ARTICLE IN PRESS

Soil Biology & Biochemistry 39 (2007) 2769–2776

www.elsevier.com/locate/soilbio

Effect of Colletia hystrix (Clos), a pioneer actinorhizal plant from the

Chilean matorral, on the genetic and potential metabolic diversity of the
soil bacterial community
Julieta Orlando, Mónica Chávez, Lorena Bravo, Rafael Guevara, Margarita Carú
Departamento de Ciencias Ecológicas, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile
Received 23 January 2007; received in revised form 24 May 2007; accepted 31 May 2007
Available online 21 June 2007

Abstract

Colletia hystrix are dominant shrubs in the sclerophyllous matorral, a natural ecosystem in the central valley of Chile affected by
erosion, soil with low fertility and limiting nitrogen. The soil microbial communities associated to these pioneer plants have received little
attention even though they may have an important role in the ability of these to colonize the nutrient-poor soils from these semi-arid
ecosystems. T-RFLP profiles using 16S rDNA were used to compare the bacterial community structure from soil samples (enriched and
unenriched) associated to C. hystrix and neighboring soil without plant cover (bulk soil). Additionally, the microbial communities from
both habitats were compared at the metabolic profile level using the Biolog EcoPlateTM system. Our results showed that the bacterial
community from samples of soil associated to these plants formed a separate cluster from samples derived from the neighboring soil.
These data suggest that soil associated to C. hystrix is a different microhabitat to bulk soil. When an enrichment step was performed on
the samples, the T-RFLP profiles obtained showed few T-RFs suggesting that only some species were recovered. The enriched samples
exhibited a low similarity between them and are clearly separated from the unenriched samples. On the other hand, the comparison of the
unenriched samples from both habitats based on sole-carbon-source utilization profiles was unable to differentiate the samples according
to their habitat.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: Bacterial community; Biolog EcoPlate; T-RFLP; Actinorhizal plant; Sclerophyllous matorral

1. Introduction communities associated to pioneer plant species involved in

the early stages of plant succession, which are able to
Soil microbial communities and the biogeochemical colonize nutrient-poor soils in semi-arid ecosystems.
processes they mediate are critical for the correct function- We focus our study on the effect of Colletia hystrix
ing and maintenance of the terrestrial ecosystems (Kent (Clos) on the microbial communities; these plants are
and Triplett, 2002). To better understand the role of dominant shrubs in the Chilean sclerophyllous matorral, a
microbial communities in the processes occurring in the natural ecosystem in the central valley affected by erosion,
soil and how they respond to environmental changes, it is soil with low fertility and limiting nitrogen (Fuentes et al.,
essential to determine the structure and function of these 1986). C. hystrix is called an ‘‘actinorhizal’’ plant because
microbial assemblages. The majority of studies on the of its ability to form root nodules with Frankia, a N2-fixing
effect of the plant on microbial community structure have actinomycete (Carú, 1993; Carú and Cabello, 1999). These
been done on grassland or agricultural plant species (e.g. plants form vegetational patches in the matorral and are
Smit et al., 2001; Fang et al., 2005; Steenwerth et al., 2005). considered pioneer species that contribute to primary plant
However, little information is available on the microbial successions, stabilizing the soil and increasing its nitrogen
content (Silvester et al., 1985), and which could form
Corresponding author. islands of fertility in these ecosystems. Erickson et al.
E-mail address: [email protected] (M. Carú). (2005) found that under vegetational patches of Ceanothus,

0038-0717/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.soilbio.2007.05.024
 ARTICLE IN PRESS
2770 J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776

Table 1
Edaphic parameters of the soil samples

pH OM (%) N (mg kg1) P (mg kg1) K (mg kg1)

Soil associated to C. hystrix 7.2770.09a 8.7971.71 48.9075.41 88.40718.77 402.37788.94

Bulk soil 7.1270.23 7.2171.99 20.7779.29 61.4779.05 253.00756.20

OM, organic matter; N, nitrogen; P, phosphorus; K, potassium.

a
Values are means7S.D. (n ¼ 5).
 Statistically significant differences (ANOVA) between soil associated to C. hystrix and bulk soil Po0.05 (n ¼ 5).

another actinorhizal plant, there is a higher amount of sample was stored at 4 1C for chemical analysis, metabolic
nitrogen than in bare soils and coniferous forests. There- assays and culture. Edaphic parameters were determined as
fore, we hypothesized that the C. hystrix plant cover could follows: total nitrogen by the Kjeldahl method, phosphorus
provide a microhabitat in this soil that defines a bacterial by the molybdenum blue method, potassium by flame
community that is different to that associated to neighbor- photometry, organic matter content by colorimetric
ing soil without plant cover. determination following wet potassium dichromate diges-
The aim of this work was to determine if there are tion and pH–H2O by potentiometric determination in a soil
detectable differences between the composition of the suspension in water (Forster, 1995). Only the potassium
bacterial communities from soil associated to C. hystrix and nitrogen content showed a statistically significant
and those related to the neighboring soil without plant variation between the sites studied (Po0.05 using standard
cover (bulk soil). The genetic diversity was determined ANOVA test) (Table 1).
by T-RFLP (terminal restriction fragment length poly-
morphism) (Liu et al., 1997) profiling of the 16S rDNA
from the bacterial community derived from environmental 2.2. DNA extraction and T-RFLP profiles of unenriched
soil samples with and without an enrichment step. and enriched samples
Additionally we determined the community level metabolic
profile (Garland and Mills, 1991) to evaluate differences in Total DNA from unenriched soil samples was obtained
sole-carbon-source utilization profiles in both microbial from 0.25 g of material using the ‘‘UltraClean Soil DNA’’
communities. kit (MoBio Laboratoriess). The enrichment culture frac-
tions were prepared in liquid bacto nutrient broth medium
(NB medium Difco) inoculated with a soil suspension and
2. Materials and methods incubated at 28 1C for 48 h in a shaker. The samples from
the enrichment culture step were identified with an
2.1. Sampling site additional ‘‘e’’: Ce for soil associated to C. hystrix and
Be for the bulk soil. The colony forming units per gram of
The samples were collected in April 2005 in the locality dry weight soil (cfu g1 dw ) were enumerated by serial
of ‘‘El Romeral’’, Cajón del Maipo (331480 S, 701140 W). dilutions of soil samples on NB-agar incubated at 28 1C
The study site selected is located in the Andean pre- for 48 h and the values were in the order of 106–107. The
mountain range of Central Chile dominated by C. hystrix enrichment culture was centrifuged, washed with sterile
growing in natural conditions. distilled water and total DNA was purified directly from
The soil samples were collected from adult plants of the homogeneate following lysis in a thermocycler under
C. hystrix with an average height of 1–2 m and a canopy the following program: 98 1C for 1 min and 4 1C for 1 min,
with a diameter of approximately 1.5 m. These plants have for six cycles. The universal primers fD1 and rP2
a deep rooting system like a tap root, and many smaller (Weisburg et al., 1991) were used to amplify the ribosomal
branch roots may grow from it. The plants selected are 16S gene. For the T-RFLP analysis, the primers were
found distributed in an area of 5 m  15 m forming labeled with 50 carboxyfluorescein (Applied Biosystemss,
vegetational patches in the matorral. The plant spatial Oster City, USA) and used in independent assays. The
distribution is described in greater detail in Chávez and amplicons were purified with the UltraClean PCR Clean-
Carú (2006). A soil sample associated to C. hystrix was up Kit (MoBio Laboratoriess) followed by digestion with
collected 10 cm from the stem of five different plants and a the restriction enzyme HaeIII (Invitrogen). The digestion
soil sample without plant cover was collected approxi- fragments were separated by capillary electrophoresis in
mately 1.0 m from each plant stem. The samples were an ABI PRISM 3100 Avant-Genetic-Analyzer Prisms
designated as C for soil associated to C. hystrix and B for sequencer. The sizes of the T-RFs were determined using
the bulk soil. Each soil sample collected from the upper GeneScan 3.1 (ABI) and the T-RFs with a peak height of
10 cm with 6 cm diameter corers was homogenized and greater or equal to 30 fluorescence units were analyzed by
sieved to 2 mm aggregate size. One sample fraction was aligning fragments to the size standard. Peaks greater than
stored at 20 1C for DNA extraction and the rest of the or equal to 30 UF were selected since they were clearly
 ARTICLE IN PRESS
J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776 2771

distinguishable from the basal fluorescence. Smaller peaks (GeoMem, Blairgowrie, UK) to confirm the groupings
may not represent real T-RFs. To avoid the inclusion of observed in the UPGMA cluster analysis.
remnant primers, the terminal fragments smaller than 30 bp
were excluded from the analysis. Reproducibility of 3. Results
patterns was confirmed by repeated T-RFLP analysis
using the same DNA extracts. Patterns from different 3.1. T-RFLP profiles
samples were normalized to identical total fluorescence
units by an iterative normalization procedure (Dunbar The frequency of the most abundant T-RFs obtained
et al., 2001). Communities were characterized by the from the enriched (Ce and Be) and unenriched samples
numbers and heights of the peaks. The relative abundance (C and B) are shown in Table 2. If we consider those T-RFs
of T-RFs was expressed in percentage. with a relative abundance of more than 2.5%, the
unenriched samples showed between 10 and 20 T-RFs
2.3. Community-level physiological profiles while the enriched samples had between 3 and 7 T-RFs. In
the majority of cases, these T-RFs represent more than
To compare the potential metabolic diversity of the 80% of the terminal fragments detected for both kinds of
microbial communities from both habitats, we used the samples (unenriched and enriched). Fig. 1 shows the
Biolog EcoPlateTM system (Garland and Mills, 1991), relative abundance of the most important fragments in
which is based on the ability of micro-organisms to utilize a the different samples studied. The T-RF profiles from the
sole-carbon-source. Each Biolog EcoPlate has 31 different unenriched samples differ in at least eight T-RFs, where
carbon substrates with triplicates providing a metabolic five of them are present in the soil associated to the plant
pattern of the microbial community. Microbial suspensions and three are unique to bulk soil, suggesting that bacterial
were prepared from 1 g of fresh soil in sterile PBS buffer assemblages from these two habitats are different. On the
and left in a shaker overnight. In order to reduce the effect other hand, bacterial communities obtained from the
of inoculum density, we applied a normalization of the cell enriched samples from soil associated to C. hystrix (Ce)
numbers per well according to the recommendation or bulk soil (Be) were represented by simple DNA profiles
indicated by Preston-Mafham et al. (2002) who suggested with fewer T-RFs than the samples from the enrichment-
that each well be inoculated with the same number of independent fractions. In the C samples, 11 fragments were
viable bacteria. The soil suspensions were plated on solid found which were not detected in the Ce samples, likewise
NB medium and incubated at 28 1C for 48 h to determine 10 T-RFs were observed in the B samples which were not
the number of viable micro-organisms in each sample. The present in the Be samples. Although some T-RFs are
Biolog EcoPlates were inoculated with 100 ml of the soil common to all the samples such as 212, 622 and 804 bp,
suspensions adjusted to obtain 105 culturable micro- they are generally more abundant in the enriched fractions.
organisms per well and incubated at 28 1C for a week. In addition, in the enriched samples from bulk soil there
The color development in the individual wells was
registered every 12 h. Plates were read using an ELISA Table 2
plate reader at 590 nm. The absorbance in the water blank Frequency of the T-RFs obtained from unenriched and enriched samples
was subtracted from the absorbance readings of all the
Samples No. of T-RFs % T-RFs
other wells.
(42.5%) (42.5%)

2.4. Data analysis Unenriched C1 20 92.03

samples C2 15 82.64
C3 11 75.75
For the comparison of the T-RFLP, a binary matrix was
C4 14 81.9
constructed considering the presence or absence of a peak C5 12 80.16
in the profile. For the metabolic profiles, the dendrogram B1 10 85.74
was constructed using a binary matrix considering the B2 11 92.03
utilization or non-utilization of a C-source during the week B3 12 88.59
B4 16 93.26
of incubation. The readings were taken every 12 h and a
B5 12 88.06
presence/absence matrix was constructed including all the
readings for the week of incubation. The similarity between Enriched C1e 6 91.66
samples C2e 5 93.67
the samples was estimated using the simple matching
C3e 4 87.93
coefficient (Sneath and Sokal, 1973). The dendrograms C4e 3 83.08
were constructed using the UPGMA (unweighted pairs C5e 6 93.34
group method with arithmetic average) algorithm with B1e 7 91.52
bootstrap values over 1000 replicas using the TREECON B2e 5 95.68
B3e 6 91.43
program (Van de Peer and De Wachter, 1994). A principal
B4e 6 96.04
components analysis (PCA) was performed with the B5e 6 93.11
multivariate statistical package (MVSP) version 3.12 h
 ARTICLE IN PRESS
2772 J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776

804
100% 0.3 0.2 0.1
795
760 C1
C3
625 100 C4
622 C2
80% C5
617
B2
611 98 B3
100 68 B4
605
B5
relative abundance

60% 581 B1
77 68
563 B3e
64 B4e
325 C1e
257 69 B5e
61 C4e
40% 223
C2e
212 C3e
187 C5e
100 B1e
124 B2e
20% 87
81 1.1
C1e
60 B3e
0.7 B4e
40 B5e
0.4 C2e B5
0% C4e B3 B4
34 C3e C4
C B Ce Be B1
B2 C5
C1
C3C2
0.0
samples
14.79%

-0.4 C5e
Fig. 1. Percentual average of the UF values of the most representative T-
RFs in each set of samples. The different textures represent different T- -0.7
RFs in base pairs (bp) (C, soil associated to C. hystrix; B, bulk soil; Ce,
soil associated to C. hystrix after an enrichment culture step; Be, bulk soil -1.1
after an enrichment culture step). B2e
-1.5
B1e
-1.8
are three T-RFs (60, 605 and 617 bp), which are not present -1.8 -1.5 -1.1 -0.7 -0.4 0.0 0.4 0.7 1.1
in the unenriched samples (B). In the enriched samples 25.37%
from soil associated to C. hystrix there are also three
Fig. 2. (a) Clustering using the simple matching coefficient and the
unique T-RFs present (187, 617, and 760 bp) that are not UPGMA algorithm in the TREECON program. Bootstrap values for
observed in the unenriched samples (C) (Fig. 1). nodes are presented for only those clusters of samples which occurred
Similarity between the samples is shown in the dendro- more than 70% of the time over 1000 replicates (C, soil associated to C.
gram of Fig. 2a. The results of the cluster analysis show hystrix; B, bulk soil; Ce, soil associated to C. hystrix after an enrichment
culture step; Be, bulk soil after an enrichment culture step) and (b)
that the profiles of the bacterial community obtained from
principal components analysis using the MVSP version 3.12 h program
the unenriched samples clustered separately from the (solid square: soil associated to C. hystrix; empty square: soil associated to
profiles obtained from the enriched fractions with a high C. hystrix after an enrichment culture step; solid triangle: bulk soil; empty
bootstrap value. Another observation is that within the triangle: bulk soil after an enrichment culture step).
cluster formed by the enrichment-independent fractions,
the soil samples associated to C. hystrix (C) clustered On the other hand, the enriched fractions (Ce and Be)
together and clearly separated from the group formed by showed a low similarity between them according to the
samples from bulk soil (B). The C samples have about 95% simple matching coefficient values. The cluster analysis
similarity between them. In the case of the B samples, shows that the enriched fractions do not separate according
although they exhibited more variability between them to their origin nor do they reflect the effect of the plant.
compared to the C samples, they still formed a well-defined The enrichment step in both types of samples would favor
cluster which separated them from all the other samples. the growth of some bacterial groups thus modifying the
The grouping obtained with the unenriched samples T-RFLP profiles. This result is expected because it is well
indicate that the structure of the bacterial community known that the culture conditions alter the structure of the
from soil associated to C. hystrix is different to that present bacterial community.
in bulk soil, suggesting that the plant has an effect on the The PCA applied to the T-RFLP profiles of the enriched
soil bacterial communities. and unenriched fractions is shown in Fig. 2b. The first
 ARTICLE IN PRESS
J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776 2773

component accounted for 25.4% of the variation in to the cluster analysis and PCA of T-RFLP patterns from
the data and the second for 14.8%. The distribution of unenriched samples. On the other hand, bacterial commu-
the samples in the PCA confirms the groups observed in the nities from enriched samples do not cluster according to
dendrogram. the habitat of origin. They were represented by simple
In conclusion, our results show that the difference DNA profiles with a few T-RFs (Table 2) with low
between the bacterial communities from soil associated to similarity between them, suggesting that only a few species
C. hystrix and the neighboring soil without plant cover is were able to grow under the enrichment conditions
only observed when the analysis was conducted with assayed. In fact, bacteria found in low abundance and
unenriched samples. not detected in the unenriched fraction profile could
become abundant in enriched fractions and thus explain
3.2. Metabolic pattern the presence of unique T-RFs such as 617 bp in the
enriched samples (Fig. 1). For example, using traditional
The metabolic profiles of the bacterial communities were cultural methods, the Firmicutes group was considered
investigated by examining the ability of the community to most abundant in all types of soil; however, they are
utilize the 31 different carbon sources found in the Biolog difficult to detect using DNA–PCR-based methods (Jans-
EcoPlates. The results obtained were analyzed in order to sen, 2006). On the other hand, although the enriched
detect differences between the samples derived from both fractions are a subset of the environmental samples, they
habitats: soil associated to C. hystrix and neighboring bulk are unlikely to represent the in situ genetic diversity since
soil. Cluster analysis showed an unexpected grouping the majority of bacteria are not readily cultured on
which was not coincident with the habitats of origin. standard laboratory media (Ward et al., 1990; Hugenholtz
Additionally, these groupings do not correspond to the et al., 1998). Traditional approaches to the study of
paired distribution of the samples, for example C3 and B3, microbial communities in soils have entailed the use of
which are found in separate groups. The dendrogram culture-based protocols, but these approaches are limited
shows a cluster, which groups seven metabolic profiles with in their applicability in the development of microbial
a high similarity between them (Fig. 3a). A possible ecology because cultivation-based methods are very
explanation could be that even though the communities selective and only a small proportion of soil micro-
have a different phylotype composition, they could have organisms, less than 1% of the bacteria present in soils,
similar physiological abilities to metabolize the carbon may be isolated and cultured (Amann et al., 1995; Torsvik
sources assayed. The effect of the 1-week incubation of the et al., 2002; Rappé and Giovannoni, 2003). Therefore, an
Biolog EcoPlates, however, cannot be discarded since it enrichment-independent approach that involves total
could modify the community structure of the naturally community genomic DNA extraction followed by ampli-
occurring bacterial assemblages in these habitats minimiz- fication of 16S rRNA-encoding DNA (rDNA) could better
ing the original differences. reflect the naturally occurring communities. As a result,
The distribution observed in dendrogram was confirmed T-RFLP profiles of the unenriched samples showed that
by PCA in which the first component accounted for the soil associated to C. hystrix is a well-defined micro-
56.4% of the variation in the data and the second for habitat different to the neighboring bulk soil revealing the
13.8% (Fig. 3b). effect of the plant on the microbial community structure.
Another observation from the dendrogram was that the
4. Discussion variability of T-RFLP profiles within the unenriched
samples was lower for soil associated to plant than for
To our knowledge, this is the first study carried out to bulk soil (Fig. 2a), suggesting that the plant may also
investigate the bacterial community associated to the contribute to minimizing a possible heterogeneous habitat.
actinorhizal plant Colletia hystrix. There is little informa- We also determined the community level metabolic
tion available on the soil microbial communities associated profile with the commercially available Biolog EcoPlateTM
to the pioneer plant species involved in the colonization of system. This approach has been commonly applied to
nitrogen-poor soils in semi-arid ecosystems such as compare the potential metabolic activity of microbial
C. hystrix. Therefore, we determined the genetic and communities (Buyer and Drinkwater, 1997; Konopka et
metabolic profiles of the soil bacterial communities al., 1998; Preston-Mafham et al., 2002). In this case, our
associated to C. hystrix and the neighboring bulk soil, objective was to determine if the sole-carbon-source
using the molecular biology-based T-RFLP technique and utilization profiles are able to group the samples according
Biolog EcoPlateTM system. The T-RFLP analysis of the to their original habitat as was observed with the genetic
naturally occurring bacterial communities in both habitats T-RFLP profiles. Our results revealed that the metabolic
was carried out using both enriched and unenriched profiles do not reflect the effect of the plant on the
samples, while the Biolog EcoPlate assay was done using naturally occurring bacterial communities. These results
only unenriched samples. suggest that communities with a different genetic composi-
Bacterial community structure of soil associated to tion could have similar physiological abilities, that is to say
C. hystrix was different from that in bulk soil according some degree of functional redundancy. The single carbon
 ARTICLE IN PRESS
2774 J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776

0.4 0.3 0.2 0.1

C2

95 C3

93 B4

B5
100
B2
100
B1

100 C5

C1
99

B3

C4

6.6

5.3

3.9
B3

2.6
13.76%

C1
1.3

C4
C2CB2
3
0.0 B4
B5
B1

-1.3

C5
-2.6
-2.6 -1.3 0.0 1.3 2.6 3.9 5.3 6.6
56.35%

Fig. 3. (a) Clustering using the simple matching coefficient and the UPGMA algorithm in the TREECON program. The reliability of the phenogram was
determined by bootstrapping over 1000 replicates. Values for nodes are presented for only those clusters of samples which occurred more than 70% of the
time (C, soil associated to C. hystrix; B, bulk soil) and (b) principal components analysis using the MVSP version 3.12 h program (solid square: soil
associated to C. hystrix; solid triangle: bulk soil).

source utilization profiles would be expected to detect only would not necessarily have to be correlated to the T-RFLP
drastic changes in the microbial community. On the other profiles since the Biolog EcoPlate profiles describe, at the
hand, molecular techniques such as DGGE, ARDRA and phenotypic level, the metabolic diversity of the hetero-
T-RFLP are capable of detecting changes at the phylotype trophic bacterial fraction present in the unenriched sample
level and therefore are likely to detect smaller shifts in the (Garland, 1997). Other authors have pointed out that
community (Kirk et al., 2004). Although the existence of molecular methods are more discriminatory than physio-
functional redundancy is a plausible explanation for our logical and biochemical approaches (Ramsey et al., 2006;
results, we cannot discard the culture effect produced by Singh et al., 2006).
the incubation of the Biolog EcoPlates which could modify The plant may affect the microbial communities
the original community structure. In fact, some authors associated to its roots in different ways. First, the root
criticize the use of the Biolog EcoPlate assay because it is itself is the main nutrient source for the soil micro-
based on bacterial culturability (Konopka et al., 1998). The organisms such that during the colonization of plant roots
carbon source utilization profiles in the unenriched samples the soil bacteria undergo a selective enrichment in the plant
 ARTICLE IN PRESS
J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776 2775

rhizosphere in response to different components of the root Binkley, D., Giardina, C., 1997. Nitrogen fixation in tropical forest
exudates (McCully, 1999; Knee et al., 2001). On the other plantations. In: Nambiar, E., Brown, A. (Eds.), Management of Soil,
hand, plants can affect available soil carbon, temperature, Nutrients and Water in Tropical Plantation Forests, ACIAR/CSIRO/
CIFOR Monograph No. 43, Canberra, Australia (Chapter 9).
and water content, which each have the potential to affect Buyer, J.S., Drinkwater, L.E., 1997. Comparison of substrate utilization
microbial community composition and function (Waldrop assay and fatty acid analysis of soil microbial communities. Journal of
and Firestone, 2006). Additionally, the quality litter and Microbiological Methods 30, 3–11.
cover can provide nutrients and improve soil fertility thus Carú, M., 1993. Characterization of native Frankia strains isolated from
affecting the bacterial assemblages (Binkley and Giardina, Chilean shrubs (Rhamnaceae). Plant and Soil 157, 137–145.
Carú, M., Cabello, A., 1999. Infectivity and effectivity of some Frankia
1997; Sharma et al., 1998). The soil microbial communities strains from the Rhamnaceae family. Arid Soil Research and
frequently limited by carbon (Zak et al., 1994) or nitrogen Rehabilitation 13, 53–59.
(Zak et al., 1990) obtain their main source of substrates Chávez, M., Carú, M., 2006. Genetic diversity of Frankia microsymbionts
from the leaf litter (Bardgett et al., 2005). in root nodules from Colletia hystrix (Clos.) plants by sampling at a
C. hystrix is a species that is well adapted to semi-arid small-scale. World Journal of Microbiology & Biotechnology 22,
813–820.
zones which, due to their deep rooting system, can extract Davison, J., 1988. Plant beneficial bacteria. Biotechnology 6, 282–286.
water from greater depths and improve the soil moisture Dawson, J.O., 1986. Actinorhizal plants: their use in forestry and
(Gutteridge and Akkasaeng, 1985). In addition, the effect agriculture. Outlook Agriculture 15, 202–208.
of C. hystrix on the composition and relative abundance of Denarie, J., Debelle, F., Rosenberg, C., 1992. Signaling and host
soil microbial species could be supported in part by the range variation in inoculation. Annual Review of Microbiology 46,
487–531.
ability of the plants to form root nodules with Frankia, a Dobbelaere, S., Vanderleyden, J., Okon, Y., 2003. Plant growth-
nitrogen-fixing actinomycete (Carú, 1993, Schwencke and promoting effects of diazotrophs in the rhizosphere. Critical Reviews
Carú, 2001) increasing the content of nitrogen and other in Plant Sciences 22, 107–149.
nutrients in soil (Table 1) and generating a different Dunbar, J., Ticknor, L.O., Kuske, C.R., 2001. Phylogenetic specificity and
bacterial community to that in bulk soil. Johnson (1995) reproducibility and new method for analysis of terminal restriction
fragment profiles of 16S rRNA genes from bacterial communities.
and Erickson et al. (2005) found that soils under Applied and Environmental Microbiology 67, 190–197.
Ceanothus, another actinorhizal plant, contained more Erickson, H.E., Soto, P., Johnson, D.W., Roath, B., Hunsaker, C., 2005.
total nitrogen than soils under adjacent patches of non- Effects of vegetation patches on soil nutrient pools and fluxes within a
actinorhizal plants and soil without cover. On the other mixed-conifer forest. Forest Science 51, 211–220.
hand, the microbial community can also improve the Fang, M., Kremer, R.J., Motavalli, P.P., Davis, G., 2005. Bacterial
diversity in rhizospheres of nontransgenic and transgenic corn.
ability of plants to colonize nitrogen-poor sites and become
Applied and Environmental Microbiology 71, 4132–4136.
established as fertility islands in the matorral. The micro- Forster, J.C., 1995. Soil sampling, handling, storage and analysis. In: Alef,
organisms associated to the fertility islands are important K., Nannipieri, P. (Eds.), Methods in Applied Soil Microbiology and
for plant growth since they favor the assimilation of Biochemistry. Academic Press Ltd., London, UK, pp. 49–121.
nutrients (Davison, 1988), produce growth-promoting Fuentes, E.R., Hoffman, A.J., Poiani, A., Alliende, M.C., 1986.
Vegetation change in large clearings: patterns in the Chilean matorral.
hormones (Denarie et al., 1992), fix nitrogen (Torrey,
Oecologia 68, 358–366.
1978; Dawson, 1986), suppress pathogens (Shippers et al., Garland, J.L., 1997. Analysis and interpretation of community-level
1987; Dobbelaere et al., 2003) and allow the dissolution of physiological profiles in microbial ecology. FEMS Microbiology
minerals (Nakas and Klein, 1980), among others. Ecology 24, 289–300.
Further studies on the composition of the bacterial Garland, J.L., Mills, A.L., 1991. Classification and characterisation of
heterotrophic microbial communities on the basis of patterns of
communities of environmental unenriched samples with
community-level sole-carbon-source utilization. Applied and Environ-
clone libraries could reveal the bacterial groups associated mental Microbiology 57, 2351–2359.
to these actinorhizal plants. Gutteridge, R.C., Akkasaeng, R., 1985. Evaluation of nitrogen fixing trees
in northeast Thailand. Nitrogen Fixing Tree Research Reports 3,
46–47.
Acknowledgments Hugenholtz, P., Goebel, B.M., Pace, N.R., 1998. Impact of culture-
independent studies on the emerging phylogenetic view of bacterial
This work was supported by Fondecyt Project No. diversity. Journal of Bacteriology 180, 4765–4774.
Janssen, P.H., 2006. Identifying the dominant soil bacterial taxa in
1040880. The authors thank Prof. M. Lamborot for critical
libraries of 16S rRNA and 16S rRNA genes. Minireviews. Applied
reading of the manuscript. We thank Mrs. Juana Leal for Environmental and Microbiology 72, 1719–1728.
assisting in the fieldwork. Johnson, D.W., 1995. Soil properties beneath Ceanothus and pine stands
in the eastern Sierra Nevada. Soil Science Society of America Journal
59, 918–924.
References Kent, A.D., Triplett, E.W., 2002. Microbial communities and their
interactions in soil and rhizosphere ecosystems. Annual Review of
Amann, R.I., Ludwig, W., Schleifer, K.H., 1995. Phylogenetic identifica- Microbiology 56, 211–236.
tion and in situ detection of individual microbial cells without Kirk, J.L., Beaudette, L.A., Hart, M., Moutoglis, P., Klironomos, J.N.,
cultivation. Microbiological Reviews 59, 143–169. Lee, H., Trevors, J.T., 2004. Methods of studying soil microbial
Bardgett, R.D., Bowman, W.D., Kaufmann, R., Schmidt, S.K., 2005. diversity. Journal of Microbiological Methods 58, 169–188.
Linking aboveground and belowground ecology: a temporal approach. Knee, E.M., Gong, F.C., Gao, M., Teplitski, M., Jones, A.R., Foxworthy,
Trends in Ecology and Evolution 20, 634–641. A., Mort, A.J., Bauer, W.D., 2001. Root mucilage from pea and its
 ARTICLE IN PRESS
2776 J. Orlando et al. / Soil Biology & Biochemistry 39 (2007) 2769–2776

utilization by rhizosphere bacteria as a sole carbon source. Molecular by physiological, biochemical and molecular fingerprinting methods.
Plant–Microbe Interactions 14, 775–784. European Journal of Soil Science 57, 72–82.
Konopka, A., Oliver, L., Turco, R.F., 1998. The use of carbon substrate Smit, E., Leeflang, P., Gommans, S., Van Den Broek, J., Van Mil, S.,
utilization patterns in environmental and ecological microbiology. Wernars, K., 2001. Diversity and seasonal fluctuations of the
Microbial Ecology 35, 103–115. dominant members of the bacterial soil community in a wheat field
Liu, W.T., Marsh, T.L., Cheng, H., Forney, L.J., 1997. Characterization as determined by cultivation and molecular methods. Applied and
of microbial diversity by determining terminal restriction fragment Environmental Microbiology 67, 2284–2291.
length polymorphisms of genes encoding 16S rRNA. Applied and Sneath, P.H.A., Sokal, R.R., 1973. Numerical Taxonomy. The Principles
Environmental Microbiology 63, 4516–4522. and Practice of Numerical Classification, vol. XV. Freeman, San
McCully, M.E., 1999. Roots in soil: unearthing the complexities of roots Francisco, CA, 573pp.
and their rhizospheres. Annual Review of Plant Physiology and Plant Steenwerth, K.L., Jackson, L.E., Calderón, F.J., Scow, K.M., Rolston,
Molecular Biology 50, 695–718. D.R., 2005. Response of microbial community composition and
Nakas, J.P., Klein, D.A., 1980. Mineralization capacity of bacteria and activity in agricultural and grassland soils after a simulated rainfall.
fungi from the rhizophere–rhizoplane of a semi-arid grassland. Soil Biology & Biochemistry 37, 2249–2262.
Applied and Environmental Microbiology 39, 113–117. Torrey, J.G., 1978. Nitrogen fixation by actinomycete-nodulated angios-
Preston-Mafham, J., Boddy, L., Randerson, P.F., 2002. Analysis of perms. BioScience 28, 586–592.
microbial community functional diversity using sole-carbon source Torsvik, V., Ovreas, L., Thingstad, T.F., 2002. Prokaryotic diversity
utilisation profiles—a critique. FEMS Microbiology Ecology 42, 1–14. magnitude dynamics, and controlling factors. Science 296,
Ramsey, P.W., Rillig, M.C., Feris, K.P., Holben, W.E., Gannon, J.E., 1064–1066.
2006. Choice of methods for soil microbial community analysis: PLFA Van de Peer, Y., De Wachter, R., 1994. TREECON for Windows: a
maximizes power compared to CLPP and PCR-based approaches. software package for the construction and drawing of evolutionary
Pedobiologia 50, 275–280. trees for the Microsoft Windows environment. Computer Applications
Rappé, M.S., Giovannoni, S.J., 2003. The uncultured microbial majority. in the Biosciences 10, 569–570.
Annual Reviews of Microbiology 57, 369–394. Ward, D.M., Weller, R., Bateson, M.M., 1990. 16S rRNA sequences
Schwencke, J., Carú, M., 2001. Advances in actinorhizal symbiosis: host-plant- reveal uncultured inhabitants of a well-studied thermal community.
frankia interactions, biology, and applications in Arid Land Reclama- FEMS Microbiology Reviews 75, 105–116.
tion—a review. Arid Land Research and Management 15, 285–327. Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S
Sharma, S., Rangger, A., von Lützow, M., Insam, H., 1998. Functional ribosomal DNA amplification for phylogenetic study. Journal of
diversity of soil bacterial communities increases after maize litter Bacteriology 173, 697–703.
amendment. European Journal of Soil Biology 34, 53–60. Waldrop, M., Firestone, M., 2006. Seasonal dynamics of microbial
Shippers, B., Bakker, A.W., Baker, A.H.M., 1987. Interactions of community composition and function in oak canopy and open
deleterious and beneficial rhizophere micro-organisms and the effect grassland soils. Microbial Ecology 52, 470–479.
of cropping practices. Annual Review of Phytopatology 25, 339–358. Zak, D.R., Grigal, D.R., Gleeson, S., Tilman, D., 1990. Carbon and
Silvester, W.B., Balboa, O., Martı́nez, J.A., 1985. Nodulation and nitrogen cycling during old-field succession: constraints on plant and
nitrogen fixation in members of the Rhamnaceae (Colletia, Retanilla, microbial biomass. Biogeochemistry 11, 111–129.
Talguenea, Trevoa) growing in the Chilean matorral. Symbiosis 1, Zak, D.R., Tilman, D., Parmenter, R.R., Rice, C.W., Fisher, F.M., Vose,
29–38. J., Milchunas, D., Martin, C.W., 1994. Plant production and soil
Singh, B.K., Munro, S., Reid, E., Ord, B., Potts, J.M., Paterson, E., micro-organisms in late-successional ecosystems: a continental-scale
Millard, P., 2006. Investigating microbial community structure in soils study. Ecology 75, 2333–2347.