Active Bacterial Flora Surrounding Foraminifera (Xenophyophorea) Living on the Deep-Sea Floor

120663 (060)

Biosci. Biotechnol. Biochem., 77 (2), 120663-1–4, 2013 Note Active Bacterial Flora Surrounding Foraminifera (Xenophyophorea) Living on the Deep-Sea Floor

y Sayaka HORI, Masashi TSUCHIYA, Shinro NISHI, Wataru ARAI, Takao YOSHIDA, and Hideto TAKAMI

Institute of Biogeosciences, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa 237-0061, Japan

Received August 29, 2012; Accepted October 31, 2012; Online Publication, February 7, 2013 [doi:10.1271/bbb.120663]

Bacteria form unique ecosystems by coexisting with 7000II (Dive 381) with a push-core sampler from a large organisms. Here we present the first evidence of depth of 7,111 m on the Izu-Ogasawara Trench slope active flora surrounding xenophyophorea revealed (3247.650N, 14152.590E). Surface sediment was col- through clone analyses of environmental ribosomal lected at approximately 50 cm from the collected RNA gene sequences. The flora included eight phyla in xenophyophore (Fig. 1A). Both samples were immedi- the xenophyophorean cells with agglutinated test. The ately stored at 80 C. Since the xenophyophore was major operational taxonomic units were unique from fragmented in the sampler, we isolated the top 3.5 cm that in the near-surface sediment. This flora appears to from the push-core sample as a ‘‘xenophyophore be formed by coexistence with xenophyophores. sample,’’ a mixture of xenophyophorean cells and the loosely agglutinated test. We used the upper 3.5 cm KeyAdvance words: microbial flora; deep sea; xenophyophore; View of the reference push-core sediment as a ‘‘reference foraminifera sample.’’ Total RNAs were extracted from the frozen samples individually using an RNA PowerSoil Total Bacteria form unique ecosystems on benthic struc- RNA Isolation Kit (Mo Bio Laboratories Inc., Solana tures, substrates and coexist with large organisms on the Beach, Calif.). The amounts of total recovered RNA deep-sea floor. For example, giant foraminifera (xen- are 69.4 ng of total RNA/g (xenophyophore sample), ophyophorea) can have a considerable influence on and 73.5 ng of total RNA/g (reference sample). The nearby microflora. In areas deeper than the carbonate total RNAs were treated with DNase I, Amp Grade compensation depth, foraminifera construct fragile (Invitrogen, Carlsbad, CA) and with RNase inhibitor agglutinated tests (technical term for shells) composed SUPERaseIn (Ambion, Austin, TX). cDNAs were of benthic sediment and organic cement secretion.1,2) synthesized usingProofs SuperScript III (Invitrogen, Carlsbad, Xenophyophores are the largest foraminifera (about CA) and a bacterial consensus primer (1492R).6) 25 cm in size) making tests, and are widely distributed at Bacterial 16S rRNA sequences were PCR-amplified high density on the deep-sea floor.1–4) Their large, thick, using bacterial consensus primers (27F and 1492R) from morphologically complex tests provide a substrate and the cDNAs.6) Reactions without reverse transcriptase source of food for smaller species, and can contribute to were performed to exclude DNA contamination. PCR the maintenance of high local species diversity in areas products were ligated into pGEM-T vectors and cloned where they are abundant.4) into Escherichia coli DH5-competent cells. We se- In the previous study, xenophyophores are made up of quenced 309 xenophyophore sample and 210 reference protoplasm, stercomes, and agglutinated tests.5) Previous sample clones using 27F and 1492R primers. The former study has indicated that bacterial lipid compositions average sequence lengths were 1,413 bp and the latter differ between lysed xenophyophorean cells with an were 1,411 bp. Both sequences were grouped into agglutinated test and an reference sediment, suggesting operational taxonomic units (OTUs) defined at 97% that a unique flora develops after incorporation into the sequence similarity.7) The coverage values were 88.5% xenophyophore.5) However, the types of bacteria inhab- (xenophyophore sample) and 83.8% (reference sam- iting the xenophyophores and the difference in phyla ple).8) The OTUs were aligned using Greengenes align- between the xenophyophore sample and the reference ment9) and were phylogenetically analyzed at the sample were undefined.5) Here we present the first phylum level using ARB software.10) The closest related comparison of bacterial flora in a new xenophyophore species were selected by BLAST. and nearby sediment by 16S rRNA clones obtained from We collected a xenophyophore that was rippled total RNA.6) Our findings revealed the composition of approximately 7 cm in diameter (Fig. 1A). This is the active flora surrounding the xenophyophores. first report of a xenophyophore from the Izu-Ogasawara A research cruise was conducted by the research Trench. A molecular phylogenetic tree based on the vessel (R/V) Kairei in 2007 (KR07-04). A xenophyo- partial small subunit (SSU) rDNA sequence using S8F phore with a test and a reference sediment were and sB primers11) by neighbor-joining (NJ) analysis and collected by the remotely operated vehicle (ROV) Kaiko the maximum likelihood (ML) method confirmed the

y To whom correspondence should be addressed. Tel/Fax: +81-3-5269-7362; E-mail: [email protected] Abbreviations: BLAST, Basic Local Alignment Search Tool; NJ analysis, neighbor-joining analysis; ML, maximum likelihood; NCBI, National Center for Biotechnology Information; OTU, operational taxonomic unit 120663-2 S. HORI et al. A

B 95 / 92 Pararotalia nipponica (AJ879137) 75 / 69 Heterostegina depressa (AJ879132) Cibicides refulgens C173.1 (DQ195540) 92 / 100 0.02 / 97 Cassidulinoides parkerianus 3924 (DQ408639)

89 / Islandiella sp. 2643 (DQ408638) 100 / 92 59 Uvigerina earlandi 2187 (DQ408640) 99 / 100 Uvigerina phlegeri U239 (DQ408641) Advance69 / Trochammina View hadai 95 (DQ408637) 58 / 62 Eggerelloides scabrum (AJ318228) 99 / 59 Glandulina antarctica (EU672990) 100 / 99 Shinkaiya lindsayi (EU649778) Xenophyophore sp. Boso_2008 (AB694014) 98 / 95 Syringamina corbicula (EU672993) 98 / 92 Rhizammina algaeformis (EU649779)

Borelis schlumbergeri 191 (AJ404295) Xenophyophore 100 / 100 100 / 100 Peneroplis sp. 69 (AJ132368)Proofs 85 / Marginopora vertebralis 499 (AJ404312) 51 Reticulomyxa filosa (AJ132367) Allogromia sp. (X86093)

Fig. 1. Phylogenic Analysis of the Collected Xenophyophore. A, Sampling at the sea floor. Collected new xenophyophore (arrowhead) and reference sediment (arrow) are shown. Hollow arrowheads show other xenophyophores. Scale bar, 10 cm. B, Molecular phylogenetic tree based on partial SSU rDNA sequences and representative foraminifera inferred by NJ analysis and the ML method. An alignment scoring 975 nucleotide positions (gaps excluded) was created using Clustal X. Bootstraps support values higher than 50% (1,000 replicates for NJ analysis, 100 replicates for the ML method), and are represented as NJ/ML. Scale bar indicates estimated number of base changes per nucleotide sequence position. monophyly of xenophyophores (Fig. 1B). Nucleotide the reference sample, suggesting that no difference BLAST analysis showed the highest similarity to existed between active flora in the xenophyophore and Shinkaiya lindsayi (NCBI accession no. EU649778) in the reference sediments at the phylum level. the conserved regions, but the sequence similarities were At the OTU level, high genetic diversity was less than 94%, suggesting that the two xenophyophores observed. The OTUs within the Proteobacteria sub- were different species. These results suggest that the classes showed the highest diversity and the largest bacterial clones obtained from the total RNA associated proportions (Fig. 2A and B). Although OTU0001 in with the xenophyophore described as a genetically new Alphaproteobacteria was the most abundant, its pop- phylotype. ulation comprised less than 6%, suggesting that no Bacterial 16S rRNA clones obtained from the total dominant OTU was present in this flora. Most of the RNA of the xenophyophore sample showed that the OTUs were unique to one of the two samples relative to bacterial OTUs belonged to eight phyla, Proteobacteria, the other and were detected in all the phyla name above Gemmatimonadetes, Acidobacteria, Chloroflexi, Bacter- (Figs. 2 and 3), with the exception of Chlorobi, a minor oidetes, Actinobacteria, Planctomycetes, Chlorobi, and phylum including only one OTU (Fig. 3D). Alphapro- candidate divisions (Figs. 2 and 3), suggesting that the teobacteria OTU008 was detected in both samples, but active flora were comprised of these bacteria. The same showed a significantly lower population (p < 0:05, phyla and candidate divisions were detected in the Fisher’s exact test) in the xenophyophore sample bacterial 16S rRNA clones obtained from total RNA of compared to the reference sample (Fig. 2A). This Flora Surrounding Foraminifera 120663-3

A Alphaproteobacteria B Deltaproteobacteria

D Gemmatimonadetes

C Acidobacteria Advance View

E Chloroflexi F Candidate divisionsProofs

Xenophyophore Reference

Fig. 2. Clone Analysis of Bacterial OTUs of Proteobacteria, Acidobacteria, Gemmatimonadetes, Chloroflexi, and Candidate Divisions. Clonal frequencies of the xenophyophore (OTU no. xeno) and reference (OTU no. Ref) samples are shown by black and gray bars respectively. Common OTUs (OTU no. Com) include both clones. Minor OTUs contain fewer than two OTUs. suggests that population of OTU008 decreased after OTU005 and OTU073 did not show significant dif- incorporation into xenophyophorean cells and/or agglu- ferences between the two samples (Fig. 2A). tinated test. Unfortunately, we could not identify the Our results suggest that active phylotypes in surface localization patterns of the OTUs unique to xenophyo- sediment change greatly after coexisting with xeno- phore sample by microscopy, because xenophyophore phyophores. A previous study suggested that xenophyo- cells are fragile,1–3,5) and were the totally destroyed phores contain higher amounts of bacteria than the during sampling. However, we can at least say that reference sample, and that the bacterial ratio patterns OTU008 is not an obligatory symbiotic bacterium, between the two samples are different.5) Further analysis because OTU008 is active in the sediment without the of the amounts, localization, ecology of OTU008 and xenophyophore.12) other unique OTUs, and individual differences between OTU008 forms a clade with uncultured bacterium xenophyophore sp. Boso 2008 should clarify the inter- clone S26-96 (EU287396) collected from the Arctic and action mechanism between the bacteria and the xeno- with clone Ulrdd 36 (AM997500) collected from the phyophores. South Atlantic Ocean13) with 99% sequence homology In conclusion, this study provides an initial view of the (Fig. 3F), suggesting that three clones are same species. microflora surrounding xenophyophores in the deep sea, They are found in deep-sea sediments, but their and is the first report of a molecular ecological study of metabolisms are unknown. Close clades containing this flora. Description of the bacterial ecology in a micro 120663-4 S. HORI et al.

A Actinobacteria B Planctomycetes

D Chlorobi C Bacteroidetes

E Other bacteria

Xenophyophore Reference AdvanceF View

Fig. 3. Clone Analysis of Bacterial OTUs of Actinobacteria, Planctomycetes, Bacteroidetes, Chlorobi, and Other Bacteria. A, Clonal frequencies of the xenophyophore (OTU no. xeno) and reference (OTU no. Ref) samples are shown by black and gray bars respectively. Common OTUs (OTU no. Com) include both clones. Minor OTUs contain fewer than two OTUs. B, Phylogenic analysis of OTU008. Molecular phylogenetic tree based on 16S rDNA sequences inferred by NJ analysis. Scale bar indicates estimated number of base changes per nucleotide sequence position. Proofs habitat, such as co-localization with large organisms, is 3) Gooday AJ, Bett BJ, and Pratt DN, Deep Sea Res. Part I, 40, still rare, especially based on RNA analysis correlating 2131–2143 (1993). biological activity. Active microbial flora in the xen- 4) Pawlowski J, Holzmann M, Fahrni J, and Richardson SL, J. Eukaryot. Microbiol., 50, 483–487 (2003). ophyophore test can provide clear evidence of ecology of 5) Laureillard J, Me´janelle L, and Sibuet M, Mar. Ecol. Prog. Ser., minor and dormant microorganisms in the environment. 270, 129–140 (2004). The collected xenophyophore and bacterial rRNA 6) Nogales B, Moore ER, Llobet-Brossa E, Rossello-Mora R, sequences obtained in this study were deposited in the Amann R, and Timmis KN, Appl. Environ. Microbiol., 67, GenBank nucleotide sequence databank under accession 1874–1884 (2001). nos. AB694014 to AB694522. 7) Brosius J, Palmer ML, Kennedy PJ, and Noller HF, Proc. Natl. Acad. Sci. USA, 75, 4801–4805 (1978). 8) Good IJ, Biometrica, 40, 237–264 (1953). Acknowledgments 9) DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, Huber T, Dalevi D, Hu P, and Andersen GL, Appl. We thank Mr. T. Murashima and the captain and crew Environ. Microbiol., 72, 5069–5072 (2006). of the R/V Kairei KR07-04 cruise for providing core 10) Ludwig W, Strunk O, Westram R, Richter L, Meier H, samples. We also thank Mr. T. Maeda for technical Yadhukumar, Buchner A, Lai T, Steppi S, Jobb G, Fo¨rster W, support in phylogenetic analyses, and Dr. T. Maruyama Brettske I, Gerber S, Ginhart AW, Gross O, Grumann S, for informative discussion. This work was supported by Hermann S, Jost R, Ko¨nig A, Liss T, Lu¨ssmann R, May M, Nonhoff B, Reichel B, Strehlow R, Stamatakis A, Stuckmann N, a Grant-in-Aid for Young Scientists (Start-up, Grant Vilbig A, Lenke M, Ludwig T, Bode A, and Schleifer KH, no. 20810047). Nucleic Acids Res., 32, 1363–1371 (2004). 11) Lecroq E, Gooday AJ, Tsuchiya M, and Pawlowski J, Zool. J. References Linn. Soc., 156, 455–464 (2009). 12) Werren JH, Baldo L, and Clark ME, Nat. Rev. Microbiol., 6, 1) Gooday AJ, Palaios, 9, 14–31 (1994). 741–751 (2008). 2) Gooday AJ and Tendal OS, ‘‘The Illustrated Guide to the 13) Schauer R, Bienhold C, Ramette A, and Harder J, ISME J., 4, Protozoa,’’ eds. Lee JJ, Leedale GF, and Bradbury P, Allen 159–170 (2010). Press, Lawrence, pp. 1086–1097 (2000).