The diverse liver viromes of Australian geckos and skinks are dominated by hepaciviruses and picornaviruses and reflect host taxonomy and habitat

Abstract

Lizards have diverse ecologies and evolutionary histories, and represent a promising group to explore how hosts shape virome structure and virus evolution. Yet, little is known about the viromes of these animals. In Australia, squamates (lizards and snakes) comprise the most diverse order of vertebrates, and Australia hosts the highest diversity of lizards globally, with the greatest breadth of habitat use. We used meta-transcriptomic sequencing to determine the virome of nine co-distributed, tropical lizard species from three taxonomic families in Australia and analyzed these data to identify host traits associated with viral abundance and diversity. We show that lizards carry a large diversity of viruses, identifying more than thirty novel, highly divergent vertebrate-associated viruses. These viruses were from nine viral families, including several that contain well known pathogens, such as the Flaviviridae, Picornaviridae, Bornaviridae, Iridoviridae, and Rhabdoviridae. Members of the Flaviviridae were particularly abundant across species sampled here, largely belonging to the genus Hepacivirus: fourteen novel hepaciviruses were identified, broadening the known diversity of this group and better defining its evolution by uncovering new reptilian clades. The evolutionary histories of the viruses studied here frequently aligned with the biogeographic and phylogenetic histories of the hosts, indicating that exogenous viruses may help infer host evolutionary history if sampling is strategic and sampling density high enough. Notably, analysis of alpha and beta diversity revealed that virome composition and richness in the animals sampled here was shaped by host taxonomy and habitat. In sum, we identified a diverse range of reptile viruses that broadly contributes to our understanding of virus-host ecology and evolution.

Keywords: Hepacivirus; evolution; meta-transcriptomics; metagenomics; one health; viral ecology.

Figure 1.

Vertebrate-infecting viruses in the sampled lizard species. (A) Relative abundance (left plot) and overall abundance (right) of the vertebrate-infecting virus families present in each library. Libraries are plotted in taxonomic sequence, with host relationships indicated by a cladogram and host infraorder indicated in grey bars. Library names are as follows: Cam_M, Carlia amax (collected from a mesic environment); Cmun_M, Carlia munda (mesic), Csex_M, Carlia sexdentata (mesic); Cmet_M, Cryptoblepharus metallicus (mesic); Gnan_A, Gehyra nana (arid); Gnan_M Gehyra nana (mesic); Garn_M, Gehyra arnhemica (mesic); Hbin_A, Heteronotia binoei (arid); Hbin_M, Heteronotia binoei (mesic); Hplan_A, Heteronotia planiceps (arid); Omar_M, Oedura marmorata (mesic). (B) Viruses found in each library are represented by circles colored by viral genus (UC = unclassified genus), with lines connecting them to the libraries in which they were found, represented by lizard silhouettes. Libraries are grouped vertically by host family (indicated by grey bars above; Diplod. = Diplodactylidae), and horizontally by habitat (trees: Cmet_M, Garn_M; rocks: Cam_M, Gnan_A, Gnan_M, Hplan_A, Omar_M; habitat generalist [GEN]: Cmun_M, Hbin_A, Hbin_M; riparian: Csex_M) which is further indicated by the color of the lizard silhouette. Numbers in parentheses beside virus names indicate the number of variants of that virus detected (where >1). Note that all the Amnoonviridae belong to currently unclassified genera, but may represent more than one genus.

Figure 2.

Maximum likelihood phylogeny of the RdRp of Flaviviridae in Australian lizards. Taxon names are colored according to apparent host. Viruses discovered in this study are indicated by bold and italicized taxa names and lizard silhouettes beside taxa names, colored by host genus. Squares next to the taxa names indicate the sampling location/environment (env.) for viruses discovered in this study. An asterisk beside the taxa name for viruses detected here indicates that the sequence is not the complete length of the alignment. The accession number is indicated in the taxon name for all sequences. Circles at the nodes represent the branch support as estimated using the SH-like approximate likelihood ratio test. Trees are mid-point rooted for clarity. Scale bars indicate the number of amino acid substitutions per site. To achieve greater resolution, the Hepacivirus phylogeny was estimated from a sequence alignment of this genus only.

Figure 3.

Maximum likelihood phylogenies of the RdRp of positive-sense RNA viruses in Australian lizards (with the exception of the Flaviviridae, shown in Fig. 2). Taxon names are colored according to apparent host. Viruses discovered in this study are indicated by bold and italicized taxa names and lizard silhouettes beside taxa names, which are colored by host genus. Squares next to the taxa names indicate the sampling location/environment (env.) for viruses discovered in this study. An asterisk beside the taxa name for viruses detected here indicates that the sequence is not the complete length of the alignment. The accession number is indicated in the taxon name for all sequences. Circles at the nodes represent the branch support as estimated using the SH-like approximate likelihood ratio test. Trees are mid-point rooted for clarity. Scale bars indicate the number of amino acid substitutions per site. Host family and genus names are indicated for the Lepidosauria in the Astroviridae phylogeny to demonstrate virus-host co-divergence within the Lepidosauria.

Figure 4.

Maximum likelihood phylogenies of the RdRp of negative-sense RNA viruses in Australian lizards. Taxon names are colored according to apparent host. Viruses discovered in this study are indicated by bold and italicized taxa names and lizard silhouettes beside taxa names, which are colored by host genus. Squares next to the taxa names indicate the sampling location/environment (env.) for viruses discovered in this study. An asterisk beside the taxa name for viruses detected here indicates that the sequence is not the complete length of the alignment. Accession numbers are indicated in the taxon name for all sequences. Circles at the nodes represent the branch support as estimated using the SH-like approximate likelihood ratio test. Trees are mid-point rooted for clarity. Scale bars indicate the number of amino acid substitutions per site.

Figure 5.

Maximum likelihood phylogenetic tree of the major capsid protein of the Iridoviridae in Australian lizards. Taxon names are colored according to apparent host. The virus discovered in this study is indicated by a bold and italicized taxon name and a lizard silhouette beside the taxon name, which is colored by host genus. The square next to the taxon name indicates the sampling location/environment (env.) for the virus discovered in this study. The accession number is indicated in the taxon name for all sequences. Circles at the nodes represent the branch support as estimated using the SH-like approximate likelihood ratio test. The tree is mid-point rooted for clarity, and the scale bar indicates the number of amino acid substitutions per site.