Equine Encephalosis Virus in India, 2008

A virus isolated from a sick horse from India in 2008 was confirmed by next-generation sequencing analysis to be equine encephalosis virus (EEV). EEV in India is concerning because several species of Culicoides midge, which play a major role in EEV natural maintenance and transmission, are present in this country.

E quine encephalosis is an arthropodborne, noncontagious disease of equids, characterized by fever and lassitude (1). For some affected horses, lack of appetite and some degree of edema have been reported as secondary complications (2). Equine encephalosis virus (EEV; family Reoviridae, genus Orbivirus) was first isolated from horses in South Africa in 1967 and was considered endemic to southern Africa until an outbreak was reported in Israel in 2008 (3). Reports of EEV circulation in Israel indicated that the virus is not limited to a particular region and raised concerns about the risk for its spread to other countries where competent hosts and vectors are present (3). A recent study examining EEV seroprevalence in Israel, Palestine, and Jordan has emphasized the potential risk for invasion of pathogens into new ecologic niches (4).
Two species of midge, Culicoides imicola (species complex) and C. bolitinos, have been implicated as EEV vectors (1). Although EEV infection results in high (60%-70%) morbidity rates among equids, deaths from this infection are rare (1). Diagnosis of EEV infection is confirmed by several techniques, including virus isolation in baby hamster kidney cells or suckling mice and demonstration of EEV antigen by antigen capture ELISA and serum virus neutralization tests. In addition, the TaqMan minor groove binder probe reverse transcription PCR has been used for rapid detection of EEV strains and differentiation from African horse sickness virus (5).
Recent reports of next-generation sequencing analysis assert an unbiased identification of novel viruses in clinical samples (7,8). We investigated a virus isolated from necropsy samples received in 2008 from a horse in Pune, Maharashtra, India. Data analyzed from the MiniSeq (https:// sapac.illumina.com/?langsel=/in/) output led to identification and generation of the complete EEV genome.

The Study
In July 2008, the National Institute of Virology in Pune received necropsy samples from a horse in Pune, including blood and organ tissue (lung, liver, kidney, and spleen). In addition, nasal swab, blood, and serum samples were collected from 13 horses on the same farm, reported to have clinical signs similar to those of the dead horse (fever, nasal discharge, loss of appetite, and weakness). These 13 horses recovered. No further specific details could be retrieved after this virus was identified.
All samples from the 14 horses were negative for equine influenza, Japanese encephalitis, and West Nile virus RNA according to PCR (9). We isolated virus by using horse tissue suspensions and blood samples as inocula. Each specimen was individually inoculated at a volume of 0.1 mL into 24-well plates containing a subconfluent monolayer of Vero CCL-81 cells. After incubating for 1 h at 37°C, the inoculum suspensions were removed and the cells rinsed twice with phosphate-buffered saline. We subsequently added Eagle minimum essential medium supplemented with 2% fetal calf serum and incubated the cells at 37°C. Postinfection cytopathic effects were observed by using an inverted light microscope (Nikon, Melville, NY, USA), and cellular morphologic changes were recorded. We determined viral titers in Vero CCL-81 cells to be 10 7.67 50% tissue culture infective dose/mL by using the method of Reed and Muench (10). We checked 8 vertebrate and 1 invertebrate cell lines for susceptibility to this virus. Most cells, with the exception of bat embryonic cells and A549 (human lung carcinoma), displayed postinfection cytopathic effects (Table 1).
Because the initial identification attempts in 2008 yielded negative results, we performed a pathogen discovery protocol by using our recently acquired next-generation sequencing instrument. We performed DNA and RNA extractions separately on 1 mL of tissue culture supernatants or cell pellets of virus-infected cells by using a QIAamp To assemble contiguous sequences (contigs) by using de novo assembly, we used paired-end reads for DNA and RNA. DNA reads for blood and lung samples gave 166 and 113 contigs, with average lengths of 1,112 and 6,413 bp, respectively; 29 and 1,077 contigs were generated from the assembly of RNA reads for blood and lung samples, with average lengths of 1,376 and 1,034 bp, respectively. Figure 1 depicts data from the RNA reads, classified by using Taxonomer (11). BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) analysis of RNA contigs led to sample identification as EEV, and DNA contigs failed to identify any virus.
After the infectious agent was identified, we conducted reference sequence mapping for the RNA reads by using the EEV isolate Potchefstroom with the help of a referenceguided assembly program (CLC Genomics Workbench). We recovered the full-genome sequence of EEV by using both contigs generated from de novo assembly and the reference mapping of the reads. A total of 7 fragments encoding the structural genes VP1-7 and 3 fragments encoding the genes NS1-3, together comprising ≈19,290 nt of EEV, were generated. We compared the obtained genome sequence with other publicly available EEV sequences (Table 2; Figure 2, panels A, B). NS3 gene analysis revealed that the EEV isolate from India shared the highest identity with the Potchefstroom EEV strain from South Africa, belonging to serotype 6. However, the VP2 gene had the highest resemblance to the Bryanston EEV strain, belonging to serotype 4.

Conclusions
The complete genome of the EEV isolated from a dead horse confirmed the presence of EEV in India. We compared the resultant full-genome sequence of this isolate with other EEV sequences available in GenBank. Phylogenetic analysis of the VP2 gene sequences, which is used for genotype classification of reoviruses, revealed that the isolate from India groups with the Bryanston EEV strain, belonging to serotype 4.   of EEV in India is concerning because several species of Culicoides midge, which play a major role in EEV natural maintenance and transmission, are present in this country. This information about EEV has been provided to the Department of Animal Husbandry, Dairying, and Fisheries of the Ministry of Agriculture and Farmers Welfare in New Delhi, India. The Department of Animal Husbandry should soon initiate a survey to provide information about EEV presence in equine centers and commercial farms and to screen samples from sick horses. The concern is that this virus may be widely circulating in India without having been noticed earlier.