Wide diversity of methane and short-chain alkane metabolisms in uncultured archaea

Abstract

Methanogenesis is an ancient metabolism of key ecological relevance, with direct impact on the evolution of Earth's climate. Recent results suggest that the diversity of methane metabolisms and their derivations have probably been vastly underestimated. Here, by probing thousands of publicly available metagenomes for homologues of methyl-coenzyme M reductase complex (MCR), we have obtained ten metagenome-assembled genomes (MAGs) belonging to potential methanogenic, anaerobic methanotrophic and short-chain alkane-oxidizing archaea. Five of these MAGs represent under-sampled (Verstraetearchaeota, Methanonatronarchaeia, ANME-1 and GoM-Arc1) or previously genomically undescribed (ANME-2c) archaeal lineages. The remaining five MAGs correspond to lineages that are only distantly related to previously known methanogens and span the entire archaeal phylogeny. Comprehensive comparative annotation substantially expands the metabolic diversity and energy conservation systems of MCR-bearing archaea. It also suggests the potential existence of a yet uncharacterized type of methanogenesis linked to short-chain alkane/fatty acid oxidation in a previously undescribed class of archaea ('Candidatus Methanoliparia'). We redefine a common core of marker genes specific to methanogenic, anaerobic methanotrophic and short-chain alkane-oxidizing archaea, and propose a possible scenario for the evolutionary and functional transitions that led to the emergence of such metabolic diversity.

Figure 1

Placement of nine MAGs described in this study in the reference phylogeny of Archaea. NM2 was not included due to low completeness. Bayesian phylogeny (PhyloBayes, CAT+GTR+ Γ4) based on concatenation of 40 conserved phylogenetic markers (8,564 amino acid positions) and 156 genomes/MAGs (see Supplementary Table 5 and Supplementary Table 6 for detail). Node supports refer to posterior probabilities, and for reasons of readability only values above 0.8 are shown. The tree is rooted according to Raymann et al.. The scale bar represents the average number of substitutions per site. Black arrows point to the 9 obtained MAGs and accolated pie charts indicate their estimated completeness. Colors indicate that genomes of these lineages encode an MCR/MCR-like complex, Class I/II methanogens are in green, methyl-dependent hydrogenotrophic lineages are in red, methanotrophs are in orange (some being within Class II), potential or validated shot-chain alkane users are in in blue. NM1 could also have a methane metabolism (see text for discussion).

Figure 2

Phylogeny of the MCR/MCR-like complex and conservation of important positions in the catalytic site. A) Unrooted Bayesian phylogeny (CAT+GTR+Γ4) based on a concatenation of McrABG/McrABG-like subunits (1,187 amino acid positions) from 109 genomes/MAGs (see Supplementary Table 6 for details). Node supports refer to posterior probabilities, and for reasons of readability only values above 0.8 are shown. The scale bar represents the average number of substitutions per site. The color code is similar to that in Fig. 1 with the exception of NM1 which have both an MCR-like (in blue) and a canonical MCR (in purple) (see text for discussion). B) Conservation of 17 residues previously described to interact with CoM, CoB, F₄₃₀ cofactors, making part of the substrate cavity wall, or having post-translational modifications,,. Replacement of conserved amino acids associated to a negative value in the Blosum45 matrix are indicated by white on black background, those with a null or positive value in the Blosum45 matrix are in bold, “.” indicate conserved positions and “-“ indicate missing positions in the sequence due to sequencing incompleteness.

Figure 3

Predicted methane and short-chain alkane metabolism of the MAGs described in this study, with the exception of NM1, which is presented in Fig. 4, and NM2, which has a low completeness. Colored arrows correspond to reactions modifying or transferring the C1 carbon group of the substrate. Details on the annotation of the enzymes are presented in Supplementary Table 1. MFR, methanofuran; H₄MPT, tetrahydromethanopterin; Fd, ferredoxin; F₄₂₀, coenzyme F₄₂₀; LCFA, Long Chain Fatty Acids; MQ, menaquinone; Mp, methanophenazine; Mhc, c-type multiheme cytochromes; DIET, Direct Interspecies Electron Transfer. Grey color indicates the absence of the enzyme, complex, reaction or compound. Comparisons of with other methane-cycling or short-chain alkane oxidizers, which are discussed in the text, are presented in Supplementary Figs 1 and 3. The percentages between brackets indicate the estimated completeness of the corresponding MAGs.

Figure 4

Predicted methane, short-chain alkane and long/medium-chain fatty acid metabolism of the two MAGs NM1a (“Ca. Methanoliparum thermophilum” completeness of 92.5%) and NM1b (“Ca. Methanolliviera hydrocarbonicum” completeness of 90.2%) belonging to the candidate class “Ca. Methanoliparia”. Colored arrows correspond to reactions present in both MAGs which modify or transfer the carbon group(s) of the substrate. Predicted possible metabolisms are the utilisation of short-chain alkanes and long/medium-chain fatty acids (L/MCFA) (blue), methanotrophy (orange) and methanogenesis from short-chain alkanes or L/MCFA (purple). Details on the annotation of the enzymes are provided in Supplementary Table 1. MFR, methanofuran; H₄MPT, tetrahydromethanopterin; Fd, ferredoxin; F₄₂₀, coenzyme F₄₂₀; L/MCFA, Long/medium chain fatty acids; MQ, menaquinone; Mhc, c-type multiheme cytochromes; DIET, Direct Interspecies Electron Transfer. Grey color indicates the absence of the corresponding enzyme, complex, reaction or compound in both MAGs.