen and phosphorus.
Microbial nitrogen fixation could perhaps be harnessed for supply to algal culture systems,
but the only source of phosphorus is from the mining of natural deposits and supplies are
running out quickly (this threatens all forms of agriculture). The challenge will be, there-
fore, to develop sustainable systems with almost perfect systems for recycling nutrients after
harvesting.

Microalgal culture is also used for production of high-value human and animal foodstuffs
and additives. Global pressures for the supply of vegetable oils—a major essential compo-
nent of human and animal diets—leads to extensive land use, often resulting in destruction
of natural habitats, such as felling of tropical forests for palm oil production. Microalgae
are particularly important as a source of essential polyunsaturated fatty acids (PUFA) and
carotenoids in fish and shellfish aquaculture and as a component of food supplements and
nutraceuticals.

Genetic modification has been shown to enhance production of triacyl glycerides, omega 3
oils and triterpenoids and shows potential to increase productivity and lower production costs
in microalgal culture. However, scale-up to industrial application will require careful risk
assessment of environmental concerns.

Marine microbes are a rich source of biomedical products

Natural products provide many compounds used in medicine and health promotion.
Throughout much of the twentieth century, the pharmaceutical industry was engaged in
a continual search for new compounds with biological activities that can be exploited
for treatment or prevention of disease. Many of our most successful drugs are secondary
metabolites obtained from plants or bacteria and fungi isolated in the terrestrial envi-
ronment, especially soil. However, in the last few decades, the effort necessary to iso-
late valuable new compounds has become increasingly disproportionate to the returns.
Companies moved away from natural product-based drug discovery and came to rely more
on the use of combinatorial chemistry to synthesize libraries of compounds, and struc-
ture–function analysis to chemically modify already existing compounds. Now, there is
a resurgence of interest in natural products from marine habitats because of the develop-
ment of new tools for underwater exploration and the discovery of the enormous diversity
of marine organisms. Thousands of bioactive compounds isolated from marine microbes
have been investigated for their biological activities, including anti-microbial, anti-cancer,
anti-inflammatory, and antioxidant activities, and a selection of these is shown in Table
14.4. Cyanobacteria and Actinobacteria have been especially rich sources of potentially
useful pharmaceutical agents, with many patents issued. Many of these are non-ribosomal
peptides or complex polyketides, but there is enormous diversity in structure, so that dif-
ferent species and strains within species produce a variety of compounds with differences
in pharmacological properties.

Many bioactive compounds from marine

invertebrates are produced by microbes
For many years, marine invertebrates (especially sponges, bryozoans, and ascidians) have been
investigated as a source of bioactive compounds, and many antibiotics and antitumor agents
have been isolated. Many of these soft-bodied animals have evolved chemical defenses against
predators or to prevent colonization or disease. Different species of these animals are host to
a wide range of microbes, many of which are symbionts (see Chapter 10). It now seems likely
that many of the compounds discovered are produced by the microbes inhabiting the ani-
mals, rather than by the host’s own metabolism. However, although compounds isolated from
sponges and bryozoans often share structural similarity with known microbial metabolites,
394 Chapter 14

Table 14.4 Examples of pharmaceutical compounds from marine microbes

Agent Producing organism Activity

Abyssomicin Verrucispora sp. Inhibits para-amino-benzoic acid synthesis; broad-spectrum antibiotic.

Apratoxins Lyngbya spp. Interfere with signalling and transcription in tumor cells.

Bryostatins “Ca. Endobugula sertula” Inhibits protein kinase C; anti-tumor; treatment of leukemia and
esophageal cancer; may be useful in treatment of Alzheimer’s disease.

Coibamide A Leptolyngbya sp. Induces apotosis in tumor cells.

Cryptophycins Nostoc spp. Inhibits polymerization of tubulin in tumor cells.

Curacin Lyngbya majuscula Blocks mitosis in tumor cells.

Largazole Symploca sp. Inhibits histidine decarboxylase; anti-tumor; anti-epileptic; prevents mood
swings.

Psymberin Uncultivated sponge symbiont Cytotoxic.

Salinosporamide A Salinospora sp. Inhibits protein breakdown in the proteasome; effective against a range of
cancers; anti-malarial; may relieve some symptoms of Alzheimer’s disease.

it is not easy to prove that they are indeed of microbial origin in the symbiosis. One notable
example is the family of compounds known as bryostatins, isolated from the bryozoan Bugula
neritina (Figure 14.2A). Bryostatins are cytotoxic macrolides, which are very likely synthe-
sized by an as-yet-uncultured member of the Gammaproteobacteria named “Ca. Endobugula
sertula.” Bryostatin-1 shows great promise for the treatment of certain types of leukemia and
esophageal cancer in clinical trials and has also been found to have promising potential in
the treatment of Alzheimer’s disease. The compound ecteinascidin is a highly complex and
potent antitumor agent that was isolated from the tunicate Ecteinascidia turbinata (Figure
14.2B) that proved extremely difficult to obtain in sufficient quantities. After many years, a
metagenomic approach identified the biosynthetic gene cluster that is responsible for identify-
ing the 25 enzymes needed for synthesis of ecteinascidin. Sequence analysis confirmed that
these genes belong to an uncultured symbiotic bacterium “Ca. Endoacteinascidia frumen-
tis.” Eribulin is a synthetic analogue of halichondrin B originally isolated from the sponge

Figure 14.2 Examples of successful

pharmaceuticals derived from symbi-
onts of invertebrates. A. Bugula neri-
tina and bryostatin-1. B. Ecteinascidia
turbinata and trabectedin. Image
credits: A: Reprinted from Mans
(2016), CC-BY-4.0. B: Sean Nash,
CC-BY-2.0 via Flickr.
 Marine Microbial Biotechnology 395

Halichondria okadai. It is proving very successful for treatment of breast cancer and liposar-
coma. Obtaining sufficient supplies of the compound from its natural source became impossi-
ble due to shortage of supply, and laboratory synthesis of the complex molecule was eventually
achieved using 67 separate synthetic steps. There is strong evidence that it is produced by a
sponge-associated microbe, but the organism responsible has not yet been identified. If the
phenomenon of production of useful metabolites by symbionts is found to be widespread, it
overcomes the major problem of the need to harvest scarce marine animal life and the conse-
quent disruption of ecosystems. Exploitation of microbial symbionts for pharmaceutical pro-
duction could involve isolation or culture of the microbes, but since many cannot be cultured,
it is likely to be more productive to identify and clone genes for recombinant expression.

Sponges are a particularly good source for new microbial compounds. As noted in Chapter 10,
sponges contain highly diverse microbial communities comprising up to half of the sponge vol-
ume, and there appear to be sponge-specific microbiota that have coevolved for ~600 million
years. Cultivation of sponge-associated microorganisms that produce bioactive compounds is
being actively pursued by several research groups using a range of culture media and conditions;
if successful, this approach has obvious advantages for large-scale inexpensive production of
compounds. Several species of sponge-associated actinomycetes, vibrios, and fungi have been
successfully isolated and shown to produce active compounds with antitumor, antibacterial,
antifungal, anti-inflammatory, immunosuppressive, neuroactive, and other activities. However,
despite employing various strategies to improve isolation, such as the addition of sponge tissue
extracts to media, only a small proportion of the microbes present in sponges have been cultured.
It is possible that many sponge microbes are obligate symbionts and will prove very difficult or
impossible to culture as they may require specific metabolic intermediates from the host sponge.

Metagenomic approaches to identify and clone appropriate genes are increasingly used.
Many of the most interesting secondary metabolites with biotechnological potential are
produced by the action of polyketide synthases. These are giant modular enzymes com-
posed of multiple protein units that contain a coordinated group of active sites. The complex
polyketide molecules are synthesized in a stepwise manner from simple 2-, 3-, or 4-carbon
building blocks. Bacterial polyketide synthases contain many catalytic domains organized
into repeated sets of modules—each module incorporates one carbon unit into the growing
polyketide chain. Several of these enzyme complexes have now been cloned, including that
from “Ca. E. sertula,” responsible for bryostatin synthesis, and a gene cluster responsible for
synthesis of psymberin (a highly potent antitumor agent) from the metagenome of a complex
microbial community in the sponge Psammocinia aff. bulbosa.

With so much potential from the sea,

why are there so few new drugs?
Thousands of potential drugs from marine sources have shown great promise in initial trials,
but only a handful have been fully developed by pharmaceutical companies for approved use in
medicine. After demonstration of in vitro activity, agents must be rigorously tested for toxicity
and potential adverse effects in cell cultures and laboratory animals. Probably only 1 in 1000
agents reaches the next stage of phase 1 clinical trials in human volunteers. The drug must then
go through progressively larger phase 2 and phase 3 clinical trials involving thousands of patients
before it is approved and licensed by government authorities for use as a drug. The process is very
expensive and can take 10–15 years, so at each stage of development the company must evalu-
ate the costs of continuing the process against the potential market returns — this cost–benefit
analysis is especially important as patent protection on new drugs may last less than 20 years.
The estimated cost of bringing a drug to market for human use is estimated at over USD 2 billion,
which explains why so few potential agents reach the final stages of approval for general use.

Study of complex microbial communities

may lead to new antibiotics
The spread of antimicrobial resistance (AMR) among pathogenic microbes is one of the
greatest threats to human health. Many of the antimicrobial antibiotics used in the treatment
of microbial infections were derived from terrestrial fungi (e.g. penicillins) and bacteria (e.g.
396 Chapter 14

tetracyclines, aminoglycosides, and macrolides). Cephalosporin was obtained from the fun-
gus Cephalosporium, isolated near a coastal sewage outfall. Drug-discovery companies have
begun a renewed search for novel microbes in the marine environment. Large-scale rapid-
throughput screening programs are underway in several parts of the world, investigating
marine sediments, sponges, corals, and other habitats. It is likely that the intense competitive
pressure found in dense mixed microbial communities, such as those on algal surfaces and
in corals, will select for microbes producing antimicrobials. Biofilms are likely to be a rich
source, and attention should be given to using new types of bioassays to detect metabolites
that interfere with cell signaling or chemotaxis, as well as those that cause outright growth
inhibition in the standard detection method.

Marine microbes provide various

health-promoting products
A wide range of substances used for health promotion—often referred to as nutraceuticals—
and includes functional foods, probiotics, and nutritional supplements. This is an expand-
ing global market, valued at over USD 380 billion, growing at ~7% each year. New marine
microbial products, especially those from microalgae, are likely to contribute to this area. For
example, the polyunsaturated long-chain omega-3 fatty acid docosahexaenoic acid (DHA)
promotes cardiovascular health and is especially important in fetal brain development; it is
therefore taken by many consumers as a food supplement and added to infant formula feed
and omega-3-enriched eggs and milk. With growing pressure on fish stocks and the pos-
sibility of contamination by methylmercury (see p.377), DHA is produced commercially by
the marine heterotrophic dinoflagellate Crypthecodinium cohnii. A spin-off from this work
was the development of DHA-enriched diets for larval stages in aquaculture. Newly isolated
deep-sea psychrophilic bacteria and thraustochytrids (see p.183) may also be a good source
of polyunsaturated fatty acids and other compounds with health benefits. These sources are
also favored by vegetarians, and demand for non-animal sources of foods and food supple-
ments is growing rapidly. It is necessary to prove that microorganisms with no history of use
in food products are nontoxic and nonpathogenic and to develop suitable methods for large-
scale cultivation.

A major part of the healthfood industry is concerned with antioxidative effects. Research
with bacteria from coral reefs exposed to very high levels of visible light and ultraviolet irra-
diation indicates that they have novel mechanisms of reversing the resultant oxidative dam-
age. This could lead to products that overcome some aspects of the aging process. Marine
bacteria and fungi produce a diverse range of polysaccharides, some of which have been
shown to have anti-inflammatory properties and ability to boost the immune system. These
have potential applications as probiotics for humans and animals (see p.403).

Marine microbes have applications in biomimetics,

nanotechnology, and bioelectronics
In materials science and technology, biomimetics is the term given to the process of “taking
good designs from nature.” Nanotechnology involves the construction of materials and func-
tional objects assembled from the basic molecular building blocks, which offers the potential
of new products ranging from new computer technology to microscopic machines. Marine
microbes are proving to be a rich source for these new technologies. Bacterial and archaeal
S-layers (see p.69) have applications in nanotechnology because of the ordered alignment of
functional groups on the surface and in the pores, which allow chemical modifications and
the binding of molecules in a very precise fashion. Isolated S-layer subunits can self-assemble
as monolayers on solid supports such as lipid films, metals, polymers, and silicon wafers,
which can be used to assemble different “building blocks” such as proteins, nucleic acids, lip-
ids, and glycans. These have a wide range of applications in colloid and polymer science and
the electronics industry. Examples include carriers for vaccine or drug delivery, “biochip”
sensors and biocatalysts. The uniform size and alignment of pores in S-layers also makes
them suitable for use as ultrafiltration membranes. Fusion proteins incorporating a range of
specific properties such as antibodies or enzymes can be developed to allow highly precise
construction and alignment of functional membranes for use in biosensors.
 Marine Microbial Biotechnology 397

As described in Chapter 6, different species of diatoms construct their silica shells in a huge
diversity of forms, with different shapes and surface architecture. Understanding the molecu-
lar basis by which diatoms achieve the construction of their frustules is leading to great
advances in nano-assembly of materials into desired structures. Diatom silica has important
applications in drug delivery, biosensors, tissue engineering, and energy storage devices. The
calcite scales (coccoliths) of coccolithophores can also be modified by adsorption of proteins
or incorporation of metal ions and genetic modification can lead to production of material
with customized surface architecture.

The microscopic rotary motor of bacterial flagella has attracted the interest of engineers for
some years, with suggestions that isolated basal bodies could form the basis of self-propelled
micromachines (e.g. for targeted drug delivery systems). The recent discovery of chemotaxis
and ultrafast swimming in marine bacteria (p.98) could lead to advances in this field, as fur-
ther research leads to an understanding of the molecular basis of these processes.

Magnetotactic bacteria (p.43) produce intracellular magnetic crystals with a stable single-
magnetic domain with uniform structure and purity that is difficult to achieve with chemi-
cal processes. These could have important applications in the electronics industry and for
biomedical applications such as the production of magnetic antibodies, enzyme immobiliza-
tion, cell separation, and in magnetic resonance imaging. Understanding the mechanisms
by which bacteria construct the magnetosomes and the introduction of changes via genetic
modification could lead to the ability to display particular proteins. These bacteria are very
difficult to culture, but there is active research into optimization of culture media and incuba-
tion conditions to improve yields.

Biomolecular electronics relies on the use of native or genetically modified biological mol-
ecules such as proteins, chromophores, and DNA. One of the best examples to date is the
use of bacteriorhodopsin isolated from Halobacterium salinarum (p.155). Bacteriorhodopsin
changes its structure every few milliseconds to convert photons into energy. A chromophore
embedded in the protein matrix absorbs light and induces a series of changes to the opti-
cal and electrical properties of the protein. Bacteriorhodopsin can store many gigabytes of
information in three-dimensional films (holographic memories), and genetic modification
produces proteins with various desirable properties. Bacteriorhodopsin has also been used to
construct artificial retinas, and to make nano-sized solar cells and motion sensors.

Microbial biotechnology has many

applications in aquaculture
Intensive aquaculture now supplies well over half of all fish and shellfish consumed and the
industry continues to grow rapidly as natural stocks are depleted. As discussed in Chapter
11, the health of marine animals depends on interactions between the host, the pathogen, and
the environment. In aquaculture, the most important practical measures to prevent or limit
diseases are those that reduce stress and maintain good hygiene and overall health of the
stock (Table 14.5). These factors are largely a matter of good husbandry and management
practice. However, microbial biotechnology has played a key role in the massive expansion
of intensive global aquaculture over the past few decades, especially in the control of water
quality, development of fast and effective diagnostic procedures, control of disease with anti-
biotics and vaccines, and the development of nutrients and probiotics for stimulating growth
and feed conversion.

Antimicrobial agents are widely used in aquaculture

Although many antimicrobial agents are effective against bacterial pathogens in the labora-
tory, the number of effective treatments for fish or shellfish disease is quite limited for several
reasons. (Note: the term antimicrobials includes true antibiotics, which have a biological
origin, as well as synthetic agents). Firstly, antimicrobials must be proven to be active against
the pathogen but should produce minimal side effects in the host. The best chemotherapeutic
antimicrobials work by targeting a process present in bacteria that is absent or different in
their eukaryotic host. For example, penicillins (such as amoxicillin) target peptidoglycan
398 Chapter 14

Table 14.5 Important factors for optimizing the health of fish and shellfish in aquaculture. Items in bold type show
practices dependent on microbial biotechnology

Health factor Practices

Design and operation of culture systems Separation of hatchery and growing-on facilities
Good management practices and record keeping

Hygiene Disinfection of nets

Protective clothing and equipment
Prompt removal of moribund and dead fish
Bioremediation for improvement of water quality

Nutrition Careful monitoring of optimal growth rates at all stages of the life cycle
Immune stimulants as feed additives
Probiotics for growth promotion and disease prevention

Minimizing stress Avoid netting, grading, overcrowding

Maintain good water quality
Avoid feeding before handling
Use anasthesia during handling
Breed “domesticated” lines of fish

Breaking the pathogen’s life cycle Disinfection of tanks and equipment

Separate fish of different ages
Fallowing sites for 6 months to 1 year

Diagnosis of disease Development of rapid methods (e.g. antibody tests and PCR assays)

Eliminating vertical transmission Production of specific pathogen free brood stock, testing eggs and sperm
for pathogen (e.g. PCR assays)

Preventing geographic spread Licensing system for egg and larval suppliers
Notifiable disease legislation
Movement restrictions from infected sites

Eradication Slaughter policy for notifiable diseases

Government compensation

Antimicrobial treatment Antibiotics and synthetic antimicrobial agents

Sensitivity testing
Limit use to prevent evolution of resistance

Vaccination Immersion, oral, and injectable vaccines

Ensure strains used for vaccines are appropriate for local disease
experience
Well-designed tests for evaluation of efficacy in appropriate species
Assess the need for re-immunization (boosters)
DNA vaccines

Genetic improvement of stock Select for disease resistance traits

Transgenics—incorporation of genes for disease resistance and growth
promotion

synthesis whilst oxytetracycline targets the 30S ribosomal subunit; both of these are unique
to bacteria. Secondly, the agent must reach the site of infection in adequate concentrations to
kill the pathogen or, more usually, inhibit its growth sufficiently to allow the host’s immune
system to eliminate the pathogen. The rate of uptake, absorption, transport to the tissues, and
excretion rates vary greatly among different fish species. Very few agents are transported
across eukaryotic membranes, which is why intracellular pathogens such as Renibacterium
salmoninarum or Piscirickettsia salmonis are particularly difficult to control. Furthermore,
because fish are poikilothermic, uptake and absorption of antimicrobials are very dependent
on temperature. Therefore, the efficacy of a particular compound should be evaluated for
each host–pathogen interaction under various environmental conditions. A third factor is the
need to evaluate the rate of elimination of the drug to ensure that there are no unacceptable
residues in the flesh of fish intended for human consumption. Again, because fish are poiki-
lothermic, the rate of excretion and degradation depends on temperature, so it is necessary
 Marine Microbial Biotechnology 399

to calculate a “degree-day” withdrawal period between the last administration of the antimi-
STEMMING THE TIDE
crobial and the slaughter of the fish for the market. For example, the withdrawal period for
oxytetracycline is 400 degree-days (e.g. 40 days at 10°C or 20 days at 20°C). With the growth i OF ANTIMICROBIAL
RESISTANCE (AMR)
of aquaculture, government agencies and large retailers in many countries now test farmed
IN AQUACULTURE
fish for antimicrobial residues, in the same way that they test meat or milk.
Data for the quantity of antimicro-
Antimicrobial agents are most commonly administered in medicated feed. These are often bials used in aquaculture is scarce,
given to the whole population, but reduction in feeding is often one of the first signs of because of unregulated use in
disease, so infected fish may not receive the appropriate dose of antimicrobial agents from many countries, with large quanti-
medicated feed. Antimicrobials may sometimes be given by immersion of infected fish in a ties used prophylactically with
little regard for sensitivity testing
bath containing the agent, especially for gill and skin infections. Problems arise with both
or proper withdrawal periods. This
routes of administration because of wastage and contamination of the environment. Injection has been especially problematic
is rarely used, except for brood stock and aquarium fish. during the rapid growth of shrimp
and salmon aquaculture; for
Government regulatory authorities require a considerable amount of testing before licensing example, 700 g of antimicrobials
a drug for use, and the high costs of testing deter pharmaceutical companies from introducing are used to produce one tonne of
new agents. The range of treatments is therefore very limited. Regulatory control is strict in shrimp in Vietnam and one tonne
the EU, USA, Canada, and Norway, with only a small number of agents officially approved. of salmon in Chile requires 1500 g.
The routine use of antibiotics for disease prevention or growth promotion has been banned By contrast, only 4.8 g of antimi-
in many countries because of the problems of resistance and residues discussed below, but crobials are required to produce
the same amount in Norway
illegal use continues and in some countries with the most intensive aquaculture (especially in
(Cabello et al., 2016). Aquaculture
Asia and South America), there are no effective controls at all on antimicrobial usage.
leads to dynamic hotspots for
generation of resistance genes and
it is essential that we analyze the
Antimicrobial resistance (AMR) is transfer of AMR between farmed
a major problem in aquaculture fish, the microbial community, and
the environment as our reliance on
Bacteria possess three main strategies for AMR, as shown in Table 14.6. Individual bacterial aquaculture grows (Watts et al.,
isolates often possess more than one resistance mechanism, and individual antimicrobials 2017). Santos and Ramos (2018)
may be affected by different resistance mechanisms in different bacteria. Bacteria possess argue that urgent global action
intrinsic resistance to certain agents because of inherent structural or metabolic features of is required through the “One
the bacterial species—this is almost always expressed by chromosomal genes. This type Health” approach—integrating
of resistance is relatively easy to deal with, but acquired resistance causes major problems human, animal, and environmental
in all branches of veterinary and human medicine. The use of almost every antimicrobial health—to prevent a major crisis in
sooner or later leads to the selection of resistant strains from previously sensitive bacterial the treatment of bacterial infec-
populations. This occurs via spontaneous mutations in chromosomal genes, which occur tious diseases in both human and
veterinary populations
with a frequency of about 10 −7, or by the acquisition of plasmids or transposons. AMR genes
carried on conjugative plasmids (R-factors) may spread rapidly within a bacterial population
and may transfer to other species. Plasmid-borne AMR is frequently encountered in bacteria
pathogenic to fish and shellfish. Emergence of a resistant strain at a fish site renders particular
antimicrobials useless, and the resistance can easily spread until it is the norm for that spe-
cies. Antimicrobials do not cause the genetic and biochemical changes that make a bacterium
resistant, but they do select for strains carrying the genetic information that confers resis-
tance. The more an antimicrobial is used, the greater the selection pressure for resistance to
evolve. If an antimicrobial agent is withdrawn from use, the incidence of strains resistant to it
sometimes declines, because the resistant bacteria now have no advantage and the additional

Table 14.6 Examples of the biochemical basis of acquired bacterial resistance

to antimicrobials used in aquaculture

Strategy Example Mechanism

Modification of the Penicillins Altered penicillin-binding membrane proteins

target binding site
Quinolones Altered DNA gyrase

Enzymic degradation Penicillins β-lactamase production

Reduced uptake or Tetracyclines Altered membrane transport proteins →

accumulation active efflux

Metabolic bypass Sulfonamides Hyperproduction of substrate

(p-aminobenzoic acid)
400 Chapter 14

Figure 14.3 Representation of

pathways of selection and transfer of
antimicrobial resistance (AMR) genes
in an aquaculture system. Sediments
provide an interface for microbial
interactions, leading to transfer of
AMR genes by horizontal gene trans-
fer (HGT).

burden of extra genetic information makes them less competitive. However, this is not always
the case, as plasmids frequently confer resistance to several antibiotics. AMR causes con-
siderable problems in aquaculture. For example, at times between 20 and 30% of cases of
furunculosis in Scottish salmon culture have been due to strains of Aeromonas salmonicida
that are resistant to three or more antimicrobials.

Besides the obvious economic losses caused by inefficiency in disease control, there is great
concern about the risks of antimicrobial usage in aquaculture to human health and environ-
mental quality. Several studies have shown a buildup of resistant strains in sediments under-
neath sea cages in sites with poor water exchange, due largely to the accumulation of uneaten
food and excretion of antimicrobials in the fish feces (Figure 14.3). Transfer of resistance
genes to marine bacteria is known to occur, and AMR bacteria have been isolated from fish
that have escaped from facilities where these agents are used excessively. Experimental stud-
ies have shown transfer to human pathogens and related bacteria, and this raises concerns
about risks of transfer of resistance genes into the gut microbiota of consumers or fish-farm
workers. Antibiotic residues could also prove toxic or cause allergies in some people who
eat the fish. These concerns about aquaculture are part of a general awareness of the folly of
indiscriminate use of antimicrobial agents in medicine, agriculture, and everyday products,
with AMR emerging as a major global threat.

Vaccination of finfish is widely used in aquaculture

Teleost fish possess an efficient immune response and respond to the administration of
microbial antigens by the production of antibodies (B-cell response) and cell-mediated
immunity (T-cell response). The most common method of administering vaccines to
small fish (up to about 15 g) is via brief immersion in a dilute suspension. Particulate
antigens (such as bacteria) probably stimulate immunity after passage across the gills.
Intraperitoneal injection is necessary for reliable protection with some vaccines, especially
viral vaccines and bacterial vaccines that contain soluble components (e.g. most furuncu-
losis vaccines). Injection vaccines are usually administered with an oil adjuvant, which
ensures slow release of the antigen and a heightened immune response. Injection vaccines
have high efficacy but have some drawbacks. Despite devices to convey fish from the water
onto an injection table and the use of repeater syringes (Figure 14.4), injection vaccines
incur high labor costs and they cannot be used on fish weighing less than about 15 g. Unless
carefully managed, the stress associated with crowding, removal from the water, and injec-
tion causes mortalities and may even precipitate infection. The most desirable form of
vaccine is one that can be administered orally. The main difficulty with this approach is
 Marine Microbial Biotechnology 401

Figure 14.4 Vaccination of salmon.

A. Fish are pumped from hold-
ing tanks and anesthetized before
intraperitoneal (ip) injection with a
bacterin suspension, using a repeat-
ing syringe. B. close up of ip injec-
tion. Credit: Kathy Taylor, Aqualife
Services Ltd.

that microbial antigens are degraded in the fish’s stomach and fore-gut before reaching the
gut-associated lymphoid tissue in the hindgut, where the immune response occurs. This is
overcome by microencapsulation of the vaccine in biodegradable polymers such as poly
dl-lactide-co-glycolide. Many commercial vaccines using this, or similar, approaches are
now available.

The simplest type of bacterial vaccine is a bacterin, which consists of a dense culture of bac-
terial cells killed by formalin treatment. Although technically simple, careful attention must
be given to the quality control of media composition and incubation conditions, to ensure that
the bacterin contains the appropriate protective antigens. One of the earliest successes in vac-
cine development was the Vibrio anguillarum vaccine effective against vibriosis in salmon.
The protective antigen in this case seems to be lipopolysaccharide in the bacterial outer
membrane. Commercial vaccines usually incorporate two or more serotypes to allow for
antigenic variation and they often work well in a range of situations. Development of effective
bacterins for Vibrio [now Aliivibrio] salmonicida and Vibrio ordalii also proved relatively
straightforward, but the early success with the vibrios was not repeated with other diseases.
For example, the breakthrough in development of an effective, long-lasting furunculous
vaccine against Aeromonas salmonicida was only achieved after recognition of the crucial
role of extracellular proteases and iron-regulated outer membrane proteins (see p.312), and
manipulation of the culture and formulation conditions to ensure the correct blend of particu-
late and soluble antigens.

Inactivated vaccines can also be produced to protect against viral diseases. Viruses can only
be propagated in cell culture and the cost of production is very high, although some commer-
cial vaccines have been developed against infectious salmon anemia virus (ISAV), infectious
pancreatic necrosis virus (IPNV), and salmon pancreas disease virus (SPDV). To ensure
good protection, careful attention must be paid to the presence of specific antigens and it may
be necessary to administer at a specific life stage of the fish.

Recombinant DNA technology is used to produce vaccines

for diseases caused by viruses and some bacteria
Viruses can only be propagated in cell culture and some bacterial pathogens, notably
Renibacterium salmoninarum and Piscirickettsia salmonis, are slow-growing and difficult
to culture. The cost of production of inactivated vaccines using cell culture is prohibitively
expensive, so attention turned to the use of recombinant DNA technology. Genes important
in virulence—such as viral capsid proteins, bacterial toxins, or surface antigens—can be
cloned and expressed in a recombinant host to produce a subunit vaccine. The most com-
mon method of achieving this is to fuse the gene of interest to a gene for β-galactosidase
or maltose-binding protein, leading to production of a fusion or hybrid protein in a high-
expression system such as E. coli. The fusion protein is produced in large amounts, some-
times up to 30% of total protein. If desired, the carrier protein can be removed and the
cloned microbial antigen can be purified using chromatography, but in practice this is often
not necessary as crude cell lysates make efficient vaccines, thus reducing processing costs.
Sometimes, yeast or insect cell cultures (using a baculovirus vector) are used, especially
if glycosylation of the protein is required. Additional genes for T-cell epitopes (e.g. from
402 Chapter 14

tetanus toxin or measles virus) may also be introduced to enhance the immunostimulatory
potential of a vaccine by stimulating function of the T-cells in the immune system. This
approach has been used to generate vaccines against several bacterial diseases, including
furunculosis (based on the A. salmonicida serine protease), piscirickettsiosis (P. salmonis
OspA membrane protein), and bacterial kidney disease (BKD; R. salmoninarum hemolysin
and metallo-protease). Subunit viral vaccines against ISAV and IPNV are based on capsid
proteins.

Live attenuated vaccines are effective

but not widely used
Live vaccines, in which the virulence of the pathogen is attenuated, are very effective
because the bacterium or virus replicates within the host and delivers antigens over a pro-
longed period. They are also better at stimulating mucosal and cell-mediated immunity
without the need for adjuvants, and more suitable for oral delivery. Many human viral vac-
cines are based on this principle, although live bacterial vaccines have been less favored
because their more complex genomes lead to the possibility of incomplete attenuation or
subsequent reversion to virulence. Recombinant DNA technology allows the deletion and
replacement of specific genes necessary for virulence and survival in vivo in a more con-
trolled and targeted approach to attenuation. Such an approach has been used to construct
a live A. salmonicida vaccine by deletion of the aroA gene, which encodes an essential
amino acid biosynthesis pathway, not present in animals. Allelic replacement of this gene
and subsequent further attenuation guards against the possibility of reversion. In trials,
this vaccine was highly effective, and the vaccine strain could be engineered to deliver
other antigens. A similar approach was used for P. damselae subsp. piscicida by removing
a siderophore gene and for S. iniae by removing the gene for a surface virulence protein.
Live vaccines for IPN have been successful in freshwater salmonid culture, although they
have not proved effective against marine birnaviruses. Even with good evidence of pro-
tection and no evidence of reversion, live vaccines have not found favor with licensing
ARE DNA VACCINES
?
authorities for use in fish, largely prompted by concerns about deliberate release into the
THE KEY TO environment. One exception is the use in Chile of a commercial live vaccine against BKD
FUTURE SUCCESS? in Atlantic salmon, based on administration of Arthrobacter davidanieli. This bacterium
Despite hundreds of experimen- is closely related to R. salmoninarum but is non-pathogenic. Good cross-protection has
tal DNA vaccines, only four have been reported.
been licensed for veterinary use to
date. This is partly because some
have given limited protection, but DNA vaccination depends on fish cells
also because public opinion and expressing a protective antigen
licensing authorities are wary of
the safety of genetically modi- Unlike conventional vaccination, which depends on the administration of an antigen, DNA
fied organisms (GMOs) in the vaccination, also known as genetic immunization, is based on the delivery of naked DNA
environment or in food. However, containing the sequence for a region (epitope) of the protective antigen. The DNA containing
the DNA is not introduced into this sequence is incorporated into a plasmid with appropriate promoters; if this is done cor-
the germ line and the plasmids rectly, the gene will be expressed in the fish host cells. Thus, the fish makes foreign antigens
and the vaccinated host are not internally and then mounts an immune response to them. An alternative strategy is based
considered to be GMOs (Collins
on the expression of an antibody fragment inside the cell that can bind to and inactivate
et al., 2019). DNA vaccines are
the pathogen. Fish cells efficiently express foreign proteins encoded by eukaryotic expres-
thought by most experts to have
very low risks and with better sion vectors. The first experiments using this technology were carried out using glycopro-
understanding and acceptance tein genes of the rhabdoviruses causing viral hemorrhagic septicemia (VHS) and infectious
they hold great potential in hematopoietic necrosis (IHN). These were highly successful, and large-scale trials have
aquaculture, especially for control confirmed their efficacy in some, but not all, fish species tested. A DNA vaccine against
of diseases caused by viruses, salmon pancreas disease virus (SPDV, alphavirus) has also recently been approved for use
eukaryotic parasites, and hard-to- in aquaculture. However, DNA vaccines for other fish viruses such IPNV, ISAV, and halibut
culture bacteria. With advances nodavirus) have produced mixed results, and refinement of the vector plasmids and delivery
in genomics, transcriptomics, system are required. In experimental trials, many of these vaccines have been effective after
and proteomics, it will become
intramuscular injection. An alternative method of gene delivery into the skin or muscle is to
increasingly possible to better
use a “gene gun,” which fires tiny gold particles coated with the DNA into the tissue. Issues
identify the key correlates of pro-
tective host responses for targeted of longevity of protection, stimulation of the correct immune responses, and safety need
vaccine design (Dalmo, 2018). further investigation.
 Marine Microbial Biotechnology 403

Probiotics, prebiotics, and immunostimulants WHAT ARE

are widely used in aquaculture
Interest in this area has occurred largely because of the need to reduce antibiotic usage
? PROBIOTICS AND
PREBIOTICS?
due to concerns over antimicrobial resistance and residues. Probiotics have been very suc- The word “probiotic” is derived
cessful in other branches of agriculture such as poultry and pig farming and commer- from the Latin “for life.” In its
cial production is now of considerable economic importance to the animal feed industry. most common usage, probiotics
Numerous studies in various species of finfish have shown that addition of certain bacteria are defined as live microbial feed
to feeds leads to disease resistance or other benefits. In general, probiotic strains have been supplements that stimulate health
isolated from naturally occurring microbiota of fish. They are distinguished from vaccines by inducing beneficial changes in
the intestinal microbiota. Merrifield
because they do not require participation of the host immune system and may act directly
et al. (2010) propose a wider defi-
by inhibition of pathogens via production of antimicrobial metabolites or bacteriocins, nition in aquaculture, stating that
or by competition for nutrients or sites for colonization of the gut mucosa. Besides direct a probiotic effect may be based on
effects on pathogens or enhancement of nutrition, some probiotic products (in the broadest direct benefits to the animal host
sense of the term) have immunostimulatory effects. A range of preparations incorporat- (e.g. immunostimulation, produc-
ing lipopolysaccharides, glucans, and other microbial components have been shown to tion of inhibitors against patho-
enhance host defenses. Whilst many studies lack scientific rigor and proof of the mecha- gens, reduced stress response, and
nisms of disease prevention by specific challenge, evidence of their efficacy is provided by improved gastrointestinal mor-
major reductions in antibiotic usage. Some statistically rigorous studies have shown that phology) as well as benefits to the
probiotics improve the survival and growth of larvae and lessen the survival of pathogenic fish farmer or consumer (such as
improved fish appetite, growth, or
bacteria in the gut.
quality). Prebiotics are dietary com-
ponents that stimulate the growth
Many probiotic agents show antagonistic activity against pathogens in vitro, but this is no of beneficial bacteria. A range of
guarantee that they have this effect in vivo. Most antagonistic compounds are secondary carbohydrates, such as inulin, and
metabolites produced during the stationary phase in culture, and it is unknown whether they complex oligosaccharides are used
are produced in significant amounts in vivo. Electron microscopy has been used to show how and many studies show health
probiotic bacteria may compete with pathogens for colonization sites. Some probiotics may benefits (Ringo et al., 2010).
produce beneficial dietary compounds such as vitamins, antioxidants, and lipids or may pro-
vide their host with digestive enzymes, enabling the degradation of complex nutrients such
as chitin, starch, or cellulose. These factors are becoming very important in the development
of sustainable aquaculture, as there is a need to use alternative nutrient sources to replace the
industry’s reliance on fishmeal.

The gastrointestinal tract of fish and shellfish larvae becomes colonized by bacteria present
in the water; therefore, control of water quality in the early larval stages is very important to
prevent colonization by pathogens. Agents may be added to the water, incorporated into pel-
leted feeds, or used to enrich live feed such as rotifers or Artemia. A variety of commercial
products incorporating dried endospores of Bacillus spp. have proved effective in limiting
disease. These are cheap to manufacture, stable, and easy to administer. Other agents, such
as lactic acid bacteria, Roseobacter spp., Pseudoalteromonas, Pseudomonas, nonpathogenic
Vibrio spp., yeasts, and microalgae are effective, but these are more difficult to distribute
commercially as live cultures. There does seem to be a specific indigenous intestinal microbi-
ome in marine fish, composed mainly of Gram-negative, aerobic, anaerobic, and facultatively
anaerobic bacteria. The most common genera isolated by culture from the gut of marine fish
are Vibrio, Photobacterium, Pseudomonas, and Acinetobacter. Recent application of molec-
ular methods shows that various other organisms are important and that the microbiome is
very variable dependent on genetic, nutritional, and environmental factors.

There is a highly developed industry supplying a great range of products promoted as immu-
nostimulants for use in shrimp, prawn, and lobster culture. Some products are promoted as
vaccines, although this is usually considered an incorrect use of the term since it should
only be applied to an agent that induces protection through long-lasting immune memory
that is dependent on primary challenge with an antigen and the stimulation of clones of
specific lymphocytes. Invertebrates lack this system, although crustaceans do possess active
cells such as hemocytes that are responsible for inflammatory-type reactions, phagocytosis,
and the killing of microbes via oxidative burst or microbicidal proteins. Immune stimula-
tion can occur by enhancing nonspecific complement activation, phagocytosis, or cytokine
production. A range of treatments, including live or killed bacteria, glucans, lipopolysac-
charide, extracts of yeast, and algal cell walls, have been used, and promising effects have
been reported.
404 Chapter 14

BOX 14.1 RESEARCH FOCUS

“The enemy of my enemy is my friend”—phage therapy for marine diseases

Rediscovery of an old idea. Soon after phages were discovered in strains were sensitive to two phages. In an experimental infection,
1915 (p.212), it was proposed that they might be useful for treat- 100% of fish survived intraperitoneal injection with L. garviae
ing bacterial infections of humans (reviewed by Wittebole et al., when phage was injected at the same time, whereas the survival
2014). Over many years, phage therapy (PT) research institutes in rate was only 10% in fish injected with the bacteria alone. Partial
Tbilisi and Wroclaw developed methods for isolating and propa- (50%) protection was observed if 24 h elapsed between bacterial
gating phages and using them to treat a variety of human infec- infection and injection of the phage. Imbeault et al. (2006) showed
tions. Phages were also employed for the treatment of cholera in that addition of phages to the water delayed the onset of furun-
India between 1925 and 1934, when vibriophages were admin- culosis caused by Aer. salmonicida in brook trout (Salvelinus
istered at the start of cholera outbreaks and released into com- fontinalis). However, Verner-Jeffreys et al., 2007) showed that
munity drinking water for prophylaxis, resulting in significant Atlantic salmon (Salmo salar) injected with phage immediately
reduction in death rates (Nelson et al., 2009) and there is now after injection of Aer. salmonicida died more slowly, but even-
interest in reviving this approach (Bhandare et al., 2018). In the tual mortality was the same as the control treatment and there
1930s, pharmaceutical companies in the USA began to produce was some evidence of evolution of resistance to specific phage
phages commercially, but there were problems with standardiza- strains. These authors concluded that PT may not be suitable for
tion and poor design of clinical trials. Furthermore, the discovery such a highly virulent pathogen, although the injection challenge
and rapid success of antibiotics meant that the idea of PT was used in this study is not representative of the natural infection
quickly forgotten in Western countries in the 1940s, although it processes. There have been several promising studies showing the
continued to be used successfully in Georgia, Russia, and Poland. potential of phages for treatment or prevention of Vibrio diseases
In the 1980s, the problems of antimicrobial resistance (AMR) (Kalatzis et al., 2018). Higuera et al. (2013) showed that juve-
prompted scientists to reassess the potential of PT and carefully nile Atlantic salmon were protected from experimental infection
designed trials have proved that it can be very successful in treat- with V. anguillarum in the laboratory and in fish farm conditions
ment of bacterial infections of mammals. Although numerous when phages were added to the tank water. The most promising
US companies are developing PT and there is increased open- application is probably the addition of the phages to the water
ness towards PT among Western medical professionals, much to prevent infection during larval rearing, when vaccines would
more research and robust clinical trials are needed before it will be ineffective because the immune system is not fully devel-
be widely accepted for human use (Górski et al., 2018). In con- oped. Phages can also exclude the pathogenic bacteria from the
trast, there are many encouraging developments in the use of water without harming the beneficial microbiota and remain in
PT in agriculture and aquaculture (Gon Choudhury et al., 2017), the water for a week or more, providing a prophylactic measure
although the lack of standardized research approaches and incon- (Higuera et al., 2013). Rørbo et al. (2018) showed that a broad-
sistent reporting make it difficult to compare results from various host-range vibriophage reduced mortality of cod and turbot lar-
studies (Richards, 2014). vae challenged experimentally with four different V. anguillarum
strains. For fry and juvenile fish, continuous delivery of phages
Advantages and disadvantages of phages as therapeutic and bio- via coated feed pellets would provide an easy method of routine
control agents. Phages usually have a narrow host range directed administration. Christiansen et al. (2014) showed effective uptake
against specific pathogens, and they should therefore not affect the of Flavobacterium psychrophila phages after oral administration
normal host microbiota. They are self-perpetuating in the presence to rainbow trout (Oncorhynchus mykiss). Huang and Nitin (2019)
of susceptible bacteria, so repeated administration is not necessary. recently developed an edible protein-based biopolymer coat-
Also, phages coevolve with their host bacteria, so even if a bacterial ing for feed pellets, which proved to be a cost-effective delivery
strain acquires resistance, it should be easy to isolate new phages method for phages.
that are infective. There is also some evidence that accumulated
phage resistance may lead to dominance of less virulent phenotypes PT in aquaculture of crustaceans. The need to prevent cata-
(Kalatzis et al., 2018). However, this variability means that if the strophic infections by Vibrio spp., which are the primary opportu-
exact strain of bacterium causing an infection is unknown, it is nec- nistic pathogens of shrimps and prawns, could make the effort of
essary to inoculate with a “cocktail” of phages. Additional prob- maintaining collections of different phage cocktails commercially
lems can occur because animals may mount an immune response to worthwhile. Although there are a number of anecdotal reports
the phage proteins and the phage virions may not reach all parts of since the 1990s, there are few published reports of rigorously
the body. Precautions are also needed to avoid the use of temperate conducted field trials in scientific journals. In India, Vinod et al.
phages, which could raise the potential of transfer of virulence or (2006) isolated several lytic phages of V. harveyi from farm or
antibiotic resistance genes (Flegel et al., 2005). hatchery water used to rear Penaeus monodon shrimp. In labora-
tory microcosm experiments, addition of phages to water produced
PT in aquaculture of finfish. The first documented use of PT was a reduction in the bacterial counts from 106 to 103 CFU mL -1 and
by Nakai et al. (1999) to control a serious opportunistic disease 80% of larvae survived, compared with 25% in the control. In the
problem caused by Lactococcus garvieae affecting yellowtail same laboratory, Karunasagar et al. (2007) showed a dose-depen-
(Seriola quinqueradiata). They found that over 90% of L. garviae dent effect on the formation of V. harveyi in tanks. In Australia,
 Marine Microbial Biotechnology 405

BOX 14.1 RESEARCH FOCUS

Crothers-Stomps et al. (2010) isolated a collection of phages

from two virus families with lytic activity against V. harveyi and
showed their ability to control vibriosis in culture of larvae from
the tropical rock lobster Panulirus ornatus. Lomelí-Ortega and
Martínez-Díaz (2014) achieved successful control of experimen-
tal V. parahaemolyticus infection of whiteleg shrimp Litopenaeus
vannamei larvae.

Could phages be used to control coral disease? The threat

to coral reefs from infectious diseases, especially those caused
by bacteria, was discussed in Chapter 11. Control of disease by
large-scale use of antibiotics is impractical and the lack of an
antibody response in corals means that vaccination is not possi-
ble—although Teplitski and Ritchie (2009) suggest that it may be
possible to prime the immune defenses of certain corals. Eugene
Rosenberg and coworkers at the University of Tel Aviv, Israel,
have pioneered the concept of PT for control of bacterial coral
diseases. In aquarium experiments, Efrony et al. (2007) isolated
phages of the pathogen Thallassomonas loyaena from the Red
Sea and showed that addition of one type of phage at a density
of 10 4 –10 6 PFU mL -1 protected the coral Favia favus against
infection caused by incubation in water taken from an aquarium
containing diseased coral. The phages increased in number, Figure 14.5 Representation of hypothetical phage treat-
and levels remained high even after water in the aquarium was ment of diseased reef (see text for details).
changed, indicating that the phage replicates and remains associ-
ated with coral tissue. The T. loyaena BA3 phage was character- Would it really be technically feasible to treat a disease outbreak
ized and shown to prevent transmission of the disease to healthy to stop it spreading to surrounding parts of the reef? Figure 14.5
corals (Efrony et al., 2009). Similarly, the coral Pocillopora shows that to treat an area of diseased coral of 10 4 m 2 in water
damicornis was protected against infection and reinfection by V. 10 m deep (a water volume of 105 m 3) with an effective level
coralliilyticus using phages specific for this bacterium (Efrony et of phage of ~105 mL -1—this is 100 times the protective level in
al., 2007). Some coral biologists argue that the proposal to treat the Atad et al. trial—would require only 10 L of phage lysate,
reefs in this way is impractical and environmentally unsound. an amount easily obtainable from a laboratory fermenter. In the
Thus, although additional laboratory experiments have shown USA, Kellogg (2007) suggested that prospects are good for using
that phage infection have the potential to control bacterial infec- phages to treat diseased corals in the seriously impacted reefs of
tions in laboratory-reared corals (Cohen et al., 2013; Jacquemot the Florida Keys, although this idea has not been implemented.
et al., 2018), there appears to be only one example of environ- Whilst generally supportive of the concept, Teplitski and Ritchie
mental application of PT for corals. A small-scale pilot field (2009) argue that a better understanding of the role of vibrios as
study on a threatened reef infected in the Red Sea gave promising commensal bacteria in corals is needed before PT can be used in
results (Atad et al., 2012). Treatment of diseased F. fava corals the field, especially the effect of phages on other members of the
with phage BA3 inhibited transmission of T. loyaena infection to coral holobiont. In a situation in which some reefs are in critical
nearby corals—only 5% of the healthy corals became infected danger of being lost forever (see Chapter 11) because of bleaching
when placed near phage-treated diseased corals, whereas and disease, further field research is urgently needed to decide
61% of healthy corals were infected in the no-phage control. whether PT in corals is a feasible and worthwhile approach.
406 Chapter 14

Conclusions
The untapped hidden treasure available from microbes in the sea is immense. Advances in
molecular biology, especially high-throughput sequencing, powerful bioinformatics tools,
and functional screening methods mean that many new potential products and processes
for industry and medicine will be discovered in the coming years. Despite the great oppor-
tunities for biotechnology, it is disappointing that only a tiny fraction of the discoveries has
been successfully exploited. Promising research leads from laboratory studies are often not
fully exploited because of the high costs of developing new products and processes and the
risk-averse nature of many commercial enterprises and government authorities. Furthermore,
most of the successful developments have involved isolation of marine microbes and exploi-
tation of their products and processes in land-based industries. Large-scale direct application
of marine microbial biotechnology in the ocean setting should play a major role in ensuring
new sources of energy and sustainable aquaculture to feed the world’s growing population,
and it could help to mitigate some of the environmental damage caused by human activi-
ties. We will face important economic. ethical and environmental issues to ensure that the
exploitation of these discoveries can be properly harnessed for the benefit of humanity and
our planet.

Lian, J., Wijffels, R.H., Smidt, H., & Sipkema, D. (2018) The effect of
References and further reading the algal microbiome on industrial production of microalgae. Microb.
Biotechnol. 11: 806–818.
Enzymes
Milano, J., Ong, C., Masjuki, H.H., et al. (2016) Microalgae biofuels as an
Beygmoradi, A. & Homaei, A. (2017) Marine microbes as a valu- alternative to fossil fuel for power generation. Renew. Sust. Energy Rev.
able resource for brand new industrial biocatalysts. Biocatal. Agric. 58: 180–197.
Biotechnol. 11: 131–152.
Shuba, E.S. & Kifle, D. (2018) Microalgae to biofuels: ‘promising’ alterna-
Bonugli-Santos, R.C., dos Santos Vasconcelos, M.R., Passarini, M.R.Z., et tive and renewable energy, review. Renew. Sustain. Energy Rev. 81: 743–755.
al. (2015) Marine-derived fungi: Diversity of enzymes and biotechnologi-
Vinayak, V., Manoylov, K.M., Gateau, H., et al. (2015) Diatom milking: A
cal applications. Front. Microbiol. 6: 269.
review and new approaches. Mar. Drugs 13: 2629–2665.
de Freitas, R.C., Marques, H.I.F., da Silva, M.A.C., et al. (2019) Evidence
Xue, Z., Yu, W., Liu, J., et al. (2018) Edible oil production from microal-
of selective pressure in whale fall microbiome proteins and its potential
gae: A review. Eur. J. Lipid Sci. Technol. 120: 1700428.
application to industry. Mar. Genom. 45: 201–227.
Goffredi, S.K., Johnson, S.B., & Vrijenhoek, R.C. (2007) Genetic diversity
and potential function of microbial symbionts associated with newly dis-
Bioactive compounds and pharmaceuticals
covered species of Osedax polychaete worms. Appl. Environ. Microbiol. Bibi, F., Faheem, M., Azhar, E.I., et al. (2017) Bacteria from marine
73: 2314–2323. sponges: A source of new drugs. Curr. Drug Metab. 18: 11–15.
Goffredi, S.K., Orphan, V.J., Rouse, G.W., et al. (2005) Evolutionary Dias, D.A., Urban, S., & Roessner, U. (2012) A historical overview of nat-
innovation: A bone-eating marine symbiosis. Environ. Microbiol. 7: ural products in drug discovery. Metabolites 2: 303–336.
1369–1378.
Gerwick, W.H. & Fenner, A.M. (2013) Drug discovery from marine
Miyamoto, N., Yoshida, M., Koga, H., & Fujiwara, Y. (2017) Genetic microbes. Microb. Ecol. 65: 800–806.
mechanisms of bone digestion and nutrient absorption in the bone-eating
ul Hassan, S.S., Anjum, K., Abbas, S.Q., et al. (2017) Emerging biophar-
worm Osedax japonicus inferred from transcriptome and gene expression
maceuticals from marine actinobacteria. Environ. Toxicol. Pharmacol.
analyses. BMC Evol. Biol. 17: 17.
49: 34–47.
Parte, S., Sirisha, V.L., & D’Souza, J.S. (2017) Biotechnological applica-
Mans, D.R. (2016) Exploring the global animal biodiversity in the search
tions of marine enzymes from algae, bacteria, fungi, and sponges. Adv.
for new drugs–marine invertebrates. J. Transl. Sci. 2: 170–179.
Food Nutr. Res. 80: 75–106.
Popovic, A., Tchigvintsev, A., Tran, H., et al. (2015) Metagenomics as Nanotechnology and bioelectronics
a tool for enzyme discovery: Hydrolytic enzymes from marine-related
metagenomes. In: Prokaryotic Systems Biology. Springer, Cham, pp. 1–20. Ali, I., Peng, C., Khan, Z.M., & Naz, I. (2017) Yield cultivation of mag-
netotactic bacteria and magnetosomes: A review. J. Basic Microbiol. 57:
Smith, C.R., Glover, A.G., Treude, T., et al. (2015) Whale-fall ecosystems: 643–652.
Recent insights into ecology, paleoecology, and evolution. Ann. Rev. Mar.
Sci. 7: 571–596. Charrier, M., Li, D., Mann, V.R., et al. (2019) Engineering the S-Layer of
Caulobacter crescentus as a foundation for stable, high-density, 2D living
Biofuels and edible oils materials. ACS Synth. Biol. 8: 181–190.
Beacham, T.A., Sweet, J.B., & Allen, M.J. (2017) Large scale cultivation Farjadian, F., Moghoofei, M., Mirkiani, S., et al. (2018) Bacterial com-
of genetically modified microalgae: A new era for environmental risk ponents as naturally inspired nano-carriers for drug/gene delivery and
assessment. Algal Res. 25: 90–100. immunization: Set the bugs to work? Biotechnol. Adv. 36: 968–985.
Gupta, A., Barrow, C.J., & Puri, M. (2012) Omega-3 biotechnology: Gordon, R., Losic, D., Tiffany, M.A., et al. (2009) The glass menagerie:
Thraustochytrids as a novel source of omega-3 oils. Biotechnol. Adv. 30: Diatoms for novel applications in nanotechnology. Trends Biotechnol. 27:
1733–1745. 116–127.
 Marine Microbial Biotechnology 407

Li, Y.-T., Tian, Y., Tian, H., et al. (2018) A review on bacteriorhodopsin- Pérez-Sánchez, T., Mora-Sánchez, B., & Balcázar, J.L. (2018) Biological
based bioelectronic devices. Sensors 18: 1368. approaches for disease control in aquaculture: Advantages, limitations and
Mishra, M., Arukha, A.P., Bashir, T., et al. (2017) All new faces of dia- challenges. Trends Microbiol. 26: 896–903.
toms: Potential source of nanomaterials and beyond. Front. Microbiol. 8: Ringø, E., Olsen, R.E., Gifstad, T.Ï., et al. (2010) Prebiotics in aquaculture:
1239. A review. Aquac. Nutr. 16: 117–136.
Patil, M.P. & Kim, G.-D. (2018) Marine microorganisms for synthesis of Rodiles, A., Rawling, M.D., Peggs, D.L., et al. (2018) Probiotic appli-
metallic nanoparticles and their biomedical applications. Colloids Surf. B: cations for finfish aquaculture. In: Probiotics and Prebiotics in Animal
Biointerfaces 172: 487–495. Health and Food Safety. Springer, Cham, pp. 197–217.
Ragni, R., Cicco, S.R., Vona, D., & Farinola, G.M. (2018) Multiple routes Salonius, K., Siderakis, C., MacKinnon, A.M., & Griffiths, S.G. (2005)
to smart nanostructured materials from diatom microalgae: A chemical Use of Arthrobacter davidanieli as a live vaccine against Renibacterium
perspective. Adv. Mater. 30: 1704289. salmoninarum and Piscirickettsia salmonis in salmonids. Dev. Biol. 121:
Saeedi, P., Moosaabadi, J.M., Sebtahmadi, S.S., et al. (2012) Potential 189–197.
applications of bacteriorhodopsin mutants. Bioengineered 3: 326–328. Santos, L. & Ramos, F. (2018) Antimicrobial resistance in aquacul-
Skeffington, A.W. & Scheffel, A. (2018) Exploiting algal mineralization for ture: Current knowledge and alternatives to tackle the problem. Int. J.
nanotechnology: Bringing coccoliths to the fore. Curr. Opin. Biotechnol. Antimicrob. Agents 52: 135–143.
49: 57–63. Sayes, C., Leyton, Y., & Riquelme, C. (2018) Probiotic bacteria as an
Sleytr, U.B., Schuster, B., Egelseer, E.-M., & Pum, D. (2014) S-layers: healthy alternative for fish aquaculture. Antibiotic Use in Animals. doi:
Principles and applications. FEMS Microbiol. Rev. 38: 823–864. 10.5772/intechopen.71206.

Vala, A.K. (2014) Exploration on green synthesis of gold nanoparticles by Watts, J.E.M., Schreier, H.J., Lanska, L., & Hale, M.S. (2017) The rising
a marine-derived fungus Aspergillus sydowii. 34: 194–197. tide of antimicrobial resistance in aquaculture: Sources, sinks and solu-
tions. Mar. Drugs 15: e158.
Vargas, G., Cypriano, J., Correa, T., et al. (2018) Applications of magneto-
tactic bacteria, magnetosomes & magnetosome crystals in biotechnology Phage therapy
and nanotechnology: Mini-review. Molecules 23: 2438.
Atad, I., Zvuloni, A., Loya, Y., & Rosenberg, E. (2012) Phage therapy of
Aquaculture vaccines, probiotics the white plague-like disease of Favia favus in the Red Sea. Coral Reefs
and antimicrobial resistance 31: 665–670.
Bhandare, S., Colom, J., Baig, A., et al. (2018) Reviving phage therapy for
Adams, A. & Thompson, K.D. Biotechnology offers revolution to fish
the treatment of cholera. J. Infect. Dis. 219: 786–794.
health management. Trends Biotechnol. 24: 201–205.
Christiansen, R.H., Dalsgaard, I., Middelboe, M., et al. (2014) Detection
Cabello, F.C., Godfrey, H.P., Buschmann, A.H., & Dölz, H.J. (2016)
and quantification of Flavobacterium psychrophilum-specific bacterio-
Aquaculture as yet another environmental gateway to the development
phages in vivo in rainbow trout upon oral administration: Implications for
and globalisation of antimicrobial resistance. Lancet Infect. Dis. 16:
disease control in aquaculture. Appl. Environ. Microbiol. 80: 7683–7693.
e127–133.
Cohen, Y., Pollock, F.J., Rosenberg, E., & Bourne, D.G. (2013)
Charoonnart, P., Purton, S., Saksmerprome, V., et al. (2018) Applications
Phage therapy treatment of the coral pathogen Vibrio coralliilyticus.
of microalgal biotechnology for disease control in aquaculture. Biology
Microbiologyopen 2: 64–74.
(Basel) 7: 24.
Crothers-Stomps, C., Høj L, Bourne, D.G., et al. (2010) Isolation of lytic
Collins, C., Lorenzen, N., & Collet, B. (2019) DNA vaccination for finfish
bacteriophage against Vibrio harveyi. J. Appl. Microbiol. 108: 1744–1750.
aquaculture. Fish Shellfish Immunol. 85: 106–125.
Efrony, R., Atad, I., & Rosenberg, E. (2009) Phage therapy of coral white
Dadar, M., Dhama, K., Vakharia, V.N., et al. (2017) Advances in aquacul-
plague disease: Properties of phage BA3. Curr. Microbiol. 58: 139–145.
ture vaccines against fish pathogens: Global status and current trends. Rev.
Fish. Sci. Aquac. 25: 184–217. Efrony, R., Loya, Y., Bacharach, E., & Rosenberg, E. (2007) Phage ther-
apy of coral disease. Coral Reefs 26: 7–13.
Dalmo, R.A. (2018) DNA vaccines for fish: Review and perspectives on
correlates of protection. J. Fish Dis. 41: 1–9. Flegel, T.W., Pasharawipas, T., Owens, L., & Oakey, H.J. (2005) Evidence
for phage-induced virulence in the shrimp pathogen Vibrio harveyi. Dis.
Dittmann, K.K., Rasmussen, B.B., Castex, M., et al. (2017) The aquacul-
Asian Aquac. V: 329–337.
ture microbiome at the centre of business creation. Microb. Biotechnol.
10: 1279–1282. Gon Choudhury, T., Tharabenahalli Nagaraju, V., et al. (2017) Advances
in bacteriophage research for bacterial disease control in aquaculture. Rev.
Hølvold, L.B., Myhr, A.I., & Dalmo, R.A. (2014) Strategies and hurdles
Fish. Sci. Aquac. 25: 113–125.
using DNA vaccines to fish. Vet. Res. 45: 21.
Górski, A., Międzybrodzki, R., Łobocka, M., et al. (2018) Phage therapy:
Merrifield, D.L. & Carnevali, O. (2014) Probiotic modulation of the gut
What have we learned? Viruses 10: 288.
microbiota of fish. In: Aquaculture Nutrition. John Wiley & Sons, Ltd.,
Chichester, UK, pp. 185–222. Higuera, G., Bastías, R., Tsertsvadze, G., et al. (2013) Recently discovered
Vibrio anguillarum phages can protect against experimentally induced
Merrifield, D.L., Dimitroglou, A., Foey, A., et al. (2010) The current sta-
vibriosis in Atlantic salmon, Salmo salar. Aquaculture 392–395: 128–133.
tus and future focus of probiotic and prebiotic applications for salmonids.
Aquaculture 302: 1–18. Huang, K. & Nitin, N. (2019) Edible bacteriophage based antimicrobial
coating on fish feed for enhanced treatment of bacterial infections in aqua-
Munang’andu, H. (2018) Intracellular bacterial infections: A challenge
culture industry. Aquaculture 502: 18–25.
for developing cellular mediated immunity vaccines for farmed fish.
Microorganisms 6: 33. Imbeault, S., Parent, S., Lagacé, M., et al. (2006) Using bacteriophages to
prevent furunculosis caused by Aeromonas salmonicida in farmed brook
Nayak, S.K. (2010) Probiotics and immunity: A fish perspective. Fish
trout. J. Aquat. Anim. Health 18: 203–214.
Shellfish Immunol. 29: 2–14.
408 Chapter 14

Jacquemot, L., Bettarel, Y., Monjol, J., et al. (2018) Therapeutic poten- Vinod, M.G., Shivu, M.M., Umesha, K.R., et al. (2006) Isolation of Vibrio
tial of a new jumbo phage that infects Vibrio coralliilyticus, a widespread harveyi bacteriophage with a potential for biocontrol of luminous vibriosis
coral pathogen. Front. Microbiol. 9: 2501. in hatchery environments. Aquaculture 255: 117–124.
Kalatzis, P., Castillo, D., Katharios, P., et al. (2018) Bacteriophage inter- Wittebole, X., De Roock, S., & Opal, S.M. (2014) A historical overview of
actions with marine pathogenic vibrios: Implications for phage therapy. bacteriophage therapy as an alternative to antibiotics for the treatment of
Antibiotics 7: 15. bacterial pathogens. Virulence 5: 226–235.
Karunasagar, I., Shivu, M.M., Girisha, S.K., et al. (2007) Biocontrol of
pathogens in shrimp hatcheries using bacteriophages. Aquaculture 268: Other articles
288–292. Kumar, J., Singh, D., Tyagi, M.B., & Kumar, A. (2019) Cyanobacteria:
Kellogg, C.A. (2007) Phage therapy for Florida corals? U.S. Geological Applications in biotechnology. In: Cyanobacteria, ed. by Mishra, A.K.,
Surv. Fact Sheet 2007–3065. Tiwari, D.N. and Rai, A.N., Academic Press, pp. 327–346,
Lomelí-Ortega, C.O. & Martínez-Díaz, S.F. (2014) Phage therapy against Luna, G.M. (2015) Biotechnological potential of marine microbes.
Vibrio parahaemolyticus infection in the whiteleg shrimp (Litopenaeus In: Springer Handbook of Marine Biotechnology. Springer, Berlin,
vannamei) larvae. Aquaculture 434: 208–211. Heidelberg, pp. 651–661.
Nakai, T., Sugimoto, R., Park, K.H., et al. (1999) Protective effects of bac- Martín-Rodríguez, A.J., Babarro, J.M.F., Lahoz, F., et al. (2015) From
teriophage on experimental Lactococcus garvieae infection in yellowtail. broad-spectrum biocides to quorum sensing disruptors and mussel repel-
Dis. Aquat. Organ. 37: 33–41. lents: Antifouling profile of alkyl triphenylphosphonium salts. PLoS One
10: e0123652.
Nelson, E.J., Harris, J.B., Morris, J.G., et al. (2009) Cholera transmission:
The host, pathogen and bacteriophage dynamic. Nat. Rev. Microbiol. 7: Pessôa, M.G., Vespermann, K.A.C., Paulino, B.N., et al. (2019) Newly
693–702. isolated microorganisms with potential application in biotechnology.
Biotechnol. Adv. 37: 319–339.
Richards, G.P. (2014) Bacteriophage remediation of bacterial pathogens
in aquaculture: A review of the technology. Bacteriophage 4: e975540. Rahman, P.K.S.M., Mayat, A., Harvey, J.G.H., et al. (2019) Biosurfactants
and bioemulsifiers from marine algae. In, The Role of Microalgae in
Rørbo, N., Rønneseth, A., Kalatzis, P.G., et al. (2018) Exploring the effect
Wastewater Treatment. Springer, Singapore, pp. 169–188.
of phage therapy in preventing Vibrio anguillarum infections in cod and
turbot larvae. AntibiotIcs 7: 42. Romano, S. (2018) Ecology and biotechnological potential of bacte-
ria belonging to the genus Pseudovibrio. Appl. Environ. Microbiol. 84:
Teplitski, M. & Ritchie, K. (2009) How feasible is the biological control of
e02516-17.
coral diseases? Trends Ecol. Evol. 24: 378–385.
Satheesh, S., Ba-akdah, M.A., & Al-Sofyani, A.A. (2016) Natural anti-
Verner-Jeffreys, D.W., Algoet, M., Pond, M.J., et al. (2007) Furunculosis
fouling compound production by microbes associated with marine macro-
in Atlantic salmon (Salmo salar L.) is not readily controllable by bacterio-
organisms—A review. Electron. J. Biotechnol. 21: 26–35.
phage therapy. Aquaculture 270: 475–484.
Chapter 15
Concluding Remarks

The overarching theme of the book has been our ever-expanding realization of the diversity
of microbes and their adaptations to the different physical, chemical, and biological condi-
tions in their world. We have explored exciting new discoveries about how microbial activities
underpin the biogeochemical processes that shape the function of our oceans and our planet.
Novel types of microbes, biochemical pathways, ecological interactions, and biogeochemical
effects are being discovered at ever-increasing rates and we are beginning to understand how
the microscale effects of microbial processes can affect global processes in the oceans. The
oceans undoubtedly hold many more surprises yet to be revealed.

Technological advances have driven most of the advances in our knowledge—we can now
explore the properties and activities of marine microbes at every level from single cells to
global assemblages. The next few years will undoubtedly see further technological prog-
ress and will be dominated by increasing advances and application of metagenomics, tran-
scriptomics, and proteomics, plus major advances in single-cell biology. In particular, we
can expect to see improved methods for exploring the function of the archaea, protists, and
fungi in ocean processes, knowledge of which has lagged behind that of bacteria. However,
sequence data alone is not enough to gain a true understanding of microbial processes and
community interactions. We have seen several examples of how the renaissance of culture-
based microbiology—guided by ‘omics and advanced imaging—is dispelling the dogma that
most marine microbes are uncultivable. It is difficult, but definitely worth the effort. This
will be especially important to gain a proper understanding of viral ecology—we now have a
great wealth of metagenomic data on ocean viromes, but only know the hosts of these viruses
in a fraction of cases. Advances in this approach will allow us to devise hypothesis-driven
approaches to investigate microbial communities. Soon, further progress towards a real
understanding of the ecology of marine microbes and their role in diverse activities will be
enhanced by automated sampling, robotic processing in the field, and real-time remote sens-
ing, coupled with artificial intelligence approaches to the analysis and coordination of data.

Marine microbiology research has also made tremendous contributions to general biologi-
cal knowledge, especially to our understanding of evolution. The discovery of giant viruses
and the ancient archaeal lineages is bringing us ever closer to understanding the origins of
eukaryotes during the early evolution of life on our planet. The study of marine viruses has
revolutionized our understanding of the importance of viral infection of cells on the creation
and dissemination of new genetic information—a major driver in the generation of diversity
and selection processes of evolution. The many wondrous examples of symbiotic partner-
ships between microbes and marine animals have revealed the mechanisms of coevolution
over hundreds of millions of years. This field has also led to discoveries of cellular and
molecular interactions between microbes and their hosts that have direct applications for the
advancement of human health.

Our planet is in in trouble. Climate change, pollution, and overexploitation of finite resources
threaten the future of humanity and many of the spcies with which we share our planet, but
410 Chapter 15

we need to remember that microbial life has adapted to enormous changes in atmospheric
and ocean chemistry that have occurred periodically over the past 3 billion years. We can
only make informed guesses about how microbes will adapt to the extreme changes over the
short timescale of human influence on the planet. But we can be sure they will adapt—what-
ever we do, or don’t do, in response to these changes—and we ignore their influence at our
peril. The atmospheric concentration of CO2 has now passed 400 ppm—the highest it has
been since the mid Pliocene, ~3 MYA. Even if it were possible for humanity to succeed in
limiting future CO2 emissions immediately and to develop technological solutions to reduce
the atmospheric levels, the most optimistic prediction is that it would take 100–200 years to
reduce atmospheric CO2 below 400 ppm and hold the temperature increase below 1.5°C in
the long term. We know that global warming caused by increasing atmospheric concentra-
tions of CO2, N2O, CH4, and other anthropogenic pollutants is causing shifts in ocean chemis-
try, stratification, and currents—altering microbial community composition and the balance
of biogeochemical cycles. Additional anthropogenic factors such as eutrophication and the
effect of huge quantities of microplastics on the sinking rates of particles are also deeply
troubling and make predictions of future trends even more difficult. Better understanding
will come from improved knowledge of the rates of primary production in different size
classes and taxonomic groups of phytoplankton, and from better quantification of the activi-
ties of heterotrophic microbes in the remineralization of organic matter and its sequestration
in the deep ocean. We also need better knowledge of the decay rates of marine viruses, and
how they will be affected by climate change and altered climatic conditions. Furthermore,
climate change and pollution are undoubtedly major contributors to the rise in diseases of
marine organisms—we are witnessing complete shifts in marine ecosystem functioning due
to dysbiosis and microbial infections induced by environmental change. Direct impacts on
human health are increasing.

Surely human ingenuity can find methods to mitigate some of the environmental damage
caused by our activities? Large-scale direct application of marine microbial biotechnology in
the ocean setting could play a major role in ensuring new sources of energy and sustainable
aquaculture to feed the world’s growing population. New initiatives in bioremediation, geoen-
gineering, and methods to control marine diseases and conserve organisms and ecosystems.
need urgent investigation, and—after proper evaluation and international approval—we must
be brave enough to attempt them. Microbiologists have a major role to play in investigating
these developments and in educating the public and politicians about the need for urgent and
large-scale international action to ameliorate the problems we face. I hope this book will play
some part in helping the next generation of students and scientists to exert this influence.
Index
A host–pathogen interaction, 398 oxidizing sulfides, 81
penicillins, 397 pathways to fix CO2, 83–84
Acantharea, 185 AOA, see Ammonia-oxidizing archaea transformation, 4
Acropora palmata, 300 AOB, see Ammonia-oxidizing bacteria Bacterial aquarium
Acyl homoserine lactone (AHL), 313 AOIF, see Artificial ocean iron fertilization Alphaproteobacteria, 121
Adaptive bleaching hypothesis, 279 Applied Biosystems SOLiD sequencer, 51 Caulobacterales, 121, 123
Adenosine triphosphate (ATP), 73, 74, 85 Aquaculture ocean diversity, 121
Adhesins, 339 AMR, 399–400 Bacterial diversity
Aequorea victoria, 35 antimicrobial agents, 397–399 Actinobacteria, 143–144
Aerobic anoxygenic phototrophs (AAnP), 77 factors, 398 Aquificae, 141–142
Aeromonas salmonicida, 314–315 gastrointestinal tract, 403 Chloroflexi, 141
Aerosols, 362 immunostimulants, 403 Firmicutes, 142–143
Agarases, 390 microbial biotechnology, 397 genomic methods, 115, 118
AHL, see Acyl homoserine lactone prebiotics, 403 high alpha diversity, 119–120
Aliivibrio salmonicida, 313 probiotics, 403, 405 HTS, 116–117
ALOHA, 30 vaccination, 400–401 lumpers/splitters, 118
A Long-term Oligotrophic Habitat Assessment, ARB, see Antibiotic-resistant bacteria nutrient cycling, polymers, 140–141
see ALOHA Archaea, 3–4, 22 OTUs and ASVs, 118–119
Alphaproteobacteria, 121–123 Archaeoglobus and Ferroglobus, 155 phylogenetic and genomic techniques, 114
Alvin, 106 cell forms and structural features, 66 sequences for bacterial phyla, 117
Alzheimer’s disease, 394 cell structure and function, 150 Thermotoga, 142
Ammonia Crenarchaeota, hydrothermal vents, 156–157 zobellia genes, 144
anaerobic chemolithoautotrophy, 82 Euryarchaeota, 151, 153, 155–156 Bacterial growth efficiency (BGE), 93
oxidation, 81–82 evolutionary pathway, 152–153 Bacterial kidney disease (BKD), 316–317, 402
Ammonia mono-oxygenase (AMO), 160 features of, 67–68 Bacterial shell disease, 307
Ammonia-oxidizing archaea (AOA), 243 global warming, 151 Bacterial taxa, 118
Ammonia-oxidizing bacteria (AOB), 160, 243 hyperthermophiles, 154 Bacteriocins, 360
Ammonification, 243 lifestyles, 69–72 Bacteriocytes, 261
Amnesic shellfish poisoning (ASP), 347 lipid structure, 66 Bacteriophages, 200
AMO, see Ammonia mono-oxygenase methanogens, 154 Bacteriorhodopsin, 397
Amoebozoa, 184–185 Nanoarchaeota, 161–162 Bacteroidales, 366
Amplicon sequence variants (ASVs), 54, 282 nitrification, 81–82 Baculovirus penaei virus (BPV), 309
Amplified fragment length polymorphism pathways to fix CO2, 83–84 Baltimore classification scheme, 196
(AFLP), 47–48 phylogenomic methods, 150–151 Basidiomycota, 187
AMR, see Antimicrobial resistance S-layers, 69 Bathing waters, 367, 369
AMZ, see Anoxic marine zone Thaumarchaeota, ammonia-oxidizing, BBD, see Black band disease
Anaerobic ammonia oxidation (anammox) 157–159 Bdellovibrionales, 132–133
ecological importance, 82 transformation, 4 Betaproteobacteriales, 124
microbiologists, 83 Artificial ocean iron fertilization (aOIF), 239 Biased random walk movement, 97
Anaerobic methanotrophic Archaea ASP, see Amnesic shellfish poisoning Biodegradation, 373
(ANME), 89, 153 Aspergillus sydowii, 295, 298 Biodeterioration, 358–359
Anaerobic oxidation of methane (AOM), Atlantic Circumpolar Current, 7 Biofouling
88–89, 153–154 Atomic force microscopy (AFM), 33, 34 extending shelf-life, 360–361
Anaerobic respiration, 86 Autonomous underwater vehicles (AUVs), 32 MIC, 357–358
Anammoxosome, 82 Clio, 32 microbial biofilms, 356–357
Anguibactin (Ang), 312 recovery, 33 seafood (see Seafood products)
ANME, see Anaerobic methanotrophic AZAs, see Azaspiracids stages, 356
Archaea Azaspiracids (AZAs), 347 Biogenic Si (BSi), 253
Anoxic marine zone (AMZ), 246 Biogeochemical processes, 409
Anthropogenic pollutants, 410 Bioinformatics, 390–391, 406
B
Antibiotic-resistant bacteria (ARB), 369 BIOLOG® system, 40, 42
Antimicrobial resistance (AMR), 395, Bacteria, 3–4 Bioluminescence, 176–177
399–400, 404 cell forms and structural features, 66 antioxidative mechanism, 284
Antimicrobials CO2 fixation, 77 camouflage, 286–287
antibiotics, 399 features of, 67–68 extracellular bacteria, 284
eukaryotic membranes, 398 lifestyles, 69–72 fish symbionts, 285
feeding, 399 nitrification, 81–82 luciferase enzymes, 284
412 Index

Bioluminescent bacteria, 108 CFP, see Ciguatera fish poisoning culture media, 38–40
Biomagnification, 348 CFU, see Colony-forming units enrichment culture, 40
Bioremediation, 375–376, 378, 391 Chain reaction, 45 microbial cell components, 42
Biotechnology Chemolithoautotrophy, 267 phenotypic testing, 40, 42
antibiotics, 395–396 Chemosynthetic bacterial endosymbionts uncultured, 41
aquaculture (see Aquaculture) alphaproteobacterial phylotypes, 264 Culture-based methods, 365–366
DNA polymerases, 390, 391 autotrophic and heterotrophic routes, 264 Culture-based microbiology, 409
drugs, 395 crustaceans, 265 Cyanobacteria, 17, 19, 67–68, 133, 348
invertebrates (see Marine invertebrates) DNA sequence evolution, 263 antenna pigments, 76
in vitro activity, 395 electron microscopy, 260 filamentous, 139
live vaccines, 402 endogenous autotrophic metabolism, 261 genome-based classification, 134
metagenomics and bioinformatics, 390–391 filter-feeding mechanism, 260 nitrogen fixing, 136–139
microalgae, 391–393 fuel ecosystem-engineering bivalves, Prochlorococcus, 88, 134–136
microbes (see Marine microbes) 271–272 5-Cyano-2,3-dilotyl tetrazolium chloride
polymers, 391 Gammaproteobacteria, 264 (CTC), 35
recombinant DNA technology, 401–402 genome analysis, 261 Cyanophages, 200
BKD, see Bacterial kidney disease hydrothermal vents and seeps, 264 Cytophaga–Flavobacterium–Bacteroides
Black band disease (BBD), 295, 298, 299 invertebrates, 261 (CFB), 305
Black patch necrosis, 315 Kiwa hirsuta, 266 Cytoskeleton, 166
Black smokers, 20 LPS, 261
Blue Flag classification system, 367 mucus secretions, 265 D
Bodo saltans virus (BsV), 214 nutrients/enzymes, 264
Bootstrapping technique, 49 scaly-foot snail, 264, 265 DA, see Domoic acid
Botulism, 341–342 shallow-water sediments, 266–268 Darwin’s theory, 5
BPV, see Baculovirus penaei virus stable isotope analysis, 260 Deep sea
Bray–Curtis dissimilarity index, 54 Chemotaxis, 311 brine pools, 21–22
Brevetoxin, 346 Chitin, 335 hydrothermal vents, 19, 20, 21
Bryostatins, 394 Chlorophylls, 76 microbial activity, 19–21
Bryozoans, 270 Choanoflagellates, 172–173 Deep Sea, Hypersaline, Anoxic Basins
BSi, see Biogenic Si Cholera, 332 (DHABs), 22
BsV, see Bodo saltans virus Chromophoric/light-absorbing DOM Deep-sea FISHing, 160–161
(CDOM), 381 Deepwater Horizon (DWH), 373–375
C Chrysomallon squamiferum, 264, 265 Denaturing gradient gel electrophoresis
Chytridiomycota, 187 (DGGE), 49–50, 51
Cable bacteria, 130–131 Chytrids, 187 Depuration, 370
Cafeteria roenbergensis virus (CroV), 171, 210 Ciguatera fish poisoning (CFP), 347–349 Desulfobacterota, SRB, 129
Calvin–Benson–Bassham (CBB) cycle, 83 Ciliates, 177–178 DHA, see Docosahexaenoic acid
Candidate Phyla Radiation (CPR), 115 CLAW hypothesis, 249 4′,6′-Diamido-2-phenylindole (DAPI), 34
Candidate taxonomic unit (CTU), 117 Climate change, 351 Diarrhetic shellfish poisoning (DSP), 346–347
Canine distemper virus (CDV), 322 CLOD, see Coralline lethal orange disease Diatoms
Carbohydrate active enzymes (CAzymes), 144 Clostridium perfringens, 366 antipredator defense mechanisms, 349–350
Carbon Coastal Zone Color Scanner (CZCS), 61 ASP, 347
and nitrogen fixation CO2 concentrating mechanisms (CCMs), 83 biotechnology, 182
Archaea and Bacteria, 83–84 Cold seeps, 21, 22 genome sequencing, 182
building cellular material, 84–85 Cold-shock proteins (Csp), 105 locomotion, 182
CBB cycle, 83 Colony-forming units (CFU), 367 mixotrophic metabolism, 182
sources of energy Community fingerprinting, 49 nutrient upwelling, 182
ATP, 73 Continuous Plankton Recorder (CPR), 336 ocean food chains, 212
heterotrophs, 74 Copiotrophs, 92 reproductive cycle, 181
microbes, 73–74 Coral diseases silica deposition, 181
mixotroph, 74 A. sydowii, 295 stramenopiles, 181
Carboxysomes, 67 BBD, 295, 298 toxins, 343–345
Carotenoids, 393 microbial infection, 292–294 Diazotrophs
Catalyzed reporter deposition (CARD), 38 tissue necrosis and skeletal erosion, 299–300 acetylene reduction assay technique, 242
Caulobacter crescentus, 97 vibrios, 293–295 denitrification, 243
CDOM, see Chromophoric/light-absorbing DOM viruses, 300–301 isotope tracer techniques, 242
CDV, see Canine distemper virus WPD, 298–299 microbes, 241
Cell growth and function Coralline lethal orange disease (CLOD), 325 UCYN-A, 242
Bacteria and archaea, 66 Coral Probiotic Hypothesis, 294 DIC, see Dissolved inorganic carbon
cell envelope, 68–69 Coral reefs, 25 Digital hybridization, 117
cytoplasmic membrane, 66–67 Core microbiome, 280 Dilution-to-extinction method, 39
genome size and organization, 69–72 Coriolis effect, 8, 221, 224 Dimethyl sulfide (DMS), 248–249
microbes, 73 Coxsackieviruses, 363 Dimethylsulfoniopropionate (DMSP),
organelles, microcompartments and CPR, see Candidate Phyla Radiation; 249, 250, 251
inclusion bodies, 67–68 Continuous Plankton Recorder Dinoflagellates
Cellulophaga lytica, 40 CroV, see Cafeteria roenbergensis virus bioluminescence, 176–177
CeMV, see Cetacean morbillivirus Crustaceans crustaceans, 309
Cephalosporin, 396 epizootics, 306–307 and diatoms, 343–345, 349–350
Cetacean morbillivirus (CeMV), 322 viral diseases, 307–309 diel vertical migration, 175
CFB, see Cytophaga-Flavobacterium- CTU, see Candidate taxonomic unit marine systems, 173, 175
Bacteroides Cultivation of microorganisms PSP (see Paralytic shellfish poisoning (PSP))
 Index 413

DIP, see Dissolved inorganic phosphorus hyperthermophiles, 389 morphological diversity, 187, 188
Diplonemids, 171 proteases and lipases, 389 Nucletmycea, 186–187
Direct light microscopy, 32 psychrophiles, 390
Dissimilatory nitrate reduction (DNRA), 245 pullulanases, 390 G
Dissolved inorganic carbon (DIC), 226 thermophiles, 389
Dissolved inorganic phosphorus (DIP), 252 Epifluorescence microscopy, 170, 351 Gaffkya homari, 307
Dissolved organic carbon (DOC), 381 Epizootics, 326 Gammaproteobacteria, 124–127
Dissolved organic matter (DOM), 91, 220, 227, Epsilonproteobacteria, 129, 132 Gastroenteritis, 362
228, 231–232 ESBL, see Extended-spectrum β-lactamase GBR, see Great Barrier Reef
Dissolved organic phosphorus (DOP), 252 Escherichia coli (E. coli), 6 3GC, see Third-generation cephalosporin
Dissolved silicate (DSi), 253 PR protein, 78 GCAT, see Glycerophospholipid:cholesterol
DMS, see Dimethyl sulfide ETNP, see Eastern tropical North Pacific acyltransferase
DMSP, see Dimethylsulfoniopropionate ETSP, see Eastern tropical South Pacific Genes and proteins, 85
DMV, see Dolphin morbillivirus Eukarya, 3–4 Genetic fingerprinting, 42, 366
DNA and RNA analysis, methods Eukaryotic flagella, 169 Genetic immunization, 402
amplification and sequencing rRNA, 42–45 European Community Bathing Water Gene transfer agents (GTAs), 215
in bacterial and archaeal taxonomy, 48 Directive, 367 Genomes, 56–57
cultured bacteria, 43 Eutrophication, 232–233, 410 Genomes OnLine Database (GOLD), 56
DGGE and TRFLP, 49–50 Excavata, 171 Genomic fingerprinting, 47–48
DNA sequencing, advances, 50–52 Extended-spectrum β-lactamase (ESBL), 369 GeoChip 5.0 microarray, 59
full genome sequence, microbes, 52–53 Global ocean conveyor belt, 8, 9
genomes, 56–57 F Global warming, 24
genomic fingerprinting, 47–48 Glycerophospholipid:cholesterol acyltransferase
isolation of genomic, 45 FAT, see Fluorescent antibody techniques (GCAT), 315
metabarcoding and metagenomics, 53–55 Fecal indicator organisms (FIOs) Gonyautoxins, 344
nucleic acid-based, 42 ARB, 369 Gracilaria, 38
omics technologies, 55–56 bathers’ perceptions of illness, 368 Grains of sand, microbes, 15
PCR, 45–47 controlled surveys, 368 Gram-negative bacteria, 68–69, 109
sequence data, 48–49 limitations, 366 Gram-positive bacteria, 68–69
DNA-DNA hybridization (DDH), 115 phages, 366 Great Barrier Reef (GBR), 296
DNA polymerases, 390, 391 public health risks, 363 Green fluorescent protein (GFP), 35
DNA vaccination, 402 seawater (see Seawater monitoring) Gribbles, 359
DNRA, see Dissimilatory nitrate reduction Fecal streptococci, 364 GTAs, see Gene transfer agents
DOC, see Dissolved organic carbon Filtration or fluorescence-activated cell sorting
Docosahexaenoic acid (DHA), 396 (FACS), 45 H
Dolphin morbillivirus (DMV), 322 Fimbriae, see Pili
DOM, see Dissolved organic matter FIOs, see Fecal indicator organisms HABs, see Harmful algal blooms
Domoic acid (DA), 320 Fish Halobacterium, 78
DOP, see Dissolved organic phosphorus aquaculture and natural populations, Halobacterium salinarum, 100
Driftwood, 359 309–310 Halomonas aquamarina, 92
DSi, see Dissolved silicate A. salmonicida, 314–315 Haptophytes (prymnesiophytes)
DSP, see Diarrhetic shellfish poisoning birnavirus, 319 calcite, 179
DWH, see Deepwater Horizon disease signs, 310–311 coccolithophores, 178, 180
flexibacteriosis, 315 coccoliths, 179–180
E Gram-positive bacteria, 316–317 DMS, 181
ISA, 318 E. huxleyi, 179–180
Eastern tropical North Pacific (ETNP), 245 lymphocystis virus, 319 haptonema, 178
Eastern tropical South Pacific (ETSP), 245 pasteurellosis, 313–314 phytoplankton, 178
Echinoderms, 303–304 protists, 319–320 prymnesiophytes, 180
Ecotoxicological testing, 377 streptococcosis, 317 Harmful algal blooms (HABs)
EhV, see Emiliania huxleyi and Coccolithovirus VHSV, 318 bacteria influence, 349
ELISA, see Enzyme-linked immunosorbent Vibrios (see Vibrios) Chrysochromulina, 206
assay virulence mechanisms, 311 coastal waters, 351
Emiliania huxleyi, 179–180 viruses, 318 health effects, 343
Emiliania huxleyi and Coccolithovirus (EhV), VNN, 319 toxin-associated diseases, 350–351
208–209 Flagella, 97, 98–99 Heavy metals, 376
Endoplasmic reticulum, 166 Florida Reef Rescue Project, 303 Hepatitis A virus, 363
Endosymbiosis theory, 4 FlowCam, 32 Hepatopancreatic parvovirus (HPV), 308
Enterococci, 364–365 Flow cytometry, 35–37, 170, 351 Herpesviruses, 322
Enterococcus, 364 Fluorescence microscopy, 34 Heterotrophic bacteria, 295
Environmental genomics, 42 Fluorescent-activated cell sorter (FACS), 35 Heterotrophic denitrification, 232
Environmental Sample Processor, 32 Fluorescent antibody techniques (FAT), 310 Heterotrophic metabolism
Enzyme-linked immunosorbent assay Fluorescent in situ hybridization (FISH), 15, anaerobic respiration, 86
(ELISA), 345 35–38 energy by fermentation, 85
Enzymes Foraminifera, 185–186 nitrate reduction and denitrification, 86
agar-degradation, 390 Francisella, 316 sulfate reduction, 86–87
amylase, 389–390 Freeze-thawing, 45 Heterotrophic microflagellates (HMF), 172
extracellular proteases, 388 Fungi Heterotrophic nanoflagellates (HNF), 170
extremophilic microorganisms, 389 carbon cycling, 191 Heterotrophic ocean protists, 174–175
food processing, 389 habitats, 188 HGT, see Horizontal gene transfer
glucanases, 388 marine microbiome, 187–189 High-nutrient, low-chlorophyll (HNLC), 236, 237
414 Index

High-pressure liquid chromatography Intestinal enterococci (IE), 364, 367 Mammals

fluorescence detection Invertebrates, 275, 284–285, 286 bacteria and fungi, 323
(HPLC-FLD), 345 Ion Torrent system, 51 dinoflagellate and diatom toxins, 320–321
High-throughput sequencing (HTS), 44, 52–53, IPNV, see Infectious pancreatic necrosis virus mass mortalities, cetaceans and
167, 282 Iron pinnipeds, 321
HISH-nanoSIMS, 59 aOIF experiments, 239–240 viruses, 321–323
HMF, see Heterotrophic microflagellates hydrothermal vents and glacial melting, 238 MAR-FISH, 58
HNF, see Heterotrophic nanoflagellates hypothesis, 239 Marine bacteria
HNLC, see High-nutrient, low-chlorophyll ocean microbes, 237 chemotaxis, 99
Hologenome hypothesis, 294 OIF, 239 sodium, 90
Hopanoids, 67 sources, 237–238 Marine birnaviruses (MABVs), 306
Horizontal gene transfer (HGT), 4, 209 whales and seabirds, 241 Marine columnaris, 315
Host–environment–microbe interactions, 326 Iron-sequestering mechanisms, 339 Marine flexibacteriosis, 315
Host–symbiont interactions, 262–263, 274, ISA, see Infectious salmon anemia Marine invertebrates
278–280 ISAV, see Infectious salmon anemia virus bacterial polyketide synthases, 395
HPLC-FLD, see High-pressure liquid ITS, see Intergenic transcribed spacer bioactive compounds, 393
chromatography fluorescence bryostatins, 394
detection J eribulin, 394
HPV, see Hepatopancreatic parvovirus metagenome, 395
HTS, see High-throughput sequencing JOD, see Juvenile oyster disease sponges, 395
Human Genome Project, 52 Juvenile oyster disease (JOD), 304 symbionts, 394–395
Human ingenuity, 410 Marine microbes, 2
Hydrocarbons, 372 K bacterial flagella, 397
Hydrogen, as electron donor, 81 beneficial activities, 388
Hydrothermal vent communities, 21 KIFES, see Korean Iron Fertilization beneficial properties, 2
Experiment in the Southern Ocean biomedical products, 393
I Kill the winner (KtW) hypothesis, 202 biomimetics, 396
Kinetoplastids, 171 biomolecular electronics, 397
ICTV, see International Committee on Kiwa hirsuta, 266 coccolithophores, 397
Taxonomy of Viruses Kleptoplastidy, 172 diatom silica, 397
IE, see Intestinal enterococci Korean Iron Fertilization Experiment in the enzymes (see Enzymes)
Igniococcus hospitalis, 72 Southern Ocean (KIFES), 240 health-promoting products, 396
IHHNV, see Infectious hypodermal and Kyoto Encyclopedia for Genes and Genomes magnetotactic bacteria, 397
hematopoietic necrosis virus (KEGG), 56 nanotechnology, 396
IHN, see Infectious hematopoietic necrosis size variation
Illumina sequencing technology, 51 L Brownian movement, 6
Imaging techniques Ostreococcus tauri, 6, 8
confocal laser scanning microscopy, 35 Labyrinthulids, 183–184 surface area to volume ratio (SA/V), 5, 7
epifluorescence light microscopy, 33–35 Lateral gene transfer (LGT), see Horizontal gene Thiomargarita namibiensis, 6, 8
FISH, 36–38 transfer ultramicrobacteria, 5
flow cytometry, 35–36 Leptospirosis, 323 Marine microbiology
light and electron microscopy, 32–33 Liebig’s Law of the Minimum, 236 application, 2
Immobilization, 373 Light-harvesting mechanism, 78 modern science, 2
Infectious hematopoietic necrosis (IHN), 402 Limulus polyphemus, 69 Marine planktonic bacteria, 92–93
Infectious hypodermal and hematopoietic Lipopolysaccharide (LPS), 68, 69, 261 Marine snow, 12–14
necrosis virus (IHHNV), 308 Liquid chromatography/mass spectrometry Marnaviridae, 211
Infectious pancreatic necrosis virus (IPNV), (LSMS), 345 MBV, see Monodon baculovirus
319, 401 Live–Dead kit, 35 MCP, see Major capsid protein
Infectious salmon anemia (ISA), 318 Living organisms, 25 Melosira arctica, 24
Infectious salmon anemia virus (ISAV), 310, 401 LSMS, see Liquid chromatography/mass Melting point of DNA, 48
Inipol™ EAP22, 376 spectrometry Mercury, 377
In situ activity, microbial communities Luciferin-luciferinase assay, 366 Mesocosm experiments, 61
mesocosm experiments, 61 Lymphocystis virus, 319 Messenger RNA (mRNA), 73
metatranscriptomics, metaproteomics and Lysogeny Metabarcoding, 53–55
metabolomics, 59–60 antibiotic mitomycin C, 203 MEtaGenome Analyzer (MEGAN) software, 56
microarrays, 59 bacterial production, 204 Metagenome assembled genomes (MAGs), 56, 60
microelectrodes and biosensors, 57 immunity, 204 Metagenomic Illumina tags (miTags), 54
microfluidics, 60 prophage, 201, 203, 204, 215 Metagenomics, 53–55, 57, 390–391, 409
NanoSIMS, 58–59 pseudolysogeny, 204 Methane, microbial production/oxidation
radioisotopes, 57–58 specialized transduction, 203 AOM, 88–89
remote sensing, 61–62 temperate viruses, 201 by bacterial cleavage of phosphonate, 88
SIP, 58 Lysosomal fusion, 268 marine microbes, 90
Integrated Ocean Discovery Programme methanogenesis, 87–88
(IODP), 18 M Methane monooxygenase (MMO), 90
Intergenic transcribed spacer (ITS), 186 Methane seeps, 21, 22
International Census of Marine Microbes MABVs, see Marine birnaviruses Methanogens, 366
(ICoMM), 120 Macrofouling, 356 Methanotrophs, 90
International Committee on Taxonomy of Magnetospirillum sp., 123 MGNV, see Midgut gland necrosis virus
Viruses (ICTV), 196 Magnetotactic bacteria, 124 MIC, see Microbially induced corrosion
International Maritime Organization, 350, 356 Magnetotaxis, 124 Microalgae, 166, 237
International Society of Protistologists, 168 Major capsid protein (MCP), 214 autotrophic, 392
 Index 415

biofuels, 391–393 Microbially induced corrosion (MIC), 357–358 N

chlorophytes, 392 Microbial mats, 16
cyanobacteria, 392 chemosynthetic, 19 Nanophytoplankton, 225
diatoms, 392 and evolution, 19 Nanoscale secondary ion mass spectrometry
flocculation/sedimentation, 392 Microbial processes (nanoSIMS), 58–59
genetic modification, 393 light and temperature, 9, 10 NASA SeaWiFS satellite-borne sensor, 223, 224
nitrogen fixation, 393 living world, 5 NCLDVs, see Nucleocytoplasmic large DNA
oil production, 392 oxidation and reduction, 16 viruses
photobioreactors, 392 Microbial source tracking, 365–366 Negative-sense RNA, 196
prymenesiophytes, 392 Microbial toxins Neurotoxic shellfish poisoning (NSP), 346
vegetable oils, 393 ASP, 347 Next-generation sequencing, 51
Microautoradiography (MAR), 58 botulism, 341–342 Nicotinamide adenine dinucleotide phosphate
Microbes brevetoxin, 346 (NADP), 74
full genome sequence, 52–53 CFP, 347–349 Nif genes, 85
microscopic cellular organisms and non- diatoms (see Diatoms) Nitrification, 243
cellular viruses, 2 dinoflagellates (see Dinoflagellates) Nitrogen cycle
obtaining energy, 73–74 DSP and AZAs, 346–347 ammonification and nitrification, 243
regulating cellular activities, 73 fugu poisoning, 342–343 ammonium and nitrate fuels, 244
in sea ice form, 22–23 HABs (see Harmful algal blooms (HABs)) anammox, 245
sea of gradients, 60 PSP, 344–345 denitrification, 244–245
solar power, 78–79 scombroid fish poisoning, 341 diazotrophs, 241, 242–243
source of carbon, 74 TTX, 343 metabolic reactions, 241
vicinity, 30 Microbiologists, 410 OMZs, 245–247
Microbial carbon pump, 221, 222 Microbiology, 2 oxidation-reduction transformations, 241
Microbial communities, 395–396, 409, 410 Microcystin, 349 sediments, 247
omics technologies, 55–56 Microelectrodes and biosensors, 57 Nitroplast, 138
in situ activity Microeukaryotes, 2 Nitrosococcus oceani, 81
mesocosm experiments, 61 Microfilaments, 166 NMDS, see Non-metric multidimensional
metatranscriptomics, metaproteomics Microfluidics, 60 scaling
and metabolomics, 59–60 Microorganisms, 2–4, 380–381 Nocardia spp., 317
microarrays, 59 auxotrophic, 91 Non-metric multidimensional scaling
microelectrodes and biosensors, 57 cultivation (see Cultivation of (NMDS), 172
microfluidics, 60 microorganisms) Norovirus, 370
NanoSIMS, 58–59 nutritional categories, 75 Norwalk agent, 362
radioisotopes, 57–58 Microphotography, 351 NSP, see Neurotoxic shellfish poisoning
remote sensing, 61–62 Microphytoplankton, 225 Nucleic acids, isolation, 42
SIP, 58 Microplastics, 379, 380–381, 410 Nucleocytoplasmic large DNA viruses
Microbial eukaryotes, 166 Microsatellite markers, 48 (NCLDVs), 196, 206–207
Microbial growth Microscale bacterial swimming, 98–99 Nutraceuticals, 396
nutrient acquisition and Microtox® system, 377 Nutrient limitation
acquisition of iron, 92 Microtubules, 169 iron (see Iron)
antagonistic interactions, 102 Midgut gland necrosis virus (MGNV), 309 phytoplankton, 236
copiotrophic marine bacteria, 93–94 Milky sea, 61, 128 surface waters, 236–237
flagella, 97 MinION sequencer, 52, 57
gliding and twitching motility, 100–101 MLST, see Multilocus sequence typing O
marine bacterioplankton, 92–93 Molluscs
microbes, response, 100 bacteria, 304–305 OA, see Ocean acidification
microbial metabolism, 90–92 protistan disease, 305–306 O-antigen, 69
motility, 95–97 Monodon baculovirus (MBV), 309 Ocean acidification (OA), 12, 228, 336
organic material, growth rate, 93 Moon, surface of, 33 Ocean carbon cycling
pili, 102 Moritella viscosa, 313 abiotic and biotic processes, 221
quorum sensing, 102–104 Most probable number (MPN), 200, 365, biological pump, 222, 228
starvation, bacteria, 94 370–371 Coriolis effect, 221
surface colonization, 101 Motility eutrophication, 232–233
VBNC, 95 bacteria, 95–97 mesocosm, 223
physical effects and survival gliding, 100 microbes, 220
bacterial bioluminescence, 108 with microfluidics, 98 microbial loop (see Microbial loop)
high hydrostatic pressure, 106–107 run and reverse, 98 net community production, 223
in hydrothermal systems, 105–106 run-reverse-flick, 98–99 oxygen concentration, 223
marine microbes, 105 twitching, 101 phytoplankton cells, 222
osmotic damage, prevent, 108–109 MPN, see Most probable number polymeric substances, 222
ultraviolet irradiation, 107–108 Multicellular bacteria, 68 primary production
Microbial infections Multidimensional scaling (MDS), 54 carbon fixation, 226
toxins (see Microbial toxins) Multilocus sequence typing (MLST), 45, 340 geographical and seasonal variations,
Vibrios (see Vibrios) Mussels, 305–306 224–226
Microbial iron pump, 238 Mycobacterium phytoplankton, 222–224
Microbial loop M. fortuitum, 317 solubility pump, 221
carbon cycling system, 222 M. marinarum, 340 Ocean habitats
DOM, 220, 230 M. marinum, 317 deep-water circulation systems, 8
food chain, 226–227 M. tuberculosis, 317 gelatinous biofilm, sea surface, 14
protists, 229–230 Mycology, 190–191 interconnected water system, 6–8
retention of dissolved nutrients, 227–228 Myxobacteria, 68, 132 light and temperature, 9, 10
416 Index

microbes, 10 PhyloChip assay, 59 PT, see Phage therapy

seawater, 10–13 Phytoplankton, 10 PUFA, see Polyunsaturated fatty acids
upper surface, 8 Picoeukaryotes, 2, 169 Pulsed-field gel electrophoresis (PFGE), 48
Ocean microbial diversity, 55 Picophytoplankton, 225, 229 Pure culture method, 39
Oceanographers, 223 Picoplankton, PR-based phototrophy, 78 Purple phototrophic bacteria, 76
Oceanospirillales, 127–128 Picornaviridae, 363 Pycnocline, 12
ODZ, see Oxygen depleted zone Piggyback-the-Winner (PtW) hypothesis, 203 Pyrococcus furiosus, 94
Oil pollution Pili, 102 Pyrolysis mass spectrometry (PyMS), 42
bioremediation, 373, 375–376 Piscirickettsia salmonis, 316
marine environment, 372 Planctomycetes, 139–140 Q
microbes, 372–373 Plankton, 10
physical and biological processes, 373 classification by size, 10, 11 QS, see Quorum sensing
Olavius algarvensis, 267 method of collecting, 30 Quantitative real-time PCR (Q-PCR), 47
Oligotrophy, 41 Plastic pollution, 378–379 Quorum quenching (QQ), 105
OMZs, see Oxygen minimum zones POC, see Particulate organic carbon Quorum sensing (QS), 102, 313, 333
O-nitrophenol-β-galactopyranoside (ONPG), 363 Polyaromatic hydrocarbons (PAHs), 373 Aliivibrio fischeri, bioluminescence, 107
ONPG, see O-nitrophenol-β-galactopyranoside Polychlorinated biphenyls (PCBs), 377, 378 gene regulation, 104
Open reading frames (ORFs), 391 Polycystina, 185 LuxI/LuxR system, 103
Operational taxonomic units (OTUs), 54, 118 Polymerase chain reaction (PCR), 15, 365, 390 signals, 104
Optical tweezers, 39 amplification, 44 in situ, 104
Oral rehydration therapy, 333 bias and limitations, 47 symbiotic/pathogenic interactions, 104
ORFs, see Open reading frames initial stages, 46 Vibrio fischeri, 103
Organic pollutants, 377–378 multiplex, 47
Organisms, 292 positive and negative controls, 46 R
OTUs, see Operational taxonomic units primer, 46
Oxygen depleted zone (ODZ), 246 products, 46 Radioisotopes, 57–58
Oxygenic photosynthesis, 76, 77 success with, 45 Radiolarians, 185–186
Oxygen minimum zones (OMZs), 245–247 Taq polymerase, 46 Random amplified polymorphic DNA (RAPD)
Oysters, 305–306 Polyunsaturated fatty acids (PUFA), 393 technique, 47
Pompeii worm, 265 Raphidophytes, 183
P Porins, 68 RDOM, see Recalcitrant DOM
Positive-sense RNA, 196 Recalcitrant DOM (RDOM), 221
PacBio SMRT, 52, 57 Poxviruses, 322 Recombinant DNA technology, 401–402
PAHs, see Polyaromatic hydrocarbons Prasinophytes, 184 Red Queen hypothesis, 202
Pangenome, 117 Principal components analysis (PCA), 54 Reductive tricarboxylic acid (rTCA), 267
Papillomaviruses, 322 Prochlorococcus, 37 Remotely operated vehicles (ROVs), 22, 32
Paralytic shellfish poisoning (PSP) Prochlorococcus marinus, 71 Remote sensing, 351
beach notice, 345, 346 Programmed cell death (PCD), 209 Renibacterium salmoninarum, 316, 317
dipstick test kits, 345 Proteomics, 409 Reoviridae, 363
equivalents, 344 Proteorhodopsin (PR), 78 Rep-PCR, 48
muscular and respiratory paralysis, 344 benefits to cell, 79 Roseovarius oyster disease (ROD), 305
receptor-binding assay, 345 picoplankton, 78 Restriction fragment length polymorphism
STXs, 344 Protists, 2 (RFLP) analysis, 47
toxins, 344 amoebozoa, 184–185 Reverse transcriptase PCR (RT-PCR), 47
Parsimony, 49 ciliates, 177–178 Rhynchomonas nasuta, 101
Particulate organic carbon (POC), 226 classification systems, 167–168 Ribosomal RNA (rRNA), 3, 15, 42–45
Pasteurellosis, 313–314 diatoms (see Diatoms) Ribotyping, 47
PCBs, see Polychlorinated biphenyls dinoflagellates (see Dinoflagellates) ROD, see Roseovarius oyster disease
PCD, see Programmed cell death diversity, 167–168 Roseophages, 200
PCR, see Polymerase chain reaction fish, 319–320 rTCA, see Reductive tricarboxylic acid
PDV, see Phocine distemper virus genome sequences, 168, 169 RuBisCO, 67, 83
Phage therapy (PT), 404–405 haptophytes, 178–181 Rusticles, 358
Phagocytosis, 166 heterotrophic flagellated
Pharmaceutical agents, 393, 394 feeding mechanisms, 170–172 S
Phocine distemper virus (PDV), 322 grazing, 169–170
Phosphorus cycle microbial loop, 229–230 SAGs, see Single amplified genomes
limiting/co-limiting nutrient, 252 mixotrophic, 172 Salmon pancreas disease virus (SPDV), 401, 402
low and variable levels, 252–253 omics approaches, 168–169 Sampling methods
Photic zone, 9 photosynthetic prasinophytes, 184 marine environment, 30–32
Phototrophy and chemotrophy phototrophic, 189 robotics and artificial intelligence systems, 32
aerobic anoxygenic phototrophy, 77 picoeukaryotes, 169 Sanger sequencing, 44, 51
ammonia, 82 radiolarians and foraminifera, 185–186 SAR86, 78
anaerobic anoxygenic photosynthesis, 76–77 RNA viruses, 211–212, 214 Saxitoxins (STXs), 344
bacterial and archaeal nitrification, 81–82 SEM, 167 SBP, see Sphingolipid biosynthesis pathway
chemolithotrophs stramenopiles, 183 Scanning electron microscopy (SEM), 33, 167,
hydrogen as electron donor, 81 thraustochytrids and labyrinthulids, 183–184 380
inorganic electron donors, 79 unicellular eukaryotic microbes, 166 SCG, see Single-cell genomics
light energy to chemical energy, 74, 76 Pseudomurein, 69 SCTLD, see Stony coral tissue loss disease
oxygenic photosynthesis, 76, 77 Pseusoalteromonas haloplanktis, 98–99 Seafood-associated gastroenteritis, 370
rhodopsins as light-harvesting pigment, 77, 79 PSP, see Paralytic shellfish poisoning blood agar, 339
SOB, 80–81 Psychrophilic microbes, 105 chromosomes, 340
 Index 417

estuarine and coastal environments, 339 Solubility pump, 221 Thioploca spp., 128–129
fish and marine mammals, 340–341 “Somnicell” (sleeping cells), 95 Third-generation cephalosporin (3GC), 369
necrotizing fasciitis/septicemia, 339 Spawner mortality virus (SMV), 308 Thraustochytrids, 183–184
serotypes, 340 SPDV, see Salmon pancreas disease virus Thylakoids, 76
virulence factors, 339 Specific pathogen-free (SPF), 308 Time-lapse photography, 16
Seafood products SPF, see Specific pathogen-free TonB-dependent transporters (TBDTs), 91
microbial activities, 361 Sphingolipid biosynthesis pathway (SBP), 208 Toxin coregulated pili (TCP), 334
spoilage organisms, 359–360 Sponges, 301 Toxin-related genes (TRGs), 263
Seagrasses, 324 algal polysaccharides, 274 Transcriptomics, 409
Sea ice forms, 22 antibiotics and antitumor compounds, 273 Transmission electron microscopy (TEM), 33, 198
density of diatoms, 23 asexual and sexual reproduction, 273 Transparent exopolymer particles (TEPs), 12, 229
frost flowers, 23 biotechnological exploitation, 273 Tree of life, 3, 5
melting, 24 denitrification/anammox, 274 TRGs, see Toxin-related genes
structure, 23 ecosystem functions, 274, 275 TRH, see Thermolabile related toxin
Sea squirts, 270, 271–272 mesohyl, 273 Tributyl tin (TBT), 356
Sea star associated densovirus (SSaDV), 303 metagenomics, 274 Tricarboxylic acid (TCA), 83, 373
Sea star wasting disease (SSWD), 303 metaproteomics, 274 Trimethylamine, 359
Sea surface microlayer (SML), 14, 15, 31 metatranscriptomics, 274 Tripartite ATP-independent transporters
Sea surface temperatures (SST), 336–337 microbes, 273 (TRAP), 91
Sea turtles, 323–324 microbial diversity, 273, 274 T3SS, see Type III secretion systems
Sea-viewing Wide Field-of-View Sensor microbiome, 274 T6SS, see Type VI secretion systems
(SeaWiFS) system, 61 phagocyte cells, 273 TSV, see Taura syndrome virus
Seawater SRB, see Sulfate-reducing bacteria TTX, see Tetrodotoxin
carbon dioxide, 12 16S rRNA gene, 54, 59 Tuberculosis, 323
concentration, 10–12 SSaDV, see Sea star associated densovirus Tupanvirus, 211
density, 12 SST, see Sea surface temperatures Type III secretion systems (T3SS), 315, 339, 340
as homogeneous fluid, 12 SSU, see Small ribosomal subunit Type VI secretion systems (T6SS), 335, 340
marine particles, 13 SSWD, see Sea star wasting disease
marine snow, 12–14 Stable isotope probing (SIP), 58, 191 U
minor ions, 11 Stony coral tissue loss disease (SCTLD), 302
oxygen, 12 Stromatolites, 17, 19 Ultra-microbacteria, 94
polymers, 12 STXs, see Saxitoxins Ultraviolet (UV) radiation, 107–108
remineralization, 13 Sulfate-reducing bacteria (SRB), 80, 86–87, 129, US Environment Protection Agency (USEPA),
salty, 10 295, 357, 358 367, 368
Seawater monitoring Sulfide-oxidizing bacteria (SOB), 80–81, USEPA, see US Environment Protection Agency
coliforms, 363–364 126–127, 295
E. coli, 363–364 Sulfur cycle V
enterococci, 364–365 DMS, 248–249
Seaweeds, 324–326 oceans and sediments, 247–248 Vaccination, 400–401
Sedimentation rate, 17 surface waters, 248 VAMGs, see Virus-encoded auxiliary metabolic
Sediments Symbiodiniaceae, 277–279 genes
sampling, 30, 31 Symbiosis VBNC, see Viable but nonculturable
sulfur cycle, 80 autotrophic and heterotrophic, 269–270 VGSLs, see Virally encoded glycosphingolipids
and surface chemosynthetic bacteria (see VHS, see Viral hemorrhagic septicemia
biofilms and mats, 16–17 Chemosynthetic bacterial VHSV, see VHS virus
deep marine sediments, 16 endosymbionts) VHS virus (VHSV), 318
deep subsurface, 18–19 definition, 260 Viable but nonculturable (VBNC), 95, 332
microbes, 14–16 fish and invertebrates, 284–285, 286 Vibrios, 40, 127, 293–295
“Selfish” bacteria, 92 holobiont/metaorganism, 260, 280–281 extracellular enzymes, 313
SEM, see Scanning electron microscopy Symbiosome, 276 genetic elements, 333–335
Sewage-polluted seawater, 362–363 Symbiotic nitrogen fixation, 86 invertebrate diseases, 311
Shellfish Synechococcus spp., 136 IROMPs, 312
classification, 370, 371 iron uptake, 312
human consumption, 370–371 T marine and estuarine environments, 332
pathogens, 371 QS, 313
sewage-polluted waters, 369–370 Taura syndrome virus (TSV), 309 V. abguillarum, 311, 312
Shewanella putrefaciens, 98 TBT, see Tributyl tin V. anguillarum, 313
Silicon cycle, diatoms TCA, see Tricarboxylic acid V. cholerae
blooms, 254 TCP, see Toxin coregulated pili autochthonous aquatic bacterium, 332
economic process, 253, 254 TDH, see Thermostable direct hemolysin ctxA and ctxB genes, 334
eutrophication, 254 Teichoic acids, 68 ecology, 333, 334
Single amplified genomes (SAGs), 56, 174, 226 TEM, see Transmission electron microscopy human colonization and virulence, 333
Single-cell genomics (SCG), 174 TEPs, see Transparent exopolymer particles non-O1 and non-O139, 335, 337
Single-stranded RNA, 362–363 Terminal restriction fragment length V. coralliilyticus, 296–297
SIP, see Stable isotope probing polymorphism (TRFLP), 49–50 V. ordalii, 313
Skeletal Eroding Band, 300 Tetracyclines, 360 V. parahaemolyticus, 94, 97, 100 (see also
Small ribosomal subunit (SSU), 114 Tetrodotoxin (TTX), 342, 343 Seafood-associated gastroenteritis)
SMV, see Spawner mortality virus Thermohaline circulation system, 8, 9 V. viscosus, 313
SOB, see Sulfide-oxidizing bacteria Thermolabile related toxin (TRH), 339 V. vulnificus, 313
Sodium dodecyl sulfate and separated by Thermostable direct hemolysin (TDH), 339 biotypes, 338–339
polyacrylamide gel electrophoresis Thermotolerant coliforms, 363 pathogenicity, 338
(SDS-PAGE), 42 Thermus aquaticus, 46 wounds, 338
418 Index

Vibriophages, 200 NCLDVs, 206–207 WHO, see World Health Organization

Viral ecology, 409 nucleic acid/capsid, 205 Whole genome amplification (WGA), 56
Viral hemorrhagic septicemia (VHS), phages Whole genome sequencing (WGS), 118
309, 318, 402 bacterial and archaeal cells, 200–201 Winter ulcer disease, 313
Virally encoded glycosphingolipids (vGSLs), 208 life cycle, 201 Withering syndrome, 305
Viral nervous necrosis (VNN), 319 lysogeny, 201, 203–204 World Health Organization (WHO), 367
Viral shunt, 221, 222, 230–231, 232 power-law function, 199 WPD, see White plague disease
Virions, 196, 213, 215 RNA, 211–212, 214 WPX, see White pox
Virus dilution (reduction) method, 205 virus-induced mortality, 205–206 WS, see White syndrome
Virus-encoded auxiliary metabolic genes VLPs, 198
(vAMGs), 213 Virus-like particles (VLPs), 198, 300 X
Viruses VLPs, see Virus-like particles
bacterioplankton, 199 VNN, see Viral nervous necrosis XplOrer64™ system, 365
biological pump, 206
capsomeres, 196 W Z
enumeration, 34
flow cytometry, 199, 213 Wastewater Zetaproteobacteria, 133
genetic diversity, 214 coastal pollution, 361–362 Zoonoses, 340–341
giant, heterotrophic protists, 210 effluent, 369 Zooplankton, 10, 91
lytic phages, 198, 199 WGS, see Whole genome sequencing Zooxanthellae
in marine ecosystems, 4–5 Whispovirus, 308 corals, 276–278
marine organisms, 196, 197 White plague disease (WPD), 298–299 genetic diversity and host specificity,
microbes, 196 White pox (WPX), 299 275–276
microbial communities, 212 White syndrome (WS), 293, 296, 297 giant clams, 281