Association between genetic clades and cancer prevalence suggested by French-wide study of oncogenic small ruminant β-retrovirus diversity

Abstract

Introduction: ENTV (Enzootic Nasal Tumor Virus) and JSRV (Jaagsiekte Sheep Retrovirus) are β-retroviruses responsible for respiratory cancers in sheep and goats. In this study, we analyzed the genetic features of the sheep and goat β-Retroviruses (29 JSRV and 24 ENTV strains) circulating in France to identify molecular signatures associated with disease severity in flocks.

Methods: We developed a highly specific PCR to amplify and sequence exogenous targeted regions or near full length proviruses based on limited discriminating motifs along their genomes.

Results: The phylogenetic reconstructions based on the Long Terminal Repeat (LTR) and env regions suggest that one major strain is circulating on the French territory for ENTV-1 and ENTV-2 while not clustering with already published Spanish, Canadian or Chinese strains. JSRV strains circulating in French sheep flocks were distributed in 2 distinct genetic clades clustering with sequences originating from North America, Africa and United-Kingdom. JSRV clade I was found to be associated with a higher incidence of cancer in French flocks. Specific motifs spanning the entire JSRV genome particularly in the LTRs and in the intracytoplasmic domain of the envelope were detected between the two genetic subtypes.

Discussion: This work represents the first nationwide study describing the circulation of the three closely related β-oncogenic retroviruses JSRV, ENTV-1 and ENTV-2 in French sheep and goat flocks. Better characterization of strain genetics is a critical step in monitoring circulating - retroviruses, especially those associated with higher cancer incidence in small ruminants.

Keywords: ENTV-1; ENTV-2; JSRV; genomic epidemiology; virulence factors.

Figure 1

Complexity of distinguishing between the exogenous retroviruses and their endogenous counterparts in small ruminants. Phylogenetic analysis of ENTV-1, ENTV-2, and JSRV sequences and their related endogenous sequences from sheep and goats, available in public databases. Maximum likelihood tree with an ultrafast bootstrap parameter of 10,000 iterations using EF680307.1 (enJSRV10) as the root sequence for (A) the full-length proviral sequence and (B) the env region. The last two letters of each sequence define the geographic origin. CH, China; UK, United Kingdom; CA, Canada; IN, India; SA, South Africa; SP, Spain; EU, Europe; UN, unknown. Bootstrap values greater than 80% are represented by a gray dot and the branch length is indicative of the number of substitutions per site. The scale bar indicates the number of nucleotide substitutions per site. (C) Alignment of deduced amino acid sequences of the ENTV-2 and gERV Env TM region. The position of the YXXM motif, the molecular marker for the transforming exogenous strains of JSRV, ENTV-1, and ENTV-2, is indicated in red. “1” defines the sequences with ENTV-2 as a taxonomic assignment in GenBank (Taxonomy IDs: 239365, 2913605, and 2584748) and with a complete YXXM motif. “2” defines the sequences with ENTV-2 as a taxonomic assignment in GenBank but without the YXXM motif. “3” defines the sequences with gERV as a taxonomic assignment in GenBank (Taxonomy ID: 2762664). Dots represent identical amino acids to NC_004994.2 and dashes indicate gaps in the alignment.

Figure 2

French JSRV and ENTV strains are distributed in distinct phylogenetic clades. (A, B) Maximum likelihood tree with an ultrafast bootstrap parameter of 10,000 iterations for (A) env (1851 bp) and (B) reconstituted LTR regions (458 bp). The French sequences described in this study are in bold. CH, China; UK, United Kingdom; CA, Canada; IN, India; SA, South Africa; SP, Spain; EU, Europe; KE, Kenya; US, United States; UN, unknown. Bootstrap values greater than 80% are represented by a gray dot. The scale bar indicates the number of nucleotide substitutions per site. Identical sequences from the same flock were removed from the phylogenetic analysis.

Figure 3

Variability ENTV-2 and JSRV. Nucleotide identity was determined along the reconstituted full-length proviruses of (A) ENTV-2 and (B) JSRV. Each sequence is from a different flock.

Figure 4

Amino acid sequence diversity of JSRV ORFs. (A–D) Calculated pairwise identity in deducted aa along the different ORFs of JSRV calculated with a sliding window of 10 aa and steps of 2 aa. The different protein/domains encoded by the ORF are shown above each graph. The YXXM motif is represented with a gray triangle. Stop codons are indicated by asterisks. (A) Gag, (B) Pro, (C) Pol, and (D) Env.

Figure 5

JSRVs clade I and clade II differ in the C-Terminal region of Env. Alignment of the deduced amino acid sequence of two representative JSRV clade I (2369 FR) and JSRV clade II (1481 FR) sequences. The localization of the YXXM motif is represented by a black square. The dots represent identical amino acids between the two sequences, and the asterisk represents the stop codon. YXXM motif starts at the 590th amino acid of JSRV Env.

Figure 6

Differential transcription factor binding site prediction on JSRV LTR clade I and clade II. Predicted with the EMBOSS tool TF scan on LTR clade I and clade II sequences, transcription factor binding sites unique to each LTR were presented. Transcription factor binding site prediction was performed on consensus LTR sequences for clade I or clade II. The frequency of each site was tested for both clade using the EMBOSS tool TF scan on each individual LTR sequence (9 sequences for clade I and 13 sequences for clade II). U3, unique 3′; R, repeat; U5, unique 5′.

Figure 7

JSRV genotyping by PCR. Agarose gel migration of PCR amplicons corresponding to a region of the JSRV LTR measured at 153 bp for clade I and at 245 bp for clade II. DNA extracted from JSRV-induced tumors of known clade: clade I (1751, 2369, and 2586) and clade II (1169, 1481, and 1762) were used for this proof of concept. IDO5 is an ovine dermal fibroblast cell line used as a JSRV primer specificity control. Thyroglobulin: amplicon corresponding to a region of the thyroglobulin gene in the sheep genome, measured at 163 bp.

Figure 8

Phylogenetic relationships of the French JSRV and ENTV strains with other endogenous and exogenous small ruminant β-retrovirus. Maximum likelihood tree of the env region using 10,000 ultrafast bootstrap repeats with EF680307.1 (enJSRV10) as the rooted sequence. CH, China; UK, United Kingdom; CA, Canada; IN, India; SA, South Africa; SP, Spain; EU, Europe; UN, unknown. Bootstrap values greater than 80% are represented by a gray dot and the branch lengths indicate the number of substitutions per site.