RESEARCH ARTICLE
Antagonistic interactions between heterotrophic bacteria as a
potential regulator of community structure of hypersaline
microbial mats
Richard A. Long1,2, Damien Eveillard3, Shelli L.M. Franco1, Eric Reeves4 & James L. Pinckney1,2
1
Department of Biological Sciences, University of South Carolina, Columbia, SC, USA; 2Marine Science Program, University of South Carolina,
Columbia, SC, USA; 3Computational Biology Group, LINA UMR 6241, Université de Nantes, EMN, CNRS, Nantes, France; and 4Department of
Microbiology, Clemson University, Columbia, SC, USA
Correspondence: Richard A. Long,
Department of Biological Sciences, University
of South Carolina, 715 Sumter Avenue,
Columbia, SC 29208, USA.
Tel.: +803 777 2531; fax: +803 777 4002;
e-mail: [email protected]
Received 30 January 2012; revised 2 July
2012; accepted 8 July 2012.
Final version published online 6 August 2012.
DOI: 10.1111/j.1574-6941.2012.01457.x
MICROBIOLOGY ECOLOGY
Editor: Wietse de Boer
Keywords
amensalism; consortia; antibiotics;
Gram-positive; salinity stress.
Abstract
Microbial mats are laminae of self-sustaining microbial communities with a
high level of competition for resources. We tested the hypothesis that chemically mediated antagonism is a potential mechanism for structuring the bacterial community. In the co-culturing assay, 57% of the isolates expressed
antagonistic behavior toward one or more isolates and 5% of the isolates
inhibited more than 80% of the isolates. We observed greater levels of antagonism between isolates from adjacent laminae than within. The bacterial isolate
library derived from the mat was predominately Gram-positive, and inhibition
within this group was greater than against the few Gram-negative isolates. Microdiversity of 16S rRNA gene was observed for Bacillus marisflavi isolates,
which represented 23 of the 75 isolates in the library. Within this and other
groups, the patterns of inhibition and sensitivity varied greatly, suggesting
rapid gain and loss of the ability to produce antagonistic secondary metabolites
and resistance toward such molecules. Our observations are consistent with the
hypothesis that antagonistic interactions are a potential mechanism in addition
to physiochemical properties that regulate the vertical distribution of aerobic
heterotrophic bacteria in hypersaline microbial mats.
Introduction
Microbial mats are composed of a diverse array of
microorganisms that form self-sustaining communities
(Cloud, 1976; Paerl & Pinckney, 1996). Strong physical
and chemical gradients over a few millimeters provide an
assortment of microniches filled by a diverse assemblage
of oxygenic phototrophic, anoxygenic phototrophic, chemoautotrophic, and heterotrophic bacteria (Stal, 1995;
Paerl & Pinckney, 1996; Ley et al., 2006). These gradients
are heavily influenced by light, photosynthesis, and the
production of oxygen (Paerl & Pinckney, 1996). As such,
microbial mat communities contribute significantly to the
biogeochemistry of terrestrial and shallow water habitats
(Canfield & Des Marais, 1993; Stal, 1995; Priscu et al.,
1998).
Microbial mats can be found in various climates
throughout the world, but tend to be found more
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
frequently in extreme environments. In temperate
environments such ecosystems are thought to be N
limited (Pinckney et al., 1995), while in extreme environments, both nutrient and physiochemical parameters
regulate the consortia of microorganisms and their activity, for example, tropical hypersaline ponds with salinity
45–> 200 and polar microbial mats in subfreezing
temperatures (Paerl & Yannarell, 2010). Desiccation in
any environment can also limit productivity and
affect community structure in these surface thin-layer
environments.
Antagonistic interactions between bacteria have been
documented in soils (Wright, 1956; Thomashow et al.,
1990), as well as in marine free-living and particle-associated bacteria (Long & Azam, 2001; Grossart et al., 2004;
Gram et al., 2010). The production of antagonistic molecules may result from competitive pressure for scarce
resources within a community (Burgess et al., 1999; Long
FEMS Microbiol Ecol 83 (2013) 74–81
75
Microbial mat bacterial antagonisms
& Azam, 2001; Grossart et al., 2004). Such interactions
have been observed to alter bacterial community structures, that is, inhibiting the growth of some species and
opening niches for other species on marine particles
(Long et al., 2003). However, the occurrence of chemically mediated antagonism in marine microbial mats has
not been studied.
We addressed the role of antagonistic interactions of
heterotrophic bacteria from a hypersaline pond in
structuring the vertical distribution of bacterial species
within sediments. Salt Pond is a tropical hypersaline lake
on San Salvador Island, Bahamas, that experiences both
seasonal fluctuation of salinity between 60 and 340 and
desiccation stress (Yannarell et al., 2006). Dense microbial
mats blanket the sediments in this shallow hypersaline
pond and are the dominant biotic feature within this ecosystem. Salt Pond is not extensively influenced by terrigenous, atmospheric, or anthropogenic nutrient inputs. By
far, the largest perturbation the pond experiences is the
freshwater and sediment inputs attributable to tropical
storms and hurricanes. These periodic disturbances can
result in re-colonization events and re-establishment of
consortia (Yannarell et al., 2007). We tested the hypothesis that antagonistic interactions are a mechanism that
influences heterotrophic bacterial species distribution and
composition in microbial mats such as those found in
Salt Pond.
Material and methods
Sample collection
Salt Pond is a hypersaline lake (24°05′N, 74°30′W) on
San Salvador Island, Bahamas, with seasonal fluctuation
of salinity of 60–340 (Yannarell et al., 2006). Dense
microbial mats blanket the shallow sediments in this
hypersaline pond and are the dominant biotic feature.
The Salt Pond is virtually unimpacted by terrigenous,
atmospheric, or anthropogenic nutrient inputs. By far,
the largest perturbations the lake experiences are the
freshwater and sediment inputs attributable to tropical
storms and hurricanes in late fall.
A core of microbial mat measuring 1 cm in diameter
and 2 cm in depth was collected from the eastern shore of
Salt Pond. The core had five distinct laminae ranging from
1 to 3 mm in depth distinguished by their color; a surface
green lamina would be dominated by cyanobacteria and
green sulfur bacteria; a pink lamina by sulfur and nonsulfur purple bacteria; a brown lamina by sulfate reducers
and fermenters; a white lamina, which was sand deposition from a hurricane and mostly like housed bacteria
from the brown and black laminae; at the bottom of the
core, a black lamina by iron reducers and fermenters.
FEMS Microbiol Ecol 83 (2013) 74–81
Laminae were thin sectioned with a sterile razorblade; a
fresh blade was used at each interface. Each lamina was
placed into a sterile microcentrifuge tube with autoclaved
0.2-lm-filtered seawater (salinity adjusted to 72) and vortexed to disrupt the matrix and dislodge bacteria. Aliquots
were spread on ZoBell 2216 Marine agar plates (with
salinity adjusted to 72, referred to here in as ZoBell-72;
90% seawater, 0.5% peptone, 0.1% yeast, 1.5% Bacto agar,
10% Milli-Q water, salinity 72). All isolates that grew
within 3 days at 25 °C were harvested. Isolates were restreaked three times to ensure a single genotype. Isolates
were then stored at 80 °C in 20% glycerol stocks.
Inhibition assays
Bacteria–bacterial antagonism was screened by a modified
Burkholder agar inhibition assay in a 75-by-75-isolate
array against one another (Long & Azam, 2001). Isolates
were grown overnight on ZoBell-72 agar plates. Individual colonies were transferred into 2 mL of ZoBell-72
broth and grown overnight. Lawns of target bacteria were
laid by seeding 3 mL of molten ZoBell-72 soft agar (0.6%
at 42 °C) with 100 lL of the isolate suspension
(OD600 = 1), vortexed, and immediately poured onto a
ZoBell-72 agar plate. The lawn was allowed to solidify,
and 10 lL of potential producer was spotted onto the
lawn of target cells in a 4 9 4 grid (i.e. 16 isolates per
plate). The bacteria were incubated at room temperature
(24 ± 1 °C) and were examined for zone of inhibition
(> 2 mm) after 1, 2, and 6 days. Two complete sets of
assays were performed, and any isolates that showed
ambiguous results (i.e. a positive and a negative) were
tested a third time.
DNA isolation and amplification
Isolates from frozen stock were streaked onto a 1.5%
ZoBell-72 agar plate, inoculated into ZoBell-72 broth,
grown overnight at 25 °C, and finally pelleted into a
microcentrifuge tube and frozen. Samples were treated
with 5 lL of Lyse-N-GoTM PCR Reagent (Pierce) prior
to direct amplification of the 16S rRNA gene; except for
isolates that turned out to be Bacillus firmus, in which
DNA was isolated using MO BIO UltraCleanTM Microbial DNA Isolation Kit followed by PCR to amplify the
16S rRNA gene for subsequent analysis. The primers of
Weisberg et al. (1991) were used for PCR and sequencing
reactions: 27F (forward) 5′-AGAGTTTGATCMTGGCTCAG-3′ and 1492R (reverse) 5′-GGYTACCTTGTTACGACTT 3′ amplifying a 1465-bp segment. Cycling
parameters for the Lyse-N-Go included an initial lysis
cycle of 30 s at 65 °C, 30 s at 8 °C, 90 s at 65 °C, 3 min
at 97 °C, 60 s at 8 °C, 180 s at 65 °C, 60 s at 97 °C, and
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
R.A. Long et al.
76
60 s at 65 °C. This was followed by the addition of PCR
Mastermix containing 10 lL of Taq Mastermix from Qiagen, 1 lL of each primer at a 10 mM working solution,
and 3 lL of RNAse-free PCR water to bring the final volume to 20 lL per reaction. PCR parameters were 5 min
at 95 °C and 30 cycles of 3 min at 94 °C, 60 s at 65 °C,
and 90 s at 72 °C, followed by five additional cycles of
3 min at 94 °C, 60 s at 50 °C, and 60 s at 72 °C with a
final extension for 5 min at 72 °C. For those samples
where DNA was extracted first, an additional 4 lL of
RNAse-free PCR water replaced the Lyse-N-Go with the
Mastermix and 1 lL of extracted DNA template per reaction. PCR products were validated by agarose gel electrophoresis and then cleaned up using a Sephadex spin
column (Amersham Pharmacia Bioteck AB). The purified
PCR products were stored at
20 °C until use for
sequencing.
Sequencing and phylogenetic analysis
The purified PCR products were used as the template for
the sequencing reaction. Each individual reaction consisted of 2 lL BigDye® v1.1 (ABI), 1.25 lL of the 27F
forward primer, 1.75 lL RNAse-free PCR water, and
1 lL of the template for 6 lL final volume. The cycling
conditions were an initial denaturation for 2 min at 96 °C
followed by 30 cycles of 30 s at 96 °C, 15 s at 50 °C, and
4 min at 60 °C. The amplified products were cleaned up as
mentioned previously and concentrated using a Savant
SpeedVac Concentrator SVC 200. Samples were then stored
at 20 °C until sequencing. Samples were sequenced using
an ABI 3100 Capillary Sequencer.
Sequences were analyzed, and automatic base calls were
checked using Sequencher (Gene Codes Corporation). Salt
Pond bacteria were identified using the National Center
for Biotechnology Information database and the nucleotide Basic Local Alignment Search Tool (BLAST). Bacterial
sequences were aligned using CLUSTALX version 1.83, and
dendrograms were created using NJ Plot. Reference strains
were used in constructing the phylogenetic tree, but
removed in the final version to reduce the size of the tree.
Sequences have been deposited in GenBank under the
following accession numbers JX298679 to JX298752.
approach). A similarity matrix was generated by pairwise
comparison based upon Needleman & Wunsch (1970)
global alignment algorithm. This generated matrix was
then converted into a correlation matrix, which was the
base for construction of the dendrograms using Ward
representation. Similar results (not shown) were obtained
using standard discrimination approaches, which
confirms the robustness of the used classification.
Statistical analysis
Pearson chi-square analysis was conducted to test the null
hypothesis of no differences between intra- and interlamina levels of antagonism. Cross-tabulation is a combination of frequency tables arranged such that each cell in
the resulting table represents a unique combination of
specific values of cross-tabulated variables. This allows the
examination of frequencies of observations that belong to
more than one variable. By examining these frequencies,
relationships can be identified between cross-tabulated
variables.
Results and discussion
General bacterial antagonism
Extensive inhibition was observed with the Salt Pond
microbial mat bacterial isolates. In the co-culture antagonistic assay, we observed that 57% of the bacterial isolates
inhibited one or more isolate (Fig. 1). Nearly a third of
the isolates had wide levels of inhibition that affected the
growth of 10 or more of the 75 isolates. Within that
group, 12 of the isolates inhibited > 50% of the target
isolates, while five of those isolates inhibited > 80% of
Phenotype analysis and comparison of
phenotype and genotype trees
Inhibition and sensitivity patterns were assessed by computational analysis based on bioinformatics and statistical
approaches implemented in MATLAB. Inhibition and
sensitivity data for each co-culture assay were transformed into a Boolean variable to allow discrete pairwise
comparison for each isolate (i.e. a bar coding–like
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
Fig. 1. Microbial mat bacterial–bacterial interactions. Black lines
represent the inhibition patterns of the isolates; 43 isolates inhibited
one or more isolate. Gray lines represent the distribution pattern of
bacterial isolates sensitive to (or inhibited by) other isolates; all but
one isolate was inhibited by another isolate.
FEMS Microbiol Ecol 83 (2013) 74–81
77
Microbial mat bacterial antagonisms
the isolates. There was no observation of self-inhibition.
Our results are consistent with recent ecological studies
focusing on potential bacterial antagonism in marine
systems (Long & Azam, 2001; Grossart et al., 2004;
Rypien et al., 2010).
Antagonism within and between consortia
The bacterial strains were isolated from different laminae
of the microbial mat. We examined how the distribution
of isolates within the microbial mat affects its antagonistic
phenotype to place our phenotypic observations into an
ecological context and test the hypothesis that competition between mat laminae will result in more inhibition
between species belonging to functional groups in
adjacent laminae than between those within the same
lamina. Isolates from the black lamina showed the highest
level of antagonisms inhibiting roughly a third of the
isolates, while those from the remaining laminae inhibited
c. 5–15% (Fig. 2). Six of the 12 black lamina isolates
inhibited at least 60% of the isolates, while only six of 63
isolates from the remaining laminae inhibited an equivalent fraction (Supporting information, Fig. S1). In all
cases except for the pink lamina, the degree of
within-consortia inhibition was lower than that against
neighboring consortia.
Tabulation of all the intra- and interlamina (divided
into adjacent and nonadjacent) assays supported an
association between antagonism and nonantagonism and
laminae (Person chi-square; 2, N = 11250) = 114.8,
P < 0.000. Examination of adjacent interlamina interaction showed 23.2% inhibition, while there was 13.4% and
14.9% inhibition in intralamina and nonadjacent
interlamina interactions, respectively. Specific lamina
Fig. 2. Inhibition patterns between microbial mat laminae. Bacteria
strains were isolated from five distinct laminae of the microbial mat.
From surface to depth, green, pink, brown, white, and black laminae.
Aggregate values are presented from each lamina against the various
laminae, with the color of the bar corresponding to the lamina of the
inhibited bacteria, with isolates from the black laminae inhibiting the
largest percentage.
FEMS Microbiol Ecol 83 (2013) 74–81
interactions are shown in the chi-square cross-tabulation
analysis (Table 1). Interlamina inhibition was 1.9% lower
to 45% higher than expected, while for intralamina interactions, inhibition was 7.2–69.3% lower than randomly
expected. These observations are consistent with the
hypothesis that greater antagonistic interactions occur
between communities in neighboring laminae than
within.
Environmental variables including oxygen, nutrients,
sulfide, pE and pH can rapidly fluctuate within microbial
mat systems and affect bacterial composition (Des Marais,
1995; Paerl & Pinckney, 1996; Villanueva et al., 2008;
Paerl & Yannarell, 2010) . These environments host a
high level of biodiversity (Ley et al., 2006) with high
levels of functional diversity (Yannarell et al., 2007). It
should be noted that laminae sampled were demarked by
the pigment of the dominant species not necessarily physiochemical gradients, although one would expect such
gradients to establish species distributions and vice versa.
Within the context of the fluctuating physiochemical gradients that can lead to an increased zone of overlapping
functional groups within the microbial mat competing
for the same resource, for example, electron donors, carbon sources, or terminal electron acceptors, and in combination with our observations of a general pattern of
greater inhibition toward isolates from adjacent laminae
of the mat, we hypothesize that microbial mats employ
secondary metabolites to gain a competitive advantage at
these fluctuating interfaces.
Phenotype and genotype linkages
The majority of the isolates in the library were
Gram-positive bacteria in the Firmicutes, Bacillus, Halobacillus, or Exigunobacterium genera. Eight Gammaproteobacteria representatives were also isolated. There is no
published analysis of the Salt Pond microbial mat community, so we do not know whether this is reflective of
the entire heterotrophic community present. More than
likely, this is a subsample of the population that exists
within this ecosystem because of selection by a single
medium (Rappe & Giovannoni, 2003). In the Guerrero
Negro hypersaline microbial mats, Ley et al. (2006)
reported seven dominant heterotrophic phyla including
the Firmicutes and Proteobacteria; however, the Firmicutes
were not the majority of clones in the 16S rRNA gene
library. In our current study, the distribution of isolates
is distinct from recent studies of marine antagonistic
interactions in which the majority of the isolates were
Gram-negative (Long & Azam, 2001; Grossart et al.,
2004; Gram et al., 2010; Rypien et al., 2010) .
There was microdiversity within several branches of the
tree. Bacillus marisflavi accounted for 23 of 75 isolates, all
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
78
R.A. Long et al.
Table 1. Chi–squared cross-tabulation analysis of the interlamina antagonistic interactions. Inhibitions were analyzed by chi-squared crosstabulation to test the hypothesis that negative interactions were higher between laminae than within laminae. Values reported are the
percentage difference from the expected value if inhibition was homogenously distributed; values in parenthesis in column 1 are isolates from
that lamina, and values in the remainder of the columns are the number of assays for that interlamina observation. Red cells are values with
higher than expected inhibition observed (i.e. homogenous), while green cells are values less than expected inhibition. Grey cells are within ± 5%
of expected inhibition. Bold text denotes significant difference from expected
of which had > 99% identity in the 16S rRNA gene
(Fig. 3). Despite this microdiversity, no two B. marisflavi
isolates had the same patterns of inhibition or sensitivity
(Fig. S1). The B. marisflavi isolates were relatively sensitive to chemical interactions mediated by other isolates
being inhibited on average by 17.0, a range of 13–23, and
narrow standard deviation of 2.9 (Fig. S1). The similar
degree of inhibition also led to the formation of a cluster
when examining sensitivity patterns (Fig. 3 insert, the
light blue cluster). However, the ability to inhibit the
Fig. 3. Phylogenetic and sensitivity to
antagonism relationships of isolates. Phylogeny
was based upon 16S rRNA gene sequence.
The insert in the upper left is a dendrogram
based upon pairwise comparison of sensitivity
profiles of each isolate, based upon data in
Fig. S1. Colors of the text in the phylogenetic
tree were assigned based upon cluster
sensitivity patterns within the insert figure.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
FEMS Microbiol Ecol 83 (2013) 74–81
Microbial mat bacterial antagonisms
other isolates varied greatly – with an average of 13.3
isolates, a range of 0–55, and a wide standard deviation
of 16.4. We hypothesize that this observation is because
of acquisition and loss of genes responsible for the production of these secondary metabolites or genomic rearrangement that alters the expression of those genes.
Finding primarily variation in secondary metabolites
between the genomes of two sediment bacteria of the
genera Salinispora, Penn et al. (2009) suggested positive
selection for these and thereby leading to niche differentiation. It may also be a limitation of the discriminative
power of the 16S rRNA gene, as suggested by Peña et al.
(2010), who reported 10% genomic variation between
two Salinibacter ruber isolates from solar saltern with
identical 16S rRNA gene sequence. To our knowledge,
there are no previous reports of antagonistic molecules
isolated from B. marisflavi.
Not surprising, the different clades demonstrated different patterns of inhibition and sensitivity. There were
three Bacillus cereus isolates of which two inhibited 62
and 67 of the isolates, while the third inhibited an
order of magnitude of less isolates. Natural products
from B. cereus have been previously reported (Emmert
et al., 2004; reviewed in Estibaliz & Ortiz, 2011). Halobacillus had representatives that were both extremely
antagonistic and those that were not. The genus Exiguobacterium had four representatives in the library, none
of which were antagonistic against any of the other isolates (Fig. S1).
The Gram-negative isolates represented 11% of the
library, and this group was responsible for a single antagonistic interaction against another Gram-negative. While
within Gram-positive inhibition was observed in 20.2%
of the assays, Gram-positive vs. Gram-negative was only
7.7%. Our results suggest that in the Salt Pond microbial
mats, the antagonism was predominantly Gram-positive
against Gram-positive.
Chemically mediated interactions
One caveat of the screening process is that not all inhibition is necessarily due to the production of secondary
metabolites; the inhibition might be due to the competition for nutrients or alternation of local pH. However,
several compounds from isolates from this library have
been shown to be active either in crude extracts or
purified (Socha et al., 2007; Teasdale et al., 2011), suggesting that the majority of the interactions are indeed
because of secondary metabolites. A second caveat is
that our study is based upon isolates; the high level of
antagonism observed may not be representative of the
entire community. It is also possible that our results
could be underestimates of the degree of isolates with
FEMS Microbiol Ecol 83 (2013) 74–81
79
antagonistic capabilities. Factors that support this
include (1) the bioassays were conducted on rich media,
and it has been reported that some isolates produce
compounds only in low-nutrient regimes (de Boer et al.,
2007), (2) the bioassays were two-component assays,
that is, a single producer and single target, so any consortial ‘cross-talk’ resulting in antagonism would not be
observed (de Boer et al., 2007), and (3) the panel of
isolates were a subset of the community, and potential
targets were missed. In either case, the data suggest the
potential for antagonistic interactions within the mat
system.
Secondary metabolites from Bacillus endophyticus SP31,
also part of isolate library used in this study, include at
least three bacillamides (Socha et al., 2007). One bacillamide has previously been reported to exhibit algaecidal
properties and inhibiting an assortment of cyanobacteria
(Jeong et al., 2003). More recently, Churro et al. (2009)
have reported other bacillamides that have either algaecidal or algistatic properties against freshwater cyanobacteria. While we did not test interactions between
cyanobacteria and heterotrophs in our system, the isolation
of bacillamides from one isolate in our study suggests that
heterotrophic mat bacteria may alter photoautotroph
physiology through secondary metabolites and hence
the biogeochemistry of microbial mats in a manner not
previously recognized.
The nature of chemoecology in microbial mats is
complex. In extreme environments such as hypersaline
microbial mats, where microorganisms are usually
growing at suboptimal conditions because of a combination of salinity, nutrient, and desiccation stress (Paerl &
Yannarell, 2010), secondary metabolites do not need to
be bactericidal to affect the bacterial community – hence
any additional competitive advantage can have significant
impact on population and community dynamics. Sublethal concentrations of some antibiotics lead to alteration
of genetic pathway regulation (Goh et al., 2002; Davies
et al., 2006; Linares et al., 2006). An alternative chemical
mechanism in structuring species distribution and regulating activity is quorum sensing (QS; Bachofen &
Schenk, 1998); QS has been implicated in establishing
consortia and in regulating numerous cellular pathways
and subsystems in bacterial biofilms (Davies et al., 1998).
QS compounds have been reported in microbial mats
(Decho et al., 2009). A subset of our isolates (7/75) were
shown to produce antiquorum-sensing molecules in crude
extracts (Teasdale et al., 2011). Lastly, some bacterial species are capable of utilizing numerous ‘antibiotics’ as sole
carbon sources, even at concentrations that are lethal to
other species (Dantas et al., 2008). The murky picture
emerging is one of the chemically mediated conflict and
cooperation.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
R.A. Long et al.
80
Much focus has been correctly placed upon anthropogenic introduction of antibiotics into the environment and
its consequences of rapid evolution and dissemination of
antibiotics resistance via horizontal gene transfer (reviewed
in Martinez, 2008). However, recent evidence suggests that
a pool of preindustrial resistance genes, ancient vanA
genes, were extracted and identified from remote permafrost (D’Costa et al., 2011). In addition, the level of amensalism and the variation of sensitivity among closely
relative strains that we observed in a pristine Salt Pond
mat system support an in situ ecological role of secondary
metabolites.
There is a growing appreciation of the chemoecology
of microorganisms, with interactions that run the gamut
from competition and amensalism to commensalism and
mutualism. These various chemically mediated relationships all impact the activity of microbial communities
and consequentially the diversity within it, including
those in microbial mats. Laboratory work with cultures
and metagenomic analysis (Gillespie et al., 2002; TorresCortes et al., 2011) has shown these mechanisms to be
pervasive. Now the challenge is to measure the impact in
situ and at multiple scales, that is, consortia, population,
and single-cell levels to model up to ecosystem level
impact through a systems ecology approach.
We report on extensive antagonistic phenotypes from a
library of predominantly Firmicutes isolated from a
hypersaline microbial mat. This amensalism was more
pronounced between isolates from adjacent laminae,
leading us to posit that there is competition for resources
between consortia in adjacent laminae, particularly during
periods of physiochemical fluctuation within the system.
Microdiversity was observed with some of the clades in
our study, and congruence was observed for sensitivity,
but not inhibition patterns. These observations suggest
that chemically mediated antagonism is one potential biotic mechanism that can influence the bacterial community
structure. This alteration in community structure and
function can potentially lead to changes in the nature and
intensity of the biogeochemistry within microbial mats.
While this study provides further understanding of regulation of species dynamics within these complex systems,
additional studies are necessary to examine how pervasive
this mechanism is and the extent of its impact.
Acknowledgements
We thank Dr. de Boer and two anonymous reviewers for
their constructive suggestions. We thank Clint Patrick for
assistance with manuscript preparation. This work was
supported in part by NSF-MCB-0729594 to R.A.L. and
NSF-MCB-0132528 to J.L.P.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
References
Bachofen R & Schenk A (1998) Quorum sensing autoinducers:
do they play a role in natural microbial habitats? Microbiol
Res 153: 61–63.
Burgess JG, Jordan EM, Bregu M, Mearns-Spragg A & Boyd
KG (1999) Microbial antagonism: a neglected avenue of
natural products research. J Biotechnol 70: 27–32.
Canfield DE & Des Marais DJ (1993) Biogeochemical cycles of
carbon, sulfur, and free oxygen in a microbial mat. Geochim
Cosmochim Act 57: 3971–3984.
Churro C, Alverca E, Sam-Bento F, Paulino S, Figueira VC,
Bento AJ, Lobo AM, Prabhakar S, Calado AJ & Pereira P
(2009) Effects of bacillamide and newly synthesized
derivatives on the growth of cyanobacteria and microalgae
cultures. J Appl Phycol 21: 429–442.
Cloud P (1976) Beginnings of biospheric evolution and their
biogeochemical consequences. Paleobiology 2: 351–387.
Dantas G, Sommer MOA, Oluwasegun RD & Church GM
(2008) Bacteria subsisting on antibiotics. Science 320:
100–103.
Davies DG, Parsek MR, Pearson JP, Iglewski BH, Costerton
JW & Greenberg EP (1998) The involvement of cell-to-cell
signals in the development of a bacterial biofilm. Science
280: 295–298.
Davies J, Spiegelman GB & Yim G (2006) The world of
subinhibitory antibiotic concentrations. Curr Opin Microbiol
9: 445–453.
D’Costa VM, King CE, Kalan L et al (2011) Antibiotic
resistance is ancient. Nature 477: 457–461.
de Boer W, Wagenaar AM, Klein Gunnewiek PJA & van Veen
JA (2007) In vitro suppression of fungi caused by
combinations of apparently non-antagonistic soil bacteria.
FEMS Microbiol Ecol 59: 177–185.
Decho AW, Visscher PT, Ferry J, Kawaguchi T, He L,
Przekop KM, Norman RS & Reid RP (2009) Autoinducers
extracted from microbial mats reveal a surprising diversity
of N-acylhomoserine lactones (AHLs) and abundance
changes that may relate to diel pH. Environ Microbiol 11:
409–420.
Des Marais DJ (1995) The biogeochemistry of hypersaline
microbial mats. Adv Microb Ecol 14: 251–274.
Emmert AB, Klimowicz AK, Thomas MG & Handelsman J
(2004) Genetics of Zwittermicin a production by Bacillus
cereus. Appl Environ Microbiol 70: 104–113.
Estibaliz S & Ortiz A (2011) Secondary metabolites of soil
Bacillus spp. Biotechnol Lett 33: 1523–1538.
Gillespie DE, Brady SF, Bettermann AD, Cianciotto NP,
Liles MR, Rondon MR, Clardy J, Goodman RM &
Handelsman J (2002) Isolation of antibiotics turbomycin
A and B from a metagenomic library of soil microbial
DNA. Appl Environ Microbiol 68: 4301–4306.
Goh EB, Yim G, Tsui W, McClure J, Surette MG & Davies J
(2002) Transcriptional modulation of bacterial gene
expression by subinhibitory concentrations of antibiotics.
P Natl Acad Sci USA 99: 17025–17030.
FEMS Microbiol Ecol 83 (2013) 74–81
81
Microbial mat bacterial antagonisms
Gram L, Melchiorsen J & Bruhn JB (2010) Antibacterial
activity of marine culturable bacteria collected from a global
sampling of ocean surface waters and surface swabs of
marine organisms. Mar Biotechnol 12: 439–451.
Grossart HP, Schlingloff A, Bernhard M, Simon M &
Brinkhoff T (2004) Antagonistic activity of bacteria isolated
from organic aggregates of the German Wadden Sea. FEMS
Microbiol Ecol 47: 387–396.
Jeong SY, Ishida K, Ito Y, Murakami M & Okada S (2003)
Bacillamide, a novel algicide from the marine bacterium,
Bacillus sp SY-1, against the harmful dinoflagellate,
Cochlodinium polykrikoides. Tetrahedron Lett 44: 8005–
8007.
Ley RE, Harris JK, Wilcox J, Spear JR & Miller SRI (2006)
Unexpected diversity and complexity of the Guerrero
Negro hypersaline microbial mat. Appl Environ Microbiol
72: 3685–3695.
Linares JF, Gustafsson I, Baquero F & Martinez JL (2006)
Antibiotics as intermicrobial signaling agents instead of
weapons. P Natl Acad Sci USA 103: 19484–19489.
Long RA & Azam F (2001) Antagonistic interactions among
marine pelagic bacteria. Appl Environ Microbiol 67:
4975–4983.
Long RA, Qureshi A, Faulkner DJ & Azam F (2003) 2-npentyl-4-quinolinol produced by a marine Alteromonas sp
and its potential ecological and biogeochemical roles. Appl
Environ Microbiol 69: 568–576.
Martinez JL (2008) Antibiotics and antibiotic resistance genes
in natural environments. Science 321: 365–367.
Needleman SB & Wunsch CD (1970) A general method
applicable to search for similarities in amino acid sequence
of 2 proteins. J Mol Biol 48: 443–453.
Paerl HW & Pinckney JL (1996) A mini-review of microbial
consortia: their roles in aquatic production and
biogeochemical cycling. Microb Ecol 31: 225–247.
Paerl HW & Yannarell AC (2010) Environmental dynamics,
community structure and function in a hypersaline
microbial mat. Microbial Mats: Modern and Ancient
Microorganisms in Stratified Systems (Seckbach J & Oren A,
eds), pp. 421–442. Springer, Berlin.
Peña A, Teeling H, Huerta-Cepas J, Santos F, Yarza P, BritoEcheverria J et al. (2010) Fine-scale evolution: genomic,
phenotypic and ecological differentiation in two coexisting
Salinibacter ruber strains. ISME J 4: 882–895.
Penn K, Jenkins C, Nett M, Udwary DW, Gontang EA,
McGlinchey RP et al. (2009) Genomic islands linksecondary
metabolism. ISME J 3: 1193–1203.
Pinckney J, Paerl HW & Bebout BM (1995) Salinity control
of benthic microbial mat community production in a
Bahamian hypersaline lagoon. J Exp Mar Biol Ecol 187:
223–237.
Priscu JC, Fritsen CH, Adams EE, Giovannoni SJ, Paerl HW,
McKay CP et al. (1998) Perennial Antarctic lake ice:
an oasis for life in a polar desert. Science 280: 2095–2098.
FEMS Microbiol Ecol 83 (2013) 74–81
Rappe MS & Giovannoni SJ (2003) The uncultured microbial
majority. Annu Rev Microbiol 57: 369–394.
Rypien KL, Ward JR & Azam F (2010) Antagonistic
interactions among coral-associated bacteria. Environ
Microbiol 12: 28–39.
Socha AM, Long RA & Rowley DC (2007) Bacillamides from
a hypersaline microbial mat bacterium. J Nat Prod 70:
1793–1795.
Stal LJ (1995) Physiological ecology of cyanobacteria in
microbial mats and other communities. New Phytol 131: 1–32.
Teasdale ME, Donovan KA, Forschner-Dancause SR & Rowley
DC (2011) Gram-positive marine bacteria as a potential
resource for the discovery of quorum sensing inhibitors.
Mar Biotechnol 13: 722–732.
Thomashow LS, Weller DM, Bonsall RF & Pierson LS (1990)
Production of the antibiotic phenazine-1-carboxylic acid by
fluorescent Pseudomonas species in the rhizosphere of wheat.
Appl Environ Microbiol 56: 908–912.
Torres-Cortes G, Millan V, Ramirez-Saad HC, Nisa-Martinez R,
Toro N & Martinez-Abarca F (2011) Characterization of
novel antibiotic resistance genes identified by functional
metagenomics on soil samples. Environ Microbiol 13: 1101–
1114.
Villanueva L, Navarrete A, Urmeneta J, Geyer R, White DC &
Guerrero R (2008) Monitoring diel variations of
physiological status and bacterial diversity in an estuarine
microbial mat: an integrated biomarker analysis. Microbial
Ecol 54: 523–531.
Weisburg WG, Barns SB, Pelletier DA & Lane DJ (1991)
16S ribosomal DNA amplification for phylogenetic study.
J Bacteriol 173: 697–703.
Wright JM (1956) The production of antibiotics in soil. Ann
Appl Biol 44: 561–566.
Yannarell AC, Steppe TF & Paerl HW (2006) Genetic variance
in the composition of two functional groups (Diazotrophs
and Cyanobacteria) from a hypersaline microbial mat. Appl
Environ Microbiol 72: 1207–1217.
Yannarell AC, Steppe TF & Paerl HW (2007) Disturbance and
recovery of microbial community structure and function
following Hurricane Frances. Environ Microbiol 9: 576–583.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Fig. S1. Heat map of bacteria-bacterial interactions.
Please note: Wiley-Blackwell is not responsible for the
content or functionality of any supporting materials supplied by the authors. Any queries (other than missing
material) should be directed to the corresponding author
for the article.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved