Abstract

Molecular microbial ecology studies have revealed remarkable prokaryotic diversity in extreme hydrothermal marine environments. There are no comparable reports of culture-independent surveys of eukaryotic life in warm, anoxic marine sediments. By using sequence comparisons of PCR-amplified small subunit ribosomal RNAs, we characterized eukaryotic diversity in hydrothermal vent environments of Guaymas Basin in the Gulf of California. Many sequences from these anoxic sediments and the overlaying seawater represent previously uncharacterized protists, including early branching eukaryotic lineages or extended diversity within described taxa. At least two mechanisms, with overlapping consequences, account for the eukaryotic community structure of this environment. The adaptation to anoxic environments is evidenced by specific affinity of environmental sequences to aerotolerant anaerobic species in molecular trees. This pattern is superimposed against a background of widely distributed aerophilic and aerotolerant protists, some of which may migrate into and survive in the sediment whereas others (e.g., phototrophs) are simply deposited by sedimentary processes. In contrast, bacterial populations in these sediments are primarily characteristic of anoxic, reduced, hydrocarbon-rich sedimentary habitats.

Materials and Methods

Results and Discussion

From our molecular study of ribosomal RNA sequences (Fig.1) we recognized three kinds of previously uncharacterized eukaryotic diversity. The first corresponds to sequences that are unrelated to those of any other eukaryotes, and they seem to represent early branches in the eukaryotic tree, e.g., C1_E016, the clade C2_E026 plus C3_E012, the lineages C3_E014, CS_E036, the clade C1_E017 plus C2_E016, the lineage C1_E027, and the clade CS_R003 plus DH148–5-EKD18. The second type of diversity corresponds to sequences that represent deep branches within well described eukaryotic clades including the stramenopiles, apicomplexa, dinoflagellates, ciliates, acantharea, and radiolaria. Their depth of branching is oftentimes comparable to class level differences. For example, the well supported clade comprising the sequences C1_E024, C2_E014, C3_E017, and C2_E002 is basal to other stramenopiles and likely represents organisms that are nonphotosynthetic heterotrophs. We also include within this category clades represented by C1_E023, C1_E029, A1_E004, C1_E008, and C1_E037, which are closely related to sequences from taxonomically unassignable alveolates (“OLI-” and “DH-” types) obtained in molecular surveys of marine pelagic environments (2,3). Because they are bracketed by sequences from dinoflagellates, we predict that these sequences represent similar phototrophic species. Finally, we observed clades that are neither deep in the tree nor specifically related to any recognized eukaryotic clade (e.g., CS_E022 and CS_E042).

Comparable numbers of PCR amplicons were cloned from core horizons at similar depths in Cores A and C, but only 45 were selected at random for detailed sequence analysis. Sequences of core A clustered to form a smaller number of clades than did sequences from core C. In Fig.1, the larger number of lineages from core C relative to core A reflects this bias. Because culture-independent, molecular surveys do not provide reliable indicators of viability, we cannot absolutely determine whether the sequences in our survey reflect active members of protist communities or merely recent deposition of protists into the sediment. However, we can take advantage of phylogenetic affinity with well studied species to make predictions about the phenotypes and lifestyles of organisms represented in our environmental sequence database. Some of our recovered sequences are from organisms related to specialists of anaerobic environments. For example, within the ciliates, we recovered sequences (C1_E007 and A1_E033) that are closely related to sequences from the anaerobesMetopus contortus andTrimyema compressum, respectively. In addition, the detection of rRNA sequences from diverse flagellates is consistent with published accounts of viable flagellate protists from several hydrothermal vent sites, including Guaymas Basin (11). Rapid rates of reproduction allow these small flagellates to colonize temporally and spatially variable habitats such as these fine-particle, hydrothermally heated sediments.

Other recovered sequences are closely related to those from eukaryotes that are more generally distributed in marine environments. These include certain fungi, green algae, stramenopiles, alveolates, acanthareans, and radiolarians (12). For example, in the CS layer of core C, we recovered the sequence of a specific relative of the ciliateEuplotes aediculatus (a surface-associated filter feeder), described from a variety of habitats including marine and freshwater sediments and pelagic environments (13). Many of our detected rRNA sequences are unlikely to be from active members of the benthic population. Sequences of phytoplanktonic taxa, such as green algae and diatoms (14), are from organisms that require solar radiation for active photosynthesis and are, therefore, not suited to this aphotic environment. Other sequences may represent dead or encysted forms; for example, dinoflagellate cysts have been detected in marine sediments (15). Some sequences (A2_E003, A2_E011, A3_E002, C1_E024, A1_E004, C1_E001, and C1_E027) were more frequently recovered (nearly identical sequences from ≥10 clones) from our libraries. It is impossible however, to differentiate greater representation within the microbial community from biased recovery/amplification (16).

With successively deeper sections through the cores, recovery of eukaryotic SSU rRNA amplicons diminished. Below 3 cm depth in cores A and C (approximately 65°C and 45°C, respectively), we only recovered prokaryotic sequences. The decline in efficiency of recovering eukaryotic SSU rRNA amplicons deeper in the sediments likely reflects temperature and redox limitations of the eukaryotic community, with attendant degradation of nucleic acids in metabolically inactive organisms. In contrast, recovery of prokaryotic rRNA gene sequences continues well into the sediment (>12 cm). Archaeal methanogens that can grow to 110°C have been identified (17), and sulfate reduction has been measured at temperatures near 100° in these sediments (18,19).

Eukaryotic rRNA sequences recovered from the sediment encompass the majority of described lineages in the eukaryotic domain. In addition to anoxic specialists, there exists a diverse collection of eukaryotic sequences that is comparable in composition to sequences from benthic, pelagic, and near-surface water populations. In contrast, a parallel molecular survey of the same sedimentary cores describes sequences from a prokaryotic community (20) where most lineages are characteristic of anoxic, reduced, hydrocarbon-rich sedimentary habitats (Table1). This microbial community includes relatives of the green non-sulfur bacteria and members of the candidate subdivisions OP1, OP3, OP8, OP9, and OP11. Similar community compositions occur in sulfidic geothermal springs (21), deep subsurface soil (22), a methanogenic aquifer contaminated with aromatic and chlorinated hydrocarbons (23), and from enrichments on aromatic and chlorinated hydrocarbons (24). Delta-proteobacterial sequences were primarily from members of the sulfate-reducing familyDesulfobulbaceae, which specialize in propionate oxidation (25). Propionate and acetate are the predominant, naturally occurring short-chain fatty acids in the Guaymas Basin hydrothermally altered sediments (26). ɛ- and γ-proteobacterial sequences from Guaymas are related to those of sulfur-oxidizing bacterial symbionts of marine invertebrates. These symbionts require the simultaneous presence of sulfide and chemical oxidants, a characteristic of the sediment surface (27). However, evidence of nonsymbiotic filamentous ɛ-proteobacteria also has been reported in microbial mats from hydrothermal vents (28).

These results suggest that several mechanisms account for the community composition of protists in the sediment. A significant, but relatively small component of recovered eukaryotic sequences are from organisms specialized to anaerobic sediments such as those found in Guaymas Basin. Selection for anaerobic specialists is evidenced by the recovery of sequences that are related to those of several groups of strict anaerobes not typical of the water column. The community structure of the oxic–anoxic interface likely includes both microaerophilic and facultative forms, some of which may migrate into and out of the anoxic sediments. This microbial community structure is augmented by contributions of microaerophilic and aerobic, particle-attached organisms from the water column. The active members of the eukaryotic community must feed primarily on some yet-unidentified fraction of the abundant bacteria sustained by hydrothermal activity and sedimentary input of organics (see Table1).

Irrespective of events that shape eukaryotic population structures in anoxic sediments, the level of previously undescribed protist diversity is impressive. Because the sampling of clone libraries from this environment (and other marine environments) has not reached saturation (Fig.2), and because of differences in PCR primers used in this and previously reported studies, we do not know the full extent of diversity or the overlap between different marine environments. Indeed, one of the basal environmental sequences from the overlaying seawater of core C, CS_R003, is nearly identical to sequence DH148–5-EKD18 described from deep-sea Antarctic plankton (3). Nonetheless, the largely specialized prokaryotic community present at this site suggests that at least some of the many eukaryotic sequences identified here represent hydrothermal specialists of unique physiology and ecology. This remarkable diversity represents a small fraction of previously uncharacterized protist lineages that we are likely to discover in warm, anoxic, and other extreme environments.

Acknowledgments

We thank Tom Fenchel for helpful discussion on eukaryotic microbial distribution and ecology. We dedicate this study to Holger W. Jannasch, for whom this Guaymas Basin expedition was the final scientific adventure of his life. This project was supported by National Science Foundation Grant OCE 9714195 (to H. W. Jannasch and A.T.), National Aeronautics and Space Administration Exobiology Grant NRA97-OSS-11 and Astrobiology Cooperative Agreement NCC2-1054, and continuing support from the Unger G. Vetlesen Foundation (to M.L.S.). This paper is Woods Hole Oceanographic Institution contribution number 10481.

Abbreviation

SSU

small subunit

Footnotes

References