The methanogen isolate EGK was recovered from a fecal sample of an Eastern Gray Kangaroo (Macropus giganteus), as described by Volmer et al. (2023b). Briefly, the culture of EGK fecal microbiota with the highest methane concentration was used to inoculate BRN-RF10 medium containing 1% (v/v) methanol (Sigma-Aldrich; 179337) and 1% (v/v) ethanol (Sigma-Aldrich; E7023). Following inoculation, the mixture was pressurized to 150 kPa using a gas mixture of H₂:CO₂ in a 80:20 ratio. Streptomycin (600 μg/ml) and ampicillin (200 μg/ml) were also added to this culture to suppress bacterial growth, and the culture was incubated at 37°C with gentle shaking (100 rpm). After several rounds of 10-fold serial dilution in the same media with antibiotics, the resultant culture was confirmed to be bacteria-free through polymerase chain reaction (PCR) (27F/1492R) (Enticknap et al., 2006).

The bacteria-free enrichment was serially diluted 10-fold to extinction, and the highest dilution showing growth was selected for further analysis. Archaeal-specific 16S rRNA PCR (86F/1492R) (Wright and Pimm, 2003) was used to determine an initial taxonomic classification of the archaea, according to the methodology described by Volmer et al. (2023b). Archaeal amplicons were prepared following the gel extraction protocol of the Wizard SV Gel and PCR Clean-Up System, according to the manufacturer's instructions. They were then Sanger sequenced at the Australian Genome Research Facility (AGRF; https://www.agrf.org.au/). The archaeal enrichment was then inoculated onto anaerobic BRN-RF10 agar (1.5% w/v) in an anaerobic chamber (Coy Laboratory Products, MI, USA) with an atmosphere of CO₂:H₂:N₂ in the ratio of 15:5:80, respectively. These agar plates were incubated for 4 weeks at 37°C. After incubation, individual colonies were then selected and propagated using BRN-RF10 broth medium. For cryopreservation, the cultures were subsampled and diluted 1:1 dilution with a sterile, anaerobically prepared 30% (v/v) glycerol buffer, as described by Teh et al. (2021).

The biomass of M. gottschalkii strain EGK was harvested from 100 ml of BRN-RF10 culture with a headspace of 150 kPa of H₂:CO₂ (80:20) by centrifugation at 15,000 × g for 5 min. Both the DNA extraction and its preparation for whole genome sequencing followed the protocols routinely used by the Australian Center for Ecogenomics (https://www.ecogenomic.org). Approximately 200 mg of biomass was transferred into a tube containing ~0.2 g of 0.1 mm glass beads (BioSpec Products #11079101) and combined with 750 μl of Bead Solution (Qiagen #12855-100-BS), along with 60 μl of solution C1. This mixture was then vortexed. The tubes were then heated at 65°C for 10 min and subjected to bead beating at 1,000 × g for 5 min using a Powerlyser 24 homogenizer (Mo-Bio #13155). After centrifugation at 10,000 × g for 1 min, the resulting lysate was then extracted using the Qiagen DNeasy Powersoil Kit (cat #12888-100), with a final elution volume of 50 μl. The sample library was prepared, and QC was performed as described, with the addition of the epMotion (Eppendorf # 5075000301) automated platform for preparation and clean-up. As per the manufacturer's protocol, the library was sequenced using NovaSeq6000 (Illumina) with NovaSeq6000 SP kit v1.5, 2 × 150 bp paired-end chemistry. Sequenced samples were trimmed using Trimmomatic (v0.32) (Bolger et al., 2014) and assembled using Spades (v3.14.1) (Nurk et al., 2017), with contigs <1,000 bps removed. Contigs identified by the NCBI contamination screen were also removed during WGS submission. Contig 17 and 24, which contained prophage-associated genes, were removed from the genome by the NCBI contamination screen and thus have been provided as Supplementary material. The estimated completeness and contamination of each genome assembly were determined using CheckM1 (v1.1.2) (Parks et al., 2015) and CheckM2 (Chklovski et al., 2023). CheckM2 was also used to determine general genome statistics. Taxonomic classification was performed using GTDB-Tk release 214 (v2.3.0) (Chaumeil et al., 2019).

The samples were heat-fixed on glass microscopy slides and stained using the standard Gram staining technique. The Gram status was determined by visualization using a Nikon Eclipse 50i under 100 × magnification. Transmission electron microscopy (TEM) of EGK was conducted by Dr. Rick Webb at the University of Queensland Center for Microscopy and Microanalyses (https://cmm.centre.uq.edu.au/), as described by Volmer et al. (2023b). Wet mount slides of each culture were visualized using a Zeiss AX10 epifluorescence microscope at 420 nm with a cyan (47 HE) filter set to observe autofluorescence.

Representative strains of Methanobrevibacter and other closely related species were identified using the Genome Taxonomy Database (https://gtdb.ecogenomic.org) and downloaded from the NCBI genome database. Average nucleotide identity (ANI) was calculated using fastANI (Jain et al., 2018) and visualized via heatmaps (v1.0.12) in RStudio (2023.12.0). Concatenated archaeal marker gene files were produced using GTDB-Tk release 214 (v2.3.0) (Chaumeil et al., 2019). Phylogeny was determined using IQ-Tree (v2.2.6) with 1000 bootstrap replications (Kalyaanamoorthy et al., 2017; Hoang et al., 2017; Nguyen et al., 2014) and visualized using IToL (Letunic and Bork, 2021).

Comparative genomic analyses were performed on M. gottschalkii DSM11977, DSM11978, EGK, and M. sp. A27. Core and pan-genome analysis was performed using Bacterial Pan-Genome Analysis software (BPGA; v1.3). Comparative analyses of core, accessory, and unique genes were conducted in BPGA using USEARCH (v11.0.667) with a 0.8 similarity threshold. Core genes were defined as being present in all genomes, accessory genes were encoded in multiple genomes, and unique genes were encoded by a single genome. All graphs were visualized in GraphPad Prism (v10.1.0). A further in-depth analysis of differential genes was conducted using DRAM (v1.4.6) (Shaffer et al., 2020). Identified KOs were manually assigned to respective KEGG categories. DRAM annotations were also used to identify tRNAs and rRNAs for each genome. The progressiveMAUVE option in MAUVE (v2015-02-26) (Darling et al., 2004) was used to align M. gottschalkii genomes to determine pairwise genome synteny. The genomes were re-ordered according to DSM11977 as it contained the fewest contigs.

Anaerobically prepared methanogen liquid medium (Volmer et al., 2023b) was inoculated with fecal biomass from various Australian marsupials as part of an effort to study the methanogen populations of native Australian herbivores (Volmer et al., 2023b). After 24 h, cultures with the highest methane positivity from wombat, mahogany glider, and Eastern Gray kangaroo samples were subcultured into fresh media with supplementation of different carbon sources for methanogenesis, as described by Volmer et al. (2023b), and antibiotics to suppress bacterial growth. For the Eastern Gray Kangaroo, the enrichment culture supplemented with methanol, ethanol, CO₂, and H₂ produced the greatest yield, as measured by optical density and methane production.

Microscopic examination of this culture with UV transillumination confirmed the predominance of autofluorescent rod-shaped microbes. This culture was subsequently serially diluted, and the highest dilution supporting microbial growth after incubation was sampled. PCR confirmed the sample to be archaeal-positive and bacterial-negative. The 16S rRNA amplicon sequences produced from this culture had >98% identity to Methanobrevibacter gottschalkii porcine strain PG (98.87% identity) and equine strain HO (98.3% identity) reference sequences (Miller and Lin, 2002).

Methanobrevibacter gottschalkii strain EGK was confirmed to be a Gram-positive staining coccobacillus, measuring ~0.5–1.5 μm in length. Transmission electron microscopy (TEM) of strain EGK cells revealed that the archaeon possesses a singular membrane with regular short cilia-like structures covering the cell surface that may facilitate cell–cell interactions (Figure 1).

The draft genome for M. gottschalkii strain EGK and the related strains DSM11977 and 11978, and sp. A27 lacked some rRNA genes but were assessed to be of high quality (Bowers et al., 2017) based on their predicted completeness and contamination scores, as well as the recovery of tRNAs (Table 1). Interestingly, the genome size and number of predicted coding sequences are greatest for strain EGK, and its genome was also found to possess a substantially larger total number of predicted tRNAs.

Concatenated archaeal marker genes from M. gottschalkii EGK and selected Methanobacteriota reference genomes were identified using GTDB-Tk, and phylogeny was inferred using IQ-Tree (Figure 2). The resulting phylogenetic tree confirms that strain EGK represents a novel isolate of Methanobrevibacter gottschalkii within the genus Methanobrevibacter_A, alongside Methanobrevibacter smithii and M. smithii_A.

This finding was further validated by pairwise average nucleotide identity (ANI) analysis, which showed that EGK forms a distinct cluster with other M. gottschalkii genomes, sharing >99% ANI, well above the 95% ANI threshold used for species demarcation (Supplementary Table S2) (Richter and Rosselló-Móra, 2009). The deep phylogenetic divergence of Methanobrevibacter_A, supported by relative evolutionary divergence values from GTDB, suggests that Methanobrevibacter_A likely represents a unique genus separate from Methanobrevibacter, suggesting that the type strain of Methanobrevibacter, M. ruminantium, belongs to a distinct genus from M. gottschalkii. Moreover, as M. ruminantium and M. gottschalkii are associated with different methane-emitting phenotypes, this distinction highlights important phylogenetic and biological differences between these dominant ruminant methanogens. Therefore, we propose that Methanobrevibacter_A be reclassified as Methanoenterobius to better reflect the host-associated nature of species within this genus.

Based on the results presented in Figures 2, 3, the genomes of the four M. gottschalkii strains were further compared, with the “core” genome of 1,573 genes representing ~83% of the total gene counts (Figure 3A). Most of the genes that could be confidently assigned to KOs within the core and non-core (i.e., accessory and unique) components of the pan-genome were associated with metabolism (Figure 3B). Further categorization of core genes identified 22 categories, with the most populated being “general overview” (17.4%), “amino acid metabolism” (12.9%), “carbohydrate metabolism” (11.4%), “energy metabolism” (10.8%), and “translation” (9.8%; Figure 3C). Similarly, genes with KO classifications from the non-core were assigned to many of the same subcategories, although at greater proportions than core genes for membrane transport (11.5%) and xenobiotic biodegradation and metabolism (11.5%). KOs exclusively encoded by one of the four genomes were mostly assigned to nucleotide metabolism (20.6%) and metabolism of cofactors and vitamins (17.6%). These differences in the non-core components are most likely adaptations to the variations in the nutritional and/or physiological ecology of the different hosts.

Methanogenesis-associated genes were highly conserved between the strains, suggesting that all strains can perform hydrogenotrophic methanogenesis using CO₂/H₂ and formate. The only differentially encoded KO potentially involved in methane metabolism was glycerate dehydrogenase (hprA; K00018; Figure 4A), which was not encoded by DSM11978. Most differentially enriched KOs associated with genetic information processing were encoded by EGK and enriched against M. sp. A27, with DNA helicase IV (helD, K03658) and cell filamentation protein fic (K04095) encoded by EGK (Figure 4B). Three ABC transporters encoding for iron transport (K02013, K02015, and K02016) were present in all strains except M. gottschalkii DSM11978 (Figure 4D).

Additional uncharacterized KOs (Figure 4E) and numerous genes without KO annotations were differentially enriched, requiring further functional investigation to determine their effect on the phenotypic characteristics of each strain.

Given the larger genome size of M. gottschalkii EGK and its unusual animal host, further comparative analyses were performed to identify any strain-specific genomic characteristics. Sequence alignment of the M. gottschalkii strains showed high synteny among the genomes, which is expected given that they represent different strains of the same species. However, both DSM11978 and EGK contain multiple strain-specific regions. EGK contains two contigs (~88 and ~40 Kbp, respectively), with only one of the contigs containing a small region of synteny with DSM11978 (Figure 5). These regions contain many predicted tRNAs, which may explain the larger total numbers compared to the other strains (Table 1). In total, 92% (117/127) and 97% (43/44) of the genes for the two EGK contigs were hypothetical. The only annotated genes in these regions were predicted phage proteins (head, baseplate, and tail proteins), endonucleases, phage lysozymes, and invasins, suggesting that EGK may have acquired these genes through prophage integration, and DSM11978 may contain a prophage from a different acquisition event.

Further comparison of these genes showed EGK encoded for dihydrofolate reductase (K00287) and thymidylate synthase (K00560), along with two copies of lysozyme (K01185), anaerobic ribonucleoside-triphosphate reductase activating protein (K04068), and protease IV (K04773; Figure 4C), all located in the ~80Kbp region identified (Figure 4). Tetracycline resistance protein tetA (K08151) and a putative prophage DNA-invertase (K14060) were additionally encoded by EGK (Figures 4D, E). These genes were also encoded on a ~10 Kbps contig along with phage and virus-associated proteins, suggesting that this contig may represent additional viral genes.

We suggest the reclassification of Methanobrevibacter_A to Methanoenterobius (Entero-, referring to the intestinal tract of animals; methano-, referring to methane production; and -brevibacter, referring to cellular morphology of the organisms) to identify Methanobrevibacter_A as a unique genus and reflect its prevalence in gastrointestinal-associated microbiomes of animals.