Abstract
Over the past decade, environmental metagenomics and PCR-based marker gene surveys have revealed that several lineages beyond just a few well-established groups within the Euryarchaeota superphylum harbor the genetic potential for methanogenesis. One of these groups are the Archaeoglobi, a class of thermophilic euryarchaeotes that have long been considered to live non-methanogenic lifestyles. Here, we enriched Candidatus Methanoglobus hypatiae, a methanogen affiliated with the family Archaeoglobaceae, from a hot spring in Yellowstone National Park. The enrichment is sediment-free, grows at 64-70 °C and a pH of 7.8, and produces methane from mono-, di-, and tri-methylamine. Ca. M. hypatiae is represented by a 1.62 Mb metagenome-assembled genome with an estimated completeness of 100% and accounts for up to 67% of cells in the culture according to fluorescence in situ hybridization. Via genome-resolved metatranscriptomics and stable isotope tracing, we demonstrate that Ca. M. hypatiae expresses methylotrophic methanogenesis and energy-conserving pathways for reducing monomethylamine to methane. The detection of Archaeoglobi populations related to Ca. M. hypatiae in 36 geochemically diverse geothermal sites within Yellowstone National Park, as revealed through the examination of previously published gene amplicon datasets, implies a previously underestimated contribution to anaerobic carbon cycling in extreme ecosystems.
Introduction
Methanogenesis is one of the most ancient metabolic pathways and plays a major role in the biogeochemical carbon cycle. Phylogenomic reconstructions and geological evidence suggest that methanogenesis was among the earliest metabolisms to evolve and that the last common ancestor of all extant archaea likely was a methanogen (1–9). Therefore, the study of methanogens is essential for understanding the co-evolution of life and the biosphere. Methanogenic archaea are the primary producers of biogenic methane (CH4) and contribute approximately 60% to the estimated 576 Tg of annual global methane emissions to the atmosphere (10, 11). Methanogenic pathways are classified by their carbon and electron sources (12–14). All methanogenic pathways converge at the terminal methane-forming step catalyzed by the methyl-coenzyme M reductase (MCR) complex. MCR and its homologs also catalyze the reverse reaction in the anaerobic oxidation of alkanes in alkanotrophic archaea (15, 16). MCR is uniquely present in all methanogens and is commonly used to identify potential methane and/or alkane cycling archaea in sequencing surveys (12, 17).
The physiology and biochemistry of methanogens has near-exclusively been investigated in axenic cultures of microorganisms belonging to the Euryarchaeota superphylum (12, 17–19). These predominantly grow by acetoclastic or CO2-reducing hydrogenotrophic methanogenesis, with only rare observations of Euryarchaeotal methyl-reducing methanogens (12, 20, 21). As a result, despite the dominance of methyl-based methanogenesis in anoxic environments with high salt and/or high sulfate concentration (e.g., anoxic marine sediments, coastal wetlands, hypersaline lakes), methylotrophic methanogenesis has in the past often been considered to be of comparatively limited environmental distribution. The extensive use of environmental metagenomics has led to the discovery of metagenome-assembled genomes (MAGs) encoding MCR from new lineages that are prevalent in anoxic environments, both within and outside the Euryarchaeota (2, 12, 22–26).
The majority of MAGs affiliated with archaeal phyla outside the Euryarchaeota are predicted to be methyl-reducing methanogens, with the exception of Candidatus (Ca.) Nezhaarchaeota (25, 27) and Ca. Methanomixophus affiliated with the order Archaeoglobales, which have been hypothesized to be CO2-reducing hydrogenotrophic methanogens (12, 25, 28). This result is consistent with the observation that methylated methanogenic substrates, including methylamines and methanol, are prevalent in the environment, although their concentrations in hot springs is currently unknown. Further, methyl-reducing methanogenesis is considered the predominant mode of methanogenesis in anoxic marine, freshwater, and hypersaline sediments (reviewed in (20)).
Members of the class Archaeoglobi have long been considered non-methanogenic with isolates characterized as dissimilatory sulfate reducers brought into culture as early as 1987 (29). To date, only nine species of the class Archaeoglobi have been obtained in axenic culture, and all were sourced from marine hydrothermal systems or off-shore oil reservoirs (30). The discovery of both MCR (25, 31, 32) and methyl- H4M(S)PT:coenzyme M methyltransferase (MTR) complexes in genomes of the Archaeoglobaceae have suggested that members of this family may live by methanogenesis (28).
Very recently, important progress towards experimental verification of methanogenesis by members of this family has been made. Liu et al. reported the in situ expression of genes related to hydrogen-dependent methylotrophic methanogenesis and heterotrophic fermentation within populations of Archaeoglobi in a high-temperature oil reservoir (28). Lynes, Krukenberg et al. reported that Archaeoglobi can be enriched in hot spring mesocosms under methanogenic conditions (33). Wang et al. reported that mcrABG and other methanogenesis marker genes encoded by two Archaeoglobales MAGs were highly expressed in hot spring microcosms incubated at 65 °C and 75 °C (34). Importantly, one of these Archaeoglobales MAGs represented the only Mcr-encoding archaeon that expressed mcrABG genes in methanogenic microcosms performed without substrate addition or with the addition of 10 mM methanol at 75 °C. This indirectly demonstrated the methanogenic nature of this archaeon (34). Last, Buessecker et al. reported the establishment of a methanogenic enrichment culture of Ca. Methanoglobus nevadensis from Great Boiling Spring (NV, USA) (35). The culture yields up to 158 µM methane after two weeks of incubation at its optimal growth temperature of 75 °C. Ca. M. nevadensis is represented by a 63% complete MAG obtained from the culture and a 98% complete MAG obtained a decade earlier (35).
Here, we report on the enrichment cultivation of Ca. Methanoglobus hypatiae LCB24, a methanogen affiliated with the family Archaeoglobaceae, from a hot spring in Yellowstone National Park (YNP). Using a combination of targeted cultivation, growth experiments, microscopy, stable isotope tracing, metagenomics, and metatranscriptomics, we demonstrate that Ca. M. hypatiae lives by methylotrophic methanogenesis and converts different methylamines to methane. By examining previously published datasets for the presence of Mcr-encoding Archaeoglobi, we demonstrate that these archaea are distributed in geothermal features of YNP, where they likely contribute to anaerobic carbon cycling. Our study presents direct evidence of methanogenesis within the Archaeoglobaceae and adds to the growing body of evidence demonstrating that methanogenesis is widely spread within the Euryarchaeota superphylum.
Materials and Methods
All chemicals used in this study were sourced from Sigma Aldrich unless otherwise specified.
Sample Collection, Enrichment, and Cultivation
In November 2021, a slurry of sediment and water (1:9) was collected from an unnamed hot spring in the Lower Culex Basin of Yellowstone National Park (YNP), WY, USA. In our previous survey of Mcr-encoding archaea in YNP (33), this hot spring was given the identifier LCB024 (44.573294, −110.795388; pH 7.8, 67 °C). A mixture of surface sediment (∼1-2 cm deep) and hot spring water was collected into a glass bottle and sealed headspace-free with a thick butyl rubber stopper. Collected material was transported back to the lab and stored at room temperature. Using this material as inoculum, 30 mL enrichments were established in February 2022 in 60 mL serum bottles. Material was homogenized by mixing and was then diluted 1:10 (v/v) with anoxic medium in an anoxic glove box (N2/CO2/H2; 90/5/5%).
Medium was prepared anoxically as described previously (36). Basal mineral medium contained a base of (per liter): KH2PO4, 0.5 g; MgSO4·7H2O, 0.4 g; NaCl, 0.5 g; NH4Cl, 0.4 g; CaCl2·2H2O, 0.05 g; HEPES, 2.38 g; yeast extract, 0.1 g; and 0.002% (w/v) (NH4)2Fe(SO4)2·6H2O. Medium was transferred to a Duran flask with a side opening and autoclaved for 20 m at 121 °C. Medium was then further supplemented with 5 mM NaHCO3, 1 mL trace element solution SL-10, 1 mL Selenite-Tungstate solution, 1 mL CCM vitamins (37), 0.0005% (w/v) resazurin, 10 mg of coenzyme-M, 2 mg sodium dithionite, 1 mM dithiothreitol, 1 mM Na2S·9H2O, with pH adjusted to 7.8 using sodium hydroxide (NaOH, 12 N). Serum bottles were sealed with butyl rubber stoppers and aluminum crimps before the headspace was exchanged with N2 (99.999%) for 5 minutes and set to a 200 kPa N2 atmosphere. Monomethylamine (MMA) was added from a filter-sterilized methylamine-hydrochloride anoxic stock solution to a final concentration of 10 mM. The bacterial antibiotics streptomycin (50 mg/L; inhibitor of protein synthesis) and vancomycin (50 mg/L; inhibitor of peptidoglycan synthesis) were added from filter-sterilized anoxic stock solutions. The enrichments were incubated at 70 °C in the dark without shaking. Cultures were maintained by regular transfer of 10% v/v into fresh media, which contained MMA and antibiotics. A sediment-free culture was obtained after the third transfer after which it was transferred at 10% v/v to 50 mL in 125 mL serum bottles.
Stable Isotope Tracing
The conversion of 13C- or D3-MMA (13CH3-NH2, CD3-NH2) to 13CH4 or CD3H was tracked by incubating active enrichment cultures in the presence of 20% labeled substrate (98%; Cambridge Isotope Laboratories). Incubations were carried out in 30 mL culture volumes in 60 mL serum bottles with 8% v/v inoculum, 50 mg/L streptomycin, 50 mg/L vancomycin, 10 mM MMA, and N2 gas (99.999%) incubated in anoxic media (pH 7.8, 70 °C) in six replicates (SI Appendix, Fig. S3). Duplicate control incubations included (i) 12C-MMA and (ii) inoculum without MMA. Triplicate control incubations were performed with (iii) 12C-MMA plus 10 mM bromoethanesulfonate (BES) added in mid-exponential phase (day 33) to inhibit methanogenesis and (iv) 10 mM BES added at time of inoculation (day 0) without substrate. Headspace samples were collected throughout the experiment as described above and analyzed using a Shimadzu QP2020 NX GCMS equipped with a GS-CarbonPLOT column (30 m × 0.35 mm; 1.5 μm film thickness; Agilent) and operated in Selected Ion Monitoring mode. The instrument was operated using the method described in Ai et al., 2013 (38) with helium as a carrier gas. All injections were performed by a Shimadzu AOC-6000 autosampler robot. Peak areas corresponding to m/z ratios of 16 for 12CH4, 17 for 13CH4, and 19 for CD3H were used for quantification.
Metagenomic Sequencing, Assembly, and Annotation
Two metagenomes were obtained over the course of this study. A 42 mL aliquot of the fourth transfer of the enrichment (Fig. 1 T4-MG) was filtered onto a 0.22 µm filter. The filter was transferred to a lysing matrix E tube and DNA extracted immediately following filtration. Genomic DNA was extracted using the FastDNA Spin Kit for Soil (MP Biomedicals, Irvine, CA) following the manufacturer’s guidelines.
A second metagenome was recovered from one of the six culture replicates grown in the presence of CD3-NH2 and used for recruiting transcriptomic reads from the other replicates (Fig. 1 SIT-MG). A 60 mL syringe flushed with N2 gas was used to transfer 30 mL of culture to a sterilized oak ridge tube. Cells were harvested through centrifugation for 30 minutes at 10,000 rpm at 4 °C. The supernatant was removed, and DNA extracted from the pellet using the FastDNA Spin Kit for Soil (MP Biomedicals, Irvine, CA) following the manufacturer’s guidelines. Genomic DNA for both metagenomes was shipped to SeqCenter (Pittsburgh, PA) and sample libraries were prepared using the Illumina DNA Prep kit and 10 bp unique dual indices (UDI). The first metagenome (T4-MG) was sequenced on a NextSeq 2000 System (Illumina) and the second (SIT-MG) sequenced on a NovaSeq 6000 System (Illumina), each producing 2×151 bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl-convert v3.9.3. Quality of the reads were evaluated with FastQC before quality, linker and adapter trimming, artifact and common contaminant removal, and error correction were performed with the rqcfilter2 pipeline (maxn=3, maq=10, trimq=20) and bbcms (mincount=2, hcf=0.6). Resulting reads were assembled with SPAdes v3.15.13 (Nurk, 2017) (-k 33,55,77,99,127 --meta –only-assembler) and coverage was determined with bbmap v38.94 (ambiguous=random) (https://sourceforge.net/projects/bbmap) (39). In addition to the initial assembly, co-assemblies using both T4-MG and SIT-MG metagenomes were also performed (1) with reads directly fed into SPAdes with the –only-assembler option excluded; and (2) with the trimmed and error corrected reads and the same SPAdes parameters as above. The statistics of MAGs generated through various assembly and quality control methods were evaluated, and the approach that produced the highest quality MAG was chosen for subsequent analysis (Dataset S1). Quality was determined by considering factors such as the number of resulting sequences, total length, completeness, and the minimum, maximum, and average sequence lengths. Annotation of the assembled sequences was performed with Prokka v1.14.16 (40). Assembled scaffolds ≥2000 bp were binned using Maxbin v2.2.7 (41), Metabat v2.12.1 (with and without coverage) (42), Concoct v1.0.0 (43), Autometa v1 (bacterial and archaeal modes with the machine learning step) (44), followed by bin refinement with DAS_Tool v1.1.6 (45), as previously described (46). Bin completeness and redundancy were assessed with CheckM v1.2.2 (47).
RNA Extraction, Sequencing, and Transcriptomic Processing
Total RNA was extracted for transcriptomics from four of the six replicates of Archaeoglobus cultivated in the presence of labeled substrate (13CH3-NH2 or CD3-NH2) for a total of eight replicates. Each replicate culture in the exponential growth phase (day 32) was moved from the 70 °C incubator to an ice bath placed at −20 °C for 40 minutes to stop cellular activity. A 60 mL syringe flushed with N2 gas was used to transfer 30 mL of culture to a sterilized oak ridge tube and kept on ice. Cells were harvested through centrifugation for 30 minutes at 10,000 rpm at 4 °C. The supernatant was removed, and the pellet transferred to a lysing matrix E tube (MP Biomedicals, Irvine, CA) to which 600 µL of RNA lysis buffer was added. Samples were homogenized in a MP Bioscience FastPrep instrument for 40 seconds at a speed setting of 6.0 m/s followed by centrifugation for 15 minutes at 14,000 rpm. RNA was extracted using the Quick-RNA miniprep kit (Zymo Research, Irvine, CA) including a DNAse treatment step and eluted in 50 µL of RNAse free water. Centrifugation steps were performed at 15,000 rpm and the final spin for elution at 10,000 rpm. Of the eight replicates extracted, six measured >50 ng/µL (3x 13CH3-NH2 and 3x CD3-NH2) and were sent for transcriptomic sequencing at SeqCenter (Pittsburgh, PA). Samples were DNAse treated with Invitrogen DNAse (RNAse free). Library preparation was performed using Illumina’s Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit and 10bp UDI. Sequencing was done on a NovaSeq 6000, producing paired end 151bp reads. Demultiplexing, quality control, and adapter trimming was performed with bcl-convert (v4.1.5). Read quality was further evaluated with FastQC v0.11.9 (48) before quality trimming and artifact, rRNA, and common contaminant removal with the rqcfilter2 pipeline (trimq=28, maxns=3, maq=20), and error correction with bbcms (mincount=2, hcf=0.6) from the BBTools suite v38.94 (39). Additional rRNA gene reads were detected and removed with Ribodetector v0.2.7 (49) and any remaining rRNA gene reads were finally removed with bbmap, using rRNA genes recovered from the metagenomes (see below) as references. The resulting mRNA reads were mapped against annotated genes from the paired metagenomes with bbmap to calculate RPKM (ambig=random).
Data Availability
All metagenomic, metatranscriptomic, and amplicon data discussed in this manuscript are available under NCBI BioProject ID PRJNA1014417. McrA gene amplicon data from YNP hot springs discussed in this manuscript has been previously published (Lynes et al., 2023) and is available under NCBI under BioProject PRJNA859922.
Results and Discussion
Cultivation
In our recent survey on the diversity of Mcr-encoding archaea in the geothermal features of YNP, mesocosms seeded with biomass from a hot spring located within the Lower Culex Basin (LCB024; pH 7-8, 56-74 °C), had shown potential to enrich for methanogenic Archaeoglobi (33). Using a sediment slurry collected from LCB024, we initiated incubations supplied with monomethylamine (MMA) and antibiotics incubated in anoxic media (pH 7.8, 70 °C) under a N2 headspace. The relative abundance, as determined by 16S rRNA gene amplicon sequencing, of Archaeoglobi-affiliated organisms in LCB024 was 0.32% in the initial slurry and had fallen to 0.02% by the time incubations were initiated a few months after samples had been collected (Fig. 1A).
Methane was detected after 36 days in the initial enrichment and the culture transferred to fresh media after reaching the late exponential phase of methane production following 70 days of incubation (447 µM; Fig. 1B). Five Archaeoglobi related 16S rRNA gene amplicon ASVs were identified in the initial enrichment, however one ASV grew to dominate the microbial community after the first transfer and reached 74.8% relative abundance after 62 days. In the transfers that followed, Archaeoglobi-related sequences became the only archaeal reads detected by 16S rRNA gene amplicon sequencing with the second most abundant organism a bacterium affiliated with the Pseudothermotoga at 6.8%. Although the CO2-reducing methanogen Methanothermobacter sp. was detected at 0.45% relative abundance in the slurry material used for inoculation, it was not detected in any subsequent transfers, nor were any known methanogens. Over subsequent transfers (238 days, T2-T5), the relative abundance of Archaeoglobi ASVs ranged from 46 to 69% and the final methane yield steadily increased from 1,844 to 2,459 µM. A sediment-free enrichment was obtained by the third transfer. Starting with the fourth transfer, the culture volume was scaled from 30 mL to 50 mL. By the sixth transfer, the culture produced 3,943 µM methane within 34 days. Metagenomic sequencing at two timepoints (day 335 of the enrichment and day 33 of the isotope tracing experiment described below) and 16S rRNA gene amplicon sequencing over recurring transfers (Fig. 1A) demonstrated that ASVs and MAGs affiliated with Archaeoglobi represented the only archaeon in culture LCB24. A single MCR complex (mcrAGCDB) belonging to the Archaeoglobi MAG was present, indicating this MAG represents the only methanogenic population.
Metagenomics and Phylogenetics
The reconstructed Mcr-encoding Archaeoglobi MAG from culture LCB24 was 1.62 Mbp in length with an estimated completeness of 100% according to checkM (SI Appendix, Table S3). This MAG was the result of a combined assembly of the T4-MG and SIT-MG metagenomes as this method yielded an improved assembly. Therefore, it was used for phylogenomic analysis against Archaeoglobi reference MAGs and genomes using 33 conserved single copy marker proteins and 16S rRNA genes (Fig. 2AB, SI Appendix, Table S4). The phylogenomic analysis showed that MAGs encoding MCR complexes clustered separately from those lacking mcr gene sequences. Consistently, 16S rRNA gene phylogeny supported this clustering with a pronounced separation of hot spring reference genomes and MAGs from current known isolates of Archaeoglobi, resulting in three main clusters: (i) those retrieved from North American hot springs (YNP and Great Boiling Spring, GBS), (ii) those originating from hot springs in China, and (iii) isolates, all of which were obtained from deep-sea marine hydrothermal systems (Fig. 2B).
LCB24 and closely related reference MAGs and isolate genomes exhibited a range of amino acid identities (AAI, 52.6-98.6%; Fig. 2C). Altogether, the LCB24 MAG was found to be highly related to previously obtained Archaeoglobi MAGs encoding the MCR complex and only distantly related to other Archaeoglobales sp. confirmed sulfate-reducing Archaeoglobales (AAI, 58.9-65%; ANI, 70.3-70.6%; 16S rRNA ANI, 91.6-93.8%; SI Appendix Fig. S1, S2). Based on AAI, MAG LCB24 was most closely related to Archaeoglobi LCB024-003 MAG (AAI, 98.6%), which we had obtained from the same hot spring in a previous study (33). The ANI and AAI values to the closest cultured methanogen, Ca. Methanoglobus nevadensis GBS, are 80.2 and 83.3%, respectively. Based on these results, we designate this archaeon Ca. Methanoglobus hypatiae strain LCB24, named after the philosopher Hypatia of Alexandria (for a protologue, see the SI Appendix, Results and Discussion). The estimated relative abundance of Ca. M. hypatiae based on the SIT-MG was 92.8%. Other community members in the LCB24 culture with >1% relative sequence abundance included members of the Pseudothermotoga (3.2%), Desulfovirgula (1.7%), and the family Moorellaceae (1.3%) (Fig. 1A, Dataset S1).
The only mcrAGCDB genes recovered from both metagenomes belong to the genome of Ca. M. hypatiae. Phylogenetic analysis of the single copy of McrA indicated its close relationship to McrA sequences found in members of the TACK superphylum (Fig. 2D). This contrasts with the placement of Ca. M. hypatiae within the Euryarchaeota based on phylogenomics (Fig. 2A), suggesting that Archaeoglobi could have obtained the MCR complex as a result of a horizontal gene transfer event from an archaeon in the TACK superphylum (7, 8). Also, it could indicate that non-methanogenic Archaeoglobi lost the capacity for anaerobic methane cycling after they had diverged from a shared methanogenic ancestor.
Methanogenic Activity of Ca. M. hypatiae
To gain insight into the activity of Ca. M. hypatiae under methanogenic and non-methanogenic conditions, a stable isotope tracing (SIT) experiment was conducted. Cultures were incubated in the presence of 10 mM of MMA; 8 mM of substrate were isotopically light, whereas the remaining 2 mM consisted of either 13C-monomethylamine (13CH3-NH2) or D3-monomethylamine (CD3-NH2). Addition of the methanogenesis inhibitor bromoethanesulfonate (BES) was used as a non-methanogenic control (Fig. 1B, 3, SI Appendix, Fig. S3). On average across six replicates, the cultured converted 13CH3-NH2 to 356 µM 13CH4 (17.8%) and 138.71 µM 13CO2 (6.9%) by day 32 (Fig. 3AC, Dataset S2). The conversion of CD3-NH2 was nearly identical yielding 355 µM CD3H (Fig. 3B). In the exponential phase of methane production, five of the six replicates were harvested for metagenomic and metatranscriptomic sequencing while the sixth replicate was allowed to grow to stationary phase. The replicate allowed to grow in each respective experiment converted the provided 13CH3-NH2 to 717.7 µM 13CH4 (35.9%) and 212.95 µM 13CO2 (10.65%) or CD3-NH2 to 394.76 µM CD3H (19.7%) by day 38 (Fig. 3ABC). These results confirmed monomethylamine was converted to methane by the LCB24 culture. The production of 13CO2 may represent the dismutation of 13CH3-NH2 to generate reducing power for methanogenesis via the methyl-branch of the Wood-Ljungdahl pathway or may be explained by other organisms in the culture catabolizing MMA. Yet, no transcriptomic evidence for this activity was present in this experiment. No methane production was observed for cultures treated with BES or in cultures incubated without MMA (Fig. 3D). When BES was added to cultures in the exponential phase, methane production ceased indicating the generation of methane is reliant on the Archaeoglobi MCR (Fig. 3E).
Visualization and Cell Enumeration
The growth of Ca. M. hypatiae was tracked in four replicates during the SIT experiment with catalyzed reporter deposition fluorescence in situ hybridization (CARD-FISH) using a general archaea-targeted probe Arch915 (50) and DNA-staining (DAPI) (Fig. 1C). As the production of methane increased throughout the experiment, there was a concurrent rise in the relative cell abundance of Ca. M. hypatiae (Fig. 3F, SI Appendix, Table S5). The initial assessment on day 22 across four replicates revealed the total cell density to be 3.45 × 107 ± 1.14 × 107 before substantial concentrations of methane had been detected in the headspace (<132 µM). By day 32, methane concentrations reached 1,777±739 µM and the total cell density increased to 6.97 × 107 ± 3.73 × 107 cells mL-1 with 54% (±9.6%) of cells labeled as Ca. M. hypatiae (Fig. 3F). All but one of these replicates were then sacrificed for further analysis. Finally on day 45, the remaining replicate reached a headspace methane concentration of 4,109 µM and a total cell density of 1.22 × 108 with 53% of cells labeled as Ca. M. hypatiae.
Visualization of the enrichment culture via scanning electron microscopy (SEM) revealed that most cells exhibited a regular to irregular coccoid morphology, with a width ranging from 0.5-1 µm (Fig. 1D). This morphology has previously been described for other Archaeoglobi species (30, 51–53).
Alternative Substrates and Temperature Optimum
We determined the substrate and temperature range of Ca. M. hypatiae by growing the culture in the presence of several substrates at 70 °C or with 10 mM MMA at 60-85 °C (Fig. 4AB). Conditions that lead to the production of methane included 10 mM trimethylamine (TMA), 10 mM dimethylamine (DMA), 10 mM MMA in media without yeast extract, and the control with 10 mM MMA and 0.01% yeast extract.
Methane production of cultures grown with MMA in the presence or absence of yeast extract were indistinguishable (5,202±606 and 5,703±410 µM CH4, respectively) indicating that yeast extract is not essential for methanogenic growth. Observed methane concentrations were higher in incubations amended with DMA (10,115±836 µM CH4) and TMA (9,524±3,626 µM CH4, with a wide range of 5,361-11,993 µM) on average more than the MMA controls, consistent with what has been observed for other methylotrophic methanogens (54). Incubations amended with 10 mM methanol (MeOH) did not produce methane after 47 days of incubation at 70 °C. Due to its use by sulfate reducing organisms as an electron donor (55), 10 mM lactate (LAC) was tested, as well as 10 mM MMA with 10 mM LAC, but none of these incubations produced methane. Production of methane has not been observed in any attempted transfers where hydrogen (99.9999% purity) was present in the headspace, or hydrogen with MMA was added.
The enrichment grew optimally at both 64 and 70 °C with relative amounts of methane produced at 5,304±451 µM and 5,202±606 µM, respectively. This deviates from the predicted optimal growth temperature of 74.4°C, which was derived from the translation of proteins in the Ca. M. hypatiae MAG using Tome (56). This is lower than the observed range of growth and optimum temperatures for type strains of non-methanogenic Archaeoglobus which have been demonstrated to grow between 50 and 95 °C with optimal temperatures between 75-83 °C in organisms sourced predominantly from deep sea vent environments (30). No methane production was detected at temperatures 77 °C or above or lower than 64 °C after 47 days of incubation (Dataset S4).
Genomic and Transcriptomic Basis for Methanogenesis
The assembled metagenome obtained at the end of the SIT experiment was used to align a total of 23,376,154 metatranscriptome mRNA reads obtained from six replicates harvested in the exponential growth phase and to create a detailed reconstruction of the metabolism of Ca. M. hypatiae (Fig. 3AB, 5, Dataset S5). A total of 22,891,651 reads, i.e., 97.8% of all recovered reads, were recruited to the Ca. M. hypatiae MAG. Only 2.1% of the total mRNA reads (484,503) were aligned with other co-enriched organisms. Among these, only 13 genes across four MAGs were expressed above 200 reads per kilobase of transcript per million mapped reads (RPKM) and just five genes exceeded >1,000 RPKM. Genes required for the conversion of methylamine to methane were among the top 2% of highest expressed genes transcribed by Ca. M. hypatiae, including genes encoding the MCR complex (mcrAGCDB; 13,046-18,098 RPKM), one of three monomethylamine methyltransferase copies (mtmB; 9,884 RPKM), dimethylamine corrinoid (mtbC; 3,677 RPKM), and methanol:coenzyme M methyltransferase (mtaA; 12,577 RPKM) (Fig. 5). Seven copies of substrate-specific methyltransferases for MMA (mtmB; 3 copies), DMA (mtbB; 2 copies), and TMA (mttB; 2 copies) were present in the genome, but methanol methyltransferase (mtaB) was not identified. These genes were differentially expressed with one copy for each type of methylamine expressed above 3,200 RPKM. In addition to mtbC, two gene copies of the trimethylamine corrinoid protein (mttC) were found in the genome but their expression was relatively low (<460 RPKM average). Monomethylamine corrinoid (mtmC) or methanol corrinoid (mtaC) proteins were not identified in Ca. M. hypatiae. Additionally, genes were expressed for pyrrolysine synthesis (pylBCD; 819, 343, 37 RPKM) and the methyltransferase corrinoid activation protein (ramA; 1,076 RPKM), both of which support methylamine methyltransferases in methylotrophic methanogenesis (57, 58). The absence of mtmC and the high expression levels of mtbC (3,677 RPKM) and mtaA (12,577 RPKM) suggest that they are responsible for the transfer of a methyl group from monomethylamine to coenzyme M (CoM) after it has been transferred by a substrate-specific methyltransferase (mtmB). Consistent with the observed methane production from DMA and TMA, Ca. M. hypatiae can use these methylamines and expressed the corresponding genes (mtbB, mttB) at comparatively high levels (JOOIALLP_01813 mtbB 3,249 RPKM; JOOIALLP_01787 mttB 5,324 RPKM; Fig. 4B, 5). It is worth noting that the expression of mtbB/mttB was detected despite the culture not having been previously exposed to DMA or TMA at the time of the transcriptomics experiment. We hypothesize that Ca. M. hypatiae could employ one of two strategies: it either (i) constitutively expresses all substrate-specific methyltransferases and corrinoid proteins as a precautionary measure to accommodate substrates potentially encountered in situ, or (ii) Ca. M. hypatiae transcriptionally co-regulates the genes responsible for these functions.
Ca. M. hypatiae expressed the methyl-branch of the Wood-Ljungdahl pathway (WLP) and the acetyl-CoA decarbonylase/synthase complex (Cdh, cdhABCDE), which is consistent with genes observed and shown to be expressed in sulfate-reducing Archaeoglobi genomes (55). This includes two paralogous copies of 5,10-methylenetetrahydromethanopterin reductase (mer) which might function as a traditional Mer, considering that these genes are also members of the large luciferase-like monooxygenase family (pfam00296)(35). The expression of genes in the WLP varied. Methylenetetrahydromethanopterin dehydrogenase (mtd), methenyltetrahydromethanopterin cyclohydrolase (mch), formylmethanofuran-tetrahydromethanopterin N-formyltransferase (ftr), formylmethanofuran dehydrogenase (fwdABC), and one copy of the mer homologs were expressed at comparatively high levels (456-2,763 RPKM), whereas FwdDEFG and the other mer copy were only minimally expressed (<180 RPKM) (Dataset S5). The high expression of the Cdh complex (cdhACDE; 3,063±362, cdhB 677 RPKM average across subunits) suggests that Ca. M. hypatiae is capable of autotrophically fixing CO2 to acetyl-CoA as has been shown for other Archaeoglobus species (59). Acetyl-CoA could also be derived from the degradation of fatty acids present in yeast extract through the process of beta-oxidation. Enzymes involved in this pathway were expressed at moderate to high levels during growth (Dataset S6). Pyruvate synthase (Por) was highly expressed providing a way for acetyl-CoA to be converted to pyruvate and subsequently be fed into major biosynthetic pathways. Specifically, Ca. M. hypatiae encodes pyruvate carboxylase (PycAB), an incomplete reductive tricarboxylic acid cycle (rTCA), phosphoenolpyruvate synthase (Pps), most enzymes needed for gluconeogenesis, and several enzymes associated with the pentose phosphate pathway in archaea, which were all expressed at varying levels (Dataset S5). Together, these pathways provide Ca. M. hypatiae the capacity to synthesize amino acids, carbohydrates, integral components of the cell wall, and vital sugars for nucleic acids.
Several complexes related to energy conservation and electron transport were moderately to highly expressed. Ca. M. hypatiae encodes a fused heterodisulfide reductase (hdrDE) that was highly expressed (1,106±120 RPKM) in addition to a fused hdrD/mvhD and four copies of hdrD that were all expressed at much lower levels (<500 RPKM). The differing levels of transcription suggest the membrane-bound HdrDE is responsible for the regeneration of coenzymes M and B through the reduction of heterodisulfide (CoM-S-S-CoB). Additionally, the absence of HdrB, which contains the active site for disulfide reduction, eliminates the possibility that disulfide reduction could occur via a HdrABC complex (60). As reported for Ca. M. nevadensis (35), a unique gene cluster was identified containing F420-non-reducing hydrogenase (MvhAGD), two HdrA copies and a QmoC fused to a HdrC. One HdrA copy (JOOIALLP_01710) was predicted by DiSCo analysis as a quinone-modifying oxidoreductase (QmoB), a protein related to the HdrA of methanogens (61, 62). This cluster was expressed at high levels (995-2,431 RPKM average), suggesting its importance for electron transfer in Ca. M. hypatiae. We hypothesize these subunits are associating together in vivo to bifurcate electrons from hydrogen (H2) to reduce both menaquinone (MQ) and ferredoxin (Fdox), as proposed recently (35, 63). Lastly, Ca. M. hypatiae moderately expressed a membrane-bound F420H2:quinone oxidoreductase (Fqo) complex (88-280 RPKM across subunits) and a V-type ATP synthase (24-442 RPKM across subunits).
The electrons required for reducing the CoM-S-S-CoB heterodisulfide could originate from two possible routes. The first possibility would rely on sourcing electrons from hydrogen, which could be oxidized by the Mvh-Qmo-Hdr complex coupled to menaquinone reduction. H2 may be produced through the activity of a group 3b [NiFe]-sulfhydrogenase (HydABDG), which was the highest expressed hydrogenase complex with an average RPKM of 4,421 across subunits (64, 65). To evolve hydrogen via HydABDG, reducing power, via NADPH, could be supplied by sulfide dehydrogenase (SudAB; SudA, 1,088 RPKM; SudB, 495 RPKM). Alternatively, NADPH could instead be provided to biosynthesis pathways and therefore be decoupled from methanogenic metabolism. H2 could also potentially be sourced from fermentative bacteria in the enrichment culture, however, the low number of hydrogenases encoded by co-enriched organisms were only very lowly expressed at the time of sampling for metatranscriptomics (<51 RPKM). At this point, the source of H2 Ca. M. hypatiae uses remains uncertain, as no H2 was added to the headspace. Second, in a hydrogen-independent electron transport system, reduced F420 and ferredoxin could be produced through the dismutation of methylated substrate to CO2 through the WLP. Reduced F420 could be oxidized by the Fqo complex and contribute to a reduced menaquinone pool that could be used by the fused HdrDE complex to reduce CoM-S-S-CoB. Reduced ferredoxin could be oxidized at a soluble FqoF to reduce F420 or at an Fqo complex lacking FqoF to reduce menaquinone (66, 67). Based on the low expression levels of the Fqo complex (171±67 RPKM) and the absence of F420-reducing hydrogenase (frh) from the genome, it is not likely the WLP runs in the reductive direction as a source of reduced F420 would be required. Resolving the exact configuration of the electron transport system encoded by Ca. M. hypatiae will require biochemical confirmation in future investigations.
Importantly, genes necessary for dissimilatory sulfate reduction typically observed in sulfate-reducing members of the Archaeoglobi, including dissimilatory sulfite reductase (dsrAB), sulfate adenylyltransferase (sat), and adenylylsulfate reductase (aprAB), were neither identified in the genome of Ca. M. hypatiae nor in the unbinned fraction of the metagenome. They were also absent from the comparatively incomplete MAG of Ca. M. nevadensis GBS (35). However, Ca. M. hypatiae encodes subunits dsrMK and dsrOP of the Dsr complex in addition to dsrC. This complex is strictly conserved in sulfate-reducing organisms (68) where it mediates electron transfer from the periplasm to the cytoplasm reducing the disulfide bond found in DsrC cysteines (69). The expression of the Dsr complex and dsrC was low (450±63 RPKM) during growth on monomethylamine suggesting it is not vital to the metabolism of Ca. M. hypatiae. The presence of the Dsr complex, DsrC, and subunits QmoC and QmoB in the genome may be explained as evolutionary remnants from ancestral Archaeoglobi, growing initially as sulfate-reducing organisms but later transitioning to a methanogenic lifestyle (7, 8). This raises the question whether intermediate of this process, Archaeoglobi capable of both methanogenesis and sulfate-reduction (and possible anaerobic oxidation of methane), still exist today (25, 28).
Collectively, the metagenomic and transcriptomic data confirmed that Ca. M. hypatiae is not only the sole archaeon but the sole methanogen in our culture. The metabolic reconstruction and metatranscriptomic results are consistent with methylotrophic methanogenesis from methylamines. The absence of genes required for sulfate reduction eliminates the possibility for this metabolism in Ca. M. hypatiae. A unique gene cluster (Mvh-Qmo-Hdr) potentially involved in energy conservation was expressed, however future studies will be required to test how Ca. M. hypatiae internally cycles electrons for methanogenesis and if it sources H2, or other reductants, from the medium or co-enriched bacteria.
Distribution of Ca. Methanoglobus Across Geothermal Features in YNP
16S rRNA and mcrA gene amplicon sequence data generated in a recent microbial diversity survey of 100 geothermal features in YNP (33) were used to analyze the distribution of Archaeoglobi related to Ca. M. hypatiae (SI Appendix, Fig. S5). 16S rRNA gene amplicons closely related to Ca. M. hypatiae (96.7-100% sequence identity) were found in seven DNA samples from six hot springs (pH 5.1-9.35, 31-78 °C) in addition to hot spring LCB024 (the source of this culture) at relative abundances ranging from 0.02-0.22%. In addition, mcrA gene ASVs affiliated with Archaeoglobi were PCR-amplified from 53 DNA samples, out of 201 total samples that had been screened by PCR. These 53 samples had been collected from microbial mats or sediments originating from 36 geothermal features distributed across various thermal regions within YNP by Lynes, Krukenberg et al. (33). Archaeoglobi-related mcrA genes were found in geothermal features with a pH range of 2.61 to 9.32 and a temperature range of 18.4 °C to 93.8 °C. Collectively, our results and the studies by Wang et al. and Buessecker et al., who reported that Mcr-encoding Archaeoglobi are present (35) and transcriptionally active in hot spring mesocosms (34), demonstrate the previously overlooked role that Archaeoglobi might play in the anaerobic carbon cycle of geothermal environments.
Conclusion
In summary, the cultivation of Ca. Methanoglobus hypatiae LCB24 provides direct experimental evidence that members of the Archaeoglobi are methanogens. Ca. M. hypatiae can use MMA, DMA, and TMA as methanogenic substrates and grows optimally at 64-70 °C, as evidenced by metagenomics, metatranscriptomics, and isotope tracing experiments. Metagenomic sequencing and phylogenomic analysis confirmed the close relationship of Ca. M. hypatiae to other Mcr-encoding Archaeoglobi and the relatedness of its mcrA to MAGs of the TACK superphylum, some of which have recently been shown to also be methanogens (70, 71). Together, this supports the idea that the capacity for methanogenesis is deeply rooted in the archaea and possibly dates to the last common ancestor of archaea (1, 3, 7, 8, 72). The wide distribution of Archaeoglobi-affiliated mcrA gene sequences and Ca. M. hypatiae-related 16S rRNA gene sequences in geothermal features across YNP suggests that members of this lineage play a hitherto unaccounted-for role in anaerobic carbon cycling in these extreme ecosystems. Future studies of Ca. M. hypatiae and other methanogens will provide valuable insights into the evolution of methane metabolism and the significance of these archaea in biogeochemical cycles across geothermal and other environments.
Author Contributions
M.M.L. and R.H. developed the research project. M.M.L., Z.J.J., A.J.K, and R.H. designed experiments. M.M.L. and A.J.K. conducted field sampling. M.M.L. performed cultivation, extracted DNA for amplicon and metagenomic sequencing, extracted RNA for transcriptomic sequencing, and conducted physiology and stable isotope experiments. A.J.K. developed GC/GCMS protocols and processed GCMS samples. Z.J.J. processed and annotated metagenomic and transcriptomic data, assembled MAGs, mapped transcripts, assigned taxonomy, constructed 16S rRNA gene phylogeny, and performed phylogenetic analysis of MAGs. M.M.L. conducted phylogenetic analysis of amplicon data, refined gene annotations, reconstructed, and interpreted the metabolic potential of Ca. M. hypatiae with insight from Z.J.J and A.J.K. R.H. was responsible for funding and supervision of the project. M.M.L. and R.H. wrote the manuscript, which was then edited by all authors.
Conflict of interest
none declared.
Supporting Information
SI Results and Discussion
Protologue
Methanoglobus hypatiae sp. nov
Me.tha.no.glo.bus. Gr. pref. methano-, pertaining to methane; L. masc. n. -globus, sphere; Gr.L. masc. n. Methanoglobus, methane producing organism spherical in shape. This genus was named by Buessecker et al. (1). Hy.pa.ti.ae. Gr. fem. hypatiae, to honor Hypatia of Alexandria, a respected and renowned philosopher of ancient Alexandria, Egypt, who made significant contributions to the understanding of mathematics and astronomy. A symbol of intellectual courage and scholarly achievement. This archaeon was cultured from an unnamed hot spring in the Lower Culex Basin of Yellowstone National Park identified as feature LCB024 (2). This archaeon is an obligately anaerobic thermophile that performs methylotrophic methanogenesis using methylamines and grows as regular to irregular coccoid cells approximately 0.5 to 1 µm in width. The type genome of this archaeon is deposited at NCBI under BioProject PRJNA1014417, accession number will be added upon publication.
SI Materials and Methods
Amplicon Sequencing and Analysis
DNA was extracted from environmental slurry samples and enrichment cultures sampled on the day of transfer using the FastDNA Spin Kit for Soil (MP Biomedicals, Irvine, CA) following the manufacturer’s guidelines. Archaeal and bacterial 16S rRNA genes were amplified with the updated Earth Microbiome Project primer set 515F and 806R (3). Amplicon libraries were prepared as previously described (2) and sequenced by the Molecular Research Core Facility at Idaho State University (Pocatello, ID) using an Illumina MiSeq platform with 2 x 250 bp paired end read chemistry. Gene reads were processed using QIIME 2 version 2022.8 (4). Primer sequences were removed from demultiplexed reads using cutadapt (5) with error rate 0.12 and reads truncated (130 bp forward, 150 bp reverse), filtered, denoised and merged in DADA2 with default settings (6). Processed 16S rRNA gene amplicon sequence variants (ASVs) were taxonomically classified with the sklearn method and the SILVA 138 database (7). The R package decontam (version 1.18.0) (8) was used to remove contaminants using the “Prevalence” model with a threshold of 0.5.
Annotation and Reconstruction of Metabolic Potential
Genes associated with methanogenesis pathways, dissimilatory sulfur metabolism pathways, coenzyme and cofactor biosynthesis, energy conservation, and beta-oxidation, were inventoried. Annotations assigned by Prokka were refined through manual evaluation using KofamKOALA, NCBI BLASTP, NCBI’s Conserved Domain Database, InterPro, the hydrogenase classifier HydDB, and DiSCo (9–14).
Phylogenetic and Phylogenomic Analyses
Average nucleotide identities (ANI) of 16S rRNA genes were calculated with blastn, with ANI and average amino acid identities (AAI) calculated by pyani v02.2.12 (ANIb) and CompareM v0.0.23 (-- fragLen 2000) (https://github.com/dparks1134/CompareM), respectively for selected Archaeoglobales genomes and MAGs (Table 1). Phylogenetic analysis of 16S rRNA genes was performed with fasttree (15) using masked alignments generated by ssu-align.
Archaeoglobales MAGs and reference genomes were screened for 54 phylogenetically informative single copy proteins (16, 17) of which a subset of 33 proteins were identified in them all (SI Appendix, Table S4). In order to maximize the number of proteins compared across references, MAGs LMO1 and LMO3 were excluded from this analysis, as they lacked 2 and 7 proteins out of the total 33, respectively. These were then aligned with muscle (18), concatenated, and phylogenomically analyzed with maximum likelihood analysis with fasttree (WAG model). McrA alignments were performed with MAFFT-LINSi v7.522 (19), trimmed with trimAL v1.4.rev22 (20) using a 0.5 gap threshold, and maximum likelihood trees were built with IQTree2 v2.0.6 (21) using LG+C60+F+G model and 1,000 ultrafast bootstraps.
Temperature and Substrate Optimum Experiments
Methane production and growth of Archaeoglobus was evaluated at different temperatures and in the presence of methylated substrates (i.e., methanol and mono-, di-, and trimethylamine), lactate, and media prepared without yeast extract. The sixth transfer of the enrichment was used to inoculate triplicate 30 mL serum bottles containing 15 mL of medium with 8% v/v inoculum, streptomycin (50 mg/L), vancomycin (50 mg/L), and 10 mM of each substrate tested. Cultures were evaluated at 60 °C, 64 °C, 70 °C, 77 °C, 80 °C, and 85 °C with 10 mM MMA. Separately, we tested whether the culture would grow on the following substrate (combinations): 10 mM dimethylamine (DMA); 10 mM trimethylamine (TMA); 10 mM methanol (MeOH); 10 mM lactate (LAC); 10 mM MMA and 10 mM LAC); 10 mM MMA with media without yeast extract; and a control in media without yeast or any methanogenic substrate. The 70 °C cultures amended with 10 mM MMA served as the control. All incubations were performed in biological triplicate.
Methane Measurements
During cultivation, 250 µL subsamples of the headspace were taken using a gas tight syringe (Hamilton) and injected into a 10 mL autosampler vial that had been sealed with grey chlorobutyl septa. Samples were taken from the autosampler vials and injected into a Shimadzu 2020-GC gas chromatograph equipped with a GS-CarbonPLOT column (30 m x 0.32 mm; 1.5 μm film thickness; Agilent) and a Rt-Q-BOND column (30 m x 0.32 mm; 1.5 μm film thickness; Restek) using helium as a carrier gas. All injections were performed by a Shimadzu AOC-6000 autosampler robot. The injector, column, and flame ionization detector (FID) were maintained at 200 °C, 50 °C, and 240 °C, respectively. Methane concentrations were calculated based on injection of a standard curve.
Fluorescence in situ hybridization and cell counts
Aliquots of enrichment cultures incubated with 13C-MMA during the SIT experiment were treated with 2% paraformaldehyde (PFA) and fixed for 1 hr at room temperature. Following fixation, cells were washed twice with 1x PBS, followed by centrifugation at 16,000 × g to remove the supernatant, resuspended in 1x PBS, and stored at 4 °C. For direct cell counts, aliquots of fixed cell suspensions were filtered onto polycarbonate filters (0.2 μm pore size, 25 mm diameter, GTTP Millipore, Germany) and air dried before filter pieces were cut and embedded in 0.2% low melting agarose. We attempted to use the Archaeoglobales-specific probe Arglo32 (22), however fluorescent signal was insufficient. Given Ca. M. hypatiae was the sole archaeon in the enrichment culture, the relative abundance of Ca. M. hypatiae cells was determined via catalyzed reporter deposition fluorescence in situ hybridization (CARD-FISH) using the general archaea-targeted 16S rRNA oligonucleotide probe Arch915 (23). Total cell counts were based on DNA-stained cells using DAPI (4,6-diamidino-2-phenylindole). CARD-FISH was performed as previously described (24). Cell wall permeabilization was achieved with a brief treatment of 0.1 M HCl (1 min, RT) followed by treatment with 0.01 M HCl (15 min, RT). Endogenous peroxidases were inactivated with 0.15% H2O2 in methanol (30 min, RT). A formamide concentration of 35% was used for all hybridization reactions (2.5 hrs, 46°C). CARD was performed using Alexa Fluor 594 labeled tyramides for 30 min at 46°C. Following signal amplification, an additional washing step in 1x PBS was included to reduce background fluorescence (15 min, RT, dark). Samples were stained with DAPI, embedded in Citifluor-Vectashield, and enumerated using an epifluorescence microscope (Leica DM4B).
Scanning electron microscopy (SEM)
An aliquot of the enrichment culture at transfer 7 (T7) was treated with 2% paraformaldehyde (PFA) and fixed for 1 hr at room temperature. Following fixation, cells were washed twice through centrifugation at 16,000 × g to remove the supernatant, resuspended in 1x phosphate buffered saline (PBS), and stored at 4 °C. Samples for imaging were prepared according to Schaible et al., 2022 (25). Briefly, a square coupon of mirror-finished 304 stainless steel (25 mm diameter, 0.6 mm thickness) was purchased from Stainless Supply (Monroe, NC). The coupon was cleaned by washing with a 1% solution of Tergazyme (Alconox, New York, NY) and rinsed with Milli-Q water. The coupon was dried under compressed air and stored at room temperature. 5 µL of fixed sample was spotted on the coupon and air-dried at 46 °C for 3 min. The coupon was then dried for 1 m each step in a successive ethanol series starting with 10% ethanol and increasing by increments of 10% with the last step 90% ethanol. SEM images were captured using a Zeiss (Jena, Germany) SUPRA 55VP field emission scanning electron microscope (FE-SEM). The microscope was operated at 1 keV under a vacuum of 0.2–0.3 mPa, with a working distance of 5.4-6.2 mm at the Imagining and Chemical Analysis Laboratory (ICAL) of Montana State University (Bozeman, MT). No conductivity coating was applied before SEM analysis as the microscope was operated at 1 keV.
SI Tables
SI Figures
Description of Available Supplementary Datasets
SI Dataset S1. Extended metagenome assembled genome (SIT-MG) and isolate genome statistics. Seqs, sequences; avg_cov, average coverage; avg_gc, average G+C content; % rel. abund., percent relative abundance.
SI Dataset S2. GCMS measurements of masses 16 (CH4), 17 (13CH4), and 19 (12CD3H) during the isotope tracing experiment. Percent of labeled methane is calculated as a fraction of provided labeled substrate. Stdev, standard deviation.
SI Dataset S3. Gas chromatograph FID measurements of 12CH4 during isotope tracing experiment. NA, not available/measured.
SI Dataset S4. Gas chromatograph FID measurements of CH4 during temperature optimum experiment. NA, not available/measured.
SI Dataset S5. Inventory of genes expressed by Ca. M. hypatiae LCB24 under methanogenic conditions and as depicted in Fig. 5. Expression levels averaged across six replicates are reported in reads per kilobase of transcript per million mapped reads (RPKM).
SI Dataset S6. Inventory of genes expressed by Ca. M. hypatiae LCB24 under methanogenic conditions belonging to the beta-oxidation pathway. Expression levels averaged across six replicates are reported in reads per kilobase of transcript per million mapped reads (RPKM).
Acknowledgements
This study was funded through a NASA Exobiology program award (80NSSC19K1633) to R.H. We thank the US National Park Service for permitting work in YNP under permit number YELL-SCI-8010. We thank George Schaible (MSU) for help with SEM imaging, Dr. Viola Krukenberg (MSU) for initial FISH methodology development, Sylvia Nupp, Dr. Andrew Montgomery, and Paige Schlegel (all MSU) for assistance with field sampling, Dr. Christopher Lemon (MSU) for allowing use of his cooling centrifuge, and Dr. Marike Palmer (UN Las Vegas) for discussing naming of this archaeon.
Footnotes
Updated manuscript and updated SI text. No new experiments are reported.