Anaerobic Oxidation of Ethane by Archaea from a Marine Hydrocarbon Seep Song-Can Chen1,2, Niculina Musat1, Oliver J

Letter https://doi.org/10.1038/s41586-019-1063-0

Anaerobic oxidation of ethane by archaea from a marine hydrocarbon seep Song-Can Chen1,2, Niculina Musat1, Oliver J. Lechtenfeld3, Heidrun Paschke3, Matthias Schmidt1, Nedal Said1, Denny Popp4, Federica Calabrese1, Hryhoriy Stryhanyuk1, Ulrike Jaekel5,8, Yong-Guan Zhu2,6, Samantha B. Joye7, Hans-Hermann Richnow1, Friedrich Widdel5 & Florin Musat1,5*

Ethane is the second most abundant component of natural gas same basic mechanism—the formation of the thioethers propyl- or in addition to methane, and—similar to methane—is chemically butyl-coenzyme M9,21,22. By contrast, the alkane-degrading bacteria unreactive. The biological consumption of ethane under anoxic couple propane or butane oxidation to sulfate reduction in the same conditions was suggested by geochemical profiles at marine cell and initiate alkane oxidation by reaction with fumarate, yielding hydrocarbon seeps1–3, and through ethane-dependent sulfate alkyl-succinates6,23. reduction in slurries4–7. Nevertheless, the microorganisms and Ethane is the second most abundant hydrocarbon in gas of thermo- reactions that catalyse this process have to date remained unknown8. genic origin, often exceeding 10% by volume24, but is the least stud- Here we describe ethane-oxidizing archaea that were obtained ied with respect to anaerobic biodegradation8. Although its anaerobic by specific enrichment over ten years, and analyse these archaea utilization has been measured as ethane-dependent sulfate reduction using phylogeny-based fluorescence analyses, proteogenomics or ethane consumption in sediment slurries4–7, the microorganisms and metabolite studies. The co-culture, which oxidized ethane and reactions that catalyse this process remain unknown8. Here we completely while reducing sulfate to sulfide, was dominated by an use continued selective enrichment to reveal the microorganisms that archaeon that we name ‘Candidatus Argoarchaeum ethanivorans’; are capable of anaerobic ethane oxidation. The identified organisms other members were sulfate-reducing Deltaproteobacteria. The are related to a lineage of uncultured archaea, initiate ethane oxidation genome of Ca. Argoarchaeum contains all of the genes that are through coenzyme M thioether formation and apparently depend on necessary for a functional methyl-coenzyme M reductase, and sulfate-reducing bacteria. all subunits were detected in protein extracts. Accordingly, ethyl- A slurry with ethane-dependent sulfate reduction6 was cultivated coenzyme M (ethyl-CoM) was identified as an intermediate by (inoculum size: one-third by volume) over 10 years at 12 °C, a tempera- liquid chromatography–tandem mass spectrometry. This indicated ture that is also suitable for anaerobic methanotrophs25. This resulted in that Ca. Argoarchaeum initiates ethane oxidation by ethyl-CoM a sediment-free enrichment culture, termed Ethane12, which reduced formation, analogous to the recently described butane activation 10 mM sulfate during a period of approximately 7 months with strict by ‘Candidatus Syntrophoarchaeum’9. Proteogenomics further dependence on ethane addition. The approximate rate of ethane oxi- suggests that oxidation of intermediary acetyl-CoA to CO2 occurs dation was 4 mmol per day per gram cell dry weight, comparable with through the oxidative Wood–Ljungdahl pathway. The identification AOM rates (Supplementary Information). The molar ratios of con- of an archaeon that uses ethane (C2H6) fills a gap in our knowledge sumed ethane to formed sulfide in duplicate experiments were 1.63 and of microorganisms that specifically oxidize members of the 1.86 (Fig. 1a and Extended Data Table 1), indicating ethane oxidation homologous alkane series (CnH2n+2) without oxygen. Detection according to: of phylogenetic and functional gene markers related to those of 10–12 2−+ Ca. Argoarchaeum at deep-sea gas seeps suggests that archaea 4C 26H(g) ++7SO14 4H →+8CO(22g) 7H S(aq.+)12H2O that are able to oxidize ethane through ethyl-CoM are widespread members of the local communities fostered by venting gaseous Δ=G ′ −.73 2kJper molethane alkanes around these seeps. Natural gas venting from deep marine horizons towards the sediment Grown Ethane12 cultures were turbid and light microscopy revealed surface is a potent source of energy and carbon for microbial commu- small single cocci (with a diameter of around 0.5 μm) as the abundant nities using various electron acceptors (or ‘oxidants’). The high abun- morphotype. Amplicon and metagenome sequencing consistently dance of sulfate in seawater (a concentration of 28 mM; by contrast, retrieved a distinct archaeal phylotype affiliated with the anaerobic the concentration of oxygen is approximately 0.3 mM) often results in methanotrophic (ANME)-2d cluster of Methanosarcinales (Fig. 1b extended anoxic subsurface zones that are shaped by the reduction of and Supplementary Table 1). This phylotype, named Eth-Arch1, sulfate to sulfide13. Here, anaerobic oxidation of methane (AOM) is a constituted on average 65% of the total number of cells (Fig. 2a, b prominent process carried out by various archaea14–16. According to and Supplementary Table 2). Cells of the Eth-Arch1 phylotype grew geochemical depth profiles, ethane, propane, n-butane and iso-butane mostly unattached. Many formed protruding smaller vesicular struc- (non-methane alkanes) may also be biodegraded1–3. Involved micro- tures that had DAPI and catalysed reporter deposition–fluorescence organisms have been identified so far for the degradation of propane in situ hybridization (CARD–FISH) signals (Fig. 2f–h and Extended and n-butane, and are either bacteria or archaea6,9,17,18. Like the AOM Data Fig. 1), suggesting division by budding, which is an alternative archaea, those that oxidize propane and butane also depend on cell-division mechanism found across major archaeal phyla26. In syntrophic interactions with sulfate-reducing bacteria9,19,20. The acti- addition, two phylotypes of Desulfosarcina-affiliated sulfate-reducing vation of propane and butane in the archaea apparently involves the bacteria (SRB) were identified, Eth-SRB1 and Eth-SRB2, each

1Department of Isotope Biogeochemistry, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany. 2State Key Laboratory of Urban and Regional Ecology, Research Center for Eco- Environmental Sciences, Chinese Academy of Sciences, Beijing, China. 3Department of Analytical Chemistry, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany. 4Department of Environmental Microbiology, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany. 5Max Planck Institute for Marine Microbiology, Bremen, Germany. 6Key Laboratory of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen, China. 7Department of Marine Sciences, University of Georgia, Athens, GA, USA. 8Present address: Department for Research Infrastructures, The Research Council of Norway, Oslo, Norway. *e-mail: [email protected]

108 | NATURE | VOL 568 | 4 APRIL 2019 Letter RESEARCH

ab5 Archaeoglobales Guaymas Basin sediment (KJ569648) Ca. A. ethanivorans Amsterdam mud volcano (FJ649528)

ANME-1 Ca. M. nitroreducens 4 Ca. Methanoperedens sp. BLZ1 )

–1 Ca. S. butanivorans ANME-2ab Ca. S. caldarius 3 GoM-Arch 87 Methanosarcinaceae ANME-2c Methanocellales M. shengliensis ANME-3 Methanosaeta 0.1 2 Methanomicrobiales Desulfobacteraceae Hydrate Ridge (AM229186) Ethane, sul de (mmol l Gulf of Cadiz (FJ813539) Seabed sediment (HE803866) Eth-SRB1 Eth-SRB2 1 c Hydrothermal seep (KP091115) Gulf of Mexico sediment (AM745215) Desulfococcus oleovorans Hxd3 Desulfococcus multivorans DSM 2059 Desulfosarcina sp. BuS5 0 Desulfosarcina cetonica JCM 12296 Butane12-HyR 050 100 150 200 250 300 350 Desulfosarcina variabilis Propane12-GMe 0.05 Time (days) Desulfatibacillum alkenivorans AK-01 Fig. 1 | Ethane oxidation with sulfate and major 16S rRNA gene diamonds). Similar results were obtained with n = 4 biological replicates. phylotypes. a, The Ethane12 culture oxidized ethane (two different starting b, Candidatus Argoarchaeum ethanivorans is affiliated with a lineage of concentrations of ethane; black and grey diamonds, 2.7 and 1.6 mmol l−1, uncultured Methanosarcinales. c, Eth-SRB1 and Eth-SRB2 share over 95% respectively) while reducing sulfate to sulfide (corresponding black and 16S rRNA gene identity and belong to the marine SEEP-SRB1 group of the grey circles). Cultures without ethane did not produce sulfide (white Desulfosarcina–Desulfococcus clade. Filled circles on branch nodes indicate circles). Ethane concentrations were stable in sterile incubations (white bootstrap values >60%; scale bars, nucleotide subsitutions per site (b, c).

− accounting for about 15% of the total number of cells (Figs. 1c, 2a, b and ethenesulfonate (C2H3O3S ), respectively. Identical mass peaks and Supplementary Table 2). Their cells were morphologically distinct, and fragments were obtained with synthetic ethyl-CoM (Fig. 3b, c). appearing as slightly curved rods (1.0–1.5 μm by 0.3 μm) and larger These findings were further corroborated by liquid chromatography– ellipsoids (2.0–2.5 μm by 1 μm), respectively; they were also found tandem mass spectrometry (LC–MS/MS) analyses, yielding identical predominantly as single cells (Fig. 2a, b). An estimated 10% of the total retention time and mass transitions for extracted metabolites and the culture biomass occurred as archaeal–bacterial aggregates (Fig. 2c–e). synthetic standard (Fig. 3d–f). Because the abundance of the archaeal phylotype suggested a key We conclude that Eth-Arch1 archaea use the MCR-like enzyme role in ethane oxidation, we examined its catabolism through metage- to activate ethane to ethyl-CoM, similar to propane and n-butane nome, metaproteome and metabolite analyses. One bin of 1.99 Mb activation in thermophilic archaea9. All analyses are thus consistent from the metagenome assembly corresponded to Eth-Arch1. Marker with Eth-Arch1 being the primary ethane-degrading microorganism; genes for Euryarchaeota and Archaea indicated a genome complete- we therefore name it ‘Candidatus Argoarchaeum ethanivorans’. The ness of 89–94% (Supplementary Table 3). The Eth-Arch1 genome bin description of the taxon is as follows: Argoarchaeum, argós (Greek): contained all genes (mcrABG) for a functional methyl-coenzyme M slow, unhurried, archaeum from archaeon (Greek): an ancient life reductase (MCR)-like enzyme. In addition, their gene products were form; ethanivorans, ethane (hydrocarbon) from aithérios (Greek): airy, detected in protein extracts (Extended Data Table 2). The large sub- gaseous, vorans (Latin): eating, devouring. The name implies a unit clustered closely to a McrA type that has been identified in Ca. slow-growing archaeon capable of ethane oxidation. Syntrophoarchaeum9; both are divergent from the McrA of methano- Further metagenomics and metaproteomics analyses of Ca. gens (Fig. 3a). Other MCR-encoding genes were not detected. Genes Argoarchaeum predict terminal oxidation of the ethane-derived C2 that encode (methyl)alkylsuccinate synthases, which would indicate a unit through cleavage of acetyl-CoA by acetyl-CoA decarbonylase/ reaction of ethane with fumarate, were not found in the genomes of Eth- synthase (ACDS), and stepwise dehydrogenation of the derived Arch1, Eth-SRB1 and Eth-SRB2, or in the whole metagenome library. C1-units (oxidative Wood–Ljungdahl pathway) (Extended Data The Fourier transform ion cyclotron resonance mass spectrometry Fig. 2, Extended Data Tables 2, 3). The reactions correspond—in (FT–ICR–MS) spectrum of extractable metabolites from Ethane12 cul- principle—to those in acetoclastic methanogenesis and AOM, respec- 27 tures revealed a peak of m/z = 168.9999 (Fig. 3b), which corresponds tively . Beta-oxidation, as noted for the butane-derived C4 unit in Ca. − exactly to a predicted ethyl-coM anion (C4H9O3S2 ). Subsequently Syntrophoarchaeum, is not needed in Ca. Argoarchaeum; its metage- produced fragments of this mass peak had m/z values of 80.9652 and nome did not show evidence of such a pathway. The reactions for the − 106.9808 and were identified as ethyl-CoM-derived bisulfite (HSO3 ) conversion of the thioether (ethyl-CoM) to the thioester (acetyl-CoA)

a b c f h

Fig. 2 | Microscopic characterization of the Ethane12 culture. microscopy images of Ca. A. ethanivorans displayed vesicular structures. a–e, Fluorescence upon specific probing of Ca. A. ethanivorans Representative of n = 20 recorded images. h, The violet colour from the (a–e, red), the Eth-SRB1 phylotype (a, green) or both Eth-SRB overlay of Ca. A. ethanivorans probe and DAPI signals indicates nucleic phylotypes (b–e, green). Images are representative of n = 50 recorded acids (selected budding cells). Representative of n = 10 recorded images. images. Aggregates were rare, of 10–20 μm in diameter and consisted Scale bars, 5 μm (a–e, h) and 500 nm (f, g). of varying proportions of archaea and bacteria (c–e). f, g, Helium ion

4 APRIL 2019 | VOL 568 | NATURE | 109 RESEARCH Letter

SCAL_001148 CH C H C H n-C H SO 2– a KT387810 KT387806 ab4 2 6 cd3 8 4 10 4 KT387805 SBU_001343 SBU_000314 SCAL_001725 MCR ‘MCR’ ‘MCR’ ‘MCR’ SBU_001328

CH -SCoM C2H5-SCoM C3H7-SCoM C4H9-SCoM SBU_000718 3 ? ? SCAL_000921 ? C2H5CO-SCoA C3H7CO-SCoA MM-CoA pathway? Beta-oxidation Coupling AEth_00344 [Methyl] CH CO-SCoA CH CO-SCoA CH CO-SCoA PHP46140 to 3 3 3 OYT62528 [Formyl] [Methyl] [CO] [Methyl] [CO] [Methyl] [CO] to to to ANME-2d SCAL_000352 [Formyl] [Formyl] [Formyl]

Methanosarcinaceae CO2 CO2 CO2 CO2 CO2 CO2 CO2 H2S Methanosaetaceae Methanomassiliicoccaceae ANME-2c ef g Methanocella ANME-1 CH4 CH4 CH4 Methanobacteriales 0.5 Methanopyrus MCR MCR MCR Methanococcales Methanomicrobiales CH3-SCoM CH3-SCoM CH3-SCoM CH3OH

Culture + ethane [Methyl] b Ethyl-CoM standard (2 × 10–2) c [Methyl] [Methyl] Sterile control + ethane to to CH CO-SCoA CH COO– [Formyl] 3 3 [Formyl] – – [C H O S]– [C4H9O3S2] [HSO3] 2 3 3 ethyl-CoM bisul te ethenesulfonate [CO]

CO2 CO2 CO2 Relative intensity Relative intensity Relative intensity Fig. 4 | Basic reactions in archaeal oxidation of gaseous alkanes compared to methanogenesis. a–d, Viewing AOM (a) as the ‘archetype’, oxidation of 168.997 168.999 169.001 169.003 80.964 80.965 80.966 106.979 106.981 106.983 m/z m/zm/z non-methane alkanes (b–d) was apparently achieved by modified methyl- coenzyme M reductases (‘MCR’) and acquisition of metabolic modules. deCulture + ethane f Ethyl-CoM standard Reactions converting alkyl-thioethers to acyl-thioesters (b–d, red) are Solvent blank presently unknown. Oxidation of propionyl-CoA and butyryl-CoA may – [C H O S ]– [C H O S ]– [C4H9O3S2] 4 9 3 2 4 9 3 2 involve the methylmalonyl-CoA pathway (c, purple; not yet validated) and beta-oxidation (d, purple), respectively. e, Methanogenesis from CO2 has [C H S]– [HSO ]– [C H O S]– 2 5 3 2 3 3 been the basis for understanding AOM and oxidation of other alkanes by archaea. b–d, f, Cleavage of acetyl-CoA (blue) and subsequent oxidation

Relative abundance Relative abundance Relative abundance of bound CO is known from acetoclastic methanogens. a, Oxidation of the methyl moiety may occur as in AOM. g, The overall reaction resembles the 345 345 345 Retention time (min) Retention time (min) Retention time (min) oxidative branch in methanogenesis from methanol. [Methyl] to [Formyl] summarizes the reaction sequence of the cofactor-bound C1 units. Fig. 3 | Phylogeny of McrA and identification of ethyl-coM. a, The McrA of Ca. A. ethanivorans (red) branches with McrA of Ca. Syntrophoarchaeum (green) and unidentified microorganisms at marine This is, to our knowledge, the first identification of an ethane- hydrocarbon-impacted settings (blue). Filled circles on branch nodes degrading anaerobe, closing the microorganism-related gap in our indicate bootstrap values >80%. Scale bar, amino acid substitutions per site. understanding of the biodegradability of members of the homologous b, c, FT–ICR–MS of Ethane12 metabolite extracts identified a mass peak alkane series in the absence of oxygen. There is a phylogenetic rela- corresponding to ethyl-CoM (m/z = 168.9999) (b) and the fragment masses tionship and metabolic similarity between Ca. Argoarchaeum and the of ethyl-CoM-derived bisulfite and ethenesulfonate (c). Similar results were clade of methanogenic and methanotropic archaea (Figs. 1 and 4). The obtained with n = 8 independent cultures. d–f, LC–MS/MS in multiple- unique MCR reaction may have provided an evolutionarily success- reaction monitoring mode confirmed the metabolite structure as ethyl- ful, mechanistic trait for the formation and cleavage of primary apolar CoM, with all three characteristic fragments at a retention time of 3.715 min. C–H bonds. Ethane is chemically most closely related to methane, and synthetic ethyl-CoM has been used as an analogue of methyl-CoM. In remain unknown. Candidatus Argoarchaeum encodes a complete such a study, C–H bond cleavage in ethane has been analysed as the canonical tetrahydromethanopterin S-methyltransferase, of which back reaction during net ethyl-CoM conversion to ethane by MCR four subunits were detected in protein extracts (Extended Data from Methanothermobacter marburgensis28. For the archaea with Table 2). Similar to Ca. Syntrophoarchaeum9, this finding suggests the genuine metabolism of ethane, propane and butane, one may expect possibility of a previously undescribed transfer of ethyl (rather than a specific substrate–enzyme fit. Structural modelling of MCR from methyl) from ethyl-CoM to a tetrahydropterin for further oxidation Ca. Argoarchaeum and Ca. Syntrophoarcheum based on the crystal to the acetyl level. structure of MCR (from Methanothermobacter29) revealed differences Candidatus Argoarchaeum lacks established sulfate-reduction between the amino acid sequences of the catalytic pocket (Extended enzymes, indicating a syntrophic interaction with the sulfate- Data Fig. 4). However, matches between the size of the hydrocarbon reducing bacteria. Both sulfate-reducing microorganisms encode molecules and the suggested pocket are partly ambiguous. Three MCR multi-haem cytochromes and type IV pili, similar to sulfate-reducing types from Ca. Syntrophoarcheum exhibited replacement of some aro- bacteria partners of anaerobic methanotrophic archaea and Ca. matic amino acids (most obviously Phe330 and Tyr333 of McrA) by less Syntrophoarchaeum9,19,20 (Extended Data Tables 4, 5); however, space-filling aliphatic residues (Gly, Ala or Thr). By contrast, another nanowire-mediated direct electron transfer is not supported by the Ca. Syntrophoarchaeum enzyme and that from Ca. Argoarchaeum did observation of predominantly planktonic growth of Ca. Argoarchaeum not show exchanges that reflect binding of substrates that are bulkier and the absence of nanowire-like structures (Extended Data Fig. 1). than methane (Extended Data Fig. 4c–g). The proposed enzymes for Instead, the high enrichment of sulfur in Ca. Argoarchaeum cells utilization of the non-methane gaseous alkanes apparently belong to an compared to the bacterial cells (as revealed by nanoscale secondary offset cluster within the MCRs, with clear separation from the enzymes ion mass spectrometry (nanoSIMS); Extended Data Fig. 3) suggests involved in methanogenesis or AOM. Still, McrA from ethane-, pro- an interspecies interaction similar to that found in cold-adapted AOM pane- and butane-degrading microorganisms are distantly related to consortia; archaea in these consortia have been proposed to foster their each other (Fig. 3a). This suggests that adaptation to higher alkanes may bacterial partners by a diffusible sulfur species16. have occurred through independent evolutionary events. The entire

110 | NATURE | VOL 568 | 4 APRIL 2019 Letter RESEARCH tree may be regarded as a family of ‘alkyl-coenzyme M reductases’, 18. Savage, K. N. et al. Biodegradation of low-molecular-weight alkanes under mesophilic, sulfate-reducing conditions: metabolic intermediates and acting according to community patterns. FEMS Microbiol. Ecol. 72, 485–495 (2010). 19. McGlynn, S. E., Chadwick, G. L., Kempes, C. P. & Orphan, V. J. Single cell activity AlkylSCoMH++SCoBa lkaneCoMS–SCoB reveals direct electron transfer in methanotrophic consortia. Nature 526, 531–535 (2015). In summary, oxidation of the non-methane gaseous alkanes in 20. Wegener, G., Krukenberg, V., Riedel, D., Tegetmeyer, H. E. & Boetius, A. Intercellular wiring enables electron transfer between methanotrophic archaea comparison to AOM requires adaptation of the activating enzyme and bacteria. Nature 526, 587–590 (2015). and additional enzymatic reactions (Fig. 4). The latter may be viewed 21. Scheller, S., Goenrich, M., Boecher, R., Thauer, R. K. & Jaun, B. The key nickel as metabolic modules, which raises the question of their evolution- enzyme of methanogenesis catalyses the anaerobic oxidation of methane. ary acquisition. Whereas ACDS for acetyl-CoA synthesis or cleavage Nature 465, 606–608 (2010). 30 22. Thauer, R. K. Anaerobic oxidation of methane with sulfate: on the reversibility of shows vertical inheritance in archaea , other modules may have been the reactions that are catalyzed by enzymes also involved in methanogenesis acquired from unrelated microorganisms. This was found to be the from CO2. Curr. Opin. Microbiol. 14, 292–299 (2011). case in Ca. Syntrophoarchaeum, in which the entire beta-oxidation 23. Callaghan, A. V. Metabolomic investigations of anaerobic hydrocarbon- impacted environments. Curr. Opin. Biotechnol. 24, 506–515 (2013). pathway was probably acquired through lateral gene transfer from 24. Milkov, A. V. & Etiope, G. Revised genetic diagrams for natural gases based on a sulfate-reducing bacteria9. From such a perspective, an analogous global dataset of >20,000 samples. Org. Geochem. 125, 109–120 (2018). modular extension of methanogenesis to the biological formation 25. Nauhaus, K., Boetius, A., Krüger, M. & Widdel, F. In vitro demonstration of anaerobic oxidation of methane coupled to sulphate reduction in sediment of ethane, propane and butane appears to be possible, through still from a marine gas hydrate area. Environ. Microbiol. 4, 296–305 (2002). unknown natural processes or metabolic engineering. In addition, the 26. Makarova, K. S., Yutin, N., Bell, S. D. & Koonin, E. V. Evolution of diverse cell present finding of Ca. Argoarchaeum further refines our functional division and vesicle formation systems in Archaea. Nat. Rev. Microbiol. 8, 731–741 (2010). interpretation of environmental sequence data. Phylotypes related to 27. Thauer, R. K. Biochemistry of methanogenesis: a tribute to Marjory Ca. Argoarchaeum have frequently been retrieved from geographically Stephenson. 1998 Marjory Stephenson Prize Lecture. Microbiology 144, distinct marine hydrocarbon seeps where ethane was abundant 2377–2406 (1998). 28. Scheller, S., Goenrich, M., Thauer, R. K. & Jaun, B. Methyl-coenzyme M (Extended Data Table 6). Together with the detection of Ca. reductase from methanogenic archaea: isotope efects on label exchange and 12 10,11 Syntrophoarchaeum and of divergent mcr genes this may indeed ethane formation with the homologous substrate ethyl-coenzyme M. J. Am. reflect in situ biological activity towards non-methane gaseous alkanes. Chem. Soc. 135, 14985–14995 (2013). 29. Ermler, U., Grabarse, W., Shima, S., Goubeaud, M. & Thauer, R. K. Crystal structure of methyl-coenzyme M reductase: the key enzyme of biological Online content methane formation. Science 278, 1457–1462 (1997). Any methods, additional references, Nature Research reporting summaries, source 30. Adam, P. S., Borrel, G. & Gribaldo, S. Evolutionary history of carbon monoxide data, statements of data availability and associated accession codes are available at dehydrogenase/acetyl-CoA synthase, one of the oldest enzymatic complexes. https://doi.org/10.1038/s41586-019-1063-0. Proc. Natl Acad. Sci. USA 115, E1166–E1173 (2018). Received: 19 October 2018; Accepted: 27 February 2019; Acknowledgements We acknowledge R. Appel for assistance in cultivation and chemical analyses, K. Nerlich for CARD–FISH, J. M. Kaesler for FT–ICR– Published online 27 March 2019. MS analyses, B. Scheer for proteomics analyses and K. T. Konstantinidis for support with metagenomics analyses. Research was funded by the Helmholtz 1. Stadnitskaia, A. et al. Molecular and carbon isotopic variability of hydrocarbon Association and by the Max Planck Society. Further financial support was gases from mud volcanoes in the Gulf of Cadiz, NE Atlantic. Mar. Pet. Geol. 23, provided by the Helmholtz Association (grant ERC-RA-0020 to F.M.), the 281–296 (2006). Strategic Priority Research Program of the Chinese Academy of Sciences 2. Mastalerz, V., de Lange, G. J. & Dahlmann, A. Diferential aerobic and anaerobic (grants XDB15020302 and XDB15020402 to Y.-G.Z.) and the Chinese oxidation of hydrocarbon gases discharged at mud volcanoes in the Nile Scholarship Council (scholarship to S.-C.C.). We thank the shipboard science deep-sea fan. Geochim. Cosmochim. Acta 73, 3849–3863 (2009). parties and submersible operation teams of the RV Seward Johnson II. 3. Sassen, R. et al. Free hydrocarbon gas, gas hydrate, and authigenic minerals in Funding for the Gulf of Mexico cruise was provided by the US National chemosynthetic communities of the northern Gulf of Mexico continental slope: Science Foundation (OCE-0085549) and the ACS Petroleum Research relation to microbial processes. Chem. Geol. 205, 195–217 (2004). Fund (PRF-36834-AC2). The US Department of Energy and the US National 4. Adams, M. M., Hoarfrost, A. L., Bose, A., Joye, S. B. & Girguis, P. R. Anaerobic Undersea Research Program provided funding for submersible operations. We oxidation of short-chain alkanes in hydrothermal sediments: potential infuences acknowledge the Centre for Chemical Microscopy (ProVIS) at the Helmholtz on sulfur cycling and microbial diversity. Front. Microbiol. 4, 110 (2013). Centre for Environmental Research for the use of their analytical facilities. 5. Bose, A., Rogers, D. R., Adams, M. M., Joye, S. B. & Girguis, P. R. ProVIS is supported by European Regional Development Funds (EFRE – Europe Geomicrobiological linkages between short-chain alkane consumption and funds Saxony). sulfate reduction rates in seep sediments. Front. Microbiol. 4, 386 (2013). 6. Kniemeyer, O. et al. Anaerobic oxidation of short-chain hydrocarbons by marine Reviewer information Nature thanks Mike Jetten, Derek Lovley, Rudolf K. Thauer sulphate-reducing bacteria. Nature 449, 898–901 (2007). and the other anonymous reviewer(s) for their contribution to the peer review 7. Suarez-Zuluaga, D. A., Weijma, J., Timmers, P. H. A. & Buisman, C. J. N. High of this work. rates of anaerobic oxidation of methane, ethane and propane coupled to thiosulphate reduction. Environ. Sci. Pollut. Res. Int. 22, 3697–3704 (2015). Author contributions S.-C.C., H.-H.R., F.W. and F.M. designed the research. 8. Singh, R., Guzman, M. S. & Bose, A. Anaerobic oxidation of ethane, propane, and S.B.J. retrieved the original sediment sample and contributed to establishment butane by marine microbes: a mini review. Front. Microbiol. 8, 2056 (2017). of the enrichment. U.J. and F.M. performed cultivation and physiology 9. Laso-Pérez, R. et al. Thermophilic archaea activate butane via alkyl-coenzyme experiments. N.M. designed CARD–FISH probes and performed CARD–FISH M formation. Nature 539, 396–401 (2016). and fluorescence microscopy. M.S., N.S. and N.M. designed and performed 10. Dombrowski, N., Seitz, K. W., Teske, A. P. & Baker, B. J. Genomic insights into helium ion microscopy analyses. O.J.L. and H.P. performed FT–ICR–MS analyses, potential interdependencies in microbial hydrocarbon and nutrient cycling in and LC–MS/MS method development and analyses. S.-C.C., D.P., Y.-G.Z. and hydrothermal sediments. Microbiome 5, 106 (2017). F.M. performed metagenomics analyses. S.-C.C. performed phylogenetic and 11. Evans, P. N. et al. Methane metabolism in the archaeal phylum Bathyarchaeota proteomics analyses. F.C., H.S. and N.M. designed and performed nanoSIMS revealed by genome-centric metagenomics. Science 350, 434–438 (2015). analyses. S.-C.C., F.W. and F.M. wrote the manuscript with contributions from all 12. McKay, L. et al. Thermal and geochemical infuences on microbial biogeography co-authors. in the hydrothermal sediments of Guaymas Basin, Gulf of California. Environ. Microbiol. Rep. 8, 150–161 (2016). Competing interests The authors declare no competing interests. 13. Bowles, M. W., Samarkin, V. A., Bowles, K. M. & Joye, S. B. Weak coupling between sulfate reduction and the anaerobic oxidation of methane in Additional information methane-rich seafoor sediments during ex situ incubation. Geochim. Extended data is available for this paper at https://doi.org/10.1038/s41586- Cosmochim. Acta 75, 500–519 (2011). 019-1063-0. 14. Boetius, A. et al. A marine microbial consortium apparently mediating Supplementary information is available for this paper at https://doi.org/ anaerobic oxidation of methane. Nature 407, 623–626 (2000). 10.1038/s41586-019-1063-0. 15. Knittel, K. & Boetius, A. Anaerobic oxidation of methane: progress with an Reprints and permissions information is available at http://www.nature.com/ unknown process. Annu. Rev. Microbiol. 63, 311–334 (2009). reprints. 16. Milucka, J. et al. Zero-valent sulphur is a key intermediate in marine methane Correspondence and requests for materials should be addressed to F.M. oxidation. Nature 491, 541–546 (2012). Publisher’s note: Springer Nature remains neutral with regard to jurisdictional 17. Jaekel, U. et al. Anaerobic degradation of propane and butane by sulfate- claims in published maps and institutional affiliations. reducing bacteria enriched from marine hydrocarbon cold seeps. ISME J. 7, 885–895 (2013). © The Author(s), under exclusive licence to Springer Nature Limited 2019

4 APRIL 2019 | VOL 568 | NATURE | 111 RESEARCH Letter

Methods sequencing was also performed on a MinION Mk1B instrument using a R9.4.1 Data reporting. No statistical methods were used to predetermine sample size. flow cell (Oxford Nanopore Technologies). The SQK-LSK 109 Ligation Sequencing The experiments were not randomized and the investigators were not blinded to Kit was used for library preparation using 1 μg genomic DNA according to the allocation during experiments and outcome assessment. manufacturer’s instructions, except for increasing the end-repair incubation time Etymology. Argoarchaeum, argós (Greek): slow, unhurried, archaeum from to 30 min at room temperature and 30 min at 65 °C and of the ligation step to archaeon (Greek): an ancient life form; ethanivorans, ethane (hydrocarbon) from 60 min. Raw sequence data were base-called using Albacore v.2.3.1 and adapters aithérios (Greek): airy, gaseous, vorans (Latin): eating, devouring. The name implies were removed using Porechop v.0.2.3. a slow-growing archaeon capable of ethane oxidation. Protein-encoding genes in the bulk assembly were predicted using prodigal Locality. Enriched from cold marine hydrocarbon seeps of the Gulf of Mexico at v.2.6.3 (-p meta option). To search for enzymes potentially involved in ethane 550 m water depth, Gulf of Mexico, USA. activation, reference databases with sequences of curated methyl-coenzyme M Diagnosis. Ethane-oxidizing archaeon, mostly single-celled cocci with a diameter reductases and (methylalkyl)succinate synthases were compiled (Supplementary of 0.5 μm; grown at 12 °C, pH 7–8. Table 4). BLASTp was performed against the reference database using protein Sediment samples, enrichment and cultivation. Sediment samples from marine sequences that were predicted from the assembled metagenome with relaxed hydrocarbon seeps of the Gulf of Mexico were collected in the Green Canyon area stringency (e = 1 × 10−5). The resulting hits were further vetted by phyloge- at 550 m water depth, at sites GC232 (27° 44.4566′ N, 91° 18.981′ W) and GC234 netic analysis: sequences were aligned with reference enzymes using MUSCLE (27° 44.7003′ N, 91° 13.3093′ W). The samples were collected during a cruise v.3.8.3140, and maximum likelihood trees were built using RAxML v.8.2.941 using with the RV Seward Johnson II (Harbour Branch Oceanographic Institution) in the PROTGAMMALG model and 100 rapid bootstraps. 42 July 2002. Sediments were stored under N2 at 4 °C until use. Incubations with Metagenomic binning was conducted with MaxBin v.2.2.4 , which classifies ethane were set up in 20-ml cultivation tubes provided with 10 ml artificial seawater scaffolds (>1,000 bp) based on tetranucleotide frequency and read coverage. 16S (ASW) prepared as described previously31,32 and 2 ml sediment as inoculum. rRNA genes were extracted from bins using RNAmmer v.1.243 and aligned to the The tubes were sealed with butyl-rubber stoppers under an anoxic atmosphere Silva database release 132 using the ARB software package44 for phylogenetic clas- of N2 and CO2 (9:1 by volume) and provided with ethane at a partial pressure of sification. Bins corresponding to Methanosarcinales and SEEP-SRB1 were selected 0.1 MPa. The tubes were incubated horizontally at 12 °C with slow horizontal shak- and further refined by two rounds of read mapping, reassembly and binning. Read ing (100 r.p.m.). Clear ethane-dependent sulfide production in ethane-amended mapping was done using the bbmap tool of the BBMap v.38.00 package, with a cultures compared to sulfide production in control incubations without added minimum identity (‘minid’ option) of 90 and 97% in the first and second round of ethane was observed after about 1 year of incubation. Subsequent transfers (30% refinement, respectively. The mapped reads were reassembled in SPAdes using the volume) over 5 years in fresh culture medium led to a sediment-free enrichment same settings as for the bulk assembly, followed by binning in MaxBin. The refined culture. The sediment-free culture was further maintained by inoculating fresh cul- SPAdes contigs were scaffolded with the nanopore long reads using npScarf45 (min- ture medium with 30% volume of a grown culture (≥15 mM sulfide production). Contig = 1000, japsa v.1.7-05b), and further polished with the Illumina reads Quantitative growth experiments were done in 80-ml bottles that were provided using Pilon v.1.2346. The completeness and/or contamination of the refined bins with 45 ml ASW medium and 5 ml inoculum from a grown culture; the bottles were estimated using CheckM v.1.0.1147 and AMPHORA248. The refined bins were flushed and sealed as described above. Bottles were supplied with either were used as draft genomes, which were annotated with RAST49, KEGG50, Pfam51 1.7 or 3.0 ml ethane, which was added to the headspace with gas-tight syringes. and EggNOG52 databases, after gene prediction using prodigal (-p single option). Inoculated medium without ethane and sterile ASW medium with ethane were Predicted genes related to the proposed ethane-oxidation pathway and electron used as controls. transfer were manually curated by comparison with genes that encode enzymes Chemical analyses. For sulfide measurements, 0.1-ml samples were withdrawn with confirmed function. Encoded proteins with a similar length (<30% devia- from cultures and controls with N2-flushed syringes, and mixed with 4 ml of acid- tion), domain composition and conserved functional sites were retained. 16S rRNA ified copper sulfate solution. The formed colloidal copper sulfide was quantified genes and tRNA sequences in draft genomes were extracted using RNAmmer v.1.2 photometrically at 480 nm, as described previously33. Ethane concentrations and tRNAscan v.2.053, respectively. in the culture headspace were quantified using a gas chromatograph (GC-14B, Phylogenetic analyses. For phylogenetic analyses, representative full-length Shimadzu) equipped with a flame ionization detector and a Supel-Q PLOT column 16S rRNA gene sequences were selected from the SILVA database release 132. (30 m × 0.53 mm; film thickness, 30 μm). The gas chromatograph was operated Alignments were generated using the SINA v.1.3.1 software35, and refined by filter- −1 54 with N2 as carrier gas (flow rate, 3 ml min ); the oven temperature was 140 °C; ing out the columns that contained more than 95% gaps using trimAl v.1.2rev59 . the injector and detector temperatures were 150 °C and 280 °C, respectively34. Maximum likelihood trees were calculated with RAxML and the GTRCAT model. Headspace samples (100 μl) were withdrawn with N2-flushed, gas-tight syringes Then 100 rapid bootstrap analyses were performed to determine the support value and injected into the injection port with a 1:10 split ratio. Given sulfide and ethane for each branch. For McrA phylogenetic analyses, sequences were aligned with concentrations (Fig. 1a) are mean values of duplicate measurements (technical MUSCLE, followed by removal of the ambiguous sites using trimAl (-automated1 replicates). option). Phylogenetic trees of McrA were constructed based on the trimmed align- Nucleic acid extraction and community sequencing. For DNA extraction, the ment using RAxML with the PROTGAMMALG evolutionary model and empirical cells from 20-ml grown cultures were collected by centrifugation (20 min, 16,000g, base frequencies. Over 200 bootstrap replicates were conducted to generate branch 4 °C; ROTINA 380R, Hettich) and suspended in 1.35 ml extraction buffer (100 mM support values, according to autoMRE bootstopping criteria. Tris-HCl, 100 mM sodium EDTA, 100 mM sodium phosphate, 1.5 M NaCl, 1% Probe design. The specific oligonucleotide probe pETARCH669 (targeting CTAB, pH 8.0). Lysis was achieved by three cycles of freezing (liquid nitrogen) and Eth-Arch1) and helpers hETARCH589 and hETARCH627 (Supplementary thawing (37 °C), followed by incubation with proteinase K (10 mg ml−1) at 37 °C for Table 5) were designed using the Probe Design tool of the ARB v.6.0.2 software 30 min, and with 20% SDS at 65 °C for 1 h. The supernatant was extracted with an package55. Probe specificity was checked against the SILVA database56 and the equal volume of chloroform:isoamyl alcohol (24:1, v/v). Nucleic acids were precipi- Ribosomal Database Project57. The stringency of the formamide concentration tated with 0.6 volumes of isopropanol (1 h, room temperature), collected by centrif- was determined in hybridization assays with increasing formamide concentrations ugation (30 min, 21,000g, 4 °C), suspended in 40 μl PCR water and stored at −20 °C. from 0 to 60% (10% increments). Microscopic inspection showed the brightest Amplicon sequencing for the 16S rRNA gene (MiSeq; 2 × 300 cycles) was done fluorescence signal for formamide concentrations between 20 and 35%; a second using the universal prokaryotic primers 341F (5′-CCTACGGGNGGCWGCAG-3′) stringency assay was performed with formamide concentrations from 20 to 45%, and 785R (5′-GACTACHVGGGTATCTAATCC-3′). The paired-end reads were at 5% increments. merged using BBMerge 34.48 (http://bbmap.sourceforge.net/) after clipping the CARD–FISH. For CARD–FISH, 1-ml samples of Ethane12 cultures were fixed adaptors and primers. The combined sequences were analysed using the SilvaNGS for 17 h at 4 °C with 1 ml of 4% paraformaldehyde (electron microscopy grade; pipeline35,36. Electron Microscopy Sciences). Volumes of 150, 250 and 500 μl were filtered on Metagenome sequencing and data analysis. Sequencing of paired-end libraries gold–palladium-coated filters (0.22-μm pore size, GTTP type, Millipore), washed was performed on an Illumina MiSeq V3 platform (2 × 300 cycles), which gen- three times with 1× PBS, dehydrated for 1 min with 80% ethanol and dried at erated about 2 million reads. The paired-end Illumina reads were demultiplexed room temperature. The filters were stored at −20 °C. Hybridizations were per- using the Illumina bcl2fastq v.1.8.4 software, and quality trimmed after removal formed as described elsewhere58. Filters were coated with 0.2% low-melting- of sequencing-adaptor remnants using Trimmomatic v.0.3337 (minimum average point agarose kept at 48 °C (Biozym Scientific) using a spin-coater type SCI-40 Phred quality score of 33; minimum read length ≥36 bp). The quality-trimmed (LOT-QuantumDesign) at 50 r.p.m. Bacteria were permeabilized with lysozyme reads were assembled using metagenome mode (-meta option) of SPAdes v.3.11.138 (10 mg ml−1 in 0.05 M EDTA pH 8.0, 0.1 M Tris-HCl pH 7.5) for 30 min at 37 °C. with the BayesHammer error-correction step and k-mer sizes of 21, 33, 55, 77, 99 Archaea were permeabilized with 0.1 M HCl for 1 min, followed by incubation with and 127. The quality of the assembly was inspected with metaQUAST v.4.6.339 proteinase K (15 μg ml−1) for 5 min at room temperature. Endogenous peroxidases and scaffolds ≥500 bases were selected for downstream analyses. Metagenomics were inactivated by incubation in 0.15% H2O2 in absolute methanol (30 min, room Letter RESEARCH temperature). The filters were hybridized for 2.5 h at 46 °C in standard hybrid- ammonium bicarbonate buffer, and resuspended in 1 ml of acetonitrile: ization buffer58. The HRP-probe concentration was 0.166 ng ml−1. Probes used methanol:water (40:40:20, v/v). Around 0.3 g glass beads (0.1-mm diameter, and corresponding hybridization conditions are shown in Supplementary Table 5. Roth) was added to each tube. Cells were lysed using a bead-based homogenizer Hybridized filters were incubated for 15 min at 48 °C in prewarmed washing buffer. (PowerLyzer 24 bench, MO BIO Laboratories) operated for 5 cycles of 50 s recip- CARD was performed for 15 min at 46 °C in the dark in standard amplification rocal shaking at 2,000 r.p.m. and a 15-s pause. The extracts were separated from buffer58 containing either 1 μg ml−1 Alexa Fluor 488- or Alexa Fluor 594-labelled cell debris and glass beads by centrifugation (10 min at 21,000g, 4 °C), and stored tyramides. Tyramides were prepared from the corresponding succinimidyl esters, at −20 °C until analysis. Alexa Fluor 488 NHS Ester and Alexa Fluor 594 NHS Ester (Thermo Fisher Mass spectrometry of cell extracts and standards. Synthetic ethyl-coM Scientific), as previously described59. The hybridized cells were further stained for (approximately 10 μg ml−1) and extracted cellular metabolites were measured 10 min with 1 μg ml−1 of 4′,6′-diamidino-2-phenylindol (DAPI). For fluorescence with ultra-high-resolution FT–ICR–MS (SolariX XR 12T, Bruker Daltonics) with microscopy, the filters were embedded in a 4:1 (v/v) mixture of low fluorescence negative electrospray ionization (Apollo II ESI source capillary voltage: 4.5 kV) glycerol mountant (Citifluor AF1, Citifluor) and mounting fluid VectaShield (Vecta in direct infusion mode (4 μl min−1). Spectra were recorded with an 8 MWord Laboratories). Hybridizations were evaluated by fluorescence microscopy using an time domain (0.84 s transient length) and 2 s accumulation time in magnitude Axio Imager.Z2 microscope (Carl Zeiss) with a 100× Plan-Apochromat objective mode between 37 and 1,000 m/z, resulting in a mass resolution of approximately (1.4 NA) and filter sets for DAPI, Alexa Fluor 594 and Alexa Fluor 488. 450,000 at m/z 200. The instrument was linearly calibrated with NaTFA cluster Dual hybridizations were performed using a combination of HRP- between 113 and 521 m/z, resulting in an average root-mean square error of the pETARCH669 + HRP-SEEP1f-153 or HRP-pETARCH669 + HRP-DSS658. For calibration masses of 32 p.p.b. (n = 4). No lock mass or internal calibration were both hybridizations, Alexa Fluor 594 was used for the hybridization of Eth-Arch1, used for the sample mass spectra, resulting in an ethyl-coM standard mass accu- whereas Alexa Fluor 488 was used for the hybridization of Eth-SRB1/2. The dual racy of <0.01 p.p.m. For each measurement of metabolite extracts, 64 spectra CARD–FISH procedures were similar to those described above except that cell- were co-added with quadrupole preselection of the mass window of ethyl-coM wall permeabilization was done sequentially for bacteria and archaea, respectively. (169 ± 10 m/z). Collision-induced fragmentation was carried out after quadrupole − Before the second hybridization, the HRP introduced in the first hybridization was preselection with 7 V collision energy. Fragment masses of 61.0117 (C2H5S ), − − inactivated by incubation for 10 min at room temperature with 3% H2O2. Standard 80.9652 (HSO3 ) and 106.9808 (C2H3SO3 ) were used as indicative for ethyl-coM, mounting and epifluorescence microscopy was used for visualization. in accordance with previous studies9,62. The mass error for the ethyl-CoM mass Helium ion microscopy. Culture samples of 1 ml were withdrawn with N2-flushed peak and its derived fragments was <0.05 p.p.m. (ethyl-CoM), <0.2 p.p.m. (ethyl- syringes and fixed for 12 h at 4 °C in 3% glutaraldehyde prepared in 0.2 M sodium CoM-derived bisulfite) and <0.05 p.p.m. (ethyl-CoM-derived ethenesulfonate). cacodylate buffer. The fixed samples were filtered on polycarbonate filters (0.22- Fragment information of the ethyl-coM standard was used to implement a μm pore size), rinsed twice with 0.2 M sodium cacodylate buffer, washed with LC–MS/MS method to confirm the presence of ethyl-coM in extracts. A triple deionized water for 3 min and post-fixed for 45 min at room temperature in 1% quadrupole mass spectrometer (Xevo TQ-S, Waters Corporation) in negative osmium tetroxide solution in 0.2 M sodium cacodylate buffer. The samples were electrospray ionization mode (capillary voltage: 1 kV) was used in multiple- dehydrated in an ethanol series (30, 50, 70, 80, 90 and 100%; 3 min each) and crit- reaction monitoring mode. All three ethyl-coM transitions (m/z 169 to 107, ical point dried for 20 exchange cycles using a LEICA EM CPD300 Critical Point m/z 169 to 81 and m/z 169 to 61) were initially optimized (cone voltage 52 V and Dryer (Leica). Filter pieces of about 2.5 mm2 were cut, glued on standard scan- collision energy 14–16 V) by direct infusion of standard solution (approximately ning electron microscopy stubs with conductive silver epoxy (ACHESON DAG 10 ng ml−1) into the mass spectrometer. The mass spectrometer was coupled to 1415, Plano) and placed on the helium ion microscope sample holder. Helium ion an ultra-high performance liquid chromatograph (ACQUITY I-Class, Waters) microscopy imaging was done with a He-ion landing energy of 25 kV and a beam equipped with a reversed phase column (ACQUITY UPLC HSS T3 1.8 μm; current of about 1.3 pA. For imaging, secondary electrons were detected using an 2.1 mm × 100 mm, Waters; 30 °C) and run with a binary gradient (eluent A: Everhard–Thorley detector; image resolution, measured directly on the sample 2 mM ammonium acetate in 95% H2O and 5% methanol; eluent B: 2 mM ammo- (edges of filter pores), was <3 nm. Acquired images were minimally post-processed nium acetate in 75% methanol, 20% acetonitrile and 5% H2O) at a flow rate of in ImageJ v.1.48 by adjusting brightness and contrast. 0.25 ml min−1. Cell extracts (1 ml) were evaporated to dryness and dissolved in Shotgun proteome analysis. For protein extraction, cells from 30-ml culture 200 μl 95% H2O and 5% methanol. For each analysis, 10 μl was injected into the volumes were collected by centrifugation (16,000g, 4 °C), washed with 100 mM UPLC. Retention time and presence of all three multiple-reaction monitoring tran- ammonium bicarbonate buffer and suspended in 30 μl ammonium bicarbonate sitions were used as quality criteria. (50 mM). Cells were lysed by three freeze–thaw cycles (liquid nitrogen and 37 °C), Homology modelling of methyl-coenzyme M reductase. McrA and McrB and incubated with 50 mM dithiothreitol for 1 h at 30 °C. The reduced proteins sequences from methanogenic and methanotrophic archaea, Ca. A. ethanivorans were alkylated with 200 mM iodacetamide in the dark for 1 h at room tempera- and Ca. Syntrophoarchaeum butanivorans were aligned with MUSCLE. ture, and digested with 0.6 μg of trypsin (Promega) for 10 h at 37 °C with shaking Tertiary structures of McrA, McrB and McrG from Ca. Argoarchaeum and Ca. (400 r.p.m.). Peptides were purified with ZipTip-C18 columns (Millipore). The S. butanivorans were predicted using SWISS-MODEL (https://swissmodel.expasy. desalted peptides were analysed using an LTQ-Orbitrap Fusion mass spectrome- org/) with the default parameters63. The models have qualitative model energy ter (Thermo Fisher Scientific) in tandem with a nanoUPLC system (nanoAquity, analysis64 scores ranging from −3.81 to −0.44 (Supplementary Table 6). The Waters)60. modelled structures were superimposed on the crystal structure of M. marbur- The MS/MS spectra were searched against the annotated metagenomes of gensis MCR. The active sites of MCR were visualized and exported as images using Eth-Arch1, Eth-SRB1 and Eth-SRB2 using the Sequest and Amanda algorithms MacPyMOL v.1.7.4 (https://pymol.org/2/). in Proteome Discoverer (v.2.2.0.388, Thermo Fisher Scientific). The mass toler- Nano-focused secondary ion mass spectrometry. Ethane12 cultures (15 mM ances of precursor- and fragment-ion were set to 3 p.p.m. and 0.1 Da (Sequest), sulfide) were transferred under anoxic conditions to Oak Ridge centrifuge tubes; and 10 p.p.m. and 0.2 Da (Amanda). Proteins were considered to be identified cells were collected by centrifugation (13,300g, 12 °C; ROTINA 380R, Hettich), sus- when two full tryptic peptides were recovered at a false-discovery rate of 0.05. The pended in fresh ASW medium and provided with ethane. Samples were collected relative abundance of proteins was calculated based on the precursor-ion intensity of after 95, 110 and 120 days of incubation, fixed with 2% paraformaldehyde and the mapped peptides compared to the sum of all identified peptides for the respective added on gold–palladium-coated polycarbonate filters without further treatments. organism, using ‘Feature Mapper’ and ‘Precursor Ions Quantifier’ modules Filter pieces (10-mm diameter) were analysed with a NanoSIMS-50L instrument in Proteome Discoverer. The maximum retention time shift between replicate (CAMECA, AMETEK) in negative extraction mode using Cs+ as the primary ion Orbitrap runs was set to 10 min. The mass spectrometry data were deposited to source. Areas of 100 × 100 μm2 were pre-implanted with a Cs+ beam of 200 pA the ProteomeXchange Consortium via the PRIDE partner repository61. and 16 keV for 10 min to equilibrate the working function for negative secondary Synthesis of authentic standards. To synthesize ethyl-coM, 4 g of sodium 2-mer- ions. Fields of view of 30 × 30 μm2 were measured at 512 × 512-pixel resolution captoethanesulfonate (coenzyme M, purity ≥ 98%; Sigma-Aldrich) was dissolved with 2 ms dwell time per pixel. Secondary ion species (12CH−, 16O−, 12C14N−, 13 14 − 32 − 31 − 31 16 − in 30 ml of 30% ammonium hydroxide solution in a serum bottle. Twice the molar C N , S , P and P O2 ) were collected in parallel. A mass resolving amount of bromoethane (purity ≥99%; Sigma-Aldrich) was added and the bottle power (M/ΔM) between 8,000 and 12,000 was achieved as previously described65, was sealed and mixed at 300 r.p.m. for 4 h at room temperature. The residual bro- with the exit slits adjusted to 40 μm. For each field of view, twelve scans were moethane was removed by sparging with N2. Ethyl-CoM (m/z = 168.9999) was accumulated, corrected for lateral drift and aligned with the Look@NanoSIMS the major mass peak identified by FT–ICR–MS; major m/z peaks indicating free software66. Regions of interest were defined for individual cells based on 12C14N− CoM, CoM dimers or bromoethane were not detected. The standard was kept at and 32S− secondary ion count maps66. Scanned areas in which identification of 4 °C without further purification. cells was ambiguous (for example, cell clusters, overlapping cells) were not con- Extraction of metabolites. Cells from 30-ml culture volumes were col- sidered. To avoid loss of low-molecular-mass compounds67 or possible alteration lected by centrifugation (10 min, 16,000g, 4 °C), washed twice with 100 mM of intracellular sulfur after CARD–FISH (that is, H2O2 treatment), assignment of RESEARCH Letter regions of interest as Ca. Argoarchaeum, Eth-SRB1 and Eth-SRB2 was based on 57. Maidak, B. L. et al. The RDP (Ribosomal Database Project) continues. Nucleic their distinct morphologies. Acids Res. 28, 173–174 (2000). Reporting summary. Further information on research design is available in 58. Pernthaler, A., Pernthaler, J. & Amann, R. Fluorescence in situ hybridization and catalyzed reporter deposition for the identifcation of marine bacteria. Appl. the Nature Research Reporting Summary linked to this article. Environ. Microbiol. 68, 3094–3101 (2002). 59. Pernthaler, A. & Pernthaler, J. Fluorescence in situ hybridization for the Data availability identifcation of environmental microbes. Methods Mol. Biol. 353, 153–164 Metagenome sequence data are archived in the NCBI database under BioProject (2007). number PRJNA495932, including the draft genomes of Ca. Argoarchaeum 60. Yang, C., Kublik, A., Weidauer, C., Seiwert, B. & Adrian, L. Reductive dehalogenation of oligocyclic phenolic bromoaromatics by Dehalococcoides ethanivorans (SAMN10235260), Eth-SRB1 (SAMN10235261) and Eth-SRB2 mccartyi strain CBDB1. Environ. Sci. Technol. 49, 8497–8505 (2015). (SAMN10235262). The 16S rRNA gene amplicon reads have been submit- 61. Vizcaíno, J. A. et al. 2016 update of the PRIDE database and its related tools. ted to the NCBI Sequence Read Archive (SRA) database under the accession Nucleic Acids Res. 44, D447–D456 (2016). number SRR8089822. The proteomics dataset has been deposited with the 62. Jariwala, F. B., Wood, R. E., Nishshanka, U. & Attygalle, A. B. Formation of the − ProteomeXchange Consortium identifier PXD011597. Source Data for the quan- bisulfte anion (HSO3 , m/z 81) upon collision-induced dissociation of anions derived from organic sulfonic acids. J. Mass Spectrom. 47, 529–538 (2012). titative growth experiments (Fig. 1a), FT–ICR–MS (Fig. 3b, c) and LC–MS/MS 63. Waterhouse, A. et al. SWISS-MODEL: homology modelling of protein structures measurements (Fig. 3d–f) are provided. All other data are available in the paper and complexes. Nucleic Acids Res. 46, W296–W303 (2018). or the Supplementary Information. 64. Benkert, P., Tosatto, S. C. E. & Schomburg, D. QMEAN: a comprehensive scoring function for model quality assessment. Proteins 71, 261–277 (2008). 31. Widdel, F. in Handbook of Hydrocarbon and Lipid Microbiology Ch. 298 (ed. 65. Stryhanyuk, H. et al. Calculation of single cell assimilation rates from Timmis, K. N.) 3787–3798 (Springer, Berlin Heidelberg, 2010). SIP-NanoSIMS-derived isotope ratios: a comprehensive approach. Front. 32. Laso-Pérez, R., Krukenberg, V., Musat, F. & Wegener, G. Establishing anaerobic Microbiol. 9, 2342 (2018). hydrocarbon-degrading enrichment cultures of microorganisms under strictly 66. Polerecky, L. et al. Look@NanoSIMS—a tool for the analysis of nanoSIMS data anoxic conditions. Nat. Protoc. 13, 1310–1330 (2018). in environmental microbiology. Environ. Microbiol. 14, 1009–1023 (2012). 33. Cord-Ruwisch, R. A quick method for the determination of dissolved and 67. Musat, N. et al. The efect of FISH and CARD–FISH on the isotopic composition precipitated sulfdes in cultures of sulfate-reducing bacteria. J. Microbiol. of 13C- and 15N-labeled Pseudomonas putida cells measured by nanoSIMS. Syst. Methods 4, 33–36 (1985). Appl. Microbiol. 37, 267–276 (2014). 34. Jaekel, U., Vogt, C., Fischer, A., Richnow, H. H. & Musat, F. Carbon and hydrogen 68. Dowell, F. et al. Microbial communities in methane- and short chain alkane-rich stable isotope fractionation associated with the anaerobic degradation of hydrothermal sediments of Guaymas Basin. Front. Microbiol. 7, 17 (2016). propane and butane by marine sulfate-reducing bacteria. Environ. Microbiol. 16, 69. Welhan, J. A. & Lupton, J. E. Light hydrocarbon gases in Guaymas Basin 130–140 (2014). hydrothermal fuids: thermogenic versus abiogenic origin. Bull. Am. Assoc. 35. Pruesse, E., Peplies, J. & Glöckner, F. O. SINA: accurate high-throughput multiple Petrol. Geol. 71, 215–223 (1987). sequence alignment of ribosomal RNA genes. Bioinformatics 28, 1823–1829 70. Orcutt, B. N. et al. Impact of natural oil and higher hydrocarbons on microbial (2012). diversity, distribution, and activity in Gulf of Mexico cold-seep sediments. Deep 36. Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data Sea Res. II Top. Stud. Oceanogr. 57, 2008–2021 (2010). processing and web-based tools. Nucleic Acids Res. 41, D590–D596 (2013). 71. Pachiadaki, M. G., Kallionaki, A., Dählmann, A., De Lange, G. J. & Kormas, K. A. 37. Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a fexible trimmer for Diversity and spatial distribution of prokaryotic communities along a sediment Illumina sequence data. Bioinformatics 30, 2114–2120 (2014). vertical profle of a deep-sea mud volcano. Microb. Ecol. 62, 655–668 (2011). 38. Nurk, S., Meleshko, D., Korobeynikov, A. & Pevzner, P. A. metaSPAdes: a new 72. Pape, T. et al. Gas hydrates in shallow deposits of the Amsterdam mud volcano, versatile metagenomic assembler. Genome Res. 27, 824–834 (2017). Anaximander Mountains, Northeastern Mediterranean Sea. Geo-Mar. Lett. 30, 39. Mikheenko, A., Saveliev, V. & Gurevich, A. MetaQUAST: evaluation of 187–206 (2010). metagenome assemblies. Bioinformatics 32, 1088–1090 (2016). 73. Heijs, S. K., Laverman, A. M., Forney, L. J., Hardoim, P. R. & van Elsas, J. D. 40. Edgar, R. C. MUSCLE: multiple sequence alignment with high accuracy and Comparison of deep-sea sediment microbial communities in the Eastern high throughput. Nucleic Acids Res. 32, 1792–1797 (2004). Mediterranean. FEMS Microbiol. Ecol. 64, 362–377 (2008). 41. Stamatakis, A. RAxML version 8: a tool for phylogenetic analysis and 74. Egorov, A. V. & Ivanov, M. K. Hydrocarbon gases in sediments and mud breccia post-analysis of large phylogenies. Bioinformatics 30, 1312–1313 (2014). from the central and eastern part of the Mediterranean Ridge. Geo-Mar. Lett. 18, 42. Wu, Y. W., Simmons, B. A. & Singer, S. W. MaxBin 2.0: an automated binning 127–138 (1998). algorithm to recover genomes from multiple metagenomic datasets. 75. Robertson, A. H. F. et al. Collision-related break-up of a carbonate platform Bioinformatics 32, 605–607 (2016). (Eratosthenes Seamount) and mud volcanism on the Mediterranean Ridge: 43. Lagesen, K. et al. RNAmmer: consistent and rapid annotation of ribosomal RNA preliminary synthesis and implications of tectonic results of ODP Leg 160 in the genes. Nucleic Acids Res. 35, 3100–3108 (2007). Eastern Mediterranean Sea. Geol. Soc. Spec. Publ. 131, 243–271 (1998). 44. Westram, R. et al. in Handbook of Molecular Microbial Ecology I: Metagenomics 76. Lloyd, K. G., Lapham, L. & Teske, A. An anaerobic methane-oxidizing community and Complementary Approaches (ed. de Bruijn, F. J.) 399–406 (Wiley-Blackwell, of ANME-1b archaea in hypersaline Gulf of Mexico sediments. Appl. Environ. New Jersey, 2011). Microbiol. 72, 7218–7230 (2006). 45. Cao, M. D. et al. Scafolding and completing genome assemblies in real-time 77. Brooks, J. M. et al. Association of gas hydrates and oil seepage in the Gulf of with nanopore sequencing. Nat. Commun. 8, 14515 (2017). Mexico. Org. Geochem. 10, 221–234 (1986). 46. Walker, B. J. et al. Pilon: an integrated tool for comprehensive microbial variant 78. Maignien, L. et al. Anaerobic oxidation of methane in hypersaline cold seep detection and genome assembly improvement. PLoS ONE 9, e112963 (2014). sediments. FEMS Microbiol. Ecol. 83, 214–231 (2013). 47. Parks, D. H., Imelfort, M., Skennerton, C. T., Hugenholtz, P. & Tyson, G. W. 79. Nuzzo, M. et al. Origin of light volatile hydrocarbon gases in mud volcano fuids, CheckM: assessing the quality of microbial genomes recovered from isolates, Gulf of Cadiz — evidence for multiple sources and transport mechanisms in single cells, and metagenomes. Genome Res. 25, 1043–1055 (2015). active sedimentary wedges. Chem. Geol. 266, 350–363 (2009). 48. Wu, M. & Scott, A. J. Phylogenomic analysis of bacterial and archaeal sequences 80. Roalkvam, I. et al. New insight into stratifcation of anaerobic methanotrophs in with AMPHORA2. Bioinformatics 28, 1033–1034 (2012). cold seep sediments. FEMS Microbiol. Ecol. 78, 233–243 (2011). 49. Overbeek, R. et al. The SEED and the rapid annotation of microbial genomes 81. Vaular, E. N., Barth, T. & Hafidason, H. The geochemical characteristics of the using subsystems technology (RAST). Nucleic Acids Res. 42, D206–D214 (2014). hydrate-bound gases from the Nyegga pockmark feld, Norwegian Sea. Org. 50. Kanehisa, M., Furumichi, M., Tanabe, M., Sato, Y. & Morishima, K. KEGG: new Geochem. 41, 437–444 (2010). perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res. 45, 82. Cruaud, P. et al. Comparative study of Guaymas Basin microbiomes: cold seeps D353–D361 (2017). vs. hydrothermal vents sediments. Front. Mar. Sci. 4, 417 (2017). 51. Finn, R. D. et al. The Pfam protein families database: towards a more 83. Simoneit, B. R. T., Lonsdale, P. F., Edmond, J. M. & Shanks, W. C. Deep-water sustainable future. Nucleic Acids Res. 44, D279–D285 (2016). hydrocarbon seeps in Guaymas Basin, Gulf of California. Appl. Geochem. 5, 52. Huerta-Cepas, J. et al. Fast genome-wide functional annotation through orthology 41–49 (1990). assignment by eggNOG-mapper. Mol. Biol. Evol. 34, 2115–2122 (2017). 84. Case, D. H. et al. Methane seep carbonates host distinct, diverse, and dynamic 53. Lowe, T. M. & Chan, P. P. tRNAscan-SE On-line: integrating search and context microbial assemblages. MBio 6, e01348-15 (2015). for analysis of transfer RNA genes. Nucleic Acids Res. 44, W54–W57 (2016). 85. Milkov, A. V. et al. Co-existence of gas hydrate, free gas, and brine within the 54. Capella-Gutiérrez, S., Silla-Martínez, J. M. & Gabaldón, T. trimAl: a tool for regional gas hydrate stability zone at Hydrate Ridge (Oregon margin): evidence automated alignment trimming in large-scale phylogenetic analyses. from prolonged degassing of a pressurized core. Earth Planet. Sci. Lett. 222, Bioinformatics 25, 1972–1973 (2009). 829–843 (2004). 55. Ludwig, W. et al. ARB: a software environment for sequence data. Nucleic Acids 86. Pop Ristova, P., Wenzhöfer, F., Ramette, A., Felden, J. & Boetius, A. Spatial scales Res. 32, 1363–1371 (2004). of bacterial community diversity at cold seeps (Eastern Mediterranean Sea). 56. Pruesse, E. et al. SILVA: a comprehensive online resource for quality checked ISME J. 9, 1306–1318 (2015). and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids 87. Zaikova, E. et al. Microbial community dynamics in a seasonally anoxic fjord: Res. 35, 7188–7196 (2007). Saanich Inlet, British Columbia. Environ. Microbiol. 12, 172–191 (2010). Letter RESEARCH

Extended Data Fig. 1 | Helium ion microscopy images showing fixed and filtered on gold–palladium-sputtered polycarbonate filters. All vesicular structures interpreted as budding of Ca. A. ethanivorans. subsequent procedures (dehydration and critical point drying) were done Budding cells remained loosely attached to each other, leading to with filter pieces. Images are representative of n = 20 recorded images of formation of small clusters or aggregates. To avoid false stacking of cells, samples from n = 3 independent cultures. centrifugation was avoided during sample preparation. The samples were RESEARCH Letter

Extended Data Fig. 2 | Proposed pathway for ethane oxidation by Ftr, formylmethanofuran:tetrahydromethanopterin formyltransferase; Ca. Argoarchaeum based on proteogenomics analyses. a, Candidatus Fwd, formylmethanofuran dehydrogenase; Hdr, heterodisulfide Argoarchaeum uses an MCR-like enzyme (alkyl-CoM reductase reductase; Mch, methenyltetrahydromethanopterin cyclohydrolase; (‘ACR’)) to activate ethane to ethyl-CoM. Reactions converting the alkyl Mer, 5,10-methylenetetrahydromethanopterin reductase; MF, thioether to the acyl thioester are currently unknown: we hypothesize methanofuran; Mnh, multicomponent Na+:H+ antiporter; MPT, an involvement of the detected methyl-transferase (mtr; see b) in methanopterin; Mtd, methylenetetrahydromethanopterin dehydrogenase; these reactions. The proposed acetyl-CoA is oxidized to CO2 through NAD, nicotinamide adenine dinucleotide; Ntp, vacuolar/archaea the reverse (oxidative) Wood–Ljungdahl pathway. The physiological (V/A)-type H+/Na+-transporting ATPase; Nuo, NADH-quinone role of acetyl-CoA synthetase (Acs) and the acetate/cation symporter oxidoreductase. b, Organization of the genes that encode enzymes for the (ActP) is presently unclear. All enzymes depicted are encoded by the pathway proposed in a. Genes for which products were detected in protein Ca. Argoarchaeum genome. Enzymes depicted in green were fully or extracts are depicted in green, all other genes are shown in yellow and partly detected in protein extracts (see b). Fpo, F420H2 dehydrogenase; genes unrelated to ethane oxidation are shown in grey. Letter RESEARCH

Extended Data Fig. 3 | NanoSIMS analysis of the Ethane12 reducing bacteria at three different incubation times (95, 110 and enrichment culture. a–f, Ion images of analysed areas (representative 120 days). The average relative sulfur content of Ca. Argoarchaeum was of n = 6 recorded fields of view). b, The cell types shown in b are Ca. more than twofold higher than in Eth-SRB1 and Eth-SRB2. Each dot Argoarchaeum (cocci; indicated by the arrows), Eth-SRB1 (curved rods; represents the S:CN ion ratio of a single cell; in total, over 650 cells were indicated by small arrowheads) and Eth-SRB2 (large, oval; indicated by analysed (517 of Ca. Argoarchaeum, 58 of Eth-SRB1 and 105 of Eth- large arrowheads). a, d, The 12C14N− ion image was used as indicator of SRB2). The box plots show the total ion ratio range (vertical line with biomass (all cell types). b, c, e, f, Images of 32S− (b, e) and 32S−:12C14N− whiskers), the clustering of 50% of all cells analysed (box) and the mean ratios (c, f) show that cells of Ca. Argoarchaeum are enriched in sulfur, value for all cells of each strain. To calculate the ratio values in g, regions compared with bacterial cells—similar to AOM consortia16. Regions of of interest were drawn in the LOOK@NanoSIMS software (not shown); interest used to define cells were drawn based on the single ion images these were smaller than those displayed in a–f, to avoid inclusion of filter (12C14N− and 32S−) and superimposed on the ratio images. g, Relative material in the calculation. Scale bars, 2 μm (a–f). abundance of sulfur in Ca. Argoarchaeum compared to the sulfate- RESEARCH Letter

Extended Data Fig. 4 | Homology models of MCR from Ca. thermolithotrophicus; M. wolfeii, Methanothermobacter wolfeii. A. ethanivorans and Ca. S butanivorans. a, Sequence alignment of the b. Crystal structure of M. marburgensis MCR (Protein Data Bank (PDB) α subunit (McrA) and β subunit (McrB) of MCR from methanogens or accession 1MRO) bound to coenzyme B (CoB) and coenzyme M (CoM). ANME-1 methanotrophs (green), Ca. A. ethanivorans (AEth_00344, c–g, Modelled active site regions in the MCR-like enzymes of Ca. red) and Ca. S. butanivorans (SBU_000314, SBU_000718, SBU_001343 A. ethanivorans (c) and Ca. S. butanivorans (d–g). The predicted structure and SBU_001328; red); the functionally conserved residues in McrA was superimposed on M. marburgensis MCR (blue wireframes in c–g). and McrB are highlighted (yellow). The numbers below the alignment Residues in McrA and McrB are indicated as green and cyan sticks, indicate the residue position in M. marburgensis MCR29. M. barkeri, respectively; CoB and CoM are shown as yellow sticks, coenzyme F430 as Methanosarcina barkeri; M. formicicus, Methanotorris formicicus; pale yellow sticks and the arginine residue coordinating CoM as grey M. kandleri, Methanopyrus kandleri; M. thermauto, Methanothermobacter sticks (b–g). thermautotrophicus; M. thermolitho, Methanothermococcus Letter RESEARCH

Extended Data Table 1 | Quantification of the anaerobic consumption of ethane, and sulfate reduction to sulfide in the Ethane12 enrichment culture

The volume of the culture medium was 50 ml. − + − *Calculated considering complete oxidation of ethane (C2H6 + 6H2O → 2HCO3 + 16H + 14e ) and the loss of ethane in the sterile control experiment. † 2− − 2− − + − Calculated considering reduction of SO4 to HS (SO4 + 8e + 9H → HS + 4H2O) and the amount of sulfate reduced in the control culture without addition of ethane. ‡Electrons from ethane divided by electrons consumed by sulfate reduction; theoretically slightly higher than 1.0, owing to minor fraction of electrons fxed as biomass. RESEARCH Letter

Extended Data Table 2 | Genes and proteins involved in ethane activation and complete oxidation by Ca. A. ethanivorans

Genes that encode enzymes that are involved in ethane activation, candidate enzymes that are involved in further conversion to acetyl-CoA and enzymes of the Wood–Ljungdahl pathway identifed in the genome of Ca. A. ethanivorans, and detection of the corresponding proteins through shotgun proteomics analyses. *Relative abundance of detected proteins in extracts of the Ethane12 enrichment culture using Sequest/Amanda search algorithms in Proteome Discoverer (see Methods for calculations). Letter RESEARCH

Extended Data Table 3 | Genes and proteins involved in the energy metabolism of Ca. A. ethanivorans

Genes that encode enzymes involved in energy metabolism identifed in the genome of Ca. A. ethanivorans, and detection of the corresponding proteins through shotgun proteomics analyses. *Relative abundance of detected proteins in extracts of the Ethane12 enrichment culture using Sequest/Amanda search algorithms in Proteome Discoverer (see Methods for calculations). RESEARCH Letter

Extended Data Table 4 | Genes that encode type IV pili in the genomes of Eth-SRB1 and Eth-SRB2

*Predicted using KEGG. †Relative abundance of detected proteins in extracts of the Ethane12 enrichment culture using Sequest/Amanda search algorithms in Proteome Discoverer (see Methods for calculations). Letter RESEARCH

Extended Data Table 5 | Genes that encode cytochromes in the genomes of Eth-SRB1 and Eth-SRB2

*Predicted using Pfam. †Predicted using PSORTb. ‡Detected in protein extracts using the Amanda search algorithm in Proteome Discoverer (relative abundance, 1.82 × 106). RESEARCH Letter

Extended Data Table 6 | Environmental distribution of phylotypes related to Ca. A. ethanivorans

Sequences or reads showing over 97% identity to the 16S rRNA gene of Ca. Argoarchaeum, and abundance of ethane at the corresponding sites12,68–87. Sequences were retrieved from the NCBI nucleotide collection and the Integrated Microbial Next Generation Sequencing (IMNGS) platforms using BLASTn. *Data were collected from studies of the same or nearest site when ethane concentrations were not reported together with the 16S rRNA gene sequences. Concentrations reported as p.p.m. or hydrocarbon ratios were converted to volume percentage. †Percentage of 16S rRNA gene sequences related to Ca. A. ethanivorans in archaeal clone libraries (NCBI database) or percentage of reads in the corresponding datasets (IMNGS database). nature research | reporting summary

Corresponding author(s): Florin Musat

Reporting Summary Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

Statistical parameters When statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main text, or Methods section). n/a Confirmed The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly The statistical test(s) used AND whether they are one- or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section. A description of all covariates tested A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Clearly defined error bars State explicitly what error bars represent (e.g. SD, SE, CI)

Our web collection on statistics for biologists may be useful.

Software and code Policy information about availability of computer code Data collection Carl Zeiss ZEN vs. 2.3, Carl Zeiss ZEN vs. 1.0, Bruker Compass ftmsControl 2.2.0, Waters MassLynx V4.1, Orbitrap Fusion Tune Application vs. 2.1, XCalibur vs. 3.0, Chromeleon vs. 6.8

Data analysis BBMerge 34.48, SilvaNGS pipeline (Web tool: https://www.arb-silva.de/ngs/), Trimmomatic v0.33, Albacore vs. 2.3.1, Porechop vs 0.2.3, prodigal v2.6.3, BBMap v38.00, npScarf, pilon v1.23, trimAl v1.2rev59, Bruker Compass DataAnalysis 5.0, Proteome Discoverer vs. 2.2.0.388, Illumina bcl2fastq vs. 1.8.4, SPAdes vs. 3.11.1, MaxBin vs. 2.2.4, metaQUAST vs. 4.6.3, MUSCLE vs. 3.8.31, RAxML vs. 8.2.9, CheckM vs. 1.0.11, MiGA (Web tool: http://enve-omics.ce.gatech.edu:3000/), AMPHORA2 (Web tool: https://pitgroup.org/ amphoranet/), RNAmmer vs. 1.2, tRNAscan v2.0, ARB vs. 6.0.2, ImageJ vs. 1.48v, LOOK@NanoSIMS, SINA v1.3.1, SWISS-MODEL (https:// swissmodel.expasy.org/), MacPyMOL v1.7.4 (https://pymol.org/2/), eggNOG-mapper (Web tool: http://eggnogdb.embl.de/#/app/ emapper), BlastKOALA (Web tool: https://www.kegg.jp/blastkoala/), pfam_scan.pl v1.5 April 2018 For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

1 Data nature research | reporting summary Policy information about availability of data All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: - Accession codes, unique identifiers, or web links for publicly available datasets - A list of figures that have associated raw data - A description of any restrictions on data availability

Metagenome sequence data are archived in the NCBI database under the BioProject number PRJNA495932, including the draft genomes of Ca. Argoarchaeum ethanivorans (SAMN10235260), Eth-SRB1 (SAMN10235261) and Eth-SRB2 (SAMN10235262). The 16S rRNA gene amplicon reads have been submitted to the NCBI Short Read Archive database under the accession number SRR8089822. The proteomic dataset has been deposited with the ProteomeXchange Consortium with the DOI 10.6019/PXD011597. Source data for the quantitative growth experiments, FT-ICR-MS and LC-MS/MS measurements are provided. All other data are available in the manuscript or the supplementary materials.

Field-specific reporting Please select the best fit for your research. If you are not sure, read the appropriate sections before making your selection. Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf

Life sciences study design All studies must disclose on these points even when the disclosure is negative. Sample size Additional sample size calculation was not needed for the applied analyses. Sample sizes are based on multiple past experiences with the analytical instruments in this and other laboratories. The sample sizes which varied for each investigation are detailed in the Methods section.

Data exclusions No data were excluded from the analysis

Replication Growth experiments were conducted with replicate cultures over ten years (n >50). Quantitative growth experiments were conducted with n = 4 independent cultures with different starting concentrations of ethane. Fluorescence hybridization experiments were conducted with samples from n = 3 independent cultures. For all cultures n = 50 images were acquired for cell counting. The percentage of the different strains in the enrichment culture is based on over 8000 cells counted in fluorescence in situ hybridization. For visualization of vesicular structures n = 10 fluorescence images were acquired. Helium ion microscopy analyses were done with samples from n = 3 independent cultures; in total, n = 20 images were acquired. FT-ICR-MS and LC-MS/MS analyses were perfomed with metabolite extracts from n = 8 independent cultures. Metaproteomic analyses were performed with protein extracts from n = 2 independent cultures; for each culture extract, three replicate measurements (technical replicates) were done, summing to n = 6 Orbitrap LC-MS runs. NanoSIMS analyses were done with n = 3 samples collected at different time points. In total, six fields of view were acquired, and over 650 cells were analysed. All attempts at replication were succesful.

Randomization The experiments were not randomized, since all analyses concerned a single enrichment culture

Blinding This study did not involve groups. There has been one original microbiological sample retrieved from a marine site. Hence, blinding is not applicable.

Reporting for specific materials, systems and methods

Materials & experimental systems Methods n/a Involved in the study n/a Involved in the study Unique biological materials ChIP-seq April 2018 Antibodies Flow cytometry Eukaryotic cell lines MRI-based neuroimaging Palaeontology Animals and other organisms Human research participants