Tibet Orbivirus, a Novel Orbivirus Species Isolated from Anopheles maculatus Mosquitoes in Tibet, China

Background The genus Orbivirus includes a number of important pathogenic viruses, including Bluetongue virus (BTV), African horse sickness virus (AHSV), and Epizootic hemorrhagic disease virus (EHDV). In this study we describe the isolation and characterization of an Orbivirus strain isolated from Anopheles maculatus mosquitoes collected in Tibet, China. Methods and Results Initial viral screening identified a viral strain (XZ0906) that caused significant cytopathic effect (CPE) in BHK-21 cells, including rounding, cell rupture, and floating. Although CPE was not observed in insect cells (C6/36), these cells supported viral replication. Polyacrylamide gel analysis revealed a genome consisting of 10 segments of double-stranded RNA (dsRNA), with a distribution pattern of 3-3-3-1. 454 high throughput sequencing of culture supernatant was used for viral identification. Complete genome sequencing was performed by Sanger sequencing in combination with 5′-RACE and 3′-RACE. Sequence analysis demonstrated that all 5′- and 3′- untranslated regions (UTRs) for each of the 10 genome segments contained a series of six highly conserved nucleotides. In addition, homology analysis and phylogenetic analysis based on amino acid sequence was completed, and all results show that virus XZ0906 was not a member of any known species or serotype of Orbivirus, indicating it to be a new species within the genus Orbivirus. Conclusions The isolated Orbivirus strain was designated Tibet Orbivirus, TIBOV to denote the location from which it was isolated. TIBOV is a novel orbivirus species which is isolated from Anopheles maculatus mosquitoes collected in Tibet, China.

Orbiviruses contain a multi-segmented, double-stranded RNA genome, consisting of 10 segments (Seg1-Seg10) of various length, which are identified according to their molecular weight [7]. Partial nucleotide sequences for each of the gene segments for many of the Orbiviruses have been published, along with complete genome sequences of some species [3,5,[8][9][10], allowing for detailed classification and phylogenetic analysis of Orbiviruses.
This study describes a viral strain (XZ0906) isolated from Anopheles maculatus specimens collected in Tibet, China. All the results of initial viral screening showed a difference between XZ0906 and Yunnan Orbivirus (YUOV), an orbivirus also isolated from China. After whole genome sequencing, amino acid homology and molecular phylogenetic analysis, XZ0906, which is designated as Tibet Orbivirus (TIBOV), is identified as a novel species of the genus Orbivirus.

Viral isolation
Mosquito samples were collected in Medog County (altitude 1000 m) in the Nyingchi area of Tibet, China during the summer of 2009, and transported to the laboratory in liquid nitrogen containers, following morphological classification and species identification on-site. All specimens were homogenized and centrifuged at 120006g for 30 min at 4uC. To isolate the virus, 150 mL of supernatant was then added to monolayers of both C6/ 36 and BHK-21 cells, and cultured at 28 and 37uC, respectively, in a 5% CO 2 incubator. Cells were monitored at 24-h intervals to identify cytopathic effects (CPE) associated with infection [11][12][13].

dsRNA-polyacrylamide gel electrophoresis
Viral RNA was isolated as described previously, and analyzed by polyacrylamide gel electrophoresis [13].

Preparation of viral DNA and RNA and 454 sequencing
Viral DNA was extracted from 200-mL aliquots of virus-infected BHK-21 cell culture supernatants using a QIAamp DNA Blood Mini Kit (Qiagen). Viral RNA was extracted from 140-mL aliquots of virus-infected BHK-21 cell culture supernatant using a QIAamp Viral RNA Mini Kit (Qiagen) according to the manufacturer's instructions. cDNA was made with a Ready-To-Go kit (GE Healthcare) using random hexanucleotide primers. Samples were then amplified as described previously [14,15]. Amplification products were pooled, adaptor-ligated, and sequenced at the Washington University Genome Sequencing Center on the 454 GS-FLX platform (454 Life Sciences, Branford, CT).
Because the nucleic acids used for sequencing contained a mixture of host cell DNA and viral RNA, sequencing reads were filtered using the custom informatics pipeline VirusHunter [16] to identify viral sequences. Sequences identified as most similar to viruses in the genus Orbivirus, as well as those that had no significant hit to any sequence in the GenBank database, were assembled with Newbler (454 Life Sciences) using the default parameters. Sequences were trimmed to remove primer sequences prior to data analysis and assembly.

Complete genome sequencing including 59-and 39untranslated regions
Reverse-transcription polymerase chain reaction (RT-PCR) was performed to fill in gaps between viral gene sequences obtained with 454 sequencing using contig-specific primers. Total viral RNA was extracted as described in Step 4, cDNA was generated by reverse transcription, and used as a template for complete genome amplification. Next, a set of specific primers was designed to amplify each segment of the viral genome and the amplification products were sequenced using the Sanger method (Table 1). 59-RACE and 39-RACE systems (Rapid Amplification of cDNA Ends), Version 2.0 (Invitrogen) were used to amplify the 59-and 39-UTRs from each of the 10 segments, respectively. 59-RACE was performed according to the manufacturer's instructions. For 39-RACE, a PolyA tail was first added to RNA using a PolyA polymerase. 39-UTR sequences were then generated by RT-PCR using sequence-specific and oligo-dT-adapter primers. Sequence assembly was performed resulting in a complete viral genome.   The standard discontinuous polyacrylamide slab gel electrophoresis was used here with a 3.5% acrylamide concentration gel and 10% acrylamide separation gel. After staining with silver nitrate,the genome of XZ0906 was visualized separated into 10 distinct bands. doi:10.1371/journal.pone.0088738.g003 6. Molecular detection of viral genes in cell culture Viral replication was detected in infected C6/36 and BHK21 cells using RT-PCR for specific regions for TIBOV segment 1 and segment 2. Total RNA was extracted from cell culture superna-tants as described in Step 4. cDNA was then generated by reverse transcription, and used as a template for RT-PCR. Gene amplification was performed using primers 6-1-5R and 6-1-5F (primers for Seg1), 6-2-2R and 6-2-2F (primers for Seg2), etc.;  detailed sequence information for all primer sequences is shown in Table 1. PCR was performed under the following conditions: one cycle of denaturation at 95uC for 5 min, 35 cycles of 95uC for 1 min (denaturation), 52uC for 1 min (annealing), and 72uC for 1 min (extension), followed by a final extension at 72uC for 10 min. Amplification products were analyzed by gel electrophoresis on a 1% agarose gel.

Nucleotide and amino acid sequence analysis
Sequences were identified by BLAST analysis (http://www. ncbi.nlm.nih.gov/BLAST/). Multiple sequence alignments were performed using the Clustal X2 software. Phylogenetic analysis of amino acid sequences for each Orbivirus gene segment were performed using the MEGA 5.04 software package (www. megasoftware.net). Amino acid sequences were analyzed using PredictProtein (http://www.predictprotein.org/). The background information for all virus strains used in this study is shown in Table 2.

Isolation of viral strains
A. maculatus mosquitoes collected from Tibet, China were homogenized, and the supernatant added to monolayers of C6/36 and BHK-21 cells. Severe CPE was observed in BHK-21 cells three days after inoculation with mosquito lysate XZ0906, characterized by cell rounding, lysis, and floating cells (Figure 1). However, no obvious pathological changes were seen in C6/36 cells cultured with the same mosquito lysate for five days, or after three consecutive passages. Despite the lack of CPE in C6/36 cells, Orbivirus Seg1 and Seg2 could be detected by RT-PCR in the supernatant of third-generation C6/36 cultures (Figure 2), indicating that virus XZ0906 could replicate in C6/36 cells.

Identification of a segmented dsRNA genome
Viral RNA was harvested from the culture supernatant of infected BHK-21 cells, and analyzed by polyacrylamide gel electrophoresis (PAGE), revealing a genome consisting of 10 dsRNA segments, whose migration pattern was 3-3-3-1 (Figure 3).
Within this pattern Seg2 migrated to the same region as Seg3; Seg5 and Seg6 were also difficult to distinguish, indicating that these segments had similar molecular weights. Segments 7, 8, and 9 were also similar in terms of molecular weights, but were easily distinguished from Seg10.

Preliminary identification of virus XZ0906 using 454 sequencing
Following random PCR amplification, samples were pooled (with barcodes) along with other samples, and sequenced using the Roche/454 FLX Titanium platform, producing a total of 24,929 reads. Sequence data were analyzed using the customized data analysis pipeline VirusHunter [16], identifying 85 unique reads which exhibited 28.1-84.9% sequence identity to viruses in the genus Orbivirus.
All individual reads with detectable similarity to Orbivirus, as well as those sharing no detectable sequence similarity with any sequence in the GenBank database, were used as inputs and assembled into contigs using the Newbler assembler. Twenty-one contigs were assembled, of 138-1342 bp in length, with the greatest similarity to BTV at a coverage depth of 1.4-20.9-fold ( Figure 4). Almost-complete RNA sequences were obtained for segments 7, 8, 10. Segments 1, 3, 4, 6 and 9 were represented by two to five contigs; a single contig was identified for segments 2 and 5.

Sequencing and analysis of virus XZ0906 and other Orbiviruses
RT-PCR amplification was used to close the gaps between contigs for each of the 10 segments. Primer walking, together with 59-and 39-RACE, were used to sequence the 59-and 39-ends of each segment. Finally, Sanger sequencing was employed to confirm sequences using primers newly designed for each of the 10 RNA segments (Table 1); complete sequences for this virus XZ0906 have been deposited in GenBank under accession number(genome segments KF746187 to KF746196).

Phylogenetic analysis of virus XZ0906 based on VP1
amino acid sequences. To better understand the taxonomic classification of virus XZ0906, the amino acid sequences of 37 VP1 proteins (Table 2) covering 14 genera within the family Reoviridae were obtained from GenBank, and used to construct a phylogenetic tree. These 37 virus strains (including different species and different serotype of one species) readily clustered into 14 evolutionary branches, with virus XZ0906 clustering within the genus Orbivirus branch ( Figure 5(A)). To further establish the taxonomic classification of virus XZ0906, VP1 amino acid sequences from 28 known Orbivirus strains were used to construct a phylogenetic tree specific to this genus ( Table 2). This analysis shows that virus XZ0906 forms a unique phylogenetic branch independent of any known Orbivirus species (Figure 5(B)).

5.2.
Phylogenetic analysis based on the T2 protein amino acid sequence. The amino acid sequence of the T2 protein is an important marker used to classify species within the genus Orbivirus. T2 amino acid sequences from 29 known Orbivirus strains, along with the equivalent region from virus XZ0906, were selected to construct a phylogenetic tree. This analysis showed that virus XZ0906 is independent of any known Orbivirus species ( Figure 5(C)). From these results, we determined virus XZ0906 to represent a novel species within genus Orbivirus. This novel species was given the name Tibet Orbivirus, TIBOV to reflect the location from which it was isolated. Orbivirus strains. These analysis employed a neighbor-joining method (using the P-distance algorithm) using the MEGA version 5.04 software package (www.megasoftware.net). Bootstrap probabilities for each node were calculated using 1000 replicates. Scale bars indicate the number of amino acids substitutions per site. In Figure 5(C), as many of the available sequences are incomplete, analysis is based on partial sequences (residues 356-567 relative to the BTV-1A sequence). Abbreviations and serotypes (or strain name) are shown in the radial tree image of Figure 5. GenBank accession numbers and further details of the sequences can be found in Table 2 [1]. All Reoviridae genomes consist of multisegmented dsRNA; for example, the genome of Seadornavirus, Rotavirus, and Orbivirus contain 12, 11, and 10 dsRNA segments, respectively [2,10,17,18]. Here we describe a novel orbivirus species isolated from mosquitoes collected in Tibet. This virus has many features characteristic of orbiviruses.
UTRs were detected at both the 59 and 39-ends of all 10 TIBOV gene segments. The lengths of these UTRs were highly variable; however, all 39-UTRs contained a stretch of six highly conserved nucleotides at the end, which is a defining molecular characteristic used in the identification of Orbiviruses [8]. For BTV, AHSV, PALV, and Equine encephalosis virus (EEV), this stretch of six conserved nucleotides is readily detected in the 39-UTRs of each gene segment [1,4]; however, no such sequences are found at their corresponding 59-ends. Among the 10 gene segments in Yunnan virus (YUOV), a recently identified Orbivirus isolated from mosquitoes in Yunnan, China, nine (Seg2-Seg10) contained a conserved seven-nucleotide sequence at the 59-UTR end, but only three conserved nucleotide sequences at the 39-end [4]. Among the 10 gene segments of Tibet Orbivirus, TIBOV, six conserved nucleotide sequences were detected in both end of the 59-and 39-UTRs (59-GUAAAA and ACUUAC-39, respectively); these sequences were distinct from those in any other Orbivirus species.
The Orbivirus RNA-dependent RNA polymerases (RdRp), which is encoded by the Seg1 gene (VP1), is an important marker for species identification [4,8]. The VP1 protein sequence similarities of TIBOV to those of other Orbivirus species were 35.3-72.9% (Table 4), indicating that TIBOV constituted a novel member of the genus Orbiviruses. In addition, the T2 protein of Orbivirus is used to classify serotypes within the genus, with a threshold .91% identity at the amino acid level [4,19,20]. Such as Middle Point orbivirus (MPOV), which is isolated from Australian bovine serum specimens in 1998, exhibited up to 99% identity with YUOV, indicating that MPOV and YUOV were different serotypes of the same virus species [21]. The T2 protein from TIBOV shared 22.9-75.9% amino acid identity with those from six representative Orbiviruses, including YUOV (37.6%), well below the 91% threshold (Table 4). Together with phylogenetic analysis of both VP1( Figure 5(A),5(B)) and T2 ( Figure 5(C)) amino acid sequences, we can draw a conclusion that TIBOV represented a new species within the genus Orbivirus, rather than a serotype of a previously described Orbivirus.
As previously described, YUOV is a new species of Orbivirus isolated from Culex tritaeniorhynchus specimens collected in Yunnan, China [4]. The genome of this virus consists of 10 dsRNA segments, and exhibits a 3-4-2-1 migration pattern when resolved in an agarose gel. Preliminary analysis of this virus showed clear CPE in Aedes albopictus cells, but no CPE or viral replication in mammalian cells (BHK and Vero cells) [4]. TIBOV isolated from mosquito specimens collected in Tibet in this study represents the second Orbivirus isolated from mosquito specimens in China. Substantial differences between YUOV and TIBOV were evident, including replication and toxicity to insect and mammalian cells, migration profiles, protein sequences, and the presence of conserved nucleotide sequences in the 59-UTR and 39-UTRs. Together, these results demonstrate that TIBOV is distinct from YUOV, and highlights the level of genetic diversity within Orbiviruses in China.
TIBOV was isolated from A. maculatus specimens collected at a pigsty in rural Tibet. It is currently unknown whether TIBOV can infect either humans or animals. In order to determine whether this virus poses a risk to public health, serological studies to define potential human and animal exposures to TIBOV are needed.