Characterization of geographically distinct bacterial communities associated with coral mucus produced by Acropora spp. and Porites spp.
ABSTRACT Acropora and Porites corals are important reef builders in the Indo-Pacific and Caribbean. Bacteria associated with mucus produced by Porites spp. and Acropora spp. from Caribbean (Punta Maroma, Mexico) and Indo-Pacific (Hoga and Sampela, Indonesia) reefs were determined. Analysis of pyrosequencing libraries showed that bacterial communities from Caribbean corals were significantly more diverse (H', 3.18 to 4.25) than their Indonesian counterparts (H', 2.54 to 3.25). Dominant taxa were Gammaproteobacteria, Alphaproteobacteria, Firmicutes, and Cyanobacteria, which varied in relative abundance between coral genera and region. Distinct coral host-specific communities were also found; for example, Clostridiales were dominant on Acropora spp. (at Hoga and the Mexican Caribbean) compared to Porites spp. and seawater. Within the Gammproteobacteria, Halomonas spp. dominated sequence libraries from Porites spp. (49%) and Acropora spp. (5.6%) from the Mexican Caribbean, compared to the corresponding Indonesian coral libraries (<2%). Interestingly, with the exception of Porites spp. from the Mexican Caribbean, there was also a ubiquity of Psychrobacter spp., which dominated Acropora and Porites libraries from Indonesia and Acropora libraries from the Caribbean. In conclusion, there was a dominance of Halomonas spp. (associated with Acropora and Porites [Mexican Caribbean]), Firmicutes (associated with Acropora [Mexican Caribbean] and with Acropora and Porites [Hoga]), and Cyanobacteria (associated with Acropora and Porites [Hoga] and Porites [Sampela]). This is also the first report describing geographically distinct Psychrobacter spp. associated with coral mucus. In addition, the predominance of Clostridiales associated with Acropora spp. provided additional evidence for coral host-specific microorganisms.
- SourceAvailable from: George Garrity[show abstract] [hide abstract]
ABSTRACT: The Ribosomal Database Project (RDP) Classifier, a naïve Bayesian classifier, can rapidly and accurately classify bacterial 16S rRNA sequences into the new higher-order taxonomy proposed in Bergey's Taxonomic Outline of the Prokaryotes (2nd ed., release 5.0, Springer-Verlag, New York, NY, 2004). It provides taxonomic assignments from domain to genus, with confidence estimates for each assignment. The majority of classifications (98%) were of high estimated confidence (> or = 95%) and high accuracy (98%). In addition to being tested with the corpus of 5,014 type strain sequences from Bergey's outline, the RDP Classifier was tested with a corpus of 23,095 rRNA sequences as assigned by the NCBI into their alternative higher-order taxonomy. The results from leave-one-out testing on both corpora show that the overall accuracies at all levels of confidence for near-full-length and 400-base segments were 89% or above down to the genus level, and the majority of the classification errors appear to be due to anomalies in the current taxonomies. For shorter rRNA segments, such as those that might be generated by pyrosequencing, the error rate varied greatly over the length of the 16S rRNA gene, with segments around the V2 and V4 variable regions giving the lowest error rates. The RDP Classifier is suitable both for the analysis of single rRNA sequences and for the analysis of libraries of thousands of sequences. Another related tool, RDP Library Compare, was developed to facilitate microbial-community comparison based on 16S rRNA gene sequence libraries. It combines the RDP Classifier with a statistical test to flag taxa differentially represented between samples. The RDP Classifier and RDP Library Compare are available online at http://rdp.cme.msu.edu/.Applied and Environmental Microbiology 09/2007; 73(16):5261-7. · 3.68 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: The development of soil archaeal community structures in relation to primary succession in bulk and rhizosphere soil was examined across the forefield of the receding Rotmoosferner glacier in Austria. Using cloning and denaturing gradient gel electrophoresis (DGGE) analysis of reverse transcription polymerase chain reaction (RT-PCR) products of extracted 16S rRNA, archaeal community structure was compared over a chronosequence representing approximately 150 years of soil development and to reference sites outside the glacier forefield, representing soil exposed for approximately 9500 years. Archaeal community composition was found to be dominated by members of the non-thermophilic or Group 1 Crenarchaeota, where a dramatic yet highly structured successional sequence was observed. Succession over the 150 years sequence could be identified as occurring in three stages, each of which had a phylogenetically distinct 1.1b crenarchaea community with those organisms present in pioneering and intermediate stages belonging to a lineage distinct from those in developed soils. Climax communities also contained organisms belonging to three other major non-thermophilic crenarchaeal lineages. Comparison of archaeal communities in the rhizosphere indicated that plant species composition was not the major driver of specific crenarchaeal populations. These results indicate the potential role of soil crenarchaea in the development of soil substrates, as well as ecological diversity within and between major Group 1 lineages.Environmental Microbiology 04/2005; 7(3):337-47. · 5.76 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: A new method called the neighbor-joining method is proposed for reconstructing phylogenetic trees from evolutionary distance data. The principle of this method is to find pairs of operational taxonomic units (OTUs [= neighbors]) that minimize the total branch length at each stage of clustering of OTUs starting with a starlike tree. The branch lengths as well as the topology of a parsimonious tree can quickly be obtained by using this method. Using computer simulation, we studied the efficiency of this method in obtaining the correct unrooted tree in comparison with that of five other tree-making methods: the unweighted pair group method of analysis, Farris's method, Sattath and Tversky's method, Li's method, and Tateno et al.'s modified Farris method. The new, neighbor-joining method and Sattath and Tversky's method are shown to be generally better than the other methods.Molecular Biology and Evolution 08/1987; 4(4):406-25. · 10.35 Impact Factor
Characterization of Geographically Distinct Bacterial Communities
Associated with Coral Mucus Produced by Acropora spp. and
B. A. McKew, A. J. Dumbrell, S. D. Daud, L. Hepburn, E. Thorpe, L. Mogensen, and C. Whitby
Department of Biological Sciences, University of Essex, Wivenhoe Park, Colchester, Essex, United Kingdom
andregion.Distinctcoralhost-specificcommunitieswerealsofound;forexample, Clostridiales weredominanton Acroporaspp.
(atHogaandtheMexicanCaribbean)comparedto Poritesspp.andseawater.Withinthe Gammproteobacteria,Halomonas spp.
therewasalsoaubiquityof Psychrobacter spp.,whichdominated AcroporaandPoriteslibrariesfromIndonesiaand Acropora
librariesfromtheCaribbean.Inconclusion,therewasadominanceof Halomonas spp.(associatedwith AcroporaandPorites
[MexicanCaribbean]), Firmicutes (associatedwith Acropora[MexicanCaribbean]andwith AcroporaandPorites[Hoga]),and
Cyanobacteria (associatedwith AcroporaandPorites[Hoga]andPorites[Sampela]).Thisisalsothefirstreportdescribinggeo-
graphicallydistinct Psychrobacter spp.associatedwithcoralmucus.Inaddition,thepredominanceof Clostridiales associated
nutrition (7), response to stress (12), and health and disease (6,
40). Coral bleaching and various coral diseases are increasing due
to changes in environmental conditions that result in either an
fore, understanding the microbial communities associated with
corals and how they vary in response to changing environmental
conditions is important for understanding the future health of
Previous studies that used both culture-dependent and cul-
ture-independent methods demonstrated that coral-associated
those dominating the surrounding reef water (4, 17, 39). Similar
cies from geographically different locations, and different bacte-
rial communities have been found on different coral species (38,
39). By using clone libraries and sequence analysis, Bourne and
Munn (4) found that the majority of clones recovered from the
coral tissue of Pocillopora damicornis were related to Gammapro-
teobacteria, while Alphaproteobacteria were dominant in the coral
mucus, thus further supporting the hypothesis that specific bac-
terium-coral associations exist.
Although many studies have supported the hypothesis that
corals harbor unique microbiota, inconsistencies across studies
have raised many questions about the specificity and dynamics of
associations between corals and microbes. One of the major lim-
ing methods do not allow for characterization of the microbial
community beyond the most dominant taxa (47). However, cur-
icroorganisms are important to coral reef ecosystems
through their roles in carbon/nitrogen cycling (43), coral
tection of rare taxa (45). These rare taxa remain largely unex-
plored, but they may be extremely important and may become
more dominant in response to environmental changes (47).
of bacterial communities across reef bioregions and environmen-
with a hemispherical shape and slow growth rates. As a result,
Porites tend to be longer lived, often for hundreds of years, and
grow to large sizes (37, 48). The Caribbean Porites asteroides may
grow up to 1 m but tends to form more numerous, smaller colo-
nies, while Porites lutea from Indonesia grows up to a few meters
pora formosa from Indonesia grows on average up to 1 m. Both
genera occur in shallow, tropical reef environments, reef slopes,
and in lagoons (48).
the bacterial community structures associated with the coral mu-
cus produced by Porites astreoides and Acropora palmata from
Received 8 December 2011 Accepted 26 April 2012
Published ahead of print 25 May 2012
Address correspondence to C. Whitby, email@example.com.
Supplemental material for this article may be found at http://aem.asm.org/.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
August 2012 Volume 78 Number 15 Applied and Environmental Microbiologyp. 5229–5237aem.asm.org
Mexican Caribbean reefs with those associated with the closely
related Porites lutea and Acropora formosa found in Indonesia
(southeast Sulawesi, Hoga and Sampela reefs). Such information
understanding of coral physiology, health, and ecosystems.
MATERIALS AND METHODS
from three regions on a colony, from triplicate living colonies) of Porites
astreoides and Acropora palmata at Punta Maroma, Mexican Caribbean
and Acropora formosa at Hoga (18 samples) and Sampela (18 samples),
mucus samples were filtered using 0.22-?m filters, and the filters were
stored at ?20°C. Overlying seawater (two 500-ml volumes) adjacent to
the coral colonies was also collected at a depth of 10 m, filtered through
0.22-?m filters, and stored at ?20°C. The water temperature during col-
lections at both sites was typically 28°C.
Total community DNA was extracted from the filtered seawater and mu-
cus samples (from pooled colony regions on each colony replicate) by
using a modified beadbeating method (26). Eubacterial 16S rRNA genes
were PCR amplified using the primers for positions 341 to 534 in Esche-
richia coli (Table 1) (28). PCRs were performed in a GeneAmp PCR sys-
(50-?l volumes): 1? buffer (Qiagen), 0.2 mM deoxynucleoside triphos-
phates (Fermentas), 0.4 ?M each primer, 2.5 U Taq DNA polymerase
(Qiagen), and approximately 25 ng of DNA (26). PCR cycling conditions
were as follows: 95°C for 5 min, followed by 30 cycles of 94°C for 1 min,
at 4°C. PCR products were analyzed using 1% (wt/vol) 1? TAE agarose
gels (40 mM Tris-acetate, 1.0 mM EDTA; pH 8.0), stained with ethidium
bromide (0.5 mg liter?1), and visualized under UV light by using the
Gel-Doc system (Bio-Rad). Denaturing gradient gel electrophoresis
silver stained (30).
Clone libraries. PCR products were obtained using the primers pA/
efficiency JM109 Escherichia coli cells (Promega) according to the manu-
a plasmid purification kit (Qiagen) according to the manufacturer’s in-
clones by Geneservice Ltd., Cambridge, United Kingdom. Partial se-
the GenBank database by using the Basic Local Alignment Search Tool
(BLAST) network service (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi)
(1). Sequences were aligned with sequences from GenBank by using the
RDP INFERNAL alignment tool (29). Analysis was performed using
PHYLIP 3.4 (16) with Jukes-Cantor DNA distance correction and neigh-
bor-joining methods (21, 42). Bootstrap analysis was based on 100 repli-
was performed using Treeview (WIN32; version 1.5.2) (31). For pyrose-
quencing, PCR products were obtained as described above using the
primers F341-GC and 534 R (Table 1) (28), except that the forward
primer had no GC clamp and a 5= modification with a 454 amplicon
adaptor followed by a unique 10-nucleotide barcode (2, 33). PCR prod-
ucts were quantified with a Nanodrop ND-1000 spectrophotometer, and
pooled sample was analyzed by pyrosequencing by the NERC Biomolec-
ular Analysis Facility.
Pyrosequence reads were analyzed using the QIIME pipeline and its
associated modules (8). All sequences were checked for the presence of
correct pyrosequencing adaptors, 10-bp barcodes, and taxon-specific
removed. In addition, sequences of ?150 bp or ?200 bp in read length,
sequences with low quality scores (?20), and sequences containing ho-
mopolymer inserts were also removed from further analysis. All pyrose-
FIG 1 Sampling sites. (A) Map of the Wakatobi National Park, Sulawesi, Indonesia, showing the Sampela and Hoga (Hoga buoy 2) reefs. (Reprinted from
reference 19 with permission of the publisher.) (B) Map showing Punta Maroma on the northeastern coast of the Yucatan Peninsula, Mexico.
TABLE 1 Summary of PCR primer sequences used in this study
aPrimer 341 contains a 40-nucleotide GC-rich sequence (GC clamp) for DGGE
McKew et al.
aem.asm.orgApplied and Environmental Microbiology
quence reads were clustered into operational taxonomic units (OTUs) by
using the UClust algorithm (14). Representative sequences from each
OTU were identified using the RDP classifier, which assigns taxonomic
identities against the RDP database by using a naïve Bayesian classifier
(49). Finally, all singletons were removed before further analysis.
Statistical analysis. Similarity between DGGE profiles was calculated
using binary data indicating the presence of particular bands (Jaccard’s
index) and a hierarchal cluster analysis constructed using Primer E soft-
ing (NMDS) ordination of distance matrices calculated from the OTU
in composition between sites were assessed using permutation-based
multivariate analysis of variance (PERMANOVA), based on the distance
Geographic location (either the Mexican Caribbean, Sampela, or Hoga)
was not explored.
Species diversity (number and relative abundance of OTUs) was cal-
corals by using a simple randomization test, based on 10,000 randomiza-
tions (46). This randomization approach was also used to compare Jac-
card index results between coral species. The randomization approach
used treats the entire community as a single data set and is an absolute
statistical measure that does not require replication to produce probabil-
where full replication of sampled communities is impossible (25). All
analyses were conducted in the R statistical language version 2.7.2 and
using the R standard libraries and the community ecology analysis-spe-
cific package Vegan (version 2007; R Development Core Team).
mitted to GenBank and assigned accession numbers HQ456683 to
Community and phylogenetic analyses. Bacterial communities
from coral and seawater samples were analyzed by 16S rRNA
PCR-DGGE analysis (see Fig. S1 in the supplemental material).
There were clear differences in DGGE profiles between corals and
their geographical regions (see Fig. S1). Profiles were, however,
similar between colony replicates, and so the replicates were
pooled for detailed community analysis by pyrosequencing of the
16S rRNA gene. Libraries comprising a total of 9,353 sequences
with low quality scores [?20], and sequences containing ho-
cus samples produced by P. astreoides and A. palmata (Mexican
Caribbean) and P. lutea and A. formosa (Hoga and Sampela, In-
donesia) were analyzed (Table 2).
NMDS ordination revealed distinct microbial communities
associated with samples from geographically distinct regions (Ta-
ble 2; Fig. 2). PERMANOVA results supported the NMDS ordi-
nation and showed that the compositions of the bacterial assem-
blages were significantly different between geographic sites
(PERMANOVA, based on 10,000 randomizations; F1,8? 2.08;
P ? 0.03), but not between coral species and seawater samples
(F1,8? 0.58; P ? 0.88). The most noticeable result was that bac-
terial communities from the corals and seawater in the Mexican
Caribbean were clearly distinct from those from Indonesia (Fig.
2). Although across all sites there was no significant difference
between coral species in the composition of the associated bacte-
rial assemblages (see above), by focusing only on data from the
ter samples (simple pairwise randomization test based on 10,000
randomizations; Jaccard’s index, ?0.89; P ? 0.01 in all cases).
Clustering and classification (Table 2) and diversity analysis
(Table 2; Fig. 3A and B) of pyrosequencing libraries revealed geo-
graphically distinct coral-bacteria associations. In general, bacte-
diverse (H=, 3.18 to 4.25), followed by samples from Hoga (H=,
terial assemblages (H=, 2.54 to 2.64) (Fig. 3B). However, most
samples had similar levels of bacterial diversity, and only in bac-
terial assemblages from the Mexican Caribbean Acropora samples
were diversity levels significantly higher than those from other
sites or corals (simple pairwise randomization test based on
10,000 randomizations; ?H=, ?0.87; P ? 0.001 in all cases) (Fig.
3B). Analyses of diversity indices were supported by analysis of
rarefied species richness, which provided quantitatively similar
results but accounted for differences in sequencing intensities be-
tween samples (Fig. 3A).
Overall, Gammaproteobacteria dominated pyrosequencing li-
from Porites (76.1%) (from the Mexican Caribbean) (Table 2).
This was in contrast to Acropora and Porites libraries from Sam-
pela, which were dominated by Alphaproteobacteria (comprising
?54.3%). Bacterial communities that differed with coral genera
(from either Indonesia or the Caribbean) were also observed.
For example, within the Gammaproteobacteria, Halomonas spp.
(49%) and Alteromonadales (9.1%) were predominant in Porites
libraries (from the Mexican Caribbean), compared to Acropora at
the same location and the corresponding Indonesian corals
(?5.6%) (Table 2). Other differences in coral-associated bacteria
included Clostridiales, which were more predominant from Acro-
Mexican Caribbean, 7.2% of sequences were related to Clostridi-
ales from Acropora, compared to 0.2% from Porites libraries. In
compared to Acropora (6.4%) and the corresponding corals from
As with Indonesia, other distinct sequences were associated
with the corals found in the Mexican Caribbean (Table 2). For
example, there was a relative dominance of Firmicutes (24.1%),
primarily bacilli (16.7%), found in Acropora libraries, but Firmic-
utes were much rarer (?2.8%) in the Porites libraries from the
spp. were more dominant in the Mexican Caribbean Acropora
libraries (6.2%) than Porites (2.0%) and the corresponding Indo-
nesian coral libraries (?0.2%). In contrast, Sulfitobacter spp.
dominated both coral libraries from Sampela (comprising
?47.3%) compared to the corresponding coral libraries from
Hoga and the Mexican Caribbean (?3.2%).
Similarities were also found with the coral-associated bacterial
communities (from either Indonesia or the Caribbean) (Table 2).
ies (with the exception of Porites from the Mexican Caribbean).
Psychrobacter spp. were also more dominant in Acropora (63.4%)
and Porites (53.4%) from Hoga (compared to the corresponding
corals in Sampela [27.1% and 30.1%, respectively] and Acropora
[26.7%] from the Mexican Caribbean) (Table 2). Similarly, cya-
Bacterial Diversity Associated with Coral Mucus
August 2012 Volume 78 Number 15aem.asm.org 5231
nobacteria (28%) were also more dominant in Porites than in
libraries from Sampela (?5.2%) and the Mexican Caribbean
Not surprisingly, there were also differences observed in sea-
ribbean (Table 2). For example, Gammaproteobacteria (70.1%)
water from Sampela (23.6%) and the Mexican Caribbean (26.7%).
Conversely, Alphaproteobacteria (67.7%) were more dominant in
seawater from Sampela, Indonesia, than in seawater from Hoga
(20.0%) and the Mexican Caribbean (42.3%). In addition to Alpha-
and Gammaproteobacteria, seawater from the Mexican Caribbean
was dominated by Sphingomonadales (31.9%), Cyanobacteria
Clone libraries. In order to obtain almost-full-length 16S
rRNA gene sequences for phylogenetic analysis, clone libraries
were generated from each coral sample (Fig. 4A and B). Libraries
16S rRNA gene sequences (of ?500 bp) were obtained (distrib-
uted relatively evenly across all samples). Generally, there was
sequences from Acropora and Porites (at Hoga) predominantly
clusters associated with Psychrobacter spp. and Halomonas spp.
In addition to Gammaproteobacteria, one clone (HP7) from
Porites (Hoga) grouped with Synechococcus spp. within the Cya-
nobacteria, and two clones (HA11 and MA13) from Acropora
(from Hoga and the Mexican Caribbean, respectively) clustered
TABLE 2 Bacterial assemblages based on 16S rRNA pyrosequencing libraries from coral mucus and seawater
Phylum, class, or order
% of sequences (total n) in phylum, class, or order from sample areaa
Fusobacteria 0.61.200 0.20000
Rhodobacterales (not including Silicibacter spp. or
Oceanospirillales (not including Halomonas spp.)
Pseudomonadales (not including Acinetobacter
spp., Psychrobacter spp.)
20.0 59.054.3 67.728.3 14.043.2
aSamples were from Acropora formosa at Hoga (HA), Porites lutea at Hoga (HP), seawater at Hoga (HSW), Acropora formosa at Sampela (SA), Porites lutea at Sampela (SP),
seawater at Sampela (SSW), Acropora palmata at the Mexican Caribbean (MA), Porites astreoides at the Mexican Caribbean (MP), and seawater at the Mexican Caribbean (MSW).
Numbers in bold indicate that the group represents ?5% of the community.
McKew et al.
aem.asm.org Applied and Environmental Microbiology
with an uncultured Gram-positive bacterium with strong boot-
dominant within the Alphaproteobacteria, followed by Gamma-
proteobacteria (Fig. 4A and B). Dominant sequences from Acro-
fitobacter spp., while dominant clone sequences from Porites
(Sampela) affiliated with Gammaproteobacteria, including Halo-
monas spp., Alteromonas spp., and Psychrobacter spp. In addition
to Gammaproteobacteria sequences, one clone (SP19) associated
with Porites from Sampela clustered with Synechococcus spp.
within the Cyanobacteria. Another clone (SP12) associated with
Porites from Sampela clustered with Exiguobacterium spp. within
In comparison to clones from Indonesian corals, the clone li-
braries from Acropora and Porites from the Mexican Caribbean
spanned several phyla, including Gamma- and Alphaproteobacte-
ria, Firmicutes, Altermonadales, and Actinobacteria (Fig. 4A and
ples from Indonesia, only three clones from Acropora (Mexican
Caribbean) clustered within the Gammaproteobacteria. In addi-
tion, two clones (MA2 and MA11) were closely related to Exiguo-
bacterium spp. within the Firmicutes with strong bootstrap sup-
port (100%) (Fig. 4B).
Gammaproteobacteria (specifically, Halomonas spp. and Al-
teromonas spp.) dominated the Porites library (Mexican Carib-
bean), followed by Alphaproteobacteria. We also found that no
clone sequences from the Porites library (Mexican Caribbean)
clustered with Psychrobacter spp. In addition, two clones (MA1
and MP1; from Acropora and Porites libraries, respectively) had
99% sequence identity to Dietzia spp. within the Actinobacteria,
coccus spp. with strong bootstrap support (100%).
Although clone libraries generally corroborated information
from pyrosequencing libraries, some differences were observed,
ing library from Porites for Sampela (54.3%), but only one clone
sequence (SP7) was found within the Alphaproteobacteria (Fig.
4B). Conversely, Vibrionales were rarer in the pyrosequencing li-
brary, while three clone sequences (HP6, HP10, and HP13) from
Porites (Hoga) grouped with Vibrio spp. with strong bootstrap
support (96%) (Fig. 4A).
The coral mucus samples yielded more diverse 16S rRNA pyrose-
found a high relative abundance of Gammaproteobacteria se-
quences associated with Porites from the Mexican Caribbean, fol-
Porites from Sampela. Similar findings have been previously re-
with Porites astreoides from Panama and Bermuda (39). Further-
more, a high relative abundance of Gammaproteobacteria in the
coral Montastrea cavernosa from the Mexican Caribbean has also
been found (17). In our study, within the Gammaproteobacteria
there was a predominance of sequences relating to Halomonas
spp. associated with Porites and Acropora from the Mexican Ca-
ribbean; in addition, this is the first study to report an abundance
of sequences relating to Psychrobacter spp. associated with both
Acropora spp. and Porites spp.
In addition to Gammaproteobacteria, another study found
Alphaproteobacteria to be the dominant microbial group within
the coral mucus of Pocillopora damicornis from the Great Barrier
Reef (4). More specifically, 36% of the clones were affiliated with
FIG 3 Bacterial community diversity based on the analysis of rarefied species
(OTU) richness (A) and the Shannon-Wiener diversity index (B). OTUs were
taxonomic identities using the RDP classifier. Samples are from Acropora
formosa at Hoga (HA), Porites lutea at Hoga (HP), seawater at Hoga (HSW),
Acropora formosa at Sampela (SA), Porites lutea at Sampela (SP), seawater at
Sampela (SSW), Acropora palmata at the Mexican Caribbean (MA), Porites
FIG 2 NMDS ordination of distance matrices calculated from the OTU
pyrosequence read matrix and using Jaccard’s index. Shown are the bacterial
communities associated with Acropora formosa at Hoga (HA), Porites lutea at
Hoga (HP), seawater at Hoga (HSW), Acropora formosa at Sampela (SA),
the Mexican Caribbean (MA), Porites astreoides at the Mexican Caribbean
(MP), and seawater at the Mexican Caribbean (MSW) based on the 454 pyro-
Bacterial Diversity Associated with Coral Mucus
August 2012 Volume 78 Number 15aem.asm.org 5233