Targeted search for actinomycetes from nearshore and deep-sea marine sediments

Abstract

Sediment samples collected off the coast of San Diego were analyzed for actinomycete diversity using culture-independent techniques. Eight new operational taxonomic units (OTUs) in the Streptomycetaceae were identified as well as new diversity within previously cultured marine OTUs. Sequences belonging to the marine actinomycete genus Salinispora were also detected, despite the fact that this genus has only been reported from more tropical environments. Independent analyses of marine sediments from the Canary Basin (3814 m) and the South Pacific Gyre (5126 and 5699 m) also revealed Salinispora sequences providing further support for the occurrence of this genus in deep-sea sediments. Efforts to culture Salinispora spp. from these samples have yet to be successful. This is the first report of Salinispora spp. from marine sediments > 1100 m and suggests that the distribution of this genus is broader than previously believed.

marine-derived actinomycetes, Salinispora, marine microbial ecology, cultivation independent

Introduction

The phylum Actinobacteria is extraordinarily diverse (Gao & Gupta, 2012) and well represented in the marine environment (Rappé et al., 1999). Of the five subclasses that comprise this phylum (the Acidomicrobidae, Actinobacteridae, Coriobacteridae, Nitriliruptoridae, and Rubrobacteridae), sequences belonging to the Acidomicrobidae are commonly observed when culture-independent techniques are applied (Jensen & Lauro, 2008). Conversely, cultured Actinobacteria often fall within the subclass Actinobacteridae and, more specifically, within the order Actinomycetales. These bacteria are commonly referred to as actinomycetes and have been targeted from marine samples for their ability to produce structurally novel secondary metabolites (Zotchev, 2012). While a number of marine actinomycete species and genera have been described (Tian et al., 2009a; Tian et al., 2009b; Zhao et al., 2009; Goodfellow et al., 2012) it is not clear how well these cultured strains represent the extant diversity present in the marine environment.

Culture-independent studies have revealed the presence of actinomycetes in seawater (Yoshida et al., ) and deep-sea marine sediments (Stach et al., 2003a, b). Less abundant taxa such as the marine actinomycete genus Salinispora (Maldonado et al., 2005) have been detected when specific primers targeting this group were applied (Mincer et al., 2005). Actinomycetes have also been detected in marine sponges, facilitating the selection of culture media and further increasing the diversity of isolates recovered (Webster et al., 2001). In a separate study of two sponges from China, a wide difference between the genera observed using actinomycete specific primers and cultivation-based methods was observed (Xin et al., 2008). Further studies on one of these sponges revealed the importance of using both culture and culture-independent methods when studying actinomycete diversity (Sun et al., 2010). While all methods suffer from inherent biases, culture-independent techniques can help establish the occurrence of bacteria in specific environments.

In a prior study of sediment samples collected off the coast of California, culture-dependent actinomycete diversity was assessed between nearshore and offshore sites (Prieto-Davó et al., 2008). The results revealed considerable, marine-specific diversity and high levels of terrestrial influence out to 125 km from shore. The present study was undertaken to provide a culture-independent assessment of the collective actinomycete diversity present in five of these samples. These studies were complimented by independent analyses of deep-sea sediment samples collected from the Canary Basin and the South Pacific Gyre (SPG).

Materials and methods

Sample collection

To further explore the diversity of actinomycetes present in marine sediments collected off the coast of California, five of eleven sediment samples previously employed for cultivation studies (Prieto-Davó et al., 2008) were used to generate 16S rRNA gene clone libraries targeting the order Actinomycetales. All of these samples were collected using an untethered coring device designed and constructed at the Scripps Institution of Oceanography (SIO). The depths and collection sites are provided in Supporting Information Table S1. Each core was divided into five or six sections as previously described (Prieto-Davó et al., 2008). Approximately 1 g of wet sediment from each section was placed in a 1.5 mL Eppendorf tube containing 1 mL of sucrose lysis buffer (50 mm Tris–HCl, 150 mm NaCl, 5 mm EDTA, 0.25 m sucrose) and immediately stored on ice for transportation to shore. Long-term storage was at −20 °C.

Samples from the Canary Basin and the SPG were analyzed specifically for the presence of Salinispora spp. The Canary Basin sample was collected as previously described (Stach et al., 2003b). The SPG samples were collected using gravity or piston cores during the KNOX-02RR expedition (D'Hondt et al., 2009). A total of 11 cores were sectioned and subsampled from 3 to 5 times at various depths from the sediment surface to the bottom of the core generating a total of 51 samples (Table S2). Subcores were taken from each section by first removing the top layer of sediment with a sterile spatula. Sterilized cutoff syringes were then pushed into the core resulting in an uncontaminated subcore. The syringe containing the subcore was stored intact in heat sealed bags at −80 °C prior to molecular analysis.

DNA extraction, PCR amplification, and cloning

Environmental DNA (eDNA) was extracted from the sediment samples collected off the coast of California using a soil DNA extraction kit (cat. No 69506) according to manufacturer's protocol (Qiagen, Valencia, CA). 16S rRNA gene primers targeting the order Actinomycetales (Stach et al., , b) and the families Streptomycetaceae and Micromonosporaceae (Monciardini et al., 2002) were used (Table S3). PCR amplification of 1–4 μL of eDNA (18–20 ng mL⁻¹) was carried out in triplicate for each sample as follows: initial denaturation at 95 °C for 10 min followed by 30 cycles of 94 °C for 45 s, 65 °C for 45 s, and 72 °C for 1 min, followed by a 10 min extension at 72 °C. Triplicate PCR products were pooled and purified using MiniElute PCR purification columns according to the manufacturer's instructions (Qiagen). Purified DNA was ligated to the plasmid vector pCR^® 2.1-TOPO^® and used to transform One-Shot^® Mach1™ -T1^® chemically competent cells using a Topo TA^® cloning kit according to the manufacturer's protocol (Invitrogen, Carlsbad, CA). Transformed clones were identified using white–blue selection and inoculated into 10 mL Falcon tubes containing 3 mL of LB broth and 50 μg mL⁻¹ kanamycin. Plasmid DNA was extracted using the QiaPrep^® MiniPrep extraction kit according to manufacturer's instructions (Qiagen), digested with BstX I (New England BioLabs, Ipswich, MA), and run on a 1% agarose gel to confirm the presence of the correct-sized insert.

eDNA was extracted from 10 g of Canary Basin sediment using an UltraClean Mega DNA soil kit (Mo Bio Laboratories, Solana Beach, CA). The primers (Supplemental Table S3) and PCR conditions are as previously described (Stach et al., , b). PCR products were purified using QIAquick gel extraction columns according to the manufacturer's instructions (Qiagen, Crawley, UK). Purified DNA was blunt-end-ligated into the plasmid vector pST-Blue-1 and used to transform NovaBlue competent cells using a Perfectly Blunt cloning kit (Novagen, Madison, WI). Cloned plasmid DNA was extracted as described above and the presence of inserts confirmed by PCR using previously described primers (Stach et al., , b).

For the SPG sediments, a modified FastDNA^® SPIN kit for soil protocol (Qbiogene, Irvine, CA), followed by ChromaSpin™ TE-100 columns (Clontech, Mountain View, CA), was used to ensure high eDNA yields and purity (Froschner et al., 2009). The eDNA samples were subjected to whole-genome amplification using the REPLI-g^® Midi kit (Qiagen) as previously described (Froschner et al., 2009) and analyzed for the presence of Salinispora spp. using Salinispora-specific 16S rRNA gene PCR primers (Table S3). The reactions included 200 ng of whole-genome amplification product and consisted of an initial 5-min denaturation step at 94 °C followed by 30 cycles of 30 s at 94 °C, 90 s at 60 °C, and 90 s at 72 °C followed by a final 10 min extension at 72 °C. The reaction products were purified using a QIAquick PCR purification kit (Qiagen) following manufacturer's guidelines and quantified using a NanoDrop Spectrophotometer (NanoDrop, Wilmington, DE). PCR amplicons were cloned using the pGEM^®-T Easy Vector System (Promega, Madison, WI) at a 1 : 1 insert to vector ratio per the manufacturer's protocol. Transformed clones were grown overnight in LB broth and the plasmids extracted using Wizard^®Plus SV Minipreps (Promega).

Terminal restriction fragment length polymorphism (T-RFLP) analysis

T-RFLP was used to probe the SPG samples for the presence of Salinispora spp. Whole-genome amplification products were PCR amplified as described above with the addition of a 6-FAM label to the forward primer. The Salinispora-specific primers were designed as part of a prior study and shown to be > 98% specific for this taxon (Mincer et al., 2005). Approximately 40 ng of fluorescently labeled PCR product was digested independently with 4 U of the restriction enzymes AluI, AvaII, and EcoO109I in 10-μL reaction volumes. Reference T-RFLs were generated for the three Salinispora species by digesting labeled PCR products from cultured strains (Supplemental Table S5). The restriction digests were incubated at 37 °C for 16 h to ensure complete digestion prior to being deactivated at 65 °C for 20 min. The digests were precipitated and resuspended in Hi-Di formamide (Applied Biosystems, Carlsbad, CA) to a final concentration of approximately 0.6 ng μL⁻¹. About 0.5 μL GeneScan Liz 1200 size standard (Applied Biosystems) was added to each sample. The samples were denatured at 95 °C for 5 min and placed immediately on ice. The terminally labeled restriction fragments (T-RFs) were visualized with the Applied Biosystems 3130xl genetic analysis system.

Sequencing and phylogenetic analyses

Plasmid inserts from the California samples were sequenced at the UCSD Rebecca and John Moore Cancer Center using the M13 primer included in the vector (Invitrogen). Sequences were analyzed using the Basic Local Alignment Search Tool (blast) function of GenBank (Altschul et al., 1990), and strains with the highest level of sequence identity were recorded as top blast matches regardless of whether or not they were associated with a formal publication. Top matches and cloned sequences belonging to the Micromonosporaceae (560 bp) and Streptomycetaceae (533 bp) were clustered into OTUs based on ≥ 99% 16S rRNA gene sequence identity using the java application Clusterer (http://www.bugaco.com/mioritic/clusterer_jlp.php). In addition, OTU representatives from cultured strains derived from all 11 samples used in a prior study (Prieto-Davó et al., 2008) were included in the clustering analyses. Sequences were then aligned using ClustalX (Larkin et al., 2007) and imported into MacClade for manual curation of the alignment (http://macclade.org).

Plasmid inserts derived from the Canary Basin PCR products were sequenced by Qiagen using the SP6 universal primer. Plasmid inserts from the SPG samples were sequenced using the 16S primers or vector primers (T7, SP6). Sequencing was performed on an Applied Biosystems 3130xl Genetic Analyser (Applied Biosystems) at the University of Rhode Island's Genomics and Sequencing Center (Kingston, RI). The resulting sequences were trimmed and contigs built using Sequencher™ (Gene Codes, Ann Arbor, MI).

Maximum-Likelihood rooted phylogenetic trees (HKY85 substitution model, 1000 bootstraps) were constructed for the Streptomycetaceae and Micromonosporaceae using PhyML (Guindon et al., 2010) via http://www.phylogeny.fr/ and http://www.atgc-montpellier.fr/phyml/ (Dereeper et al., 2008). Trees were edited using Inkscape (inkscape.org). For both phylogenetic trees, top blast matches, as well as previously cultivated OTU representatives (Prieto-Davó et al., 2008), were included.

Diversity estimations

Estimates of the diversity present in the California clone libraries were performed using the statistical package EstimateS (http://purl.oclc.org/estimates) with 1000 runs for each library.

Results

Clone libraries targeting the order Actinomycetales (actinomycetes) and the families Streptomycetaceae and Micromonosporaceae were generated from five independent sediment cores collected off the coast of California (Table S1). The Actinomycetales-specific primers yielded largely nonactinobacterial sequences and Actinobacteria in the subclass Acidomicrobidae, which is outside of the Actinomycetales. In response, primers specific for the Streptomycetaceae and Micromonosporaceae were used because these were the dominant families cultured previously from these samples. Clone libraries were generated from two of the sediment cores using the Streptomycetaceae primers, two using the Micromonosporaceae primers, and one using both primer sets (Table S4). All libraries were generated from the surface section of the cores, with the exception of sample SD06-13, from which libraries were generated from all five sections. Seventy-seven of the combined 79 clones obtained using the Streptomycetaceae primers belonged to this family confirming the specificity of that primer set. The cloning efficiency of the Micromonosporaceae primers was poor and yielded only 23 sequences, of which 22 belonged to the targeted family.

The 99 Streptomycetaceae and Micromonosporaceae clones were clustered based on ≥ 99% sequence identity (Table ⁰⁰⁰¹). This yielded a total of 41 OTUs, each of which contained 1–10 clones (average = 2). Of these, representatives of eight OTUs were cultured as part of a prior study (Prieto-Davó et al., 2008; Table ⁰⁰⁰¹, Figs ⁰⁰⁰¹ and ⁰⁰⁰²). When compared more broadly to GenBank, which includes the sequences from the prior study, 33 of the 41 OTUs from this study have cultured representatives, and thus, eight can be considered new in terms of publically available sequence data. The majority of the OTUs (37) and all of the eight new OTUs belong to the Streptomycetaceae. The OTUs in this family are scattered throughout the phylogenetic tree; however, two of the new OTUs (represented by clones SD06-09A-02 and SD06-13C-01) form a well-supported clade with the marine-derived actinomycete strain CNQ-085 (Fig. ⁰⁰⁰¹). Two more of the new OTUs, represented by SD06-07A-01 and SD06-13E-03, belong to a much larger and previously identified marine clade that includes the new species Streptomyces marinus (Prieto-Davó et al., 2008) and the cloned OTU (SD06-13A-10). One additional OTU (SD06-13B-01) belongs to another cultivated marine clade (Prieto-Davó et al., 2008), while the two remaining new OTUs (SD06-09A-06 and SD06-13A-02) are distantly related to previously observed sequences.

1

Clustering of Streptomycetaceae and Micromonosporaceae clones into OTUs based on > 99% 16S rRNA gene sequence identity over 560 bp (Streptomycetaceae) and 533-bp (Micromonosporaceae)

Family	Cloned OTUs	Cultured OTUs	GenBank cultured OTUs
Streptomycetaceae	37	5	29
Micromonosporaceae	4	3	4
Total	41	8	33

Based on a prior analysis of the same sediment samples (Prieto-Davó et al., 2008).

1

Neighbor-joining phylogenetic tree of Streptomycetaceae 16S rRNA gene sequences. Environmental sequences from California begin with SD and are followed by number of clones (in parentheses) and accession number. Cultured strains were derived from a prior study (sequences beginning with CN). blast matches are included with species and strain name followed by accession number. Bold = new OTU (shares ≤ 98% sequence identity with any sequences deposited in Genbank), italics = marine OTU (all sequences in the OTU were derived from marine sources), underlined = OTU representative in culture. Boxed clade = marine clade (all sequences in the OTU plus top blast match and cultivars were derived from marine sources). Based on the analysis of 533-bp nucleotide positions and 1000 bootstraps.

2

Neighbor-joining phylogenetic tree of Micromonosporaceae 16S rRNA gene sequences. Environmental sequences from California begin with SD and are followed by number of clones (in parentheses) and accession number. Sequences from the Canary Basin start with CB and the South Pacific Gyre with SPG. Cultured strains were derived from a prior study (sequences beginning with CN). blast matches are included with species and strain name followed by accession number. Bold = new OTU (shares ≤ 98% sequence identity with any sequences deposited in Genbank), italics = marine OTU (all sequences in the OTU were derived from marine sources), underlined = OTU representative in culture. Based on the analysis of 560-bp nucleotide positions and 1000 bootstraps.

In an effort to estimate the total diversity in the sediments sampled, a rank abundance curve was generated from the combined Streptomycetaceae clone libraries. This curve shows little duplication and a long right-hand tail, as is characteristic of a highly diverse community (Fig. ⁰⁰⁰³). This is supported by three diversity estimators (ACE, Chao 1, and Jackknife 1), which predict that, on average, 70 OTUs were present in the three samples analyzed (Fig. ⁰⁰⁰⁴). The shape of the accumulation curve provides clear evidence that additional sequencing could reveal additional diversity. The Micromonosporaceae clones formed only four OTUs, and thus, a rank abundance curve was not generated. The poor cloning efficiency and low number of sequences analyzed suggest that additional studies of this family are warranted. Although none of the Micromonosporaceae OTUs were considered new, one claded with Catellatospora spp. (clones SD06-11-02 and SD06-12-08) and a second, which included three cloned sequences, claded with Salinispora spp. (clone SD06-12-01; Fig. ⁰⁰⁰²). The other two OTUs claded with Micromonospora spp. (clones SD06-12-05 and SD06-13-01). This is the first evidence that Salinispora spp. occur in temperate marine environments (Jensen & Mafnas, 2006; Freel et al., 2012). Given that neither Catellatospora nor Salinispora spp. had previously been cultured from these samples, cultivation techniques specific for both genera were applied (Ara & Kudo, 2006; Freel et al., 2012). However, only Micromonospora spp. were recovered.

3

Rank abundance curve for the San Diego Streptomycetaceae clone library. Seventy-seven clones generated from three independent libraries were combined and grouped into OTUs based on 99% 16S rRNA gene sequence identity.

4

Diversity estimators for the San Diego Streptomycetaceae clone library. Seventy-seven clones generated from three independent libraries were analyzed.

Independent analyses of deep-sea marine sediment samples collected from the Canary Basin and the South Pacific Gyre (SPG) were also performed to test for the presence of Salinispora sequences. A clone library generated from a Canary Basin sediment sample collected at a depth of 3814 m revealed the presence of one Salinispora sequence (Table ⁰⁰⁰²). Eleven deep-sea sediment cores obtained from the SPG were subdivided into 51 sections and analyzed by T-RFLP. PCR products were obtained using Salinispora-specific primers (Mincer et al., 2005) from 25 of the 51 subsamples including samples from each of the 11 sites (Table S2). The second depth below the sediment surface (0.3–0.89 m) yielded the largest number of successful PCR amplifications. T-RFLP analyses were performed on the 25 PCR products, and all 25 yielded T-RFs of the expected size for Salinispora spp. (Supplemental Table S5). Clone libraries generated from three of the PCR-positive samples (SPG1, SPG2-1, and SPG2-3) yielded eight Salinispora sequences and confirmed the presence of the genus in sediments collected at a depth of 5699 m (Table ⁰⁰⁰²). Additional sequencing would be required to confirm the presence of the genus in the remaining 22 samples. All of these sequences clade with Salinispora spp. (Fig. ⁰⁰⁰²) and fall within a single 99% OTU. However, due to the lack of species-specific nucleotides in the region sequenced, it is not possible to assign species level identifications. Collectively, these results provide the first evidence that Salinispora spp. occur at depths > 1100 m (Mincer et al., 2005) and outside of the geographical range from which the genus has been reported using culture-dependent methods (Freel et al., 2012).

2

Salinispora clones obtained from deep-sea sediment samples

Clone	Location depth (m)	NCBI closest blastn match	Accession numbers	Similarity (%)
SD06-12-01	San Diego 1133	S. pacifica CNH-732; Salinispora sp. CNR-040	DQ2244165.1; AY040617	99
SPG2-1-1A	Pacific Gyre 5126	S. pacifica CNT-138; S. arenicola CNT-850	HQ642853.1; HQ642848.1	100
SPG2-1-1E	Pacific Gyre 5126	S. pacifica CNT-138; S. arenicola CNT-851	HQ642853.1; HQ642848.2	99
SPG2-3-1B	Pacific Gyre 5126	S. pacifica CNT-138; S. arenicola CNT-852	HQ642853.1; HQ642848.3	100
SPG2-3-3:1H	Pacific Gyre 5126	S. pacifica CNT-138; S. arenicola CNT-853	HQ642853.1; HQ642848.4	99
SPG1-1A	Pacific Gyre 5699	S. pacifica CNT-138; S. arenicola CNT-854	HQ642853.1; HQ642848.5	100
SPG1-1D	Pacific Gyre 5699	S. pacifica CNT-138; S. arenicola CNT-855	HQ642853.1; HQ642848.6	99
SPG1-2G	Pacific Gyre 5699	S. pacifica CNT-138; S. arenicola CNT-856	HQ642853.1; HQ642848.7	100
SPG1-4A	Pacific Gyre 5699	S. pacifica CNT-148; S. arenicola NH13C	HQ642899.1; FJ232412.1	99
3896_JS14B02	Canary Basin 3814	S. pacifica CNT-148	HQ642899.1	99

In cases where equivalent best matches were found to two species, both are reported.

Discussion

Cultivation-independent methods are widely recognized as providing a more accurate estimation of bacterial diversity than culture-based approaches (Wagner et al., 1993; Hugenholtz et al., 1998). As a follow-up to a prior study of cultured marine actinomycete diversity in sediments collected off the coast of California (Prieto-Davó et al., 2008), we performed culture-independent analyses on a subset of the samples. The results reveal eight new OTUs in the family Streptomycetaceae and numerous clones that expand the extant diversity associated with previously cultured marine actinobacterial lineages. Statistical analyses suggest that additional sequencing will uncover additional diversity in this family, a conclusion also reached by Stach & Bull (2005) in relation to the analysis of Canary Basin sediments. In addition, it is also clear that some of the cultured OTUs were not observed in the clone libraries (Fig. ⁰⁰⁰¹). This could be due to insufficient sequencing; however, it supports the concept that culture-dependent methods provide a method to access members of the rare biosphere that are frequently missed when culture-independent methods are applied (Shade et al., 2012), even in cases such as this where taxon-specific primers are employed.

In the case of the Micromonosporaceae, the small size of the library made it difficult to predict the diversity in this family. Despite the small number of clones sequenced, seven (32%) were identified as Salinispora spp. This result was surprising considering that this genus has not previously been reported from temperate environments (Freel et al., 2012) and that extensive cultivation efforts over many years have failed to recover this taxon from sediments collected off the coast of California. Additional cultivation efforts using the samples that yielded Salinispora clones also failed to produce these bacteria in culture suggesting that techniques other than those proven successful for tropical samples (Gontang et al., 2007) are required.

The culture-independent results obtained for the sediments collected off the coast of California suggest that the biogeographical distributions of Salinispora spp. may be broader than those previously estimated based on culture-dependent approaches (Freel et al., ). To further explore this possibility, deep-sea sediment samples collected from the Canary Basin (3814 m) and the South Pacific Gyre (5126 and 5699 m) were analyzed using culture-independent techniques. Both locations yielded Salinispora clones, providing a new depth record for the detection of this genus and additional evidence for its widespread occurrence in deep-ocean sediments. Samples collected from the SPG sites, which extend from American Samoa to New Zealand (Figure S1), consistently tested positive for Salinispora 16S rRNA gene sequences, providing evidence that the genus is broadly distributed in deep-ocean sediments. It remains unclear whether the DNA from which these sequences originated was derived from actively growing cells or spores. Conditions in the deep sea include high pressures and low temperatures, which may inhibit growth. Yet, extensive cultivation efforts applied to the California (Prieto-Davó et al., 2008), Canary Basin (Stach et al., 2003a, b), and SPG samples failed to yield Salinispora strains, suggesting they require cultivation conditions that differ from the strains that have been obtained from more tropical sites. While there was no phylogenetic evidence that the deep-sea Salinispora sequences are distinct (Fig. ⁰⁰⁰²), less-conserved phylogenetic markers may need to be analyzed before any conclusions can be reached about their population structure.

Due to the specificity of the primers, it is likely that considerable actinomycete diversity was omitted from this study. However, the families targeted are among the most prolific in terms of secondary metabolite production and include a number of new marine taxa (Jensen et al., 2005; Goodfellow et al., 2012). The results presented here suggest that additional new diversity within the Streptomycetaceae remains to be cultured from marine sediments, thus supporting further studies of deep-sea actinomycetes as a resource for natural product discovery. In addition, there is intriguing new evidence that Salinispora spp. are broadly distributed in deep-sea marine sediments. It will be important to develop cultivation techniques appropriate for these bacteria and to determine whether they have developed adaptations to life in the deep sea that may include the production of new secondary metabolites.

Acknowledgements

P.R.J. acknowledges financial support from the California Sea Grant College Program Project #R/NMP-100 (Grant #NA100AOAR4170060) through NOAA'S National Sea Grant College Program, USA Dept. of Commerce. The statements, findings, conclusions, and recommendations are those of the author(s) and do not necessarily reflect the views of California Sea Grant or the USA Dept. of Commerce. J.E.M.S. and A.T.B. acknowledge financial support from the UK Natural Environmental Research Council (Grants NER/T/S/2000/00614 and NER/T/S/2000/00616). This project was supported by a graduate fellowship from Consejo Nacional de Ciencia y Tecnología (CONACyT #157730) to AP-D. D.C.R. and D.C.S. acknowledge financial support from the National Institutes of Health (Grant #1R15AI093158-01) and the National Science Foundation (award 0752336) and thank the captain and the crew of the R/V Revelle and chief scientist S. D'Hondt. The authors declare that they have no conflict of interest.

Supporting Information

Fig. S1. Sampling locations for South Pcific Gyre (SPG) sediments.

Table S1. Sample depth and location.

Table S2. South Pacific Gyre sampling sites (see Table S1 for locations) and depths (m) below the sediment surface (in parentheses) from which the cores were sub-sampled.

Table S3. PCR primers. *New to this study.

Table S4. Culture-independent analyses of sediments collected off the coast of California.

Table S5. T-RFLs in base pairs of cultured Salinispora species.