A global virome of methanogenic archaea highlights novel diversity and adaptations to the gut environment

Abstract

Mobile genetic elements (MGEs), especially viruses, have a major impact on microbial communities. Methanogenic archaea play key environmental and economical roles, being the main producers of methane -a potent greenhouse gas and an energy source. They are widespread in diverse anoxic artificial and natural environments, including animal gut microbiomes. However, their viruses remain vastly unknown. Here, we carried out a global investigation of MGEs in 3436 genomes and metagenome-assembled genomes covering all known diversity of methanogens and using a newly assembled CRISPR database consisting of 60,000 spacers of methanogens, the most extensive collection to date. We obtained 248 high-quality (pro)viral and 63 plasmid sequences assigned to hosts belonging to nine main orders of methanogenic archaea, including the first MGEs of Methanonatronarchaeales, Methanocellales and Methanoliparales archaea. We found novel CRISPR arrays in ‘Ca. Methanomassiliicoccus intestinalis’ and ‘Ca. Methanomethylophilus’ genomes with spacers targeting small ssDNA viruses of the Smacoviridae, supporting and extending the hypothesis of an interaction between smacoviruses and gut associated Methanomassiliicoccales. Gene network analysis shows that methanogens encompass a unique and interconnected MGE repertoire, including novel viral families belonging to head-tailed Caudoviricetes, but also icosahedral and archaeal-specific pleomorphic, spherical, and spindle (pro)viruses. We reveal well-delineated modules for virus-host interaction, genome replication and virion assembly, and a rich repertoire of defense and counter-defense systems suggesting a highly dynamic and complex network of interactions between methanogens and their MGEs. We also identify potential conjugation systems composed of VirB4, VirB5 and VirB6 proteins encoded on plasmids and (pro)viruses of Methanosarcinales, the first report in Euryarchaeota. We identified 15 new families of viruses infecting Methanobacteriales, the most prominent archaea in the gut microbiome. These encode a large repertoire of protein domains for recognizing and cleaving pseudomurein for viral entry and egress, suggesting convergent adaptation of bacterial and archaeal viruses to the presence of a cell wall. Finally, we highlight an enrichment of glycan-binding domains (immunoglobulin-like (Ig-like)/Flg_new) and diversity-generating retroelements (DGRs) in viruses from gut-associated methanogens, suggesting a role in adaptation to host environments and remarkable convergence with phages infecting gut-associated bacteria. Our work represents an important step toward the characterization of the vast repertoire of MGEs associated with methanogens, including a better understanding of their role in regulating their communities globally and the development of much-needed genetic tools.

Revisiting evolutionary trajectories and the organization of the Pleolipoviridae family

“…However, our inability to detect them in other environments could be caused by the low sequence similarity values ( S1 Fig ) and the dramatic variation in gene content among pleolipoviruses. Additionally, recent work has suggested that some pleolipo-like viruses could be part of the human gut virome of methanogenic archaea [ 22 ], further expanding their habitat distribution and suggesting that the relevance of this virus group might be underestimated.…”

A 2-million-year-old microbial and viral communities from the Kap København Formation in North Greenland

“…6A), all belonging to the class Caudoviricetes (phylum Uroviricota), the dominant group of prokaryotic viruses. To further expand the reach of the homology detection, we searched the translated reads against the proteomes of viruses from the IMG/VR included in our database as well as a curated collection of mobile genetic elements (MGE) associated with methanogenic archaea 65 . After the search step, we identified which of the proteins have a higher likelihood to be present in a sample, and then selected those genomes that have over 20% of their proteins detected.…”