Seroprevalence and Risk Factors of Bluetongue Virus Infection in Tibetan Sheep and Yaks in Tibetan Plateau, China

Bluetongue (BT), caused by bluetongue virus (BTV), is an arthropod-borne viral disease in ruminants. However, information about BTV infection in yaks in China is limited. Moreover, no such data concerning BTV in Tibetan sheep is available. Therefore, 3771 serum samples were collected from 2187 Tibetan sheep and 1584 yaks between April 2013 and March 2014 from Tibetan Plateau, western China, and tested for BTV antibodies using a commercially available ELISA kit. The overall seroprevalence of BTV was 17.34% (654/3771), with 20.3% (443/2187) in Tibetan sheep and 13.3% (211/1584) in yaks. In the Tibetan sheep group, the seroprevalence of BTV in Luqu, Maqu, Tianzhu, and Nyingchi Prefecture was 20.3%, 20.8%, 20.5%, and 19.1%, respectively. The seroprevalence of BTV in different season groups varied from 16.5% to 23.4%. In the yak group, BTV seroprevalence was 12.6%, 15.5%, and 11.0% in Tianzhu, Maqu, and Luqu counties, respectively. The seroprevalence in different seasons was 12.6%, 15.5%, 15.4%, and 9.0% in spring, summer, autumn, and winter, respectively. The season was the major risk factor concerning BTV infection in yaks (P < 0.05). The date of the BTV seroprevalence in Tibetan sheep and yaks provides baseline information for controlling BT in ruminants in western China.


Introduction
Bluetongue virus (BTV), a member of the genus Orbivirus, family Reoviridae, is the causative agent of bluetongue (BT), an infectious, noncontagious, arthropod-borne viral disease, which can infect a wide range of wild and domestic ruminants [1]. The first case of BT in sheep in India was reported in 1964 [1]. This pathogen was firstly recorded in sheep in China in 1979 [2]. Because BTV can cause a severe hemorrhagic disease with high morbidity, it is listed as a notifiable disease by Office International des Epizootics (OIE) [1]. Transmission of BTV is mainly through biting of blood-feeding insect vectors of the genus Culicoides (Diptera: Ceratopogonidae). BTV infection in sheep and wild ruminants usually presents as symptoms of fever, nasal discharge, drooling of saliva, oral lesion, facial edema, depression, anorexia, and muscle weakness. In contrast, goats and cattle may be asymptomatic [3].
Recently, a large number of BTV surveys have been conducted worldwide. In China, research focused on sheep, goats, and cattle [4,5]. According to the literature published in a Chinese journal, the abortion rates of yaks were 21.39% based on an investigation of 104 farms in Qinghai Province, which could be caused by BTV and other pathogens [6]. However, no such data concerning BTV in Tibetan sheep are available, and only one case of BTV infection in yak was reported in Qinghai Province [7]. Tibetan sheep (Ovis aries) and black and white yaks (Bos grunniens) are important semiwild animals in China, and they mainly live in Tibetan Plateau which has low air pressure, lower temperature, and

Sample Collection. This study was approved by the Animal Ethics Committee of Lanzhou Veterinary Research
Institute, Chinese Academy of Agricultural Sciences. A total of 3771 blood samples from 2187 Tibetan sheep (962 from Tianzhu with the elevation above 2,000 metres, 182 from Luqu with an average elevation of 3,500 metres, 588 from

Results
Six hundred and fifty-four (17.3%) out of 3771 serum samples were seropositive for BTV infection using an indirect ELISA test. The seroprevalence ranged from a minimum of 12.6% among the white yaks to a maximum of 20.3% among the Tibetan sheep. In the Tibetan sheep group, the seroprevalence of BTV in Luqu, Maqu, Tianzhu, and Nyingchi was 20.3%, 20.8%, 20.5%, and 19.1%, respectively, and the detailed information about each group is shown in Table 1. In the yak group, BTV seroprevalence was 12.6%, 15.5%, and 11.0% in Tianzhu, Maqu, and Luqu counties, respectively, and the detailed information about each group is shown in Table 1. According to conditional forward stepwise logistic regression, a significant difference was found between Tibetan sheep and yak groups ( < 0.05), for which the OR was 1.653 (95% CI 1.383-1.976) ( Table 2). For the Tibetan sheep group, logistic regression analysis has shown that the season, gender, age, and region were not the significant risk factors ( > 0.05). For the yak group, logistic regression analysis showed that only season was significantly associated with BTV infection ( < 0.05) ( Table 2).

Discussion
In the present study, the overall BTV seroprevalence in the examined Tibetan sheep and yaks was 17.3%. This rate is higher than the 3.53% reported in yaks in Qinghai Province, China [7], and 9.3% among the domestic ruminants in Northern Kerala, India [3]. However, it is lower than the 27.9% prevalence reported in small ruminants in Nepal [2], 33.13% in sheep and goat in South Bengal [9], 43.68% in ruminants in Jharkhand, India [10], 45.20% among domestic ruminants in the highlands of Nepal [11], and 96.7% in buffaloes and cattle in selected provinces in Lao People's Democratic Republic [12]. Many factors, including the diagnostic methods, climatic conditions, geographical conditions, species/breeds, sample sizes, and sanitation, may contribute to such differences. In the present survey, Tibetan sheep showed a higher BTV seroprevalence than yaks, and the difference was statistically significant ( < 0.05). This result is consistent with a previous study which demonstrated that sheep are more susceptible to BTV [3]. It is well known that Culicoides midges are the most important transmitting vector for BTV. Moreover, the Culicoides midges are seasonal; they began to be active in spring and are most active in summer in these regions [13]. This seasonal exposure fits with the seroprevalence data in yaks in this study. Hence, the seasons were undoubtedly the risk factor for BTV infection in yaks.
The present study has shown that seasons are highly related to BTV infection in yaks ( < 0.05). Yaks had a 1.87 times higher risk for infection with BTV in summer compared to winter (OR = 1.87, 95% CI = 1.17-3.00) and a 1.85 times higher risk for infection with BTV in autumn compared to winter (OR = 1.85, 95% CI = 1.18-2.91), but it was not different significantly between spring and winter using multivariable analysis (Table 2). Such difference may be due to the fact that the Tibetan sheep and yaks were slaughtered during September and November every year, and the unhealthy and adult animals would be slaughtered first. Hence, the seroprevalence was lower in winter than in other seasons. However, the Culicoides midges would breed when the winter is passed, and they could spread the virus again. Although the season was not considered as the risk factor for Tibetan sheep, the seroprevalence in summer was higher than other seasons. These findings suggest that seasons should be considered when carrying out control programs in the investigation areas.
In the present investigation, no significant difference in BTV seroprevalence was observed among Tibetan sheep and yaks of different ages. This might be because these animals stayed in the same location.
In summary, the present study revealed that BTV infection is widespread in Tibetan sheep (20.3%) and yaks (13.3%) in Gansu and Tibet, western China. This is also the first report of BTV seroprevalence and risk factors in Tibetan sheep in China. The logistic regression analysis showed that the species was the risk factor concerning BTV infection. Season is considered as the risk factor of BTV infection in yaks. Hence, we should pay more attention to controlling Culicoides midges in warm seasons, especially in summer and autumn. These data provide baseline information for the control of BTV infection in Tibetan sheep and yaks.