Detection of an archaeal-specific viral family, previously thought to infect exclusively hyperthermophiles, in human gut metaviromes

International audience

إنجليزي. Viruses of archaea have an exceptional diversity [1]. Several of their families are shared with bacteria, typically those from the cosmopolitan Caudovirales order. However, the majority of them are archaea-specific; they have very diverse genomic contents and morphotypes, such as two-tailed (Bicaudaviridae), spindle (Fuselloviridae, Salterprovirus), rod (Rudiviridae), or filamentous (Lipothrixviridae) shapes. Most of the known archaea-specific viruses infecthyperthermophiles and halophiles. Nevertheless, the viruses of many archaeal phylogenetic groups remain poorly explored, such as those of methanogens. Six of the seven known methanogen-infecting viruses belong to the Caudovirales order [1]. Interestingly, a spherical virus infecting Methanosarcina strains has recently been isolated [2]. Named MetSV, itprobably represents a new family [1]. The smacoviruses have moreover been suggested to infect methanogens rather than human, based on metavirome analysis [3]. Finally, unusual morphotypes have been detected in metaviromes from anaerobic digestion plants, as well as contigs of unknown hosts from various archaea-specific viral families: Salterprovirus, Bicaudaviridae, Fuselloviridae, Lipothrixviridae, and Rudiviridae [4].To identify putative viruses of methanogens in metaviromes from anaerobic ecosystems, we developed a bioinformatics pipe-line based on metaSPADES [5] for assembly and Kaiju [6] for taxonomic assignation. As a first test, we applied our pipe-line to the raw reads of 4previously published human gut metaviromes [7]. A total of 12 short contigs (~1 kb-3.5 kb) were assigned by Kaiju to known families of archaeal viruses, corresponding to Rudiviridaeexclusively; this assignation was confirmed by blast analyses. Using WiSH [8], we predictedtheir possible hosts among archaea. Nine of them had best predictions for Sulfolobales members, their presently known acidothermophilic hosts, while the remaining 3 were predicted to infect methanogens (genera Methanosarcina, Methanosphaera and Methanobrevibacter). It suggests that some of these contigs could originate from viruses of methanogens, consistent with methanogens being the dominant archaea in the human gut.In the future, we plan to include a CRISPR-spacer matching tool to our pipe-line, to predict the hosts by a complementary method. Using a state-of-the art assembler [9] yielded only short Rudiviridae contigs, likely because ruvidiruses have a low abundance in the human gut.It advocates for the use of enrichment approaches to study them more specifically. Ourresults reinforce the notion that archaea-specific viruses may not be mostly restricted to extremophiles and could be more widespread in the biosphere than previously thought.[1] Krupovic, M., Cvirkaite-Krupovic, V., Iranzo, J., Prangishvili, D., Koonin, E. V. Virus Res. 244, 181-193 (2018).[2] Weidenbach, K., Nickel, L., Neve, H., Alkhnbashi, O. S., Künzel, S., Kupczok, A., Bauersachs, T., Cassidy, L., Tholey, A., Backofen, R., Schmitz, R.A. J. Virol. 91, e00955-00917 (2017).[3] Díez-Villaseñor, C. and Rodriguez-Valera, F., Nat. Commun. 10(1), p.294 (2019).[4] Calusinska, M., Marynowska, M., Goux, X., Lentzen, E., Delfosse, P. Environ. Microbiol. 18, 1162-1175 (2016).[5] Nurk, S., Meleshko, D., Korobeynikov, A., Pevzner, P. A. Genome Res. 27, 824-834 (2017).[6] Roux, S., Emerson, J. B., Eloe-Fadrosh, E. A., Sullivan, M. B. PeerJ, 5, e3817 (2017).[7] Menzel, P., Ng, K.L. and Krogh, A. Nat. Commun., 7, p.11257 (2016).[8] Manrique, P., Bolduc, B., Walk, S. T., van der Oost, J., de Vos, W.M., Young, M.J. Proc. Natl. Acad. Sci. U. S. A.113, 10400 (2016).[9] Galiez, C., Siebert, M., Enault, F., Vincent, J., Söding, J. Bioinformatics, 33, 3113-3114 (2017).