A genome-wide study of ruminants uncovers two endogenous retrovirus families recently active in goats

Abstract

Background: Endogenous retroviruses (ERV) are traces of ancestral retroviral germline infections that constitute a significant portion of mammalian genomes and are classified as LTR-retrotransposons. The exploration of their dynamics and evolutionary history in ruminants remains limited, highlighting the need for a comprehensive and thorough investigation of the ERV landscape in the genomes of cattle, sheep and goat.

Results: Through a de novo bioinformatic analysis, we characterized 24 Class I and II ERV families across four reference assemblies of domestic and wild sheep and goats, and one assembly of cattle. Among these families, 13 are represented by consensus sequences identified in the five analyzed species, while eight are exclusive to small ruminants and three to cattle. The similarity-based approach used to search for the presence of these families in other ruminant species revealed multiple endogenization events over the last 40 million years and distinct evolutionary dynamics among species. The ERV annotation resulted in a high-resolution dataset of 100,534 ERV insertions across the five genomes, representing between 0.5 and 1% of their genomes. Solo-LTRs account for 83.2% of the annotated insertions demonstrating that most of the ERVs are relics of past events. Two Class II families showed higher abundance and copy conservation in small ruminants. One of them is closely related to circulating exogenous retroviruses and is represented by 22 copies sharing identical LTRs and 12 with complete coding capacities in the domestic goat.

Conclusions: Our results suggest the presence of two ERV families with recent transpositional activity in ruminant genomes, particularly in the domestic goat, illustrating distinct evolutionary dynamics among the analyzed species. This work highlights the ongoing influence of ERVs on genomic landscapes and call for further investigation of their evolutionary trajectories in these genomes.

Keywords: ERV; Evolution; Genome annotation; Goat; LTR retrotransposon; Ruminant.

Fig. 1

ERV families in ruminant reference genomes. Maximum Likelihood phylogenetic tree reconstructed from the alignment of the consensus sequences (without LTRs) generated from the domestic and wild sheep and goat as well as cattle reference assemblies. Publicly available sequences of exogenous and endogenous retroviruses are indicated by their acronyms: GaLV, Gibbon Ape Leukemia Virus; KoRV, Koala Retrovirus; MuLV, Murine Leukemia Virus; FeLV, Feline Leukemia Virus; MMVT, Mouse Mammary Tumor Virus; ENTV1 and 2, Enzootic Nasal Tumor Virus Type 1 and 2; JSRV, Jaagsiekte Sheep Retrovirus; HERV, Human endogenous retrovirus; enFeLV, Feline endogenous retrovirus; MuERV-L, murine endogenous retrovirus. Sheep and cattle Repbase ERV references are represented in italic. The ERV family’s names include their classification in Class I or II along with an arbitrary attributed number and are indicated next to each consensus sequence cluster

Fig. 2

ERV family integration events across ruminant evolution. Each ERV family was detected in both small ruminant and cattle reference assemblies (highlighted in bold), along with 20 additional ruminant species represented on the tree (see Supplementary Material 11 - Tab. S1 for accession numbers). Red dots, which represent the integration events, were placed before the oldest node including all the species in which each family was found present. The phylogeny was produced using TimeTree [65] coupled with divergence time from [66, 67]

Fig. 3

Family proportion and divergence landscapes of ERV in ruminant reference assemblies. Class I and Class II are represented separately in A) and B) panels respectively. The left panels represent ERV family proportions in cattle and small ruminant assemblies. Over- and under-represented families were identified comparing Caprinae species (see Pearson’s χ² residuals in Supplementary Fig S4). In the right panels, for each family and Kimura-2 sequence divergence interval, the genome coverage was computed as the percentage of the total ERV insertion length on the total genome length. The divergence distribution was compared between species using the discrete Kolmogorov–Smirnov test and the Benjamini & Hochberg correction (Statistics in Supplementary Material 11 - Tab. S4)

Fig. 4

Characteristics of the family II-3 copies in small ruminant genomes. A Number of family II-3 ERV copies excluding solo-LTRs in 29 assemblies from domestic sheep and goat of different breeds (accession numbers in Supplementary Material 11 - Tab. S1). B Length distribution of the ERV insertions for each of the small ruminant reference assembly. The red line indicates the mean length. C Number of common ERV loci excluding solo-LTRs between the species. Only insertion sites flanked by at least 100 bp of sequence on both sides were retained. D Phylogenetic tree of the family II-3 copies in C. hircus reference genome excluding solo-LTRs. Copies sharing insertion sites with other species are reported with symbols: dark blue circle for wild goat, dark green triangle for wild sheep and soft green circle for domestic sheep. To clarify the tree, nodes including only domestic goat specific insertions without any synteny and nearly identical sequences were collapsed. The white squares indicate branches supported by a bootstrap higher than 80%. Branch lengths are expressed as the number of substitutions per site. E Comparison of the expected complete consensus sequence with the one from family II-3 in small ruminants. The green boxes represent the retroviral genes and the orange ones the coding sequence (ORF)

Fig. 5

Characteristics of the family II-5 copies in small ruminant genomes. A Number of family II-5 ERV copies excluding solo-LTRs in 29 assemblies from domestic sheep and goat of different breeds (accession numbers in Supplementary Material 11 - Tab. S1). B Length distribution of the ERV insertions for each of the small ruminant reference assembly. The red line indicates the mean length. C Number of common ERV loci excluding solo-LTRs between the species. Only insertions sites flanked by at least 100 bp of sequence on both sides were retained. D Phylogenetic tree of the family II-5 copies in C. hircus reference genome excluding solo-LTRs. Copies sharing insertion sites with wild goat are represented by dark blue circle. The orange stars represent the insertions with complete gag, pro, pol and env ORFs and the pink stars, the copies missing only the complete env one. To clarify the tree, nodes including only domestic goat specific insertions without any synteny and nearly identical sequences were collapsed. The white squares indicate the branch supported by a bootstrap of 100%. Branch lengths are expressed as the number of substitutions per site. E Comparison of the family II-5 consensus sequences between small ruminants. The green boxes represent the retroviral genes and the orange ones the coding sequence (ORF)