OMICS Driven Microbial Ecology Hunt for a Core Microbiome

Environmental Microbiology, Volume 14, Issue 1, January 2012 VOLUME 14 NUMBER 1 JANUARY 2012 www.env-micro.com ISSN 1462-2912 Contents environmental microbiology Correspondence 140 Microbial rhodopsins on leaf surfaces of terrestrial plants N. Atamna-Ismaeel, O. M. Finkel, F. Glaser, I. Sharon, R. Schneider, 1 Omics for understanding microbial functional dynamics A. F. Post, J. L. Spudich, C. von Mering, J. A. Vorholt, D. Iluz, O. Béjà & J. K. Jansson, J. D. Neufeld, M. A. Moran & J. A. Gilbert S. Belkin 147 Photoautotrophic symbiont and geography are major factors affecting highly Minireviews structured and diverse bacterial communities in the lichen microbiome 4 Beyond the Venn diagram: the hunt for a core microbiome B. P. Hodkinson, N. R. Gottel, C. W. Schadt & F. Lutzoni A. Shade & J. Handelsman 162 Phosphate transporters in marine phytoplankton and their viruses: environmental cross-domain commonalities in viral-host gene exchanges 13 Targeted metagenomics: a high-resolution metagenomics approach for specifi c A. Monier, R. M. Welsh, C. Gentemann, G. Weinstock, E. Sodergren, gene clusters in complex microbial communities E. V. Armbrust, J. A. Eisen & A. Z. Worden H. Suenaga 177 Complete genome of Candidatus Chloracidobacterium thermophilum, a chlorophyll-based photoheterotroph belonging to the phylum Acidobacteria Research articles A. M. Garcia Costas, Z. Liu, L. P. Tomsho, S. C. Schuster, D. M. Ward & 23 Microbial metatranscriptomics in a permanent marine oxygen minimum zone D. A. Bryant F. J. Stewart, O. Ulloa & E. F. DeLong 191 Transcriptional responses of surface water marine microbial assemblages 41 Genome content of uncultivated marine Roseobacters in the surface ocean to deep-sea water amendment microbiology H. Luo, A. Löytynoja & M. A. Moran Y. Shi, J. McCarren & E. F. DeLong 207 Phage–bacteria relationships and CRISPR elements revealed by a 52 Genomic content of uncultured Bacteroidetes from contrasting oceanic metagenomic survey of the rumen microbiome provinces in the North Atlantic Ocean M. E. Berg Miller, C. J. Yeoman, N. Chia, S. G. Tringe, F. E. Angly, P. R. Gómez-Pereira, M. Schüler, B. M. Fuchs, C. Bennke, H. Teeling, R. A. Edwards, H. J. Flint, R. Lamed, E. A. Bayer & B. A. White J. Waldmann, M. Richter, V. Barbe, E. Bataille, F. O. Glöckner & R. Amann 228 Bacterial community transcription patterns during a marine phytoplankton bloom VOLUME 14 67 Whole-genome expression analysis reveals a role for death-related genes in J. M. Rinta-Kanto, S. Sun, S. Sharma, R. P. Kiene & M. A. Moran stress acclimation of the diatom Thalassiosira pseudonana 240 Metagenomic comparison of microbial communities inhabiting confi ned and K. Thamatrakoln, O. Korenovska, A. K. Niheu & K. D. Bidle unconfi ned aquifer ecosystems 82 Comparative microbial diversity analyses of modern marine thrombolitic mats R. J. Smith, T. C. Jeffries, B. Roudnew, A. J. Fitch, J. R. Seymour, by barcoded pyrosequencing M. W. Delpin, K. Newton, M. H. Brown & J. G. Mitchell J. M. Mobberley, M. C. Ortega & J. S. Foster 254 Metagenomic analysis of a complex marine planktonic thaumarchaeal community from the Gulf of Maine 101 The genome sequence of Desulfatibacillum alkenivorans AK-01: a blueprint for B. J. Tully, W. C. Nelson & J. F. Heidelberg

anaerobic alkane oxidation NUMBER 1 A. V. Callaghan, B. E. L. Morris, I. A. C. Pereira, M. J. McInerney, 268 Unveiling microbial life in the new deep-sea hypersaline Lake Thetis. Part II: R. N. Austin, J. T. Groves, J. J. Kukor, J. M. Suﬂ ita, L. Y. Young, a metagenomic study G. J. Zylstra & B. Wawrik M. Ferrer, J. Werner, T. N. Chernikova, R. Bargiela, L. Fernández, V. La Cono, J. Waldmann, H. Teeling, O. V. Golyshina, F. O. Glöckner, 114 Iron transporters in marine prokaryotic genomes and metagenomes M. M. Yakimov & P. N. Golyshin B. M. Hopkinson & K. A. Barbeau Web alert 129 Characterization of the rumen microbiota of pre-ruminant calves using metagenomic tools 282 Microbial omics R. W. Li, E. E. Connor, C. Li, R. L. Baldwin, VI & M. E. Sparks L. P. Wackett PAGES 1–284

Forthcoming papers

JANUARY 2012 Environmental Microbiology Environmental Microbiology Reports

Minireview Minireview Marine sponges and their microbial symbionts: love and other relationships Local and regional factors inﬂ uencing bacterial community assembly N. S. Webster & M. W. Taylor E. S. Lindström & S. Langenheder Research articles Multilocus sequence analysis, taxonomic resolution and biogeography of marine Brief reports Synechococcus Repeated sampling reveals differential variability in measures of species richness S. Mazard, M. Ostrowski, F. Partensky & D. J. Scanlan and community composition in planktonic protists Massive dominance of Epsilonproteobacteria in formation waters from a Canadian J. R. Dolan & T. Stoeck oil sands reservoir containing severely biodegraded oil C. R. J. Hubert, T. B. P. Oldenburg, M. Fustic, N. D. Gray, S. R. Larter, K. Penn, Age, sun and substrate: triggers of bacterial communities in lichens A. K. Rowan, R. Seshadri, A. Sherry, R. Swainsbury, G. Voordouw, J. K. Voordouw M. Cardinale, J. Steinová, J. Rabensteiner, G. Berg & M. Grube & I. M. Head Breaking a paradigm: cosmopolitan and abundant freshwater actinobacteria A novel genus of multicellular magnetotactic prokaryotes from the Yellow Sea are low GC K. Zhou, W.-Y. Zhang, K. Yu-Zhang, H.-M. Pan, S.-D. Zhang, W.-J. Zhang, R. Ghai, K. D. McMahon & F. Rodriguez-Valera H.-D. Yue, Y. Li, T. Xiao & L.-F. Wu Correlative microscopy for phylogenetic and ultrastructural characterization Prokaryotic taxonomy in the sequencing era – the polyphasic approach revisited of microbial communities P. Kämpfer & S. P. Glaeser B. Knierim, B. Luef, P. Wilmes, R. I. Webb, M. Auer, L. R. Comolli & J. F. Banﬁ eld OMICS Driven Microbial Ecology Hunt for a core microbiome

This Journal is indexed by Index Medicus, SciSearch, ISI High-resolution metagenomics of speciﬁ c gene clusters Alerting Services, Biotechnology Citation Index, Current Contents/Life Sciences, and Current Contents/Agriculture, Biology and Environmental Sciences. Iron transporters in metagenomes

This journal is available online at Wiley Online Library. CRISPR elements in the rumen microbiome Visit http://wileyonlinelibrary.com/ to search the articles and register for table of contents e-mail alerts.

Published jointly by the Society for Applied Microbiology and Blackwell Publishing Ltd.

001_emi_v14_i1_8.7mm_OC.indd 1 12/23/2011 2:42:47 PM Environmental Microbiology (2012) 14(1), 82–100 doi:10.1111/j.1462-2920.2011.02509.x

Comparative microbial diversity analyses of modern marine thrombolitic mats by barcoded pyrosequencing

Jennifer M. Mobberley, Maya C. Ortega and Introduction Jamie S. Foster* Microbialites are carbonate build-ups that are derived Department of Microbiology and Cell Science, University from the trapping, binding and mineral precipitation activi- of Florida, Space Life Sciences Laboratory, Kennedy ties of microbial mat communities (Burne and Moore, Space Center, FL 32899, USA. 1987). Microbialites are found across the globe in a wide variety of aquatic habitats (e.g. freshwater, hypersaline,

Summaryemi_2509 82..100 marine) and represent one of the oldest known ecosystems on the planet (Canfield and DesMarais, 1993; Grotz- Thrombolites are unlaminated carbonate structures inger and Knoll, 1999). Microbialites are classified by their that form as a result of the metabolic interactions of internal micro- and mesostructure, which can range from complex microbial mat communities. Thrombolites the well-laminated stromatolites to the unlaminated, have a long geological history; however, little is clotted thrombolites (Kennard and James, 1986). One of known regarding the microbes associated with the few modern sites where both stromatolites and throm- modern structures. In this study, we use a barcoded bolites are actively forming is the island of Highborne Cay 16S rRNA gene-pyrosequencing approach coupled located in the Exuma Sound, Bahamas (Dravis, 1983; Dill with morphological analysis to assess the bacterial, et al., 1986; Reid et al., 2000). Considerable progress has cyanobacterial and archaeal diversity associated with been made on understanding the microbiological, geo- actively forming thrombolites found in Highborne logical and biogeochemical processes associated with the Cay, Bahamas. Analyses revealed four distinct micro- stromatolitic microbialites of Highborne Cay (Reid et al., bial mat communities referred to as black, beige, pink 2000; Visscher et al., 2000; Havemann and Foster, 2008; and button mats on the surfaces of the thrombolites. Baumgartner et al., 2009; Dupraz et al., 2009; Foster At a coarse phylogenetic resolution, the domain bac- et al., 2009; Foster and Green, 2011). However, far less terial sequence libraries from the four mats were progress has been made on understanding similar pro- similar, with Proteobacteria and Cyanobacteria being cesses in the adjacent, unlaminated thrombolitic microbi- the most abundant. At the finer resolution of the rRNA alites (Planavsky et al., 2009; Myshrall et al., 2010). gene sequences, significant differences in commu- Initial work by Myshrall and colleagues (2010) provided nity structure were observed, with dramatically differ- the first assessment of the microbial communities associ- ent cyanobacterial communities. Of the four mat ated with the thrombolite structures of Highborne Cay. types, the button mats contained the highest diversity This previous study generated clone libraries to the 16S of Cyanobacteria, and were dominated by two and 18S ribosomal RNA (rRNA) genes revealing that the sequence clusters with high similarity to the genus thrombolitic communities were predominately bacterial. Dichothrix, an organism associated with the deposi- Eukaryotes comprised less than 11% of the recovered tion of carbonate. Archaeal diversity was low, but operational taxonomic units (OTUs), with most sequences varied in all mat types, and the archaeal community sharing similarity to grazing, bacterivorous Nematoda was predominately composed of members of the (Myshrall et al., 2010). This previous study also identified Thaumarchaeota and Euryarchaeota. The morpho- two dominant thrombolitic mat types at Highborne Cay logical and genetic data support the hypothesis that including a highly productive, nodous mat called ‘button’ the four mat types are distinctive thrombolitic mat mat that was comprised of filamentous calcified cyano- communities. bacterial filaments (Planavsky et al., 2009; Myshrall et al., 2010); and a ‘pink’ encrusted mat that was found adjacent to the button thrombolitic mats (Myshrall et al., 2010). Ribosomal RNA gene analysis of the bacteria associated with these two mat types indicated that both thrombolitic Received 9 November, 2010; accepted 5 April, 2011. *For correspondence. E-mail jfoster@ufl.edu; Tel. (+1) 321 861 2900; Fax (+1) 321 mats types contained microbial communities that were 861 2925. distinct from the adjacent stromatolitic mats (Myshrall

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd Microbial diversity in modern thrombolitic mats 83 et al., 2010). In this previous study, however, sequencing as ‘pink’ mats that had extensive red alga colonization coverage of the microbial population, as determined by (Fig. 1C) and like the beige mat, had extensive lithifica- Good’s estimate, indicated that only 59% of the bacterial tion throughout the surface of the mat. The fourth mat community was represented in the button mat clone librar- was an irregular nodous mat referred as ‘button’ mats ies and only 49% in the pink mat type (Good, 1953; (Fig. 1D) that contained numerous patches of vertically Myshrall et al., 2010). orientated calcified filaments. Field observations of the In the present study, we expanded on this initial report spatial organization of the four mat types revealed that by using multi-domain and deep sequencing approaches the button mat types were consistently superposed over to examine and compare the microbial diversity associ- the other mat types, most commonly pink or beige mats. ated with thrombolitic microbial mat types found at High- Cross-sections of the four mat types revealed that borne Cay, Bahamas. Our objective was to examine although there was no laminated mesostructure in the bacterial and archaeal distribution within the thrombolites thrombolitic build-ups, the surface microbial mat commu- to determine whether the thrombolitic mat types represent nities did exhibit some layering (Fig. 1E–H). Higher mag- successive stages of thrombolite formation and develop- nification of the four mat types revealed that each ment. To address this goal and build on the previous study community had a dominant ecotype that morphologically we used next-generation, barcoded pyrosequencing to distinguished it from the others (Fig. 1E–P). In the black improve sequencing coverage of the bacterial populations mats, the surface was heavily colonized by clusters of a within the thrombolitic mat types. We also generated pigmented coccoid cells that ranged between 3 and 5 mm amplicon libraries for archaeal and cyanobacterial popu- in diameter (Fig. 1I and M). Coccoid cells dissected from lations to assess the impact of these taxa on thrombolite the black mats had 16S rRNA gene sequences that community structure. Pyrosequencing has emerged as a shared 97% similarity to cyanobacterial Pleurocapsa sp. robust tool for investigating community diversity and struc- isolates from stromatolites located in Shark Bay, Australia. ture in microbial ecosystems (Edwards et al., 2006; Sogin The beige mats were distinguished by the presence of a et al., 2006; Roesch et al., 2007; 2009; Miller et al., 2009). pronounced subsurface cyanobacterial layer rich in By using this high-throughput approach to sequence the coccoid cells, which shared morphological similarity to small subunit rRNA gene, we were able to more fully Gloeocapsa spp. (Fig. 1J and N). The algal cells that gave characterize and compare these thrombolitic mat commu- the pink mats their colour (Myshrall et al., 2010) were nities in parallel to assess the microbial complexity of concentrated in the upper 500 mm of the mat surface these thrombolite ecosystems. (Fig. 1K), and each cell was encased in a thick sheath (10–20 mm; Fig. 1O). The red alga were previously classified as belonging to the genus Chlorophyta (Myshrall Results et al., 2010). The fourth mat type, referred to as the button mat types, was the most widely distributed of all the mat Morphological characterization of thrombolitic mat types types, and had a unique morphology in which calcified Thrombolitic microbialites were localized to the intertidal filaments formed vertical bundles throughout the surface zone of a fringing reef complex along the eastern margin nodes of the mat (Planavsky et al., 2009). The filaments of Highborne Cay, Bahamas. The distribution of the were previously classified as belonging to the genus thrombolite structures extended over 1 km of the inter- Dichothrix (order Nostocales) based on the morphological tidal zone and their sizes ranged up to several metres in characteristics of basal heterocysts and tapered apical length and width. The surfaces of the thrombolitic struc- ends (Planavsky et al., 2009). In this study, we genetically tures had four distinctive microbial mat communities that characterized the organism using 16S rRNA gene were spatially distributed along the intertidal zone and sequencing of filaments dissected from the mats. After were characterized by the texture and pigmentation of gene sequence alignment, two distinct clades of organ- the mats (Fig. 1A–D). Those near shore thrombolites isms from the genus Dichothrix were identified, and that recently emerged from sand burial prior to collection sequences of these clades varied up to 1.5%, and both were covered with a dark pigmented mat referred to as clades shared 97% sequence similarity to Calothrix sp. ‘black’ mat (Fig. 1A), which exhibited a uniform, lithified PCC7507 and Rivularia sp. PCC7116. crust in the upper few millimetres of the mat. The second mat type labelled as ‘beige’ mats lacked any Small subunit rRNA gene diversity in thrombolitic mats obvious surface pigmentation but had extensive encrus- tation of the surface (Fig. 1B). The remaining two mat Three 16S rRNA gene amplicon libraries were generated types have been previously described (Myshrall et al., for each of the four mat types (with eight pooled DNA 2010) and were found on the most seaward side of the extractions) and included a bacterial, cyanobacterial and site. These mats included a smooth flat mat referred to archaeal library. Each of the 12 libraries was labelled with

EFGH

IJKL eps

M N OP

Fig. 1. Morphological characterization of thrombolitic mat communities. A–D. Surface view of thrombolitic mats: black (A); beige (B); pink (C); and button (D). Bar, 1 cm. E–H. Cross-section of thrombolitic mats: black (E); beige (F); pink (G); and button (H). Bar, 1 cm. I–L. Higher magnification of thrombolitic mats at 32¥: black (I); beige (J); pink (K) with overlying exopolymeric substances (EPS); and button (L). Arrows indicate morphologically dominant cells in each mat. Bar, 250 mm. M–P. Cell morphology: black mat Pleurocapsa spp. (M). Bar, 10 mm; beige mat Gloeocapsa spp. (N). Bar, 10 mm; pink mat Chlorophyta (O). Bar, 50 mm; and button mat Dichothrix spp. filaments (P) with calcium carbonate precipitation (cp) on the gelatinous sheath. Bar, 20 mm. a unique oligonucleotide barcode and pyrosequenced button mat type having the highest (Table 1). Equalized using 454 GS FLX technology. A total of 53 456 barcoded sequence data were used to generate diversity indices for pyrosequences were recovered from the combined librar- each primer set and thrombolitic mat type (see Experi- ies with an average read length of 260 bp. After screening mental procedures; Table 1). Compared with previous for quality (see Experimental procedures) 34 027 microbialitic mat diversity analyses, there was a 10-fold sequences remained with an average read length of increase in the number of bacterial rRNA gene sequences 240 bp and were sorted by barcode, aligned and analysed recovered from these thrombolitic mats (Goh et al., 2009; with mothur (Schloss et al., 2009). Each mat type was Myshrall et al., 2010; Foster and Green, 2011). represented by between 6000 and 11 000 sequencing reads, which were clustered into OTUs at 97% sequence Bacterial community composition similarity (Table 1). Based on mothur clustering, the number of OTUs in the thrombolitic mats ranged between Classification and community analysis of the domain Bac- 779 and 1152 in four thrombolitic mat types with the pink teria sequence libraries were based on a defined region of mats having the lowest number of total OTUs and the the variable 2 (V2) region of the 16S rRNA gene in all

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 01SceyfrApidMcoilg n lcwl ulsigLtd, Publishing Blackwell and Microbiology Applied for Society 2011 ©

Table 1. Bacterial and archaeal 16S rRNA amplicon diversity analyses of thrombolitic mat types of Highborne Cay, Bahamas.

Bacteria Cyanobacteria Archaea

Black Beige Pink Button Black Beige Pink Button Black Beige Pink Button

Sequences 2662 1839 1801 1665 7851 3398 4147 8702 574 1258 293 237 Equalized seqa 1248 1248 1248 1248 2548 2548 2548 2548 178 178 178 178 OTUsb 652 552 492 540 427 317 277 593 19 20 10 19 Singletsb 348 295 277 321 220 156 144 292 7 7 5 5 Doubletsb 11593708165534495310 0 OTUs Ն 1% 17 13 15 13 13 15 8 14 7 5 5 14 (Sum)b,c (41%) (22%) (35%) (28%) (77%) (68%) (78%) (67%) (96%) (95%) (98%) (98%) Chao1d,e 909 Ϯ 57 883 Ϯ 26 837 Ϯ 49 998 Ϯ 91 401 Ϯ 44 410 Ϯ 53 350 Ϯ 10 479 Ϯ 47 19 Ϯ 421Ϯ 11 11 Ϯ 321Ϯ 7 (conﬁdence) (776/1096) (769/1043) (717/1007) (858/1192) (317/543) (344/519) (287/458) (404/598) (16/39) (15/54) (10/25) (18/31) Shannond 5.17 Ϯ 0.08 5.55 Ϯ 0.03 5.23 Ϯ 0.09 5.47 Ϯ 0.04 3.07 Ϯ 0.05 3.56 Ϯ 0.01 2.72 3.80 Ϯ 0.02 1.38 Ϯ 0.12 1.07 Ϯ 0.08 1.11 Ϯ 0.12 2.21 Ϯ 0.07

niomna Microbiology Environmental (conﬁdence) (0.07) (0.06) (0.07) (0.07) (0.07) (0.07) (0.15) (0.06) (0.10) (0.08) (0.12) (0.15) Evennessd 0.84 Ϯ 0.01 0.90 0.86 Ϯ 0.01 0.88 0.58 0.66 0.52 0.68 0.51 Ϯ 0.07 0.43 Ϯ 0.05 0.50 Ϯ 0.03 0.77 Ϯ 0.04 % coveraged,f 82 Ϯ 0.01 82 Ϯ 0.01 84 Ϯ 0.01 81 Ϯ 0.01 96 95 96 95 98 Ϯ 0.01 98 Ϯ 0.02 98 Ϯ 0.01 99 Ϯ 0.01 seq lengthg 108–124 108–122 108–124 105–120 123–137 123–137 118–136 116–139 200–205 201–209 202–206 202–205 mats thrombolitic modern in diversity Microbial Base pairsh 233 236 238 239 232 233 232 232 250 247 251 260

a. Number of randomized sequences used to generate the diversity measures. b. The values were calculated based on a 97% similarity threshold. c. The number of OTUs that contain sequences that make up at least 1% of the total number of sequences for each library. In parentheses are the per cent of sequences of the total number of sequences within these OTUs. d. Mean values Ϯ standard deviation of diversity measures using three iterations of random sequence selection to standardize libraries.

, e. Values in parentheses represent the lower and upper 95% confidence interval associated with the Chao1 non-parametric estimator. 14 f. Good’s coverage estimator. 82–100 , g. Range of lengths of reads used in OTU generation in base pairs (see Experimental procedures). h. Average length of quality-trimmed reads in base pairs. 85 86 J. M. Mobberley, M. C. Ortega and J. S. Foster compared reads (Table 1). The libraries were then clus- black mats contained several unique Pleurocapsales that tered into OTUs at 97% similarity or greater. The throm- were not found in other mat types. The beige mats con- bolitic bacterial sequence libraries were composed of tained several unique Desulfobacteraceae that have been sequences from 13 different phyla with extensive variation previously associated with sulfate reduction (Miyatake in the proportional distributions of these phyla between et al., 2009). In the button library there were unique mat types (Fig. 2A and D). Proteobacteria dominated all sequences that shared similarity to Saprospiraceae mat types and comprised 45–58% of the total sequences (Bacteroidetes), members of which have been shown to from each library (Fig. S1A). Within the Proteobacteria, degrade complex organic compounds (Hosoya et al., Alphaproteobacteria were by far the most dominant class 2006; Xia et al., 2008). comprising more than 75% of the recovered Proteobac- The clustering analyses were complemented by using teria reads in the beige, pink and button mats and 62% of Fast UniFrac (Hamady et al., 2010) to compare the bac- the black mats (Fig. S1B). Bayesian analyses at a 70% terial community structure in the four mat types (Fig. 4). confidence interval (CI) indicated that 17–30% of the Since the bacterial sequences represent the V2 partial recovered Alphaproteobacteria sequences could not be 16S rRNA gene region, a guide tree was constructed in assigned to any described order. The remaining FastTree using only these sequences. A principal coordi- sequences shared sequence similarity (95–99%) to nate analysis plot generated from weighted and normal- purple non-sulfur phototrophs, specifically the Rhizobiales ized Fast UniFrac data revealed a clustering pattern of the and Rhodobacterales orders (Imhoff, 2008). Deltaproteo- four mat bacterial populations (black symbols) similar to bacterial sequences were detected in variable abun- the jackknife analysis (Fig. 2), with the beige and pink dances (6.1–14.5%) within the different mats. The beige mats more similar to each other than the black or button mats had the highest proportion of Deltaproteobacteria mats. To assess whether the two dominant phyla, and were rich in sequences belonging to the order Myxo- Alphaproteobacteria and Cyanobacteria, were signifi- coccales and to metabolically versatile sulfate-reducing cantly influencing community structure, two data subsets bacteria of the order Desulfobacterales (Selenska-Pobell, were created from the original Bacteria pyrosequencing 2002). The deltaproteobacterial sequences in the pink data, one without the Alphaproteobacteria and one and black mat sequence libraries were predominately without the Cyanobacteria. The data set without the Myxococcales. In the button mats most of the recovered Alphaproteobacteria (grey symbols) was shifted to the left Deltaproteobacteria sequences could not be reliably clas- with respect to the original communities with the greatest sified at deeper taxonomic levels. There were also differ- differences detected in the button mat type. The data sets ences in the relative abundance of Gammaproteobacteria with only the Cyanobacteria removed (white symbols) populations between mat types, with the black (27%) and resulted in a smaller shift to the right with very little change the beige (9.8%) mat types showing an enrichment of detected in the beige and button mat types. The differen- Gammaproteobacteria, but no difference in diversity. In all tial clustering of these communities also illustrated how four of the mat types BLAST results of the recovered Gam- the bacterial population structure was influenced by envi- maproteobacteria sequences indicated similarity (96– ronmental variables. Principal Component 1 (P1), which 98%) to methylotrophic bacteria, particularly in the order accounted for 43% of the variation in community distribu- Chromatiales. tion and evenness, was correlated with length or rate of Jackknife environmental clustering using the domain burial events. The black mats, which were recently unbur- Bacteria sequence data indicated that the beige and the ied at the time of collection, are often subjected to repeat pink communities were the most similar and that the black sand burial due to their geographical location in the inter- and button mats were the more distant mat types tidal zone, whereas the button mats represent the most (Fig. 2A). The clustering was complemented by an analy- seaward mat type and undergoes the fewest sediment sis of bacterial richness using the number of shared and burial events. Principal Component 2 (P2), which unique OTUs in the four mat types (Fig. 3). The four accounted for 24% of the variation, was correlated with communities shared 59 OTUs, which accounted for the extent of lithification; the button mats exhibited the 29.5% of the total pyrosequencing reads. Thirty-three of patchy areas of lithification associated with the bundles of these OTUs were classified as Alphaproteobacteria while filamentous cyanobacteria, while the black, beige and the rest were similar to members of Gammaproteobacte- pink mats exhibited uniform lithification of the thrombolitic ria, Deltaproteobacteria, Actinobacteria and Cyanobacte- mats. ria. In addition to the shared OTUs, each mat community contained OTUs that were unique to each mat type. The Cyanobacterial community composition highest number of unique OTUs occurred in the black (388; 60%) and button (311; 58%) mats, with the lowest in Cyanobacteria are key components of microbialitic mat the beige (222; 40%) and pink mat types (210; 42%). The communities (Reid et al., 2000; Baumgartner et al., 2009;

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 Microbial diversity in modern thrombolitic mats 87 A D Mat type Bacterial Phyla/Classes Black Acidobacteria Actinobacteria 99 Bacteroidetes Chloroﬂexi Cyanobacteria Beige Firmicutes 100 Fusobacteria Gemmatimonadetes 100 Planctomycetes Pink Alphaproteobacteria Betaproteobacteria Deltaproteobacteria Gammaproteobacteria Button Spirochaetes WS3 & TM6 0.05 unclassified bacteria 0% 20% 40% 60% 80% 100% Percent of pyrosequencing reads

B E Mat type Cyanobacterial Orders Black Chroococcales Nostocales Oscillatoriales Pleurocapsales unclassified cyanobacteria 100 Beige chloroplasts 100 non-cyanobacteria unclassified bacteria

100 Button

Pink 0.05 0% 20% 40% 60% 80% 100% Percent of pyrosequencing reads C F Mat type Archaeal Orders Black Cenarchaeales Halobacteriales 100 Thermoplasmatales unclassified Crenarchaeota unclassified Euryarchaeota 99 Beige unclassified bacteria unclassified

100 Pink

Button 0.05 0% 20% 40% 60% 80% 100% Percent of pyrosequencing reads

Fig. 2. Comparisons of thrombolitic mats using both clustering-based and taxonomic diversity approaches. Dendrograms show UniFrac-based jackknife environmental clustering of the four thrombolitic mat types for bacterial (A); cyanobacterial (B); and (C) archaeal populations. The per cent of bootstrap replicates (n = 1000) are indicated at the nodes. The histograms (D–F) represent the relative abundance as the per cent of the total sequences classified into a specific taxon by a Bayesian approach at a confidence interval Ն 70% against the SILVA SSU Ref v102 database with bacterial libraries (D) classified at the phyla level (class level for Proteobacteria lineages), cyanobacterial libraries (E) classified at the order level and archaeal libraries classified at the order level. ‘Unclassified’ includes sequences that could not be assigned a domain.

similarity from the cyanobacterial libraries (n = 1614 OTU, Black Button 24 098 sequences; Fig. 5). The cyanobacterial 388 311 (60%) (58%) sequences formed 16 distinct clusters, 11 of which were 22 previously described (Myshrall et al., 2010). Six of the 75 37 clusters contained sequences that were unique to the Beige Pink Highborne Cay thrombolitic mats (Fig. 5; black-filled) and 222 13 14 210 (40%) (42%) were associated with the orders Chroococcales, Oscilla- 59 toriales and Nostocales. Nearly all of the Chroococcales 42 31 sequences recovered from the pink mats formed Cluster 42 50 6, which shared similarity (99%) to the metabolically versatile Synechocystis sp. PCC6803 (Anderson and McIn- 49 tosh, 1991). Sequences forming Clusters 7, 10, 11, 12 and 14 were enriched in the button mats, including the two Fig. 3. Venn diagram representation of the OTU richness shared distinct Dichothrix genotypes within the order Nostocales. between bacterial 16S rRNA gene libraries from the four Most of the sequences in Cluster 11 came from cloned thrombolitic mat types. Total observed richness was 1603 OTUs at 97% similarity. Percentages reflect those OTUs unique to that mat 16S rRNA genes from isolated Dichothrix spp. filaments, type. whereas Cluster 12 contained the majority of the recovered 454 pyrosequencing reads. Foster et al., 2009; Foster and Green, 2011). To examine Eight of the cyanobacterial clusters also contained envi- and compare the cyanobacterial populations an amplicon ronmental clones from both Highborne Cay and Shark library was generated to variable region 4 (V4; Nübel Bay stromatolitic mats (Fig. 5 grey-filled; Papineau et al., et al., 1997). The relative abundance of cyanobacterial 2005; Baumgartner et al., 2009; Foster et al., 2009; Goh orders characterized by the V4 region of the 16S rRNA et al., 2009). Sequences of Cyanobacteria from Cluster 1 gene showed that the four thrombolitic mat communities were highly abundant in the black mats (21%) and contain contained complex and diverse populations of Cyanobac- the Pleurocapsa spp. clones that are responsible for the teria (Fig. 2B). As was detected with in the bacterial librar- black surface pigmentation. Cluster 4 shared sequence ies, the black mats were enriched with sequences with similarity to the non-heterocystous, coccoid order Pleuro- 0.3 capsales. The relative abundance of the Pleurocapsales in the four mat types decreased in the more seaward mat types (i.e. pink and buttons). In the beige and the button 0.2 Bl:No-Cy mats there was an enrichment of sequences with similar- Bl P:No-Cy ity to the filamentous, non-heterocystous Oscillatoriales. 0.1 P In both mat types around 70% of the sequences were Bg:No-Cy classified as belonging to Halomicronema, which are Bl:No-Ap Bg common constituents in many hypersaline cyanobacterial 0 Bu:No-Cy mats (Abed et al., 2002; Green et al., 2008; Allen et al., P:No-Ap Bu 2009). There were also unclassified Oscillatoriales -0.1 ecotypes in the beige (22%) and button (23%) that could lithification increasing Bg:No-Ap not be assigned to a class using a Bayesian approach

P2- 24.31% of variation explained -0.2 with a 70% CI. The most seaward button mats exhibited an enrichment of Nostocales sequences that accounted Bu:No-Ap for 9% of the total sequences in the button mats and -0.3 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 shared similarity (97%) to the Calothrix sp. PCC7507 and P1- 43.42% of variation explained Rivularia sp. PCC7116 (Fig. 5). Only a few sequences (< 0.1%) of the Nostocales sequences were found in the decreasing sediment burial other three mat types. Jackknife clustering analysis of the Fig. 4. Principal coordinate analysis of bacterial 16S rRNA gene cyanobacterial libraries at the sequence level revealed a libraries for the four thrombolitic mats using Fast UniFrac. The different clustering pattern compared with the Bacteria figure shows a plot of the first two principal coordinate axis, which represents 43.42% (P1) and 24.31% (P2) of the variations. Each domain-specific library (Fig. 2B) with the black mats being symbol represents bacterial sequences from each mat library: the most distinctive mat type. circles (Bl, black mats), triangles (Bg, beige mats), squares (P, pink The thrombolitic mat cyanobacterial diversity was mats), diamond (Bu, button mats). The shading of each symbol refers to the sequences that are present in each library: black further explored by constructing a phylogenetic tree with (complete library); grey (No-Ap, Alphaproteobacteria removed); representative sequences from each OTU formed at 97% white (No-Cy, Cyanobacteria removed).

Cluster 1 (Bl:5522, Bg:590, P:275, Bu:461) Pleurocapsa clones (4) Pleurocapsa sp. PCC 7314, AB074511 P Cluster 2 (Bl:193, Bg:150, P:118, Bu:81) Chroococcidiopsis sp. PCC 6712, AJ34457 Pleurocapsales Cluster 3 (Bl:165, Bg:29, P:20, Bu:309) * Xenococcus sp. PCC 7305, AF132931

* Cluster 4 (Bl:189, Bg:1172, P:1233, Bu:3204)

Stanieria cyanosphaera PCC 7437, AF132931 Cyanothece sp. PCC 7424, AJ000715 Cluster 5 (Bl:143, Bg:46, P:10, Bu:73) Gloeocapsa sp. PCC 73106, AB039000 Cluster 6 (Bl:164, Bg:122, P:1167, Bu:37) * Synechocystis sp. PCC 6803, AY224195 Chroococcales Snowella litoralis, AJ781040 Cluster 7 (Bl:3, Bg:1, P:2, Bu:116) Cluster 8 (Bl:141, Bg:103, P:13, Bu:122) * Halothece sp. MP1 96P605, AJ000724

Cluster 9 (Bl:11, Bg:37, P:28, Bu:407) * Phormidium sp. HBC9, EU249125 Oscilliatoriales Cluster 10 (Bl:2, Bg:2, P:2, Bu:89) Lyngbya aestuarii, AB039013 Geitlerinema sp. black band disease isolate, DQ151461 Cluster 11 (Bl:5, Bg:0, P:0, Bu:28) * Dichothrix clones (10) Nostocales Rivularia sp. PCC 7116, AM230677 Calothrix sp. PCC 7507, AM230678 B Cluster 12 (Bl:35, Bg:18, P:9, Bu:703) environmental clones Ruidera Pool stromatolites

Cluster 13 (Bl:42, Bg:3, P:0, Bu:5) Symploca atlantica PCC 8002, AB039021 Oscilliatoriales Cluster 14 (Bl:25, Bg:36, P:54, Bu:1831) * Leptolyngbya sp. PCC 7375, AB039011

P Cluster 15 (Bl:6, Bg:259, P:34, Bu:392) Cluster 16 (Bl:286, Bg:25, P:21, Bu:58) P Synechococcus sp. PCC 7336, AF448078 Chroococcales environmental clones Guerrero Negro mats environmental clones Ruidera Pool stromatolites

Chloroplast (Bl:542, Bg:576, P:125, Bu:325) 0.10 * Gloeobacter violaceus PCC 7421, AF132790 Haloferax volcanii, DQ915828

Fig. 5. Phylogenetic tree of partial cyanobacterial 16S rRNA gene sequences from pyrosequencing libraries and clones generated from Dichothrix and Pleurocapsa isolates. Clusters containing Highborne Cay (HBC) thrombolitic sequences were assigned numbers 1–16. Clusters that also contained clone sequences from Myshrall and colleagues’ (2010) study were denoted with a P if they contained sequences associated with pink mats, andaBifthey contained clones from button mats. If cluster contained clones from both pink and button mats the cluster was labelled with asterisks (*). Clusters were also labelled based on the environments the sequences were recovered from: grey-filled, HBC thrombolitic mats and other microbialitic mat environments; black-filled, specific to HBC thrombolitic mats; stripe-filled, HBC thrombolitic mats, microbialitic mats and the hypersaline mats from Guerrero Negro; white-filled clusters contained no HBC thrombolitic sequences. For each cluster the number of sequences recovered from each mat type are enclosed in parentheses (black, Bl; beige, Bg; pink, P; button, Bu). The scale bar represents 0.10 substitutions per nucleotide position.

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 90 J. M. Mobberley, M. C. Ortega and J. S. Foster similarity to Stanieria cyanosphaera and contained 2009; Goh et al., 2009; Robertson et al., 2009). Two of the between 31% and 40% of the sequences from the pink, largest clusters unique to the thrombolites, Cluster 7 and and button mats. This cluster also contained sequences 9, shared similarity to the sponge isolate Cenarchaeum that were abundant in the Highborne Cay Type 2 and 3 symbiosum. This group of sequences clustered sepa- stromatolite clone libraries (Baumgartner et al., 2009; rately from the Nitrosopumilus-like clones found in the Foster et al., 2009). Cluster 14, which contained 22% of Shark Bay stromatolites in Cluster 8 (Preston et al., 1996; the sequences from the button mats, also included Burns et al., 2004). The thrombolitic mat Cluster 3 con- Leptolyngbya-related clones from the adjacent stromato- tained Halobacteria-like sequences that were closely lites (Foster and Green, 2011) and the freshwater Ruidera related to two known halophilic isolates, Haloarcula Pool stromatolites from Spain (Santos et al., 2010). japonica and Natronomonas pharaonis (Fig. 6). No sequences derived from known methanogens were detected in the archaeal sequence libraries. Archaea community structure

The archaeal community was examined by generating an Discussion amplicon library that targeted the V5 region of the 16S rRNA gene using Archaea-specific primers (Delong, There has been extensive research on the microbialites of 1992; Barns et al., 1994). We recovered 2362 quality Highborne Cay, Bahamas; however, most of this research sequences with a Good’s estimated coverage of between has been focused on the laminated stromatolitic microbi- 98% and 99% in all four mat types. The recovered number alites (e.g. Pinckney and Reid, 1997; Reid et al., 2000; of observed OTUs was low and ranged between 5 in the Visscher et al., 2000; Dupraz and Visscher, 2005; Baum- pink and beige mats to 14 in the black mats (Table 1). The gartner et al., 2009; Foster et al., 2009). Only a handful of estimated diversity based on Chao1 values was also low studies have examined the unlaminated clotted thromb- and ranged between 11 in the pink mats and 21 in the olites of Highborne Cay and these studies have targeted beige and button mats. Despite the low number of OTUs, aspects of carbonate deposition (Planavsky et al., 2009), the diversity indices of the four of the thrombolitic mats phage diversity (Desnues et al., 2008) and geomicrobiol- were comparable to those reported in the Shark Bay ogy (Myshrall et al., 2010) within the structures. The work stromatolites (Goh et al., 2009) but less diverse than the by Myshrall and colleagues (2010) also provided the first archaea populations reported from non-lithifying microbial insight into the microbial ecology of the thrombolitic struc- communities such as the Guerrero Negro mats (Robert- tures at Highborne Cay; however, this study provided only son et al., 2009). The recovered archaeal sequences a limited coverage of the microbial communities associ- were classified to the order level and compared between ated with the thrombolites. In this study we expanded on mat types (Fig. 2C and F). Sequences similar to Cenar- this previous work by characterizing two additional throm- chaeales, belonging to the Thaumarchaeota phylum bolitic mat types and examining the archaeal diversity (Brochier-Armanet et al., 2008), were dominant in the in within all four mats, thus providing a more comprehensive the black and beige mats and decreased in abundance in and comparative sequencing analysis of the thrombolitic the more seaward pink and button mats. The Halobacte- mat communities. The results of our study provide mor- riales of the phylum Euryarchaeota dominated the pink phological and genetic evidence that there are at least mat types, but were present at lower levels in the black, four thrombolitic mat types at Highborne Cay and that beige and button mats. The highest level of diversity was these four mat types may represent successive stages of detected in the button mat types, where the Thermoplas- thrombolitic mat development. matales (50%) and unclassified Euryarchaeota (10%) sequences made up most of the button mat archaeal Differences in microbial populations between community. Hierarchical clustering revealed a different thrombolitic mat types relationship between the four mat communities (Fig. 2B). Unlike the clustering in the bacterial populations, the The pyrosequencing of 16S rRNA gene amplicon libraries archaeal populations of the near shore black and beige generated to each of the four thrombolitic mat types dra- mats were more similar to each other than those found in matically increased the sequencing coverage of the bac- the more seaward pink and button mats. terial (80–84%), cyanobacterial (95–96%) and archaeal A phylogenetic tree constructed with archaeal (98–99%) communities (Table 1) compared with the pre- sequences revealed 10 distinct clusters (Fig. 6; black- vious study by Myshrall and colleagues (2010), which filled) that were specific to the Highborne Cay thromb- relied on traditional clone libraries. The pyrosequencing olites. The closest relatives to these clusters included both results indicate the thrombolitic mat communities are sig- lithifying (Fig. 6; grey-filled) and non-lithifying mat (Fig. 6; nificantly more diverse than previously measured with an white-filled) communities (Jahnke et al., 2008; Allen et al., estimated 779–1152 bacterial OTUs in each of the mat

environmental Euryarchaeota Clade 5 clones Guerrero Negro mats

environmental Euryarchaeota Clade 4 clones Guerrero Negro mats

Cluster 1 (Bl:0, Bg:9, P:0, Bu:24) environmental Crenarchaeota clones Guerrero Negro mats Cluster 2 (Bl:1, Bg:0, P:1, Bu:17) unidentified archaeon PMC-2A209, AB019720 environmental Crenarchaeota clones nonlithifying Shark Bay mats Cluster 3 (Bl:0, Bg:0, P:0, Bu:9) environmental Euryarchaeota Clade 6 clones Guerrero Negro mats

environmental Halobacteria clones nonlithifying Shark Bay mats

Haloarcula japonica, AB355986 environmental clones lithifying Shark Bay stromatolites Halococcus hamelinensis, AB301068

environmental clones lithifying Shark Bay stromatolites

Haloferax volcanii, AY45257241 Cluster 4 (Bl:105, Bg:205, P:171, Bu:13)

environmental methanogen clones Guerrero Negro mats Methanogenium boonei, DQ177343 environmental clones nonlithifying Shark Bay mats

environmental Thermoplasmata clones Guerrero Negro & Shark Bay mats

Cluster 5 (Bl:0, Bg:0, P:0, Bu:81) unidentified archaeon PMC-2A24, AB019736 Cluster 6 (Bl:4, Bg:0, P:0, Bu:45) Thermoplasma volcanium, AJ299215 Aciduliprofundum boonei, DQ177343 Nanoarchaeum equitans, AJ318041 Cluster 7 (Bl:22, Bg:48, P:1, Bu:3) Cenarchaeum symbiosum, U51469 environmental clones lithifying Shark Bay stromatolites Nitrosopumilus maritimus SCM1, NC_010085 Cluster 8 (Bl:1, Bg:0, P:0, Bu:19) Cluster 9 (Bl:441, Bg:996, P:109, Bu:22) environmental Crenarchaeota clones Guerrero Negro & Shark Bay mats

0.10 Cluster 10 (Bl:0, Bg:0, P:11, Bu:4) Staphylothermus marinus, X99560 uncultured Korarchaeote, DQ465908 Escherichia coli K12, NC_000913

Fig. 6. Phylogenetic tree of partial archaeal 16S rRNA gene sequences from pyrosequencing libraries. The tree was constructed in ARB by inserting sequences, using a parsimony method with archaea-specific position variable filters, into a tree containing 16S rRNA gene sequences from SILVA SSU Ref v102 database. Clusters containing Highborne Cay (HBC) thrombolitic sequences are were assigned numbers 1–10. Clusters were labelled based on the environments the sequences were recovered from: white-filled, non-lithifying hypersaline mats from Guerro Negro and Shark Bay; black-filled, HBC thrombolitic mats; grey-filled, only lithifying Shark Bay stromatolites. For each cluster the number of HBC thrombolitic sequences recovered from each mat are enclosed in parentheses (black, Bl; beige, Bg; pink, P; button, Bu). The bar represents 0.10 substitutions per nucleotide position.

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 92 J. M. Mobberley, M. C. Ortega and J. S. Foster types and bacterial diversity Shannon indices ranging shown that these endolithic Pleurocapsales ecotypes are between 5.17 and 5.55 (based on equalized sequences; resistant to extensive desiccation and UV radiation Table 1). These values are higher than any previously (Krumbein and Giele, 1979; Billi et al., 2000; Dillon et al., described microbialitic mat community (Baumgartner 2002). The high relative abundance of these ecotypes in et al., 2009; Goh et al., 2009; Myshrall et al., 2010; Foster the black and beige mats relative to the pink and button and Green, 2011), and likely reflect the 10-fold increase in mats may reflect the near shore location of these mats as sequencing coverage. However, estimates of species these two mats are often fully exposed in the upper inter- richness and evenness can also be influenced by the tidal zone. Not all thrombolite cyanobacteria were unique length of the amplified region, with shorter partial reads to this environment, and some cyanobacterial taxa such having higher diversity estimates than longer 16S rRNA as those in Cluster 14 (Fig. 5) are detected in other micro- gene reads (Engelbrektson et al., 2010; Harris et al., bialite ecosystems. For example Cluster 14 contained 2010). These discrepancies could partially explain the sequences similar to the genus Leptolyngbya, with most higher diversity of the thrombolitic pyrosequencing librar- sequences recovered from the button mat types. Lep- ies relative to the clone libraries from microbialitic mats tolyngbya are non-heterocystous filamentous cyanobac- (Myshrall et al., 2010; Table 1). To prevent such an over- teria that have been found in all modern stromatolitic mat estimation of microbial diversity within the four thromb- systems to date, including both marine and freshwater olitic mats we used a pairwise alignment and similarity environments (Foster et al., 2009; Goh et al., 2009; threshold of 97% over the same exact read length as well Santos et al., 2010) and may represent a common micro- as a pre-clustering approach to account for potential bialitic mat building organism. sequencing errors (Huse et al., 2010). These approaches Although there were several clusters recovered from the have been previously shown to be effective cut-offs to thrombolitic mats that appear to be common to all micro- reduce overestimations of microbial diversity (Huse et al., bialitic mats there were six clusters that were unique to the 2007; Kunin et al., 2010; Schloss, 2010). We imple- Highborne Cay thrombolitic mats and may contribute to the mented these approaches using the mothur software differences in clotted thrombolitic microfabrics compared (Schloss et al., 2009) with the stringent quality control with laminated stromatolites. The most notable of these six trimming resulting in 28–42% of our raw pyrosequences clusters were Clusters 11 and 12 that were enriched in the discarded during pre-processing, thereby minimizing most button mat types. These two clusters were classified as the of the pyrosequencing errors and potential sources for order Nostocales and shared sequence similarity to the diversity overestimation. heterocystous, filamentous Calothrix sp. and Rivularia sp. (Rippka et al., 2001). Cluster 11 contained all of the cloned Dichothrix spp. 16S rRNA genes that were collected from Cyanobacterial diversity in thrombolitic mat communities the button mat type along with relatively few of the recov- Cyanobacteria have long been known to be a dominant ered pyrosequences, which may suggest some amplifica- functional group in microbial mat ecosystems (Canfield tion or cloning bias for this ecotype. All of the sequences and DesMarais, 1993; Reid et al., 2000; Ley et al., 2006; associated with Cluster 12 were recovered from the Foster and Green, 2011) and are critical for the accretion pyrosequencing results and were also similar to the fresh- and early lithification of stromatolitic mats (Pinckney and water Calothrix sp. and to a cluster of environmental clones Reid, 1997; Dupraz et al., 2004). To provide a deeper isolated from freshwater microbialites in Ruidera Pools in survey of the Cyanobacteria within the thrombolitic mats Spain (Santos et al., 2010). These results suggest that and overcome potential biases of using a single primer there may be two distinct populations of Dichothrix within set, cyanobacterial-specific primers that targeted the V4 the thrombolitic mats. However, the results may be the region (Nübel et al., 1997) were used in addition to uni- product of heterogeneity in the 16S rRNA gene. Extensive versal bacterial primer (Table S1). The cyanobacterial heterogeneity of the 16S rRNA gene is prevalent in strains populations in the four mats varied significantly in their of Calothrix (Berrendero et al., 2008) and previous studies order level distribution with coccoid, non-heterocystous have shown that a single morphotype of Calothrix in cyano- ecotypes abundant in the black, beige and pink mats, and bacterial mats comprised five different 16S rRNA geno- filamentous ecotypes found in the button mats (Fig. 2B). types (Hongmei et al., 2005). Further work on the There were, however, eight cyanobacterial clusters Dichothrix isolates is required to resolve this issue. detected in the thrombolitic mats that appear to be specific to microbialitic mat communities (Clusters 1–5, 9, 13, Low diversity of Archaea in Highborne Cay 14; Fig. 5). One of these clusters, Cluster 1, contained thrombolitic mats sequences similar to the non-heterocystous coccoid Pleu- rocapsales and was found at a relative abundance of at The 16S rRNA gene libraries generated in this study sug- least 10% in all four mat types. Previous studies have gested that Archaea diversity was low and accounted for

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 Microbial diversity in modern thrombolitic mats 93 only 1.7% of recovered OTUs. The observed archaeal methanogens for substrates, such as H2 and acetate diversity indices were the lowest in the beige and pink (Ehrlich, 2002; Schimel, 2004). Despite the lack of metha- mats (1.07 and 1.11 respectively) with the highest asso- nogenic archaea recovered from the thrombolitic mats, ciated with the button mats (2.21). The results are similar methanotrophic bacteria were detected in the button to community analyses of other microbialitic mats where thrombolitic mats (e.g. Nitrosococcus sp. and Thioflavic- Archaea diversity was also low (Papineau et al., 2005; occus sp.). Methane may be generated in the mats by Goh et al., 2009). All 10 of the clusters contained microbes using non-competitive osmoprotectants such as sequences associated with only the thrombolitic mats glycine betaine, choline and trehalose. Genes encoding (Fig. 6). However, as no comprehensive archaeal diver- these putative substrates have been found in the throm- sity analysis has been conducted for the adjacent stroma- bolite metagenome (J.S. Foster, unpublished). Methane tolites at Highborne Cay it is not clear whether these production from the fermentation of complex amines clusters are specific to the thrombolitic mats or to the including osmoprotectants has been shown to occur in a geographical location. The clusters were closely related to wide range of habitats including microbial mats and sedi- other lithifying and non-lithifying microbial mat ecosys- ments (King, 1984; Oh et al., 2008). tems such as Shark Bay stromatolites (Papineau et al., 2005; Goh et al., 2009) and the hypersaline mats of Guer- A model for microbial succession in rero Negro (Jahnke et al., 2008; Robertson et al., 2009). thrombolitic mat types The majority of the archaeal sequences from the black and beige mats (Clusters 7, 9) were related to a sponge Succession in microbial communities has been well docu- symbiont, C. symbiosum (Hallam et al., 2006), and shared mented in microbial mat systems (Stal et al., 1985; Stolz, similarity to a proposed mesophilic phylum, Thaumarcha- 1990; Grootjans et al., 1997) and is largely influenced by eota (Brochier-Armanet et al., 2008). The Cenarchaeum- the changes in the environment (Mackie et al., 1999; like cluster was distinct, however, from Thaumarchaeota Haruta et al., 2004; Jones and Lennon, 2010; Sato et al., found in the Shark Bay stromatolites (Goh et al., 2009), 2010). Based on the diversity analyses, we propose that which shared similarity to chemolithoautotrophic the four thrombolitic mat types represent successive ammonia-oxidizing Nitrosopumilus maritimus (Könneke stages of thrombolite development in Highborne Cay, et al., 2005). Both the C. symbiosum and N. maritimus Bahamas (Fig. 7). Although field observations of the four genomes have been shown to contain similar ammonia distinct thrombolitic mat communities have consistently oxidation and carbon fixation genes (Walker et al., 2010). showed that button mats were superposed over the pink These archaea could be contributing to nitrification and and beige mats, the 16S rRNA gene analysis supported carbon cycling in the thrombolitic mats. these patterns of thrombolitic mat transitions. No archaeal ecotypes associated with methanogenesis The jackknife environmental clustering of the bacterial were recovered from the thrombolitic mats nor have they and cyanobacterial libraries and Fast Unifrac results been recovered from other microbialites such as the stro- indicated that the black and the button mats were the matolites of Shark Bay (Papineau et al., 2005; Goh et al., most distinctive mat communities (Figs 2 and 5). We 2009). Although methanogenesis has been reported for propose that the black mats, which are the most beach- other microbial mat systems such as the hypersaline mats ward mats, represent an initial surface community after from Guerrero Negro (Smith et al., 2008; Robertson et al., thrombolites emerge from extended sand burial events. 2009) and Solar Lake (Giani et al., 1984) methanogenesis The duration of burial for the microbialites of Highborne is likely inhibited in the thrombolitic mats due to the pres- Cay can vary from a few days to several weeks (Andres ence of sulfate-reducing bacteria, which outcompete and Reid, 2006) and the geographical location of the

Fig. 7. Model for succession of Highborne Cay thrombolitic mat communities. ABButton A. Spatial distribution of thrombolitic mat Community types in situ: black (bl), beige (bg), pink (p) bu and button (bu). Bar, 0.25 m. bu B. Representation of the potential succession Burial p Pink Beige of thrombolitic mats. Black arrows indicate bg Community Community pathways associated with the gradual succession of thrombolitic mat types. Grey arrows indicate pathway that may occur in bl response to a sediment burial event. Black Community

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 94 J. M. Mobberley, M. C. Ortega and J. S. Foster black mats in the near shore intertidal zone render these that of the adjacent stromatolites (Myshrall et al., 2010). communities more susceptible to longer periods of The button mats also contained the highest cyanobacte- burial, low light and desiccation. These conditions may rial and archaeal diversity compared with the other three impose a selection pressure on maintaining the thrombolitic mats (Table 1; Fig. 2). It is possible that this Pleurocapsales-dominant community in the black mats taxonomic diversity corresponds to greater metabolic as Pleurocapsales have been found to be resistant to diversity, which could influence biologically controlled min- low light conditions (Krumbein et al., 1979; Kremer and eralization in the thrombolites (Dupraz and Visscher, Kazmierczak, 2005) and desiccation (Krumbein and 2005). The presence of calcium carbonate precipitation Giele, 1979). We propose that upon environmental dis- on the sheaths of Dichothrix cells is unique to the button turbances, such as extensive sediment burial, thromb- thrombolitic mats (Planavsky et al., 2009), and the 16S olitic mat communities shift to the black mat state rRNA genes of Dichothrix have not been recovered from (Fig. 7) and that many of the ecotypes typically enriched the adjacent stromatolites (Foster and Green, 2011). The in other mat types either die or enter a dormant state. transition towards the button mat type likely reflects mul- Dormancy has been shown to be widespread in micro- tiple environmental cues and may include factors such as bial ecosystems accounting for up to 80% of microbial eukaryotic grazing and shifts in pH. Although eukaryotic populations (Cole, 1999; Càceres and Tessier, 2003) diversity has been shown to be low in the button thromb- and may facilitate an organism’s ability to survive exten- olitic mats numerous sequences with similarity to sive environmental perturbations (Jones and Lennon, ecotypes associated with bacterivorous grazing (e.g. 2010). nematodes) have been recovered (Myshrall et al., 2010). Once the mats become unburied due to high-energy The prevalence of filamentous cyanobacteria and bacte- wave action, we propose the black communities experi- ria within the button mats may provide resistance to ence an environmental trigger that results in the transition grazing activity, as previous studies have shown that to either beige or pink mat communities (black arrows, meiofaunal grazers selectively target smaller coccoid or Fig. 7B). The beta-diversity measures such as the jack- rod-shaped ecotypes (Jürgens et al., 1999). knife environmental clustering (Fig. 2A–C) and Fast The transitions between mat types may also reflect a Unifrac analyses (Fig. 4) indicated that the bacterial popu- shift in the pH, which may result in the differences in the lations of beige and pink mats were more similar to each patterns of biologically induced biomineralization. Shifts in other than to either the black or button mats. Even when pH have been shown to trigger succession events in the two dominant taxa (Alphaproteobacteria and Cyano- complex microbial communities (Haruta et al., 2004) and bacteria) were removed from the pyrosequencing data set in the button mats there are numerous ecotypes associ- the beige and pink communities were more similar to each ated with metabolisms that influence carbonate alkalinity, other, suggesting that these two mat types are closely such as oxygenic, anoxygenic photosynthesis and sulfate related (Fig. 4). The environmental factors that may ini- reduction (Dupraz et al., 2009). Together these metabo- tiate the transition to beige and pink may the result of light lisms can initiate and promote carbonate precipitation and nutrient availability. Previous studies have shown that (Dupraz et al., 2009); however, more work is necessary to light availability can have a strong impact on the compo- monitor the specific changes in pH between the four mat sition of cyanobacterial communities (Havens et al., types to assess the impact of pH on each of the thromb- 1998). As photosynthetic mat communities experience olitic mat types. increases in light availability previous studies have shown there is an increase in uptake of dissolved organic carbon Experimental procedures (Yannarell and Paerl, 2007). The rate of uptake is influenced by the type and composition of the cyanobacterial Site description, sample collection and microscopy communities (e.g. filamentous, colony coccoid or solitary Thrombolitic mat samples were collected from the island of coccoid; Yannarell and Paerl, 2007). The transition to the Highborne Cay located in the northern Exumas, Bahamas coccoid rich communities of the beige (enriched in colony- (24°43′N, 76°49′W). The mat samples were taken from a forming Pleurocapsales) or the pink (enriched in solitary series of intertidal thrombolites located at Site 6, as desig- Chroococcales) mats may reflect seasonal differences in nated by Andres and Reid (2006), in March 2008. Eight uptake or differences in the substrate-utilization patterns replicate samples (approximately 3 g of thrombolitic mat of the two distinct mat types. material) were collected from each mat type and immediately After a transition to either beige or pink mat types, we immersed in RNALater (Ambion, Austin, TX, USA) to stabi- lize the nucleic acids, and then frozen at -20°C. Frozen propose there is a trend towards the succession of the samples were transported to the Space Life Science Labo- button thrombolitic mat communities. Besides being the ratory and stored at -20°C until processing. A corresponding most abundant mat type, the button mats are the most set of mat samples were taken for immediate microscopic productive (O2 and DIC production) with rates higher than examination in the field. Freshly collected thrombolitic mats

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 Microbial diversity in modern thrombolitic mats 95 were sectioned (0.5 cm) with a rock saw and examined using tamination was detected. The PCR products for both cell an Olympus SX12 stereoscope (Olympus, Center Valley, PA, types were purified with the UltraClean PCR Kit (MoBio, USA). Mat surface morphologies were documented then Carlsbad, CA) and cloned into the pCR 2.1 vector using the cross-sections were immersed in filtered seawater and Topo TA Cloning Kit (Invitrogen, Carlsbad, CA) following imaged using 10¥,32¥ and 1000¥ objectives. manufacturer’s instructions. Ten clones from each cell type were picked for sequencing with an ABI 3130 DNA sequencer at the University of Florida Interdisciplinary Molecular identification of cyanobacterial isolates from Center for Biotechnology Research (UF-ICBR). Consensus the thrombolitic mats sequences were generated using CLUSTAL W and all clone sequences were submitted to GenBank under Accession To identify the morphologically dominant cyanobacterial iso- No. HQ415794–HQ415799. lates from the button and black mat types, two cells types were dissected from their respective mat types using an Olympus SZX12 stereoscope scope. Bundles of a filamentous cyano- Generation of barcoded 16S rRNA gene libraries bacterium, morphologically identified as Dichothrix spp. (Planavsky et al., 2009), were dissected from the button mats To determine the bacterial, cyanobacterial and archaeal com- and clusters of previously undescribed pigmented coccoid munity composition of the four main thrombolitic mat types cells were dissected from the black mat type. Approximately barcoded 16S rRNA gene libraries were generated and 20 mg of each cell type was dissected and immediately placed sequenced with high-throughput pyrosequencing. DNA was into an Eppendorf tube containing DNAzol ES reagent isolated from three replicate samples for each mat type using (Molecular Research Center, Cincinnati, OH, USA). Each tube previously described methods (Foster et al., 2009). DNA was then underwent three cycles of freeze–thawed in liquid nitro- then PCR-amplified using a fusion 454-primer that included gen followed by grinding with a mortar and pestle. an oligonucleotide tag (i.e. barcode) located on the 3′ end. Due to the presence of non-heterocystous background Each barcode tag was unique for the four mat types and all cyanobacteria the 16S rRNA gene from the Dichothrix spp. sequences are listed in Table S1 (Hamady et al., 2008; cell type was amplified using two unique primer sets that Roesch et al., 2009). were designed based on alignments of full-length 16S rRNA The PCR reactions for the bacterial 16S rRNA library, gene sequences of closely related Calothrix and Rivularia which target the V1–2 region of the gene, contained the species recovered from NCBI and Rivularia-like sequences following final concentrations: 1¥ Clone Pfu Reaction Buffer recovered from the bacterial and cyanobacterial partial 16S (Stratagene, La Jolla, CA), 280 mM dNTPs, 2.5 mg of BSA, rRNA gene clone libraries from Myshrall and colleagues 600 nM each primer, 0.75 ng of genomic mat DNA, 1.25 U of (2010). Primers Dicho-1F (CTGGCTCAGGATGAACGCTG) cloned Pfu DNA Polymerase (Stratagene) and nuclease-free and Dicho-1R (GCCAAACCACCTACGAACGC) produce a water (Sigma, St. Louis, MO) in a volume of 25 ml. The 523 bp amplicon of the V3–5 region. Primers Dicho-2F reactions were held at 95°C for 5 min, followed by 35 cycles (TGGGGTAAAAGCGTACCAAG) and Dicho-2R (CCACGC- of 95°C for 1 min, 64°C for 1 min, 75°C for 3 min and a final CTAGTATCCATCGT) produce a 600 bp amplicon of the extension of 75°C for 7 min. Negative controls were con- V4–6 region. The PCR reactions contained the following final ducted and no contamination was detected. All PCR products concentrations: 5¥ GoTaq Flexi Reaction Buffer (Promega, were combined in equimolar amounts and purified using the Madison, WI), 1.5 mM magnesium chloride, 600 mM dNTPs, PCR Purification kit (Qiagen, Valencia, CA). 400 nM each primer, 20 mg of bovine serum albumin (BSA), The PCR reactions for the barcoded cyanobacterial 16S 10 ng of Dichothrix spp. genomic DNA, 2.5 U of GoTaq Flexi rRNA library used the Nübel primer set, which has been DNA Polymerase (Promega). The PCR conditions were 94°C shown to detect a wider range of cyanobacteria than the for 5 min, followed by 30 cycles of 94°C for 1 min, 58°C for ‘universal’ bacterial primer set used above (Foster et al., 1 min, 72°C for 2 min and a final extension of 72°C for 7 min. 2009; Wang and Qian, 2009). The primers targeted the V3–4 Due to the presence of numerous non-cyanobacterial het- region of the gene contained the following final concentra- erotrophic bacteria in the black cyanobacterial cell dissec- tions: 1¥ High Fidelity PCR Buffer (Invitrogen), 2 mM mag- tions, a nested primer approach was used to amplify the nesium sulfate, 200 mM dNTPs, 2.5 mg of BSA, 400 nM 359F 16S rRNA gene. The first round used 16S universal bacte- primer, 200 nM each of 781Ra and 781Rb primers, 5 ng of rial primers 27F (AGAGTTTGATCCTGGCTCAG) and genomic DNA, 1 U of Platinum Taq High Fidelity (Invitrogen) 1525R (TAAGGAGGTGATCCAGCC; Lane, 1991), and the and nuclease-free water (Sigma) to 25 ml. The PCR condi- second round used cyanobacterial primers, Cya359F, and tions were 94°C for 2 min, 30 cycles of 94°C for 0.5 min, 58°C an equimolar mixture of Cya781Ra and Cya781Rb (Nübel for 1 min, 68°C for 2 min and a final extension of 68°C for et al., 1997). For both rounds, the reagent concentrations 10 min. PCR products were extracted from an agarose gel were the same as in the Dichothrix PCR described above, with the QIAquick Gel Extraction kit (Qiagen). Negative con- except that in the second round 2 ng of amplicon generated trols were conducted for all PCR reactions and results indi- in the first round of PCR was used. The conditions for the cated no contamination. All purified PCR products were first round of PCR included 95°C for 2 min, 30 cycles of combined in equimolar amounts. 95°C for 30 s, 58°C for 2 min, 72°C for 2 min and a final Due to the low numbers of Archaea within the thrombolitic extension of 72°C for 10 min. The conditions for the second communities, a nested PCR approach with two rounds of round of PCR with the cyanobacterial-specific primers were amplification was used for generating the archaeal 16S rRNA the same as described for the Dichothrix PCR. Negative library that targeted the V3–5 region of the 16S rRNA gene. controls were conducted for all PCR reactions and no con- The PCR reaction concentrations were the same as the

© 2011 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 14, 82–100 96 J. M. Mobberley, M. C. Ortega and J. S. Foster cyanobacterial library except the first round used 400 nM each sequences were removed, a pairwise distance matrix was of primers 23F and 958R (Delong, 1992; Barns et al., 1994) calculated using mothur and reads were clustered into OTUs and 10 ng of genomic DNA, while the second round of PCR at 3% distance using the furthest neighbour method (Schloss used 400 nM each of primers 334F and 915R (Casamayor and Handelsman, 2005). SILVA alignments of representative et al., 2002) with 10 ng of round one amplicon as a template. sequences from each OTU for all three libraries were The PCR conditions for both rounds were similar to the cyano- extracted from the mothur generated NAST alignments and bacterial run conditions except the annealing steps were at imported into the ARB software package (Ludwig et al., 55°C for 1.5 min for the first round, and 61°C for 1 min for the 2004). The sequences were inserted into a tree composed of second round. A PCR Purification kit (Qiagen) was used to full-length 16S rRNA sequences from the SILVA SSU Ref clean up the first and second rounds of PCR. Negative controls v102 database using the parsimony option with domain- were conducted in parallel with for each round of PCR and specific position variable filters. The representative sequence results indicated no contamination. All amplicons from the from each OTU was classified using mothur with a naïve second round were combined in equimolar amounts. Bayesian approach at a confidence threshold of 70% (Wang Bacterial, cyanobacterial and archaeal 16S rRNA barcoded et al., 2007). The bacterial and archaeal reference databases libraries were sequenced from the 454 Life Sciences A primer were composed of unique, full-length sequences from the using the standard GS FLX chemistry (454 Life Sciences, SILVA SSU Ref v102 database (http://www.mothur.org/wiki/ Branford, CT) in a single run performed by the UF-ICBR. The Silva_reference_files) and used to classify reads from their raw sequence reads and quality files were deposited into the respective barcoded amplicon libraries. NCBI sequencing read archive under project number To account for the effects of different sampling depths on SRP004035. the alpha-diversity measurements three replicate subsets from each thrombolitic mat type, containing an equal number of sequences, were created for each library using a Bioinformatic analysis of barcoded 16S ribosomal random sequence selector Perl script (Paul Stothard; http:// RNA gene libraries www.ualberta.ca/~stothard/software.html). The number of randomly subsampled sequences was calculated to capture The open-source software mothur v 1.11.0 was used for 75% of the sequences in the least represented mat type processing, cluster analysis and classification of the raw bar- (i.e. 1248 for Bacteria, 2548 for Cyanobacteria and 178 for coded sequences from the 16S rRNA gene libraries (Schloss Archaea). The subsampled OTUs from each mat type were et al., 2009). The bacterial, cyanobacterial and archaeal used to generate alpha-diversity statistics including Chao1 libraries were analysed separately since they covered differ- non-parametric species richness estimate, Shannon indices, ent variable regions of the 16S rRNA gene. Those sequence Shannon-based richness estimate and Good’s coverage reads that did not contain an exact match to the primer estimate. sequences, contained any ambiguous reads and/or had an average quality score of less than 27 were considered to be poor quality and were excluded from the rest of the analysis Community comparisons between thrombolitic mat types (Kunin et al., 2010). The remaining 454 reads were subjected to two additional screening criteria. First, based on the finding Community analysis and comparison of the thrombolitic mat by Huse and colleagues (2007) that short 454 reads may be types were performed phylogenetically using Unifrac jack- poor quality, all reads below 125 bp for the bacterial and knifed environmental clustering and Fast UniFrac, a web- 200 bp for the cyanobacteria and archaeal libraries were based program designed for the comparing distances removed. Second, based on limitations of the GS-FLX stan- between communities in large data sets (Lozupone and dard chemistry reads longer than 275 bp for bacterial and Knight, 2005; Hamady et al., 2010). For the jackknifed clus- 300 for cyanobacterial and archaeal sequences were tering analysis 1000 permutations were run with a sampling removed. The corresponding mat type for each retained bar- size of 75% of the total number of sequences in the smallest coded read was recorded and the primers were trimmed from library (1249 bacteria; 2349 cyanobacteria; 178 archaea). the sequences. Since sequencing was done from the reverse The Fast Unifrac analyses were used to determine whether primer, the reverse complements were taken and exact rep- Alphaproteobacteria and Cyanobacteria contributed to com- licate reads were removed to speed downstream analyses. munity structure. Reads classified as belonging to these The remaining sequences were aligned in mothur using the phyla were removed from the bacterial 16S rRNA gene library nearest alignment space termination (NAST) algorithm to create two artificial libraries: without Alphaproteobacteria against a bacterial and archaeal SILVA 16S rRNA gene tem- (No-Ap) and without Cyanobacteria (No-Cy). The No-Ap and plate, as this template has been shown to provide quality No-Cy libraries were combined with the original bacterial alignments of variable regions (DeSantis et al., 2006; library and aligned within mothur. Phylogenetic trees for the Pruesse et al., 2007; Schloss, 2010). No sequence mask was bacterial (original, No-Ap and No-Cy), cyanobacterial and used in this analysis as it has been previously shown to archaeal libraries for UniFrac analysis were constructed from significantly reduce the diversity observed between the NAST alignments generated above using FastTree, a sequences (Schloss, 2010). The alignment was trimmed relaxed maximum-likelihood tree-building program (Price such that all reads were aligned to the same exact region. To et al., 2010). Tab-delimited sample ID mapping files were account for potential pyrosequencing errors the sequences generated for each library with data collected during mothur from each of the three libraries were analysed with the analysis. Principal coordinate analysis for the three libraries pre.cluster function in mothur, which is based on the SLP was performed on weighted and normalized data within Fast clustering algorithm (Huse et al., 2010). Once error-prone UniFrac (Hamady et al., 2010).

Acknowledgements Canfield, D.E., and Des Marais, D.J. (1993) Biogeochemical cycles of carbon, sulfur, and free oxygen in a microbial mat. The authors would like to thank Steven Ahrendt for his tech- Geochim Cosmochim Acta 57: 3971–3984. nical assistance and Stefan Green for his comments regard- Casamayor, E.O., Massana, R., Benlloch, S., Ovreas, L., ing the manuscript. This work was supported by the NASA: Diez, B., Goddard, V.J., et al. (2002) Changes in archaeal, Exobiology and Evolutionary Biology Program Element bacterial and eukaryal assemblages along a salinity gradi- (NNX09AO57G) awarded to J.S.F., and J.M.M. was sup- ent by comparison of genetic fingerprinting methods in a ported by a NASA Graduate Student Research Program fel- multipond solar saltern. Environ Microbiol 4: 338–348. lowship (NNX10AO18H). Cole, J.J. (1999) Aquatic microbiology for ecosystem scien- tists: new and recycled paradigms in ecological microbiology. Ecosystems 2: 215–225. References DeLong, E.F. (1992) Archaea in coastal marine environments. Proc Natl Acad Sci USA 89: 5685–5689. Abed, R.M.M., Garcia-Pichel, F., and Hernandez-Marine, M. DeSantis, T.Z., Jr, Hugenholtz, P., Keller, K., Brodie, E.L., (2002) Polyphasic characterization of benthic, moderately Larsen, N., Piceno, Y.M., et al. (2006) NAST: a multiple halophilic, moderately thermophilic cyanobacteria with sequence alignment server for comparative analysis of very thin trichomes and the proposal of Halomicronema 16S rRNA genes. Nucleic Acids Res 34: W394–W399. excentricum gen. nov., sp. nov. Arch Microbiol 177: 361– Desnues, C.G., Rodriguez-Brito, B., Rayhawk, S., Kelley, S., 370. Tran, T., Haynes, M., et al. (2008) Biodiversity and bioge- Allen, M.A., Goh, F., Burns, B.P., and Neilan, B.A. (2009) ography of phages in modern stromatolites and thromb- Bacterial, archaeal and eukaryotic diversity of smooth and olites. Nature 452: 340–345. pustular microbial mat communities in the hypersaline Dill, R.F., Shinn, E.A., Jones, A.T., Kelly, K., and Steinen, R.P. lagoon of Shark Bay. Geobiology 7: 82–96. (1986) Giant subtidal stromatolites forming in normal Anderson, S.L., and McIntosh, L. (1991) Light-activated het- salinity water. Nature 324: 55–58. erotrophic growth of the cyanobacterium Synechocystis sp. Dillon, J.G., Tatsumi, C.M., Tandingan, P.G., and Castenholz, strain PCC 6803: a blue-light-requiring process. J Bacteriol R.W. (2002) Effect of environmental factors on the synthe- 173: 2761–2767. sis of scytonemin, a UV-screening pigment, in a cyano- Andres, M.S., and Reid, R.P. (2006) Growth morphologies of bacterium (Chroococcidiopsis sp.). Arch Microbiol 177: modern marine stromatolites: a case study from Highborne 322–331. Cay, Bahamas. Sediment Geol 185: 319–328. Dravis, J.J. (1983) Hardened subtidal stromatolites, Barns, S.M., Fundyga, R.E., Jeffries, M.W., and Pace, N.R. Bahamas. Science 219: 385–386. (1994) Remarkable archaeal diversity detected in a Yellow- Dupraz, C., and Visscher, P.T. (2005) Microbial lithification in stone National Park hot spring environment. Proc Natl marine stromatolites and hypersaline mats. Trends Micro- Acad Sci USA 91: 1609–1613. biol 13: 429–438. Baumgartner, L.K., Spear, J.R., Buckley, D.H., Pace, N.R., Dupraz, C., Visscher, P.T., Baumgartner, L.K., and Reid, R.P. Reid, R.P., and Visscher, P.T. (2009) Microbial diversity in (2004) Microbe–mineral interactions: early carbonate pre- modern marine stromatolites, Highborne Cay, Bahamas. cipitation in a hypersaline lake (Eleuthra Island, Bahamas). Environ Microbiol 11: 2710–2719. Sedimentology 51: 1–21. Berrendero, E., Perona, E., and Mateo, P. (2008) Genetic and Dupraz, C., Reid, R.P., Braissant, O., Decho, A.W., Norman, morphological characterization of Rivularia and Calothrix R.S., and Visscher, P.T. (2009) Processes of carbonate (Nostocales, Cyanobacteria) from running water. Int J Syst precipitation in modern microbial mats. Earth Sci Rev 96: Evol Microbiol 58: 447–460. 141–162. Billi, D., Friedmann, E.I., Hofer, K.G., Caiola, M.G., Edwards, R.A., Rodriguez-Brito, B., Wegley, L., Haynes, M., and Ocampo-Friedmann, R. (2000) Ionizing-radiation Breitbart, M., Peterson, D.M., et al. (2006) Using pyrose- resistance in the desiccation-tolerant cyanobacterium quencing to shed light on deep mine microbial ecology. Chroococcidiopsis. Appl Environ Microbiol 66: 1489– BMC Genomics 7: 57. 1492. Ehrlich, H.L. (2002) Geomicrobiology. New York, NY, USA: Brochier-Armanet, C., Boussau, B., Gribaldo, S., and Fort- Marcel Dekker. erre, P. (2008) Mesophilic Crenarchaeota: proposal for a Engelbrektson, A., Kunin, V., Wrighton, K.C., Zvenigorodsky, third archaeal phylum, the Thaumarchaeota. Nat Rev N., Chen, F., Ochman, H., and Hugenholtz, P. (2010) Microbiol 6: 245–252. Experimental factors affecting PCR-based estimates of Burne, R.V., and Moore, L.S. (1987) Microbialites: orga- microbial species richness and evenness. ISME J 4: 642– nosedimentary deposits of benthic microbial communities. 647. Palaios 2: 241–254. Foster, J.S., and Green, S.J. (2011) Microbial diversity in Burns, B.P., Goh, F., Allen, M., and Neilan, B.A. (2004) Micro- modern stromatolites. In Cellular Origin, Life in Extreme bial diversity of extant stromatolites in the hypersaline Habitats and Astrobiology: Interactions with Sediments. marine environment of Shark Bay, Australia. Environ Micro- Seckbach, J., and Tewari, V. (eds). Berlin, Germany: biol 6: 1096–1101. Springer, pp. 385–405. Càceres, C., and Tessier, A.J. (2003) How long to rest: the Foster, J.S., Green, S.J., Ahrendt, S.R., Hetherington, K.L., ecology of optimal dormancy and environmental constraint. Golubic, S., Reid, R.P., and Bebout, L. (2009) Molecular Ecology 84: 1189–1198. and morphological characterization of cyanobacterial

diversity in the marine stromatolites of Highborne Cay, Hosoya, S., Arunpairojana, V., Suwannachart, C., Kanjana- Bahamas. ISME J 3: 573–587. Opas, A., and Yokota, A. (2006) Aureispira marina gen. Giani, D., Giani, L., Cohen, Y., and Krumbein, W.E. (1984) nov., sp. nov., a gliding, arachidonic acid-containing bacte- Methanogenesis in the hypersaline Solar Lake (Sinai). rium isolated from the southern coastline of Thailand. Int J FEMS Microbiol Lett 25: 219–224. Syst Evol Microbiol 56: 2931–2935. Goh, F., Allen, M.A., Leuko, S., Kawaguchi, T., Decho, A.W., Huse, S.M., Huber, J.A., Morrison, H.G., Sogin, M.L., and Burns, B.P., and Neilan, B.A. (2009) Determining the spe- Welch, D.M. (2007) Accuracy and quality of massively par- ciﬁc microbial populations and their spatial distribution allel DNA pyrosequencing. Genome Biol 8: R143. within the stromatolite ecosystem of Shark Bay. ISME J 3: Huse, S.M., Welch, D.M., Morrison, H.G., and Sogin, M.L. 383–396. (2010) Ironing out the wrinkles in the rare biosphere Good, I. (1953) The population frequencies of species and through improved OTU clustering. Environ Microbiol 12: estimation of population parameters. Biometrika 40: 237– 1889–1898. 264. Imhoff, J.F. (2008) Systematics of anoxygenic phototrophic Green, S.J., Blackford, C., Bucki, P., and Jahnke, L.L. (2008) bacteria. In Sulfur Metabolisms in Phototrophic Organisms. A salinity and sulfate manipulation of hypersaline microbial Hell, R., Dahl, C., Knaff, D., and Leustek, T. (eds). Dor- mats reveals stasis in the cyanobacterial community struc- drecht, the Netherlands: Springer, pp. 269–287. ture. ISME J 2: 457–470. Jahnke, L.L., Orphan, V.J., Embaye, T., Turk, K.A., Kubo, Grootjans, A.P., van den Ende, F.P., and Walsweer, A.F. M.D., Summons, R.E., et al. (2008) Lipid biomarker and (1997) The role of microbial mats during primary succes- phylogenetic analyses to reveal archaeal biodiversity and sion in calcareous dune slacks: an experimental approach. distribution in hypersaline microbial mat and underlying J Coast Conserv 3: 95–102. sediment. Geobiology 6: 394–410. Grotzinger, J.P., and Knoll, A.H. (1999) Stromatolites in Pre- Jones, S.E., and Lennon, J.T. (2010) Dormancy contributes cambrian carbonates: evolutionary mileposts or environ- to the maintenance of microbial diversity. Proc Natl Acad mental dipsticks? Annu Rev Earth Planet Sci 27: Sci USA 107: 5881–5886. 313–358. Jürgens, K., Pernthaler, J., Schalla, S., and Amann, R. Hallam, S.J., Konstantinidis, K.T., Putnam, N., Schleper, C., (1999) Morphological and compositional changes in a Watanabe, Y., Sugahara, J., et al. (2006) Genomic analy- planktonic bacterial community in response to enhanced sis of the uncultivated marine crenarchaeote Cenar- protozoan grazing. Appl Environ Microbiol 65: 1241– chaeum symbiosum. Proc Natl Acad Sci USA 103: 1250. 18296–18301. Kennard, J.M., and James, N.P. (1986) Thrombolites and Hamady, M., Walker, J.J., Harris, J.K., Gold, N.J., and Knight, stromatolites: two distinct types of microbial structures. R. (2008) Error-correcting barcoded primers for pyrose- Palaios 1: 492–503. quencing hundreds of samples in multiplex. Nat Methods King, G.M. (1984) Metabolism of trimethylamine, choline, and 5: 235–237. glycine betaine by sulfate-reducing and methanogenic bac- Hamady, M., Lozupone, C., and Knight, R. (2010) Fast teria in marine sediments. Appl Environ Microbiol 48: 719– UniFrac: facilitating high-throughput phylogenetic analyses 725. of microbial communities including analysis of pyrose- Könneke, M., Bernhard, A.E., de la Torre, J.R., Walker, C.B., quencing and PhyloChip data. ISME J 4: 17–27. Waterbury, J.B., and Stahl, D.A. (2005) Isolation of an Harris, J.K., Sahl, J.W., Castoe, T.A., Wagner, B.D., Pollock, autotrophic ammonia-oxidizing marine archaeon. Nature D.D., and Spear, J.R. (2010) Comparison of normalization 437: 543–546. methods for construction of large, multiplex amplicon pools Kremer, B., and Kazmierczak, J. (2005) Cyanobacterial mats for next-generation sequencing. Appl Environ Microbiol 76: from Silurian black radiolarian cherts: phototrophic life at 3863–3868. the edge of darkness? J Sediment Res 75: 897–906. Haruta, S., Kondo, M., Nakamura, K., Chanchitpricha, C., Krumbein, W.E., and Giele, C. (1979) Calciﬁcation in a Aiba, H., Ishii, M., and Igarashi, Y. (2004) Succession of a coccoid cyanobacterium associated with the formation of microbial community during stable operation of a semi- desert stromatolites. Sedimentology 26: 593–604. continuous garbage-decomposing system. J Biosci Bioeng Krumbein, W.E., Buchholz, H., Frank, P., Giani, D., Giele, C.,

98: 20–27. and Wonneberger, K. (1979) O2 and H2S coexistence in Havemann, S.A., and Foster, J.S. (2008) Comparative char- stromatolites. Naturwissenschaften 66: 381–389. acterization of the microbial diversities of an artificial micro- Kunin, V., Engelbrektson, A., Ochman, H., and Hugenholtz, P. bialite model and a natural stromatolite. Appl Environ (2010) Wrinkles in the rare biosphere: pyrosequencing Microbiol 74: 7410–7421. errors can lead to artificial inflation of diversity estimates. Havens, K.E., Phlips, E., Cichra, M.F., and Li, B. (1998) Light Environ Microbiol 12: 118–123. availability as a possible regulator of cyanobacteria Lane, D.J. (1991) 16S/23S rRNA sequencing. In Nucleic species composition in a shallow subtropical lake. Freshw Acid Techniques in Bacterial Systematics. Stackebrandt, Biol 39: 547–556. E., and Goodfellow, M. (eds). Chichester, UK: Wiley, pp. Hongmei, J., Aitchison, J.C., Lacap, D.C., Peerapornpisal, Y., 115–175. Sompong, U., and Pointing, S.B. (2005) Community phy- Ley, R.E., Harris, J.K., Wilcox, J., Spear, J.R., Miller, S.R., logenetic analysis of moderately thermophilic cyanobacte- Bebout, B.M., et al. (2006) Unexpected diversity and com- rial mats from China, the Philippines and Thailand. plexity of the Guerrero Negro hypersaline microbial mat. Extremophiles 9: 325–332. Appl Environ Microbiol 72: 3685–3695.

Lozupone, C., and Knight, R. (2005) UniFrac: a new phylo- Systematic Biology. Boon, D.R., and Castenholz, R.W. genetic method for comparing microbial communities. Appl (eds). New York, NY, USA: Springer, pp. 562–566. Environ Microbiol 71: 8228–8235. Robertson, C.E., Spear, J.R., Harris, J.K., and Pace, N.R. Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., (2009) Diversity and stratification of archaea in a hypersa- Yadhukumar, et al. (2004) ARB: a software environment for line microbial mat. Appl Environ Microbiol 75: 1801–1810. sequence data. Nucleic Acids Res 32: 1363–1371. Roesch, L.F., Fulthorpe, R.R., Riva, A., Casella, G., Hadwin, Mackie, R.I., Sghir, A., and Gaskins, H.R. (1999) Develop- A.K.M., Kent, A.D., et al. (2007) Pyrosequencing enumer- mental microbial ecology of the neonatal gastrointestinal ates and contrasts soil microbial diversity. ISME J 1: 283– tract. Am J Clin Nutr 69: 1035S–1045S. 290. Miller, S.R., Strong, A.L., Jones, K.L., and Ungerer, M.C. Roesch, L.F., Lorca, G.L., Casella, G., Giongo, A., Naranjo, (2009) Bar-coded pyrosequencing reveals shared bacterial A., Pionzio, A.M., et al. (2009) Culture-independent identi- community properties along the temperature gradients of fication of gut bacteria correlated with the onset of diabetes two alkaline hot springs in Yellowstone National Park. Appl in a rat model. ISME J 3: 536–548. Environ Microbiol 75: 4565–4572. Santos, F., Pena, A., Nogales, B., Soria-Soria, E., Del Cura, Miyatake, T., MacGregor, B.J., and Boschker, H.T. (2009) M.A., Gonzalez-Martin, J.A., and Anton, J. (2010) Bacterial Linking microbial community function to phylogeny of diversity in dry modern freshwater stromatolites from sulfate-reducing Deltaproteobacteria in marine sediments Ruidera Pools Natural Park, Spain. Syst Appl Microbiol 33: by combining stable isotope probing with magnetic-bead 209–221. capture hybridization of 16S rRNA. Appl Environ Microbiol Sato, Y., Willis, B.L., and Bourne, D.G. (2010) Successional 75: 4927–4935. changes in bacterial communities during the development Myshrall, K., Mobberley, J.M., Green, S.J., Visscher, P.T., of black band disease on the reef coral, Montipora hispida. Havemann, S.A., Reid, R.P., and Foster, J.S. (2010) Bio- ISME J 4: 203–214. geochemical cycling and microbial diversity in the modern Schimel, J. (2004) Playing scales in the methane cycle: from marine thrombolites of Highborne Cay, Bahamas. Geobiol- microbial ecology to the globe. Proc Natl Acad Sci USA ogy 8: 337–354. 101: 12400–12401. Nübel, U., Garcia-Pichel, F., and Muyzer, G. (1997) PCR Schloss, P.D. (2010) The effects of alignment quality, dis- primers to amplify 16S rRNA genes from cyanobacteria. tance calculation method, sequence filtering, and region on Appl Environ Microbiol 63: 3327–3332. the analysis of 16S rRNA gene-based studies. PLoS Oh, G., Zhang, L., and Jahng, D. (2008) Osmoprotectants Comput Biol 6: e1000844. enhance methane production from the anaerobic digestion Schloss, P.D., and Handelsman, J. (2005) Introducing DOTUR, of food wastes containing a high content of salt. J Chem a computer program for defining operational taxonomic Technol Biotechnol 83: 1204–1210. units and estimating species richness. Appl Environ Micro- Papineau, D., Walker, J.J., Mojzsis, S.J., and Pace, N.R. biol 71: 1501–1506. (2005) Composition and structure of microbial communities Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hart- from stromatolites of Hamelin Pool in Shark Bay, Western mann, M., Hollister, E.B., et al. (2009) Introducing mothur: Australia. Appl Environ Microbiol 71: 4822–4832. open-source, platform-independent, community-supported Pinckney, J.L., and Reid, R.P. (1997) Productivity and com- software for describing and comparing microbial commu- munity composition of stromatolitic microbial mats in the nities. Appl Environ Microbiol 75: 7537–7541. Exuma Cays, Bahamas. FACIES 36: 204–207. Selenska-Pobell, S. (2002) Diversity and activity in uranium Planavsky, N., Reid, R.P., Andres, M., Visscher, P.T., waste piles. Radioact Environ 2: 225–254. Myshrall, K.L., and Lyons, T.W. (2009) Formation and Smith, J.M., Green, S.J., Kelley, C.A., Prufert-Bebout, L., and diagenesis of modern marine calcified cyanobacteria. Geo- Bebout, B.M. (2008) Shifts in methanogen community biology 7: 566–576. structure and function associated with long-term manipu- Preston, C.M., Wu, K.Y., Molinski, T.F., and DeLong, E.F. lation of sulfate and salinity in a hypersaline microbial mat. (1996) A psychrophilic crenarchaeon inhabits a marine Environ Microbiol 10: 386–394. sponge: Cenarchaeum symbiosum gen. nov., sp. nov. Proc Sogin, M.L., Morrison, H.G., Huber, J.A., Mark Welch, D., Natl Acad Sci USA 93: 6241–6246. Huse, S.M., Neal, P.R., et al. (2006) Microbial diversity in Price, M.N., Dehal, P.S., and Arkin, A.P. (2010) FastTree 2 – the deep sea and the underexplored ‘rare biosphere’. Proc approximately maximum-likelihood trees for large align- Natl Acad Sci USA 103: 12115–12120. ments. PLoS ONE 5: e9490. Stal, L.J., van Gemerden, H., and Krumbein, W.K. (1985) Pruesse, E., Quast, C., Knittel, K., Fuchs, B.M., Ludwig, W., Structure and development of a benthic marine microbial Peplies, J., and Glockner, F.O. (2007) SILVA: a compre- mat. FEMS Microbiol Ecol 31: 111–125. hensive online resource for quality checked and aligned Stolz, J. (1990) Structure of microbial mats and biofilms. In ribosomal RNA sequence data compatible with ARB. Microbial Sediments. Riding, R., and Awramik, S.M. (eds). Nucleic Acids Res 35: 7188–7196. Berlin, Germany: Springer, pp. 1–8. Reid, R.P., Visscher, P.T., Decho, A.W., Stolz, J.F., Bebout, Visscher, P.T., Reid, R.P., and Bebout, B.M. (2000) Micros- B.M., Dupraz, C., et al. (2000) The role of microbes in cale observations of sulfate reduction: correlation of micro- accretion, lamination and early lithification of modern bial activity with lithified micritic laminae in modern marine marine stromatolites. Nature 406: 989–992. stromatolites. Geology 28: 919–922. Rippka, R., Castenholz, R.W., and Herdman, M. (2001) Walker, C.B., Torre, J.R., Klotz, M.G., Urakawa, H., Pinel, N., Phylum Cyanobacteria Section IV. In Bergey’s Manual of Arp, D.J., et al. (2010) Nitrosopumilus maritimus genome

reveals unique mechanisms for nitrification and autotrophy Supporting information in globally distributed marine crenarchaea. Proc Natl Acad Sci USA 107: 8818–8823. Additional Supporting Information may be found in the online Wang, Q., Garrity, G.M., Tiedje, J.M., and Cole, J.R. (2007) version of this article: Naive Bayesian classifier for rapid assignment of rRNA Fig. S1. Comparisons of proteobacterial lineages in thromb- sequences into the new bacterial taxonomy. Appl Environ olitic mats using a clustering-based approach. The histo- Microbiol 73: 5261–5267. grams represent the per cent of classified sequences that Wang, Y., and Qian, P.Y. (2009) Conservative fragments in could be assigned to a specific taxon by a Bayesian approach bacterial 16S rRNA genes and primer design for 16S ribo- at a confidence interval Ն 70% against the SILVA SSU Ref somal DNA amplicons in metagenomic studies. PLoS ONE v102 database. 4: e7401. A. Proteobacteria were classified to the order level. Xia, Y., Kong, Y., Thomsen, T.R., and Halkjaer Nielsen, P. B. Alphaproteobacteria were classified to the family level. (2008) Identification and ecophysiological characterization Table S1. Primers used for generating 16S rRNA barcoded of epiphytic protein-hydrolyzing saprospiraceae (‘Candida- library. tus Epiflobacter’ spp.) in activated sludge. Appl Environ Microbiol 74: 2229–2238. Please note: Wiley-Blackwell are not responsible for the Yannarell, A.C., and Paerl, H.W. (2007) Effects of salinity content or functionality of any supporting materials supplied and light on organic carbon and nitrogen uptake in a by the authors. Any queries (other than missing material) hypersaline microbial mat. FEMS Microbiol Ecol 62: 345– should be directed to the corresponding author for the 353. article.