Methane-Consuming Archaea Revealed by Directly Coupled Isotopic and Phylogenetic Analysis

See allHide authors and affiliations

Science  20 Jul 2001:
Vol. 293, Issue 5529, pp. 484-487
DOI: 10.1126/science.1061338


Microorganisms living in anoxic marine sediments consume more than 80% of the methane produced in the world's oceans. In addition to single-species aggregates, consortia of metabolically interdependent bacteria and archaea are found in methane-rich sediments. A combination of fluorescence in situ hybridization and secondary ion mass spectrometry shows that cells belonging to one specific archaeal group associated with theMethanosarcinales were all highly depleted in13C (to values of –96‰). This depletion indicates assimilation of isotopically light methane into specific archaeal cells. Additional microbial species apparently use other carbon sources, as indicated by significantly higher13C/12C ratios in their cell carbon. Our results demonstrate the feasibility of simultaneous determination of the identity and the metabolic activity of naturally occurring microorganisms.

Microbes critically impact global geochemical cycles. Although the general ecological importance of microbial activity is well recognized, the identity and involvement of microbes in specific biogeochemical cycles are often poorly understood. For example, the anaerobic oxidation of methane (AOM) is a widespread and geochemically well documented process [e.g., (1)], yet very little is known about the physiology, biochemistry, and identity of the microbes involved. One reason for this is that often the ecologically relevant microorganisms are difficult to isolate in pure culture. Approaches that combine phylogenetic and stable isotope analyses have considerable potential for linking microbial diversity with in situ activity (2–4). Recent studies that combine phylogenetic surveys of ribosomal RNAs (rRNAs) with structural and stable isotopic analyses of lipids have revealed new information about methane-oxidizing microbes in anoxic marine sediments (3, 5, 6). However, the stable isotopic and phylogenetic methods used in these studies were uncoupled, so identification of the specific microbes mediating AOM is based mainly on indirect lines of evidence.

Here, we report a cultivation-independent study of marine microbial assemblages in anoxic methane-rich sediments that combined microbial cell identification using ribosomal RNA-targeted fluorescent in situ hybridization (FISH) (7) with secondary ion mass spectrometry (SIMS) (8). After rRNA-targeted probes were applied to identify microbial cells, the stable isotope composition of the identified cells was determined by using SIMS. Coupled FISH-SIMS provided a measure of the stable carbon isotope composition of individual phylogenetically identified cell aggregates. Uncultured, naturally occurring microbial cells that utilize methane as the source of cell carbon could therefore be identified unambiguously.

Previous studies suggested that cell aggregates of archaea belonging to the Methanosarcinales (ANME-2 group), surrounded by sulfate-reducing bacteria related to theDesulfosarcina, were involved in AOM (5,6). To test this hypothesis, microbial cell aggregates were recovered from marine sediments (intervals 3 to 5 cm deep) at methane seeps in the Eel River Basin, California (9). The sediments contained highly 13C-depleted archaeal and bacterial lipid biomarkers (10) (Table 1). By using FISH-SIMS, δ13C (11) was profiled from the periphery to the interior of separate cell aggregates (in approximately 0.75-μm increments) by repeated sputtering of the aggregate surface with a Cs+ beam (12–15). These analyses revealed that cell aggregates binding a specific archaeal probe (ANME-2) in their inner core, and a bacterialDesulfosarcina-Desulfococcus (DSS) probe on their periphery, were composed of extremely depleted carbon with δ13C values as low as –96‰ (Figs. 1 and 2, A and B). These isotopic signatures are best explained by assimilation of carbon from a 13C-depleted source. The only plausible source with sufficiently 13C-depleted carbon is methane and, indeed, the δ13C values of methane obtained from adjacent seep sites in Eel River Basin range between –63 and –35‰ (16).

Table 1

Carbon isotopic compositions (versus PDB) and source assignments of selected extracted archaeal ether lipids and bacterial fatty acids (FAs). SRB, sulfate-reducing bacteria.

View this table:

In addition to putative syntrophic ANME- 2/DSS consortia, cell aggregates that contained the archaeal ANME-2 group, but not bacteria, were also occasionally observed (Fig. 2, C and D). Isotopic analysis of these monospecific ANME-2 aggregations also revealed extreme 13C-depletions, reaching δ13C values of –85‰, again indicating that the major portion of their biomass was derived from methane. That these δ13C values are lower than coexisting methane indicates significant isotopic fractionation of the assimilated carbon by the methane-oxidizing archaeal group, ANME-2 (Fig. 3). Although bacteria-free aggregations of the Methanosarcinales-related ANME-2 group in methane-rich marine sediments have not been previously reported (5, 6), our observations indicate that this archaeal group may sometimes exist independently of syntrophic partners.

Figure 1

A schematic diagram of an archaeal ANME-2/bacterialDesulfosarcina (DSS) aggregate, showing the direction of penetration of an ion beam, indicated by the arrow. The graph shows the δ13C profile obtained with the ion microprobe versus time (minutes of Cs+ beam exposure) through the same 10-μm aggregate that is depicted in Fig. 2, A and B.

Figure 2

Whole-cell FISH of methane-oxidizing consortia and other cell aggregates in methane seep sediments, identified microscopically and targeted with the ion microprobe. (A) Overlaid epifluorescent image of a Cy-3–labeled archaeal ANME-2 (in red) and fluorescein-labeled bacterialDesulfosarcina (DSS, in green) cell aggregate from Eel River Basin. (B) Corresponding DAPI (nonspecific stain for DNA) and average δ13C value of the aggregate obtained by SIMS. (C) Cy-3–labeled archaeal ANME-2 aggregate. (D) DAPI stain of same field showing isotopic values of both the ANME-2–targeted aggregate and a cell aggregate not targeted by either the archaeal ANME-2 or bacterial DSS rRNA probes. (E) Sulfate-reducing bacterial aggregates from a Santa Barbara hydrocarbon seep targeted with fluorescein-labeled bacterial DSS probe. (F) Corresponding DAPI stain of same field and average δ13C values obtained by SIMS. Scale bar, 10 μm in each panel.

Figure 3

Ion microprobe measured δ13C profiles for individual cell aggregates recovered from marine methane seep sediments (Eel River Basin* and Santa Barbara Basin) (12). Numbers indicate sample number and designate the start of ion microprobe analysis for each cell aggregate. In most cases, extended analyses produced isotopically lighter values, as continued sputtering penetrated through the aggregate interior where the archaeal ANME-2 microbes are concentrated. In contrast, the bacterial samples (10 through 14) exhibited relatively constant δ13C profiles with depth. The hatched area represents the range of δ13C values measured for CH4 extracted from pore water from adjacent Eel River Basin methane seeps (average –50‰; n = 28). The relatively high δ13C values for pore water CH4are likely due to isotopic enrichment from active methanotrophy.

In contrast to the archaeal-bacterial consortia, the carbon isotopic compositions of other microbial aggregates from the same Eel River Basin sample were about –20‰. This value is consistent with assimilation of organic carbon derived from photosynthetic primary productivity or from the fixation of CO2 (Fig. 2D). For comparison, we analyzed sulfate-reducing bacterial aggregates originating from a shallow water hydrocarbon seep offshore Santa Barbara, California. Microscopic surveys using FISH of the hydrocarbon-impregnated sediment sample revealed abundant bacterial aggregates phylogenetically related to the Desulfosarcina, but no ANME-2 archaea. Ion microprobe analyses ofDesulfosarcina aggregates displayed δ13C values similar to those of sedimentary organic carbon (17) and oils from the underlying Monterey Formation (Fig. 2, E and F). Carbon isotopic compositions from fatty acids extracted from the hydrocarbon seep sediment ranged from –26 to –19‰ (Table 1). No lipids with isotopic compositions indicative of microbial utilization of methane-derived carbon were detected (18). Moreover, the fatty acid distribution from the oil seep differed significantly from that of the Eel River Basin methane seep, suggesting that these two sites harbored different bacterial communities. Both lipid and whole-cell isotopic data suggest that anaerobic oxidation of methane is not a significant biogeochemical pathway at this site.

The variation in δ13C values of aggregates containing the archaeal ANME-2 was greater than that of other cell aggregates. Significant isotopic variations were observed between individual ANME-2/DSS consortia, with archaeal-bacterial aggregates falling into two isotopically distinct groups (Fig. 3). For aggregates with relatively high δ13C values, methane was probably not the exclusive carbon source. Despite the large range in isotopic values, all the ANME-2/DSS consortia exhibit lower δ13C values than Desulfosarcina cell clusters and other cell aggregates, which are characterized by average δ13C values in the range –30 to –15‰ (Fig. 3).

To date, there have been no reported studies of single-cell isotopic variation for microbes, making interpretations of the variability we observe difficult. Preliminary SIMS data obtained for reference from pure cultures of cyanobacteria revealed substantial carbon isotopic variation (Gloeothece sp. 27152: mean δ13C = –17.1 ± 2.1‰, SD = 7.5;n = 13; and Gloeocapsa sp. 29159: mean δ13C = –24.3 ± 1.3‰, SD = 6.3;n = 23), reflecting some degree of organism-specific heterogeneity probably magnified by closed-system culture conditions. Whatever the causes, the heterogeneity of isotopic values in cell populations does not appear to be an artifact of the SIMS procedure (19). In the context of the cell aggregates analyzed, the large δ13C variations we observed may reflect differences in the relative proportion of bacterial and archaeal biomass; heterogeneity in physiological status; dilution of the isotopic signal by contaminating materials, such as sediment particles (20); or isotope effects caused by methane depletion.

We also observed significant variation in δ13C between the outer and inner portions of some ANME-2/DSS cell clusters, which exhibit a distinctive isotopic trend to lower δ13C values with increasing penetration into the aggregate (Figs. 1 and 3). Notably, the average isotopic variation within single ANME-2/DSS aggregates was significantly greater (18‰; n = 8) than aggregates composed of other microbial species (5‰;n = 5). This observation suggests a highly13C-depleted inner core of methane-oxidizing archaea, surrounded by a somewhat less 13C-depleted outer shell of sulfate-reducing bacteria. These cellular data are concordant with isotopic analyses of lipid biomarkers extracted from the same samples (Table 1). The somewhat lower (by about –10 to –20‰) δ13C values of the archaeal lipids compared with the total cell carbon determined by SIMS is in agreement with previous observations of isotopic compositions of individual lipids versus biomass of Methanosarcina barkeri grown on trimethylamine (21). The isotope patterns we observed in the total carbon of individual ANME-2/DSS cell aggregates are consistent with the transfer of methane-derived intermediates (possibly acetate or CO2), in addition to hydrogen, from methane-consuming archaea to their sulfate-reducing bacterial partners (22, 23).

Although ion microprobe mass spectrometry has been used in diverse applications in the earth and planetary sciences, ranging from interplanetary dust particles (14) to microfossil analyses (13), this technique had not yet been applied to active microbial cells from environmental samples. The ability to microscopically characterize microbial cells directly with FISH-SIMS provides an effective strategy for identifying additional microbial groups participating in the anaerobic oxidation of methane, as chemotaxonomic and phylogenetic evidence indicates diverse microbial assemblages are involved in this process (6,24–26). This approach can provide direct information on the identity of environmentally relevant microorganisms, as well as their metabolic activities and ecological interactions, by using either naturally occurring or exogenously added stable isotopes as tracers.

  • * These authors contributed equally to the work.

  • To whom correspondence should be addressed. E-mail: chouse{at} and delong{at}


View Abstract

Stay Connected to Science

Navigate This Article