Survey and Characterization of Jingmen Tick Virus Variants

Abstract

We obtained a Jingmen tick virus (JMTV) isolate, following inoculation of a tick pool with detectable Crimean-Congo hemorrhagic fever virus (CCHFV) RNA. We subsequently screened 7223 ticks, representing 15 species in five genera, collected from various regions in Anatolia and eastern Thrace, Turkey. Moreover, we tested specimens from various patient cohorts (n = 103), and canine (n = 60), bovine (n = 20) and avian specimens (n = 65). JMTV nucleic acids were detected in 3.9% of the tick pools, including those from several tick species from the genera Rhipicephalus and Haemaphysalis, and Hyalomma marginatum, the main vector of CCHFV in Turkey. Phylogenetic analysis supported two separate clades, independent of host or location, suggesting ubiquitous distribution in ticks. JMTV was not recovered from any human, animal or bird specimens tested. Near-complete viral genomes were sequenced from the prototype isolate and from three infected tick pools. Genome topology and functional organization were identical to the members of Jingmen group viruses. Phylogenetic reconstruction of individual viral genome segments and functional elements further supported the close relationship of the strains from Kosovo. We further identified probable recombination events in the JMTV genome, involving closely-related strains from Anatolia or China.

Keywords: Anatolia; flavivirus; jingmen; tick; turkey.

Figure 1

Illustrative map of the locations used for specimen collection in the study (circle: tick, triangle: avian, square: canine, diamond: bovine, hexagon: human).

Figure 2

The maximum likelihood analysis of the JMTV partial segment 3 coding sequences (253 bp). The tree is constructed using the general time reversible (GTR) model, gamma distributed with invariant sites (G + I) for 500 replications. The sequences characterized in this study are given in bold and indicated with a blue triangle, GenBank accession number and specimen codes. Global virus strains are indicated by GenBank accession number, virus and strain/isolate name. Bootstrap values higher than 70 are provided.

Figure 3

The maximum likelihood analysis of the JMTV complete coding sequences ((A): segment 1, (B): segment 2, (C): segment 3, (D): segment 4). The tree is constructed using the general time reversible (GTR) model, gamma distributed with invariant sites (G + I) for 500 replications. The sequences characterized in this study are given in bold and indicated with red circles (segment 1), yellow squares (segment 2), green triangles (segment 3) or blue diamonds (segments 4), GenBank accession number and specimen codes. Global virus strains are indicated by GenBank accession number, virus and strain/isolate name. Bootstrap values higher than 70 are provided.

Figure 4

Similarity and bootscan plots of the alignments of the JMTV genome segments with evidence for recombination. The plots are prepared within a sliding window 200 base pairs (bp,) wide centered on the position plotted, with a 20 bp step size, for 1000 replications (GapStrip: On, Maximum Likelihood (Simplot), Neighbor-joining (Bootscan), T/t: 2.0). Virus strains used for analyses are indicated in the legends and in Supplementary File 3.

Figure 5

Alignment of the JMTV RNA-dependent, RNA polymerase and helicase/protease motifs, identified on viral genomic segments 1 and 3. The figures cover 576–748 and 341–472 residues of the mature nonstructural protein 1 (NSP1) and NSP2 proteins, respectively. Regions and residues corresponding to the conserved motifs of the flavivirus replicase (A, B, C, E) (brown) and with functional significance (black) are indicated. Information on virus strains used for comparison are provided in Supplementary Table S3.