Diversity and function of methyl-coenzyme M reductase-encoding archaea in Yellowstone hot springs revealed by metagenomics and mesocosm experiments

Metagenomic studies on geothermal environments have been central in recent discoveries on the diversity of archaeal methane and alkane metabolism. Here, we investigated methanogenic populations inhabiting terrestrial geothermal features in Yellowstone National Park (YNP) by combining amplicon sequencing with metagenomics and mesocosm experiments. Detection of methyl-coenzyme M reductase subunit A (mcrA) gene amplicons demonstrated a wide diversity of Mcr-encoding archaea inhabit geothermal features with differing physicochemical regimes across YNP. From three selected hot springs we recovered twelve Mcr-encoding metagenome assembled genomes (MAGs) affiliated with lineages of cultured methanogens as well as Candidatus (Ca.) Methanomethylicia, Ca. Hadesarchaeia, and Archaeoglobi. These MAGs encoded the potential for hydrogenotrophic, aceticlastic, hydrogen-dependent methylotrophic methanogenesis, or anaerobic short-chain alkane oxidation. While Mcr-encoding archaea represent minor fractions of the microbial community of hot springs, mesocosm experiments with methanogenic precursors resulted in the stimulation of methanogenic activity and the enrichment of lineages affiliated with Methanosaeta and Methanothermobacter as well as with uncultured Mcr-encoding archaea including Ca. Korarchaeia, Ca. Nezhaarchaeia, and Archaeoglobi. We revealed that diverse Mcr-encoding archaea with the metabolic potential to produce methane from different precursors persist in the geothermal environments of YNP and can be enriched under methanogenic conditions. This study highlights the importance of combining environmental metagenomics with laboratory-based experiments to expand our understanding of uncultured Mcr-encoding archaea and their potential impact on microbial carbon transformations in geothermal environments and beyond.


SI
. Location, pH, and temperature of main study sites. Data recorded at the time of sampling for metagenomics (2017) and mesocosm experiments (2019). Table 2. Aquatic geochemistry of main study sites (2017)(2018)(2019)(2020). Data shown as ranges when applicable.   Table 4). Squares indicate ultrafast bootstrap values of 100% (black) and 95-99% (gray).   Total dissolved sulfide (H2S(aq)) was measured using the amine-sulfuric acid method (2). For this, 7.5 mL of geothermal water were collected with a 10 mL syringe and added into triplicate glass tubes pre-filled with 0.5 mL of amine sulfuric acid reagent (Ricca Chemical). Approximately 0.15 mL (4 drops) of 15 M FeCl3 were immediately added and the sample was mixed by slowly inverting the test tube once. After 3-5 minutes, 1.6 mL of 3.8 M (NH4)2PO4 were added and the sample was stored in the dark until absorbance was measured at 664 nm using a BioMate 3S spectrophotometer or Ocean Optics USB2000 spectrophotometer within 24 hours of sample collection.

SI
The ferrozine assay was used to measure total and ferrous iron (Fe, Fe(II)) (3). Briefly, 5 mL of geothermal water were filtered through a 0.22 µm pore size PES syringe filter into three replicate tubes. 250 µL of a 10% w/v hydroxylamine solution were added for the Fe(II) assay before 250 µL of 4.9 mM ferrozine reagent and 625 µL ammonium-acetate buffer (3) were added for both Fe total and Fe(II) assays. After dilution to 12.5 mL with distilled water, samples were stored in the dark until absorbance was measured at 562 nm using a BioMate 3S spectrophotometer or Ocean Optics USB2000 spectrophotometer within 24 hours of sample collection.
To determine dissolved methane (CH4(aq)), geothermal water was collected using a peristaltic pump with an in-line 142 mm diameter, 0.2 µm pore size hydrophilic polycarbonate membrane filter (Millipore). 160 mL serum bottles were purged with three pour volumes of geothermal water, capped without headspace using butyl rubber stoppers, secured with aluminum rings, and stored upside-down at 4 ºC. Prior to the analysis of dissolved gases, samples were weighed before and after 30 mL of liquid was removed, headspace replaced with ambient air, and incubated on a shaker at 60 rpm for two hours at room temperature. From the headspace, 1 mL was withdrawn using a gas tight syringe and manually injected into an SRI Multiple Gas Analyzer #5 gas chromatograph equipped with a Haysep D column (6 foot), a thermal conductivity detector, and a flame ionization detector with a methanizer (carrier gas: N2 at 13 psi, 50ºC). Methane and mixed gas (CH4, CO2, CO, H2) standards (100 ppm, 10,000 ppm) were measured regularly. Concentrations of dissolved gases were calculated using Henry's law and solubility equations (2,4). LCB024-003 (Archaeoglobi) encodes two mechanisms for coenzymes M and B regeneration.
hdrBC were not detected; however, HdrD has been proposed to substitute for HdrBC in heterodisulfide reduction in other Mcr-encoding Archaeoglobi MAGs (29,30). LCB024-003 also encodes a fused HdrDE and a membrane-bound F420H2:quinone oxidoreductase (Fqo) complex. In Archaeoglobi isolates, the Fqo complex has been shown to reduce menaquinone instead of methanophenazine (31), which is characteristic of methanogens containing an Fpo complex (32).
HdrDE contains two cysteine-rich domains (CCG) and is homologous to the membrane bound complex DsrMK, which is inferred to reduce a cytoplasmic cysteine disulfide (Cys-S-S-Cys) in DsrC during dissimilatory sulfate reduction (33). Similarly, HdrDE could couple the periplasmic oxidation of menaquinol, generated from Fqo, to the cytoplasmic reduction of the disulfide bond in CoM-S-S-CoB. A replacement of HdrDE for DsrMK has been shown in Archaeoglobus fulgidus VC16 (34) and has been proposed for other Mcr-encoding Archaeoglobi MAGs (30).