Specific capture and whole-genome phylogeography of Dolphin morbillivirus

Abstract

Dolphin morbillivirus (DMV) is considered an emerging threat having caused several epidemics worldwide. Only few DMV genomes are publicly available. Here, we report the use of target enrichment directly from cetacean tissues to obtain novel DMV genome sequences, with sequence comparison and phylodynamic analysis. RNA from 15 tissue samples of cetaceans stranded along the Italian and French coasts (2008-2017) was purified and processed using custom probes (by bait hybridization) for target enrichment and sequenced on Illumina MiSeq. Data were mapped against the reference genome, and the novel sequences were aligned to the available genome sequences. The alignment was then used for phylogenetic and phylogeographic analysis using MrBayes and BEAST. We herein report that target enrichment by specific capture may be a successful strategy for whole-genome sequencing of DMV directly from field samples. By this strategy, 14 complete and one partially complete genomes were obtained, with reads mapping to the virus up to 98% and coverage up to 7800X. The phylogenetic tree well discriminated the Mediterranean and the NE-Atlantic strains, circulating in the Mediterranean Sea and causing two different epidemics (2008-2015 and 2014-2017, respectively), with a limited time overlap of the two strains, sharing a common ancestor approximately in 1998.

Figure 1

Bayesian phylogenetic tree based on the nucleotide genome sequence of DMV. The tree was inferred by MrBayes using the genomic data partitioned by gene, with the proper nucleotide substitution model, according to the bModelTest results: GTR for H, M, and PVC genes, GTR + G + I for F, N, and L genes, and JC69 + G + I for the noncoding partition. Taxa names in bold are the new sequences presented in this work. For Italian samples presented here, the province abbreviation is reported in the taxon name: SS: Sassari; ME: Messina; IM: Imperia; SA: Salerno; CH: Chieti; LE: Lecce; GE: Genoa; SV: Savona; RM: Roma. Strains mentioned in the manuscript are reported close to the related clade. Statistical support to the internal nodes is reported as posterior probability (pp).

Figure 2

Maximum Clade Credibility phylogeny based on the whole genome sequence of DMV. The phylogeny was inferred from the complete genome sequences partitioned by gene and noncoding region. Tip dates and collection site, reported as sea of origin, were included in the MCMC analysis. Branches are colored based on the most probable location. At internal nodes, the estimated tMRCA is reported. Strains are shown on the right side. The probability for the most probable location is reported for the most relevant nodes, colored according to the sea.