Biological Control 94 (2016) 25–32

Contents lists available at ScienceDirect

Biological Control
journal homepage: www.elsevier.com/locate/ybcon

Plants from the Caatinga biome harbor endophytic Trichoderma species

active in the biocontrol of pineapple fusariosis
Jorge T. De Souza a,⇑, Rafael O. Trocoli b,1, Fernando P. Monteiro a
a
Lavras Federal University, Lavras 37200-000, Minas Gerais, Brazil
b
Recôncavo da Bahia Federal University, Cruz das Almas 44380-000, Bahia, Brazil

h i g h l i g h t s g r a p h i c a l a b s t r a c t

Trichoderma species occur as natural

endophytic inhabitants of native
Caatinga plants.

Six Trichoderma species were
identified among 109 isolates from
four plant species.

Fusarium guttiforme was inhibited by
21 isolates in Petri plates and by 3 in
the greenhouse.

Greenhouse bioassays are
recommended to select Trichoderma
isolates against F. guttiforme.

Three isolates reduced fusariosis
severity by 68–84% across two field
experiments.

a r t i c l e i n f o a b s t r a c t

Article history: Fusariosis incited by Fusarium guttiforme is the main limiting factor for pineapple cultivation in Brazil and
Received 20 August 2015 other South American countries. Disease management based on biological agents is desired because resis-
Revised 10 December 2015 tant varieties are not yet widespread through the producing areas and there is no information on the
Accepted 11 December 2015
durability of the resistance. Fungicide application has led to the selection of resistant populations of
Available online 12 December 2015
the fungus in the field, besides the negative effects of these chemicals on human health and the environ-
ment. Because pineapple cultivation is expanding in the Caatinga biome, this study is an attempt to
Keywords:
employ Trichoderma species isolated as endophytes of plants from this region, which is characterized
Ananas comosus
Biological control
by low precipitation and high temperatures. A total of 109 isolates of Trichoderma were obtained from
Fusarium guttiforme the sapwood of four plant species: Ananas comosus var. bracteatus (Bromeliaceae), Astronium fraxinifolium
Screening (Anacardiaceae), Bowdichia virgilioides (Fabaceae) and Caesalpinia pyramidalis (Fabaceae). The 109 isolates
were clustered in 11 BOX groups and one isolate of each group was identified by sequencing the ITS and
tef-1a regions. The isolates were subjected to a series of experiments in vitro in Petri dishes and pineapple
stalk discs, and in vivo in a greenhouse and in the field. The number of isolates showing ability to inhibit
the pathogen and control the disease at levels higher than 80% was 26 in Petri plates, 21 in stalk discs, and
3 in the greenhouse and field assays. These three isolates were consistently able to decrease disease
severity by 68–84% across two field experiments.
Ó 2015 Elsevier Inc. All rights reserved.

⇑ Corresponding author at: Lavras Federal University, Phytopathology Department, P.O. Box 3037, Lavras 37200-000, Minas Gerais, Brazil.
E-mail address: [email protected] (J.T. De Souza).
1
Present address: IF Baiano, Estrada da Igara km 04, 48970-000 Senhor do Bonfim, Bahia, Brazil.

http://dx.doi.org/10.1016/j.biocontrol.2015.12.005
1049-9644/Ó 2015 Elsevier Inc. All rights reserved.
26 J.T. De Souza et al. / Biological Control 94 (2016) 25–32

1. Introduction In this study endophytic Trichoderma were isolated from native

Caatinga plants, screened in vitro, in the greenhouse and the most
The major world pineapple [Ananas comosus (L.) Merr.] produc- promising isolates were tested in the field against pineapple fusar-
ers are Costa Rica, Brazil, Philippines, Thailand, and Indonesia. iosis. Additionally, the diversity of the isolates was assessed by
Although Brazil is the second largest producer, with an average BOX-PCR and representatives were identified by sequencing the
of 2.478 million tons in 61,000 ha, its yield of 41 t ha1 is far lower ITS and tef-1a regions.
than that of Indonesia, which is 124.5 t ha1 (FAOSTAT, 2012).
Among the factors that contribute to these low yields are adverse 2. Materials and methods
environmental factors, inadequate cultural practices, and phy-
tosanitary problems (Matos and Reinhardt, 2009). 2.1. Isolation, cultivation and preservation of fungi
The disease known as pineapple fusariosis incited by the co-
evolved pathogen Fusarium guttiforme Nirenberg and O’Donnell Four plant species commonly found in the Caatinga biome were
(sin. Fusarium subglutinans f. sp. ananas Ventura, Zambolin and Gil- selected as sources of Trichoderma species: A. comosus var. bractea-
bertson) is considered the most damaging disease in Brazil and tus (Lindley) Coppens and Leal (Bromeliaceae) is an herbaceous
other South American countries (Matos and Reinhardt, 2009). plant similar to pineapple; Caesalpinia pyramidalis Tul. (Fabaceae)
The pathogen is known to occur only in South America and is is a shrub of up to 4 m high in rocky soil, but may reach 16 m in
pathogenic to most commercial varieties, such as Smooth Cayenne, deeper soils, and trunk diameter of 8–15 cm; Astronium fraxini-
Queen, Red Spanish and Pérola (Matos et al., 2009). This disease folium Schott (Anacardiaceae) and Bowdichia virgilioides Kunth
typically causes 30–40% fruit loss and 15–20% loss of the propaga- (Fabaceae) are trees of 8–20 m high with a trunk diameter varying
tive material (Ventura et al., 2009). The fungus is able to infect all from 30 to 50 cm measured at 1.5 m of height (Giulietti et al.,
plant parts, but symptoms are more evident on fruits. Infected 2004). All samples were collected at the municipality of Senhor
pineapple plants show bent or dead stem apices, altered phyl- do Bonfim, Bahia, in an area of approximately 20 ha. Fragments
lotaxy, shortened stems, general stunting and chlorosis (Matos of the stem of A. comosus var. bracteatus, and sapwood of the
et al., 2009). Symptoms on fruits include an initial discoloration woody species were employed in the isolations of Trichoderma.
of the infected area, fruitlet rot, and profuse pink fungal sporula- These plant parts were disinfested with 70% alcohol for 1 min
tion and gum exudation. Infection can occur in plantlets originated and 1% sodium hypochlorite for 5 min and rinsed 3 times with ster-
from infected mother plants and in all stages of fruit development ile distilled water. Fragments of each sample were transferred to
(Matos et al., 2009). Petri plates with 1/5 PDA (Potato – Dextrose – Agar). Aliquots of
Management of the disease relies on integrated tactics, includ- the last rinse water were spread onto PDA to exclude the possibil-
ing cultivar resistance (Ventura et al., 2009), fungicide application ity of isolating epiphytes. Trichoderma spp. isolates were purified
and cultural practices (Reinhardt et al., 2002). Some cultivars with and kept preserved in mineral oil and sterile water in the collection
resistance to the pathogen are available but they are not yet of Microbiology and General Biology Laboratory at the Federal
adopted in the whole country. There is also a lack of information Institute of Education, Science and Technology of Bahia State, Cam-
on the stability of their resistance to the pathogen in the long term. pus Senhor do Bonfim. Trichoderma spp. spore suspensions were
Control of the disease by spraying fungicides from the early stages prepared from cultures grown for 7 days on PDA at 25 °C and
of inflorescence development until flower closure is largely used adjusted to 108 conidia/mL.
(Matos et al., 2009), but frequently without taking into account The pathogenic isolate Fgt72CA-CNPMF of F. guttiforme was
the possible risks to health and the environment. Furthermore, obtained from the collection of Embrapa at Cruz das Almas, Bahia
pathogen populations with resistance to the fungicide benzimida- and used in all experiments. It was grown for 7 days on PDA at
zole have been reported (Ventura et al., 1992). In recent years, 25 °C and suspensions were adjusted to 105 conidia/mL.
numerous reports on promising biological control agents have
been published for different pathosystems (Chen et al., 2015;
2.2. BOX-PCR and molecular identification
Luambano et al., 2015). Fungi in the genus Trichoderma are the
most studied among the fungal agents to control plant diseases
BOX-PCR was performed to study the diversity in the collection
worldwide (Druzhinina et al., 2011). Nevertheless, there are few
of isolates used in this study. The DNA of all isolates was extracted
studies on the biocontrol of pineapple fusariosis in general and
by the protocol described by Doyle and Doyle (1990). The DNA was
none under field conditions.
amplified using the primer BOXA1R (50 -CTACGGCAAGGCGACGCT
In this study, the control of pineapple fusariosis was attempted
GACG-30 ) (Versalovic et al., 1994). PCR reactions were done in a
with Trichoderma species isolated from plants growing in the Caa-
final volume of 25 lL containing 10 mM Tris–HCl pH 8.3, 50 mM
tinga biome. The rationale for this approach is the fact that pineap-
KCl, 1.5 mM MgCl2, 200 mM of each dNTP, 1 U Taq DNA poly-
ple cultivation is expanding in areas under the influence of this
merase, 20 pM of primer and approximately 40 ng of DNA. Ampli-
biome and the fungi from these plants are supposedly adapted to
fications were performed with a cycle program consisting of an
the local conditions.
initial denaturation at 95 °C for 5 min followed by 10 PCR cycles
The Caatinga biome occupies an area of 844,443 km2, corre-
of denaturation at 94 °C for 1 min, annealing at 40 °C for 30 s and
sponding to 9.9% of the Brazilian territory (IBGE, 2004). It is located
extension at 72 °C for 1 min and additional 30 PCR cycles of
in the northeast of country, in a region with low and irregular pre-
90 °C for 1 min, 48 °C for 30 s, 72 °C for 1 min and a final extension
cipitation (300–800 mm/year), long drought periods and annual
at 72 °C for 10 min.
mean temperatures as high as 32 °C in some areas (CPRM, 2006;
Amplified fragments were separated in 2% agarose gels stained
INMET, 2015). The vegetation is predominantly composed of Legu-
with 1 lg/mL ethidium bromide in TAE buffer (90 mM Tris–Acet-
minosae and Cactaceae (Queiroz, 2002; Taylor and Zappi, 2002).
ate and 1 mM EDTA) at 50 V using a 1-kb DNA ladder (Invitrogen)
The Caatinga is not a uniform environment, but there are areas
as a marker. Presence or absence of bands with determined sizes
with characteristics of Cerrado, that vaguely resembles the African
were recorded for each isolate and the program FreeTree version
Savana and Atlantic forest, mainly when located in the proximity of
2.02 (Hampl et al., 2001) was used to group the isolates with the
rivers or in higher altitudes (Giulietti et al., 2004).
unweighted pair group mean average (UPGMA) method using Jac-
 J.T. De Souza et al. / Biological Control 94 (2016) 25–32 27

Table 1
Classification of 109 Trichoderma isolates in BOX groups, their origin and identification of one representative of each group on the basis of ITS and tef-1a sequences.

BOX group Isolates Origina Representative Species ID Accession ITS/tef-1a

1 37;64;74;75;76;80; Anc(2);Asf(1);Bov(8);Cap(11) Tc88 T. longibrachiatum KP898758/KP890335
82;86;88;91;93;97;
109;118;120;127;130;
131;132;135;136;137
2 03;04;08;16;43;45;46;56 Asf(2);Anc(3);Bov(2);Cap(1) Tc08 T. harzianum KP898749/KP890327
3 09;47;48;53;54;55;63; Asf(5);Anc(2); Tc09 T. atroviride KP898750/KP890328
78;81;84;94;95; Bov(1);Cap(7)
105;123;138
4 13 Cap(1) Tc13 T. virens KP898752/KP890330
5 01;14;33;87;89;92;139 Asf(3);Bov(1);Cap(3) Tc01 T. harzianum KP898748/KP890326
6 57;60;61;62;68;72;79 Anc(1);Asf(4);Cap(2) Tc60 T. harzianum KP898756/KP890333
7 32;59;66;67;70;73;85;90; Anc(2);Asf(5); Tc85 T. koningiopsis KP898757/KP890334
100;103;115;122;128; Bov(5);Cap(5)
129;134;140;141
8 10;42;44;50;58;69;71; Anc(1);Asf(5); Tc10 T. atroviride KP898751/KP890329
77;83;96;126 Bov(3);Cap(2)
9 11;26;28;30;31;34;35;36; Anc(2);Asf(6); Tc26 T. koningiopsis KP898755/KP757743
39;40;41;49;142 Bov(3);Cap(2)
10 20;24;111 Anc(1);Bov(1);Cap(1) Tc24 T. koningiopsis KP898753/KP890331
11 25;104;116;119;125 Anc(2);Asf(3) Tc25 T. sinuosum KP898754/KP890332
a
(Anc): Ananas comosus var. bracteatus, (Asf): Astronium fraxinifolium, (Bov): Bowdichia virgilioides, and (Cap): Caesalpinia pyramidalis.

card’s coefficient of similarity. BOX groups were defined on the 3.5 kg of a mixture composed of soil, organic matter and sand
basis of 100% similarity. (2:1:1). The bags were maintained in a greenhouse with irrigation
One representative of each BOX group was selected for identifi- and fertilization done as described by Cunha and Reinhardt (2004).
cation by sequencing a fragment of the translation and elongation After 60 days in the greenhouse the plantlets were removed
factor of the RNA polymerase (tef-1a) and internal transcribed from the bags, had their roots washed with tap water, received four
spacer (ITS). DNA extraction, amplification and sequencing were wounds 2 cm deep and spaced 1 cm from each other in the stalk
done as previously described (Samuels et al., 2012). The accession and were immersed separately for 15 min in spore suspensions
numbers for tef-1a and ITS sequences are listed in Table 1. of 109 isolates of Trichoderma. After a period of 10 min, plantlets
were immersed in a spore suspension of F. guttiforme for 5 min
2.3. Screening in vitro in Petri plates and on discs of pineapple stalks as described by Matos and Cabral (2006). Subsequently, plantlets
were replanted in the bags and kept in a greenhouse. The experi-
F. guttiforme mycelial growth was assessed in Petri plates and mental design was completely randomized with 10 replicates
on pineapple stalks to evaluate the in vitro antagonism of 109 Tri- and controls were plantlets treated with sterile distilled water or
choderma isolates. Trichoderma isolates were cultivated in 1 mL of the pathogen alone. Ninety days after the inoculations, plantlets
PDB (Potato Dextrose Broth) and incubated at 25 °C for three days. were removed from the bags and disease severity was determined.
Aliquots of 10 lL were transferred to three equidistant points on Severity was scored by making longitudinal cuts in the stalks and
the surface of PDA plates and 0.5 mm-diameter PDA discs contain- using a grading scale ranging from of 0 to 5 as described by
ing mycelium of the pathogen were transferred to the center of the Matos and Cabral (2006).
plates. The plates were organized in a random experimental design
with five replicates, incubated at 25 °C and the inhibition zones 2.5. Field experiments
were measured five days after the incubation.
Healthy pineapple plantlets cultivar ‘‘Perola” with approxi- Pineapple plantlets were obtained, kept in the greenhouse for
mately the same age were selected to extract stalk discs. Discs 60 days, wounded and inoculated as described above, except that
measuring 5 cm in diameter and 1 cm in height were removed only fifteen isolates of Trichoderma were used. After the inocula-
by cutting at approximately 3 cm from the base of the plantlets. tions, plantlets were transplanted to the field with spacing of
The stalk discs were rinsed in tap water and disinfested with 70% 0.90 m between rows and 0.35 m between plants. Irrigation began
alcohol for 2 min and 1% sodium hypochlorite for 5 min followed 3 days after planting and was done as needed. The field was located
by two rinses with sterile distilled water. The discs were treated in the municipality of Jaguarari, Bahia. The experimental design
separately with 200 lL of spore suspensions prepared from 109 was completely randomized with 10 replicates per treatment and
Trichoderma isolates and 10 min later with 200 lL of a F. guttiforme the whole experiment was done twice. Floral induction was done
spore suspension. Control discs were treated only with F. gutti- 60 days after transplanting by spraying 50 mL per plant of a mix-
forme or with distilled sterile water. All treatments were incubated ture containing 2.4% Ethephon, 2% urea and 0.03% CaCO3. One hun-
at 25 °C for five days in moist chambers containing one stalk disc. dred and sixty days after transplanting, plants were harvested to
The moist chambers were organized in a completely randomized assess fusariosis severity. Disease severity was determined as
design with ten replicates. Evaluation of the surface area colonized described above.
by F. guttiforme was done with the Image tool version 3.0 software
(Wilcox et al., 2002). The whole experiment was done twice. 2.6. Statistical analyses

2.4. Screening in the greenhouse In vitro data were recorded as percentage of pathogen inhibition
and disease inhibition in the field and in the greenhouse were
Pineapple plantlets (suckers) 30 cm high were obtained from recorded as grades on a scale varying from 0 to 5. Data on pathogen
commercial plantations located in the municipality of Jaguarari, or disease inhibition obtained in in vitro, greenhouse and field
Bahia. These plantlets were transferred to plastic bags containing experiments were subjected to the Shapiro–Wilk normality test.
28 J.T. De Souza et al. / Biological Control 94 (2016) 25–32

Because the data did not follow a normal distribution, they were 26 reduced disease severity by 72–81% in the first field experiment
analyzed by the one-way Kruskal–Wallis anova and paired means and by 68–84% in the second (Table 2). These three isolates also
were compared with the Wilcoxon’s test with Bonferroni’s correc- showed 80–100% of inhibition in greenhouse and in pineapple
tion. Spearman’s correlation was computed between pathogen stalk discs. In Petri plates, isolates 77 and 26 inhibited 100% of
inhibition in Petri plates and stalk discs, and disease control in pathogen growth, but isolate 26 inhibited 17% of pathogen growth
the greenhouse and in the field. Severity data obtained in the and was classified in group I (Supplementary Table S2). The other
greenhouse and field experiments were used to calculate the isolates displayed control activities lower than 50% and these
McKinney disease index (DI) using the formula: DI = [(scale levels were not consistent across the two field experiments
grade  frequency)/(No. of plants  maximum grade)]  100. All (Table 2). Isolate 141 showed lowest control activity in both field
analyses were done with the R software (R Core Team, 2013). experiments and also showed no activity in Petri plates, stalk discs
and in the greenhouse (Supplementary Table S2). The positive con-
trols always showed disease both in the greenhouse and in the
3. Results
field experiments, whereas there was no disease in the non-
inoculated controls (Table 2).
3.1. Diversity of Trichoderma

A total of 109 endophytic Trichoderma isolates were obtained 3.3. Relationships between origin, diversity and activity of
from four selected native plant species of the Caatinga biome. From Trichoderma isolates
these, 15.6%, 30.3%, 22.0% and 32.1% were obtained from A. como-
sus var. bracteatus (Anc), A. fraxinifolium (Asf), B. virgilioides (Bov), There was a positive Spearman’s correlation between stalk disc,
and C. pyramidalis (Cap), respectively. The 109 isolates were clus- greenhouse and field experiments, but not between Petri plate and
tered in 11 BOX groups on the basis of nine polymorphic bands. field experiments (Table 3). Correlation coefficients were higher
Most isolates (71.6%) were placed in 5 groups (1, 3, 7, 8, and 9) between greenhouse and field experiments. No clear relationship
and only one group contained a single isolate (Table 1). There was observed between the origin of the isolates and their activity
was no relation between the BOX groups and the origin of the against the pathogen in vitro or their activity against the disease
isolates. in the greenhouse and field experiments. Likewise, there was no
Six different Trichoderma species were identified among the relationship between BOX groups or species identity and activity
BOX representatives with fragments of the ITS and tef-1a regions (Supplementary Table S2).
sequenced. The species Trichoderma harzianum Rifai, Trichoderma
atroviride P. Karst. and Trichoderma koningiopsis Samuels, Suarez
4. Discussion
& Evans represented 3, 2 and 3 BOX groups, respectively. The other
three remaining species, including Trichoderma longibrachiatum
Trichoderma isolates were sought out in plants from the Caa-
Rifai, Trichoderma sinuosum Chaverri & Samuels and Trichoderma
tinga biome due to the harsh conditions these plants and their
virens (Mill, Giddens & Foster) Arx represented one BOX group
coexisting endophytes are adapted to. Pineapple cultivation is
each. The highest number of isolates belonged in the species T.
expanding in the Caatinga biome and the chance of finding co-
koningiopsis, followed by T. atroviride, T. harzianum and T. longi-
evolved biocontrol agents for F. guttiforme in this environment is
brachiatum, whereas only one isolate was T. virens and 5 were T.
theoretically higher. There are no reports on attempts to control
sinuosum (Table 1).
pineapple fusariosis with biocontrol agents under field conditions.
There seemed to be a uniform distribution of T. koningiopsis, T.
In this study, 109 endophytic Trichoderma isolates were
atroviride and T. harzianum in the four plant species. A higher num-
obtained from four native plants from the Caatinga biome. These
ber of isolates of T. longibrachiatum was recovered from the two
isolates were classified into 11 BOX groups and one representative
plant species from the Fabaceae family, B. virgilioides and C. pyrami-
of each group had its identity determined by sequencing fragments
dalis. Additionally, T. sinuosum was recovered only from A. comosus
of the ITS and tef-1a regions. A series of experiments performed in
var. bracteatus and A. fraxinifolium and T. viride only from C. pyrami-
Petri plates, pineapple stalk discs, greenhouse and finally in the
dalis (Supplementary Table S1).
field were used to select the best isolates to control F. guttiforme.
Three isolates were able to reduce disease severity in the field by
3.2. Selection of Trichoderma for pathogen inhibition and disease 68–84%.
control The Trichoderma isolates were grouped with BOX fingerprinting,
which was initially developed for bacteria (Versalovic et al., 1994),
The activity of 109 Trichoderma isolates against F. guttiforme but other authors showed its application for fungi, including Tri-
was assessed in Petri plates, pineapple stalk discs and in the green- choderma (Grosch et al., 2007). In our study, species identification
house. The levels of pathogen or disease control displayed by the was done for a limited number of isolates as an indicative of the
Trichoderma isolates were arbitrarily divided into three groups: identity of each BOX group, but it has been shown that there is a
isolates that reduced colonization or disease severity by less than good correspondence between BOX groups and species identity
40% (I), 41–79% (II), and by 80–100% (III) (Fig. 1). In Petri plate when sequencing data was used for confirmation, although in a
experiments there was an approximate even distribution of the few cases more than one species may occur in a BOX group
isolates in the different control levels. This distribution became (Grosch et al., 2007; Zachow et al., 2009). The division of T. harzia-
more concentrated in group I (less than 40% control) in the green- num and T. koningiopsis into three BOX groups each and T. atro-
house experiment (Fig. 1). The number of isolates that inhibited viride in two groups indicates that they are complexes composed
the pathogen or decreased disease severity by 80–100% (group of cryptic species yet to be described (Samuels, 2006).
III) was higher in Petri plate experiments (26 isolates) and in The number of Trichoderma species found in the sapwood of the
pineapple stalk discs (21 isolates), but sharply decreased in the Caatinga plants indicates that there is a relatively low diversity of
greenhouse experiment (3 isolates). Interestingly, the same three this genus as endophytes in these four plant species. The majority
isolates (77, 26, and 36) classified in group III (80–100% control) of the studies on Trichoderma diversity focused on soil and dead
in the greenhouse were the ones showing the highest levels of dis- plant materials as a source for the isolation of these fungi. Other
ease control in both field experiments (Table 2). Isolates 77, 36, and authors reported a higher number of species among isolates from
 J.T. De Souza et al. / Biological Control 94 (2016) 25–32 29

Fig. 1. Trichoderma isolates distributed according to the host of origin and classified in different intervals of Fusarium guttiforme control: I (0–40%), II (41–79%), and III (80–
100%) in experiments conducted in Petri plates (A), bioassays in stalk discs (B) and in a greenhouse (C). Host plants were Ananas comosus var. bracteatus, Astronium
fraxinifolium, Bowdichia virgilioides, and Caesalpinia pyramidalis. A total of 109 endophytic isolates of Trichoderma were tested in each of the experiments. Inhibition of mycelial
growth was evaluated in Petri plates and on the surface of pineapple stalk discs and fusariosis severity was evaluated in a greenhouse experiment.
30 J.T. De Souza et al. / Biological Control 94 (2016) 25–32

Table 2 the dominant endophytic genus in any plant species. For example,
Field experiments with fifteen selected endophytic isolates of Trichoderma represent- seven Trichoderma species accounted for 14% of a collection of 344
ing six BOX groups and obtained from different plant species from the Caatinga
biome. Data correspond to the control in relation to the positive, inoculated control
fungal endophytes in the sapwood and pods of Theobroma gileri
(CIF). Disease severity was scored 160 days after transplanting with a grading scale Cuatr. in Ecuador (Evans et al., 2003), while in Theobroma cacao
varying from 0 to 5. L., five species of Trichoderma accounted for 13% of 189 isolates
Isolate/ BOX Origin Experiment Experiment
(Hanada et al., 2010). In the sapwood of rubber trees [Hevea
treatment group 1 2 brasiliensis (Willd. ex A. Juss.) Muell. Arg.] sampled in Peru, two
Control (%)
species of Trichoderma composed 22% of a total of 175 isolates of
endophytes (Gazis and Chaverri, 2010). In coffee plants (Coffea ara-
CIFa – – 0.0c 0.0c
bica L.), Trichoderma species were rare and comprised 0.35% of 843
141 7 A. comosus var. 6.0NS 11.0NS
bracteatus isolates from four different countries (Vega et al., 2010), while it
122 7 B. virgilioides 21.0NS 20.0NS was not detected among 357 isolates from leaves and stems of five
129 7 B. virgilioides 21.0NS 25.0NS ornamental flowering species in China (Zhou et al., 2014), and also
63 3 A. fraxinifolium 33.0NS 18.0NS
in samples from Jatropha curcas L. (Kumar and Kaushik, 2013) and
40 9 A. fraxinifolium 28.0NS 30.0NS
80 1 C. pyramidalis 32.0NS 27.0NS
brazilwood (Caesalpinia echinata Lam.) stems and bark (Campos
54 3 A. fraxinifolium 30.0NS 30.0NS et al., 2015).
50 8 B. virgilioides 34.0NS 34.0NS Almost all species of Trichoderma recovered from the Caatinga
66 7 C. pyramidalis 37.0NS 30.0NS plants are common in tropical regions. Except for T. sinuosum, all
116 11 A. comosus var. 38.0NS 39.0*
the other species were previously found as endophytes in diverse
bracteatus
42 8 B. virgilioides 45.0NS 36.0* plant species, such as T. cacao (Evans et al., 2003), H. brasiliensis
30 9 A. fraxinifolium 47.0NS 50.0* (Gazis and Chaverri, 2010), and Camelia sinensis (L.) Kuntze (Wu
36 9 B. virgilioides 77.0* 68.0* et al., 2009).
26 9 A. comosus var. 72.0* 68.0* Another hypothesis to explain the low number of species and
bracteatus
77 8 A. fraxinifolium 81.0* 84.0*
BOX groups found in the Caatinga plants might be due to a higher
NICb – – 100.0* 100.0* specialization of these fungi to colonize the interior of plants. There
is evidence that endophytic Trichoderma from banana roots are less
NS – not significantly different from the positive, inoculated control at 5%
probability.
diverse than isolates of the same species living on the root surface
a
Positive control inoculated with Fusarium guttiforme. (Xia et al., 2011). Additionally, endophytic Trichoderma species are
b
Negative control, non-inoculated. grouped in terminal clades of phylogenetic analyses, indicating
c
Pairwise comparisons between the inoculated control and the treatments with that the capability to colonize plants endophytically is a recently
the Wilcoxon’s test corrected by Bonferroni.
* evolved trait (Chaverri and Samuels, 2013).
Significantly different from the positive control at 5%.
The results of this study indicate that there might be some host
specificity for T. longibrachiatum and T. sinuosum, but T. koningiop-
Table 3 sis, T. harzianum, and T. atroviride appear to lack specificity, as their
Comparisons of experiments with Spearman’s coefficient of correlation showing the occurrence was uniform across the four plant species in the area
number of Trichoderma isolates considered in each comparison, Spearman’s Rho studied. Trichoderma species interact with host plants producing
correlation coefficient and the significance at the 5% level. host-specific patterns of gene expression (Morán-Diez et al.,
Comparisons Number of Correlation P value 2015). These different patterns might influence the levels of endo-
isolates coefficient phytic colonization in different host plants in a manner similar to
Petri plates  stalk discs 109 0.48 <0.0001* what was demonstrated for bacteria. A recent study has shown
Petri plates  greenhouse 109 0.26 0.007* that plants choose the bacteria allowed to colonize their roots
Petri plates  field exp. 1 15 0.48 0.07NS and probably other organs (Lebeis et al., 2015). This hypothesis still
Petri plates  field exp. 2 15 0.50 0.06NS
awaits experimental confirmation in interactions between plants
Stalk discs  greenhouse 109 0.56 <0.0001*
Stalks discs  field exp. 1 15 0.65 0.008* and Trichoderma species.
Stalks discs  field exp. 2 15 0.62 0.014* The high temperatures of the Caatinga biome may influence
Grenhouse  field exp. 1 15 0.79 <0.0001* endophytic species composition, selecting the ones more adapted
Grenhouse  field exp. 2 15 0.76 <0.0001*
to these specific conditions. Although growth at 37 °C or higher
NS – non-significant at 5% probability is a trait that qualifies microorganisms to infect humans, not all
*
Significantly different at 5%. species that grow at this temperatures are human pathogens. With
the exception of T. virens, all the other species found in this study
were listed as clinically relevant in a recent study (Sandoval-Denis
soil, such as 17 species among 76 soil isolates from Russia, Nepal et al., 2014). From these, only T. longibrachiatum and some isolates
and North India (Kulling et al., 2000); 17 species among 96 isolates of the T. harzianum species complex and T. atroviride are able to
from Southeast Asia (Kubicek et al., 2003); and eight species were infect immunocompromised patients (Sandoval-Denis et al.,
reported in 42 isolates from the Canary Island of Tenerife (Zachow 2014). Pineapple fusariosis is a disease that occurs at high temper-
et al., 2009). In plant materials, 75 species of Trichoderma were atures and only decreases its incidence when temperatures are
reported among 620 isolates obtained in 14 countries in Central higher than 35 °C (Ghini et al., 2011; Matos et al., 2000). Several
and Northern Europe (Jaklitsch, 2009, 2011) and 90 species were Trichoderma species will have a competitive advantage when com-
identified among 636 isolates collected in Southern Europe pared to some other biocontrol agents because they are less
(Jaklitsch and Voglmayr, 2015). affected by high temperatures (Bettiol and Ghini, 2009), as long
Contrary to studies in soil and plant materials, where Tricho- as the risk of being human pathogens is cleared.
derma is relatively common, no large-scale studies with a high Selection of biocontrol agents is a very time-consuming activity
number of endophytic isolates is available. There are, however, because only approximately 1–2% of the antagonists tested have
several studies reporting the diversity of fungal endophytes in a some potential (Chen et al., 1996; Guetskyl et al., 2002), and our
great number of plant species and Trichoderma is among them in results showed approximately 3% of potential candidates among
most cases. Nevertheless, Trichoderma was never reported to be the Trichoderma isolates. Unfortunately, the number of promising
 J.T. De Souza et al. / Biological Control 94 (2016) 25–32 31

antagonists in plants from the Caatinga biome was not different (Brazilwood) and identification of beauvericin as a trypanocidal metabolite
from Fusarium sp. Mem. Inst. Oswaldo Cruz 110, 65–74. http://dx.doi.org/
from the expected number of potential biocontrol agents reported
10.1590/0074-02760140243.
for other systems. Chaverri, P., Samuels, G.J., 2013. Evolution of habitat preference and nutrition mode
Greenhouse experiments were better predictors of the activity in a cosmopolitan fungal genus with evidence of interkingdom host jumps and
of Trichoderma isolates against F. guttiforme in the field than major shifts in ecology. Evolution 67, 2823–2837. http://dx.doi.org/10.1111/
evo.12169.
pineapple stalk discs. Not unexpectedly, as already shown for other Chen, Y., Mei, R., Liu, L., Kloepper, J.W., 1996. The use of yield-increasing bacteria
systems (Fravel, 1988), Petri plate experiments cannot be effi- (YIB) as plant growth-promoting rhizobacteria in Chinese agriculture. In:
ciently used to select biocontrol agents against pineapple fusariosis Utkhede, R.S., Gupta, V.K. (Eds.), Management of Soil Born Disease. Kalyani
Publishers, Ludhiana, pp. 165–184.
for field applications. Due to the complexity of the natural systems, Chen, J., Li, B., Qin, G., Tian, S., 2015. Mechanism of H2O2-induced oxidative stress
bioassays in vitro without the host have little chances of success- regulating viability and biocontrol ability of Rhodotorula glutinis. Int. J. Food
fully selecting biocontrol agents as compared to in vivo bioassays Microbiol. 193, 152–158. http://dx.doi.org/10.1016/j.ijfoodmicro.2014.10.025.
CPRM, Serviço Geológico do Brasil, 2006. Levantamento da Geodiversidade Projeto
with the host (Fravel, 1988). The best approach to select Tricho- Atlas Pluviométrico do Brasil Isoietas Anuais Médias No Período de 1977 a 2006.
derma isolates to control F. guttiforme is to perform greenhouse Ministério de Minas e Energia, Brazil, <http://www.cprm.gov.br/
bioassays. publique/media/Isoietas_Totais_Anuais_1977_2006> (Accessed in 2015).
Cunha, G.A.P., Reinhardt, D.H.R.C., 2004. Orientações básicas para o cultivo do
The mechanisms of action employed by the isolates of Tricho- abacaxizeiro. Cruz das Almas: Embrapa Mandioca e Fruticultura Tropical,
derma to control F. guttiforme were not a subject of our investiga- Editora Embrapa, 4p. (Embrapa Mandioca e Fruticultura Tropical.
tion, but mycoparasitism is probably involved because this genus ComunicadoTécnico, 110).
Doyle, J.J.T., Doyle, J.L., 1990. Isolation of plant DNA from fresh tissue. Focus 12, 13–15.
is primarily mycophagous (Atanasova et al., 2013). Other mecha-
Druzhinina, I.S., Seiboth, V.S., Estrella, A.H., Horwitz, B.A., Kenerley, C.M., Monte, E.,
nisms of action, such as antibiosis, competition for space and nutri- Mukherjee, P.K., Zeilinger, S., Grigoriev, I.V., Kubicek, C.P., 2011. Trichoderma:
ents, and induction of resistance may have occurred the genomics of opportunistic success. Nat. Rev. Microbiol. 9, 749–759. http://
simultaneously and contributed to the control levels displayed by dx.doi.org/10.1038/nrmicro2637.
Evans, H.C., Holmes, K.A., Thomas, S.E., 2003. Endophytes and mycoparasites
Trichoderma against F. guttiforme. associated with an indigenous forest tree, Theobroma gileri, in Ecuador and a
In our greenhouse and field experiments the Trichoderma iso- preliminary assessment of their potential as biocontrol agents of cocoa diseases.
lates were applied as a suspension of spores where pineapple Mycol. Progr. 2, 149–160. http://dx.doi.org/10.1007/s11557-006-0053-4.
FAOSTAT, 2012. Food and Agriculture Organization of the United Nations: Statistics
plantlets were immersed. Soil application of the antagonists may Division. <http://faostat.fao.org> (Accessed in 2015).
be tested in future experiments. The activity of the isolates may Fravel, D.R., 1988. Role of antibiosis in the biocontrol of plant diseases. Annu. Rev.
be further improved by the use of an adequate formulation and Phytopathol. 26, 75–91.
Gazis, R., Chaverri, P., 2010. Diversity of fungal endophytes in leaves and stems of
spores produced in an optimized culture medium. The lack of an wild rubber trees (Hevea brasiliensis) in Peru. Fungal Ecol. 3, 240–254. http://dx.
appropriate formulation allied with the complex interactions that doi.org/10.1016/j.funeco.2009.12.001.
biocontrol agents have with plants, other microorganisms and Ghini, R., Bettiol, W., Hamada, E., 2011. Diseases in tropical and plantation crops as
affected by climate changes: current knowledge and perspectives. Plant Pathol.
the environment were probably responsible for the inconsistencies 60, 122–132. http://dx.doi.org/10.1111/j.1365-3059.2010.02403.x.
of some isolates in the field experiments. For example, isolates Giulietti, A.M., Bocageneta, A.L., Castro, A.A.J.F., Gamarra-Rojas, C.F.L., Sampaio, E.V.
Tc116, Tc42 and Tc30 (Table 2), showed different results in each S.B., Virgínio, J.F., Queiroz, L.P., Figueiredo, M.A., Rodal, M.J.N., Barbosa, M.R.V.,
Harley, R.M., 2004. Diagnóstico da vegetação nativa do bioma Caatinga. In:
experiment. These inconsistences were shown to be a common
Silva, J.M.C., Tabarelli, M., Fonseca, M.T., Lins, L.V. (Eds.), Biodiversidade da
phenomenon in biological control (Fravel, 1988). Additional Caatinga: Áreas e Ações Prioritárias para a Conservação. Ministério do Meio
knowledge on these interactions will increase the chances of devel- Ambiente, pp. 48–90.
oping biocontrol products with improved performance. Grosch, R., Lottmann, J., Rehn, V.N.C., Rehn, K.G., Mendonça-Hagler, L., Smalla, K.,
Berg, G., 2007. Analysis of antagonistic interactions between Trichoderma
In summary, the results of our screening experiments yielded isolates from Brazilian weeds and the soil-borne pathogen Rhizoctonia solani. J.
three Trichoderma isolates from the 109 tested that may be applied Plant Dis. Prot. 114, 167–175.
in the field to manage pineapple fusariosis and contribute to Guetskyl, R., Shtienberg, D., Dinoor, A., Elad, Y., 2002. Establishment, survival and
activity of the biocontrol agents Pichia guilermondii and Bacillus mycoides
reduce fungicide use and maintain the stability and durability of applied as a mixture on strawberry plants. Biocontrol Sci. Technol. 12, 705–714.
the resistance in pineapple cultivars against fusariosis. http://dx.doi.org/10.1080/0958315021000039888.
Hampl, V., Pavlicek, A., Flegr, J., 2001. Construction and bootstrap analysis of DNA
fingerprinting-based phylogenetic trees with the freeware program FreeTree:
Acknowledgments application to trichomonad parasites. Int. J. Syst. Evol. Microbiol. 51, 731–735.
http://dx.doi.org/10.1099/00207713-51-3-731.
Hanada, R.E., Pomella, A.W.V., Costa, H.S., Bezerra, J.L., Loguercio, L.L., Pereira, J.O.,
The authors thank the financial support of CAPES and CNPq for
2010. Endophytic fungal diversity in Theobroma cacao (cacao) and T.
conducting the experiments and J.T.S. acknowledges a productivity grandiflorum (cupuaçu) trees and their potential for growth promotion and
scholarship provided by CNPq. biocontrol of black-pod disease. Fungal Biol. 114, 901–910. http://dx.doi.org/
10.1016/j.funbio.2010.08.006.
IBGE, 2004. Instituto Brasileiro de Geografia e Estatística, Mapa de Biomas do Brasil,
Appendix A. Supplementary data Primeira Aproximação. <http://saladeimprensa.ibge.gov.br/ noticias?view=
noticia&id=1&busca=1&idnoticia=169> (Accessed in 2015).
INMET, 2015. Instituto Nacional de Meteorologia, Mapas do Boletim
Supplementary data associated with this article can be found, in Agroclimatológico. Ministério da Agricultura, Pecuária e Abastecimento,
the online version, at http://dx.doi.org/10.1016/j.biocontrol.2015. <http://www.inmet.gov.br/portal/index.php?r=agrometeorologia/
12.005. boletimAgroclimatologico> (Accessed in 2015) (1p.).
Jaklitsch, W.M., 2009. European species of Hypocrea part I. The green-spored
species. Stud. Mycol. 63, 1–91. http://dx.doi.org/10.3114/sim.2009.63.01.
References Jaklitsch, W.M., 2011. European species of Hypocrea part II: species with hyaline
ascospores. Fungal Divers. 48, 1–250. http://dx.doi.org/10.1007/s13225-011-
0088-y.
Atanasova, L., Le Crom, S., Gruber, S., Coulpier, F., Seidl-Seiboth, S., Kubicek, C.P.,
Jaklitsch, W.M., Voglmayr, H., 2015. Biodiversity of Trichoderma (Hypocreaceae) in
Druzhinina, I.S., 2013. Comparative transcriptomics reveals different strategies
Southern Europe and Macaronesia. Stud. Mycol. 80, 1–87. http://dx.doi.org/
of Trichoderma mycoparasitism. BMC Genomics 14, 121. http://dx.doi.org/
10.1016/j.simyco.2014.11.001.
10.1186/1471-2164-14-121.
Kubicek, C., Bissett, J., Druzhinina, I., Kulling-Grandiger, C., Szakacs, G., 2003.
Bettiol, W., Ghini, R., 2009. Impactos das mudanças climáticas sobreo controle
Genetic and metabolic diversity of Trichoderma: a case study on South East
biológico de doenças de plantas. In: Bettiol, W., Morandi, M.A.B. (Eds.),
Asian isolates. Fungal Genet. Biol. 38, 310–319. http://dx.doi.org/10.1016/
Biocontrole de Doenças de Plantas: Uso e Perspectivas. Embrapa Meio
S1087-1845(02)00583-2.
Ambiente, Jaguariúna, Brazil, pp. 29–48.
Kulling, C., Szakacs, G., Kubicek, C., 2000. Molecular identification of Trichoderma
Campos, F.F., Sales Junior, P.A., Romanha, A.J., Araújo, M.S.S., Siqueira, E.P., Resende,
species from Russia, Siberia and the Himalaya. Mycol. Res. 104, 1117–1125.
J.M., Alves, T.M.A., Martins-Filho, O.A., Santos, V.L., Rosa, C.A., Zani, C.L., Cota, B.
http://dx.doi.org/10.1017/S0953756200002604.
B., 2015. Bioactive endophytic fungi isolated from Caesalpinia echinata Lam.
32 J.T. De Souza et al. / Biological Control 94 (2016) 25–32

Kumar, S., Kaushik, N., 2013. Endophytic fungi isolated from oil-seed crop Jatropha Samuels, S.J., Ismaiel, A., De Souza, J.T., Chaverri, P., 2012. Trichoderma stromaticum
curcas produces oil and exhibit antifungal activity. PLoS One 8 (2), e56202. and its overseas relatives. Mycol. Progr. 11, 215–254. http://dx.doi.org/10.1007/
http://dx.doi.org/10.1371/journal.pone.0056202. s11557-011-0743-4.
Lebeis, S.L., Paredes, S.H., Lundberg, D.S., Breakfield, N., Gehring, J., McDonald, M., Sandoval-Denis, M., Sutton, D.A., Cano-Lira, J.F., Gené, J., Fothergill, A.W.,
Malfatti, S., del Rio, T.G., Jones, C.D., Tringe, S.G., Dangl, J.L., 2015. Salicylic acid Wiederhold, N.T., Guarro, J., 2014. Phylogeny of the clinically relevant species
modulates colonization of the root microbiome by specific bacterial taxa. of the emerging fungus Trichoderma and their antifungal susceptibilities. J. Clin.
Science. http://dx.doi.org/10.1126/science.aaa8764. Microbiol. 52, 2112–2125. http://dx.doi.org/10.1128/JCM.00429-14.
Luambano, N.D., Narla, R.D., Wanjohi, W.J., Kimenju, J.W., Kerry, B.R., 2015. Taylor, N.P., Zappi, D., 2002. Distribuição das espécies de Cactaceae na caatinga. In:
Integrated management of root-knot nematodes in a tomato-maize crop Sampaio, E.V.S.B., Giulietti, A.M., Virgínio, J., Gamarra-Rojas, C.F.L. (Eds.),
system using the biocontrol fungus Pochonia clamydosporia. Crop Prot. 71, 45– Vegetação e Flora das Caatingas. APNE/CNIP, pp. 123–125.
50. http://dx.doi.org/10.1016/j.cropro.2015.01.013. Vega, F.E., Simpkins, A., Aime, M.C., Posada, F., Peterson, S.W., Rehner, S.A., Infante,
Matos, A.P., Cabral, J.R.S., 2006. Evaluation of pineapple genotypes for resistance to F., Castillo, A., Arnold, A.E., 2010. Fungal endophyte diversity in coffee plants
Fusarium subglutinans. Acta Hortic. 702, 73–77. http://dx.doi.org/10.17660/ from Colombia, Hawaii, Mexico and Puerto Rico. Fungal Ecol. 3, 122–138. http://
ActaHortic.2006.702.8. dx.doi.org/10.1016/j.funeco.2009.07.002.
Matos, A.P., Reinhardt, D.H., 2009. Pineapple in Brazil: characteristics, research and Ventura, J.A., Zambolim, L., Chaves, G.M., 1992. Integrated management system for
perspectives. Acta Hortic. 822 (25–36), 2009. http://dx.doi.org/10.17660/ pineapple Fusarium disease control. Acta Hort. 344 (439–454), 2009. http://dx.
ActaHortic.2009.822.1. doi.org/10.17660/ActaHortic.2009.822.24.
Matos, A.P., Cabral, J.R.S., Sanches, N.F., Caldas, R.C., 2000. Effect of temperature Ventura, J.A., Costa, H., Cabral, J.R.S., Matos, A.P., 2009. ‘Vitória’: new pineapple
and rainfall on the incidence of Fusarium subglutinans on pineapple fruits. cultivar resistant to fusariosis. Acta Hort. 822, 51–56. http://dx.doi.org/
Acta Hortic. 529 (265–72), 2000. http://dx.doi.org/10.17660/ 10.17660/ActaHortic.2009.822.4.
ActaHortic.2000.529.32. Versalovic, J., Schneider, M., de Bruijn, F.J., Lupski, J.R., 1994. Genomic fingerprinting
Matos, A.P., Sanches, N.F., Teixeira, F.A., Elias Júnior, J., 2009. Integrated of bacteria using repetitive sequence-based Polymerase Chain Reaction.
management of Fusariosis in pineapple fields under integrated production Method. Mol. Cell Biol. 5, 25–40.
system. Acta Hortic. 822 (199–204), 2009. http://dx.doi.org/10.17660/ Wilcox, D., Dove, B., McDavid, D., Greer, D., 2002. Image Tool for Windows, Version
ActaHortic.2009.822.24. 3.0. University of Texas Health Science Center in San Antonio.
Morán-Diez, M.E., Trushina, N., Lamdan, N.L., Rosenfelder, L., Mukherjee, P.K., Wu, H.-Q., Su, J.-Q., Xie, M.-Y., Yang, M.-H., 2009. An endophytic Trichoderma species
Kenerley, C.M., Horwitz, B.A., 2015. Host-specific transcriptomic pattern of from Camellia sinensis: its characterization and endophytism. Mycosystema 28,
Trichoderma virens during interaction with maize or tomato roots. BMC 342–348.
Genomics 16, 8. http://dx.doi.org/10.1186/s12864-014-1208-3. Xia, X., Lie, T.K., Qian, X., Zheng, Z., Huang, Y., Shen, Y., 2011. Species diversity,
Queiroz, L.P., 2002. Distribuição das espécies de Leguminoseae na caatinga. In: distribution, and genetic structure of endophytic and epiphytic Trichoderma
Sampaio, E.V.S.B., Giulietti, A.M., Virgínio, J., Gamarra-Rojas, C.F.L. (Eds.), associated with banana roots. Microbial Ecol. 61, 619–625. http://dx.doi.org/
Vegetação e Flora das Caatingas. APNE/CNIP, pp. 141–153. 10.1007/s00248-010-9770-y.
R Core Team, 2013. R: A Language and Environment for Statistical Computing. R Zachow, C., Berg, C., Müller, H., Meincke, R., Komon-Zelazowska, M., Druzhinina, I.S.,
Foundation for Statistical Computing, Vienna, Austria, <http://www.R-project. Kubicek, C.P., Berg, G., 2009. Fungal diversity in the rhizosphere of endemic
org/>. plant species of Tenerife (Canary Islands): relationship to vegetation zones and
Reinhardt, D.H., Cabral, J.R.S., Souza, L.F.S., Sanches, N.F., Matos, A.P., 2002. Pérola environmental factors. ISME J. 3, 79–92. http://dx.doi.org/10.1038/
and smooth cayenne pineapple cultivars in the state of Bahia, Brazil: growth, ismej.2008.87.
flowering, pests, diseases, yield and fruit quality aspects. Fruits 57, 43–53. Zhou, Z., Zhang, C., Zhou, W., Li, W., Chu, L., Yan, J., Li, H., 2014. Diversity and plant
http://dx.doi.org/10.1051/fruits:2002005. growth-promoting ability of endophytic fungi from the five flower plant species
Samuels, G.J., 2006. Trichoderma: systematics, the sexual state, and ecology. collected from Yunnan, Southwest China. J. Plant Interact. 9, 585–591. http://dx.
Phytopathology 96, 195–206. http://dx.doi.org/10.1094/PHYTO-96-0195. doi.org/10.1080/17429145.2013.873959.