Global phylogeography and ancient evolution of the widespread human gut virus crAssphage

Abstract

Microbiomes are vast communities of microorganisms and viruses that populate all natural ecosystems. Viruses have been considered to be the most variable component of microbiomes, as supported by virome surveys and examples of high genomic mosaicism. However, recent evidence suggests that the human gut virome is remarkably stable compared with that of other environments. Here, we investigate the origin, evolution and epidemiology of crAssphage, a widespread human gut virus. Through a global collaboration, we obtained DNA sequences of crAssphage from more than one-third of the world's countries and showed that the phylogeography of crAssphage is locally clustered within countries, cities and individuals. We also found fully colinear crAssphage-like genomes in both Old-World and New-World primates, suggesting that the association of crAssphage with primates may be millions of years old. Finally, by exploiting a large cohort of more than 1,000 individuals, we tested whether crAssphage is associated with bacterial taxonomic groups of the gut microbiome, diverse human health parameters and a wide range of dietary factors. We identified strong correlations with different clades of bacteria that are related to Bacteroidetes and weak associations with several diet categories, but no significant association with health or disease. We conclude that crAssphage is a benign cosmopolitan virus that may have coevolved with the human lineage and is an integral part of the normal human gut virome.

Fig. 1 |. crAssphage presence or absence status over time in the human gut.

a–f, Timelines of the crAssphage status of six volunteers (a–f) between April and July 2017, in which each limb of the curve represents a week from Monday to Sunday, and subsequent months are indicated by increasingly intense colours per individual. On the circled dates, individuals were tested for crAssphage using PCR analysis of amplicons A–C; gel electrophoresis of the three amplicons was always consistent for each sample. The black and white circles indicate crAss-positive and crAss-negative samples, respectively. P values indicate the fraction of cases in which crAss status within individuals was more consistently crAss-positive or crAss-negative in 5,000 random permutations of the status labels across all individuals than those that were observed; in cases in which P < 0.0002, none of the random permutations were more consistent. g–i, Unrooted maximum-likelihood phylogenies of amplicons A (g), B (h) and C (i) show clustering of the sequences by volunteer; note that not all crAss-positive samples could be sequenced. Branches with less than 60% bootstrap support were collapsed; values of less than 100% are displayed. Scale bars indicate the average number of mutations per alignment position. Colours correspond to the individual and the month in which the sample was taken.

Fig. 2 |. Diversity of crAssphage strains in metagenomic samples

Strains for three amplicon regions A–C were detected in 2,216 metagenomes using Gretel (Supplementary File 3).

Fig. 3 |. Global locations of 2,424 crAssphage strains for amplicon A.

Pie diagrams indicate the fraction of genetically most-similar strains identified at the same site (less than 150 km apart) and at a different site. The number of strains at each location is indicated by the size of the pie diagram. The inset shows a magnification of Europe. See Supplementary Fig. 3 for amplicons B and C.

Fig. 4 |. Maximum likelihood phylogeny and dot plot showing full genomic colinearity between crAssphage and ten long contigs that were assembled from faecal metagenomes of different non-human primates.

Phylogeny based on a concatenated trimmed-protein alignment of 15 homologous open reading frames (ORFs). The tree is rooted as described previously, and candidate genera are indicated by coloured blocks. All of the branches had 100% bootstrap support, with one exception of <50%, which was collapsed. The scale bar indicates 0.3 mutations per site. nt, nucleotide. For a phylogeny of all 119 crAssphage-like genomes from Alphacrassvirinae and non-human primates, see Supplementary Fig. 5. For the dot plots, all genomes or contigs are shown in separate boxes along the x and y axes of the plot, and regions of similarity between genomes are displayed in white. Similarity is based on high-scoring segment pairs (BLASTn E < 0.001) between all contigs. The figure is to scale, numbers to the left of the dot plot indicate genome or contig lengths. Note that reverse complement sequences result in diagonals from top-right to bottom-left and circular permutation of some genomes leads to apparently broken diagonals in some dot plots.