Potential vectors of bluetongue virus in high altitude areas of Yunnan Province, China

Background Bluetongue disease of ruminants is a typical insect-borne disease caused by bluetongue virus (BTV) of the genus Orbivirus (family Reoviridae) and transmitted by some species of Culicoides (Diptera: Ceratopogonidae). Recently, the detection of BTV in yaks in high altitude meadows of the Shangri-La district of Yunnan Province, China, prompted an investigation of the Culicoides fauna as potential vectors of BTV. Methods A total of 806 Culicoides midges were collected by light trapping at three sites at altitudes ranging from 1800 to 3300 m. The species were identified based on morphology and the DNA sequences of cytochrome c oxidase subunit 1 (cox1). PCR and quantitative PCR following reverse transcription were used to test for the presence of BTV RNA in Culicoides spp. A phylogenetic analysis was used to analyze the cox1 sequences of some specimens. Results Four species dominated these collections and cox1 barcoding revealed that at least two of these appear to belong to species new to science. Culicoides tainanus and a cryptic species morphologically similar to C. tainanus dominated low altitude valley collections while C. nielamensis was the most abundant species in the high-altitude meadow. A species related to C. obsoletus occurred at all altitudes but did not dominate any of the collections. BTV RT-qPCR analysis detected BTV RNA in two specimens of C. tainanus, in one specimen closely related to C. tainanus and in one specimen closely related to C. obsoletus by barcode sequencing. Conclusions This study suggests that BTV in high altitude areas of Yunnan is being transmitted by three species of Culicoides, two of which appear to be new to science. This research may be useful in improving understanding of the effects of global warming on arboviral disease epidemiology and further study is important in research into disease control and prevention.


Background
Bluetongue disease of ruminants is caused by bluetongue virus (BTV) of the genus Orbivirus (family Reoviridae), which is a notifiable disease to the World Organization for Animal Health (OIE, Office International des Epizooties). The virus is transmitted by species of Culicoides [1]. There are 1368 extant species of Culicoides [2] but only about 30 of these have been associated with transmission of BTV [1,[3][4][5][6]. Bluetongue virus has mainly been reported in tropical and subtropical areas of Africa, Australia, North and South America, Asia and the Middle East with seasonal outbreaks reported from Mediterranean countries and Europe [7]. The unprecedented northward spread of BTV into Europe in the early 2000s has been linked to the effects of climate change and its influence on both the distribution and vector capacity of local Culicoides populations [3,8].

Open Access
Parasites & Vectors *Correspondence: htmlile@outlook.com 1 Yunnan Tropical and Subtropical Animal Virus Diseases Laboratory, Yunnan Animal Science and Veterinary Institute, Kunming, Yunnan Province, China Full list of author information is available at the end of the article and C. circumscriptus Kieffer [13]; however, solid evidence to support these assertions is lacking. A number of proven and suspected vector species are present in China including C. actoni Smith, C. brevitarsis Kieffer, C. fulvus Sen & Das Gupta, C. obsoletus (Meigen) and C. wadai Kitaoka, although some of these species are actually a complex of species and therefore require confirmation [14]. Recently, Ma et al. [15] reported the presence of BTV in yaks and sheep on the Tibetan Plateau, western China, at an altitude of over 3000 m but little is known of the species of midges living at such high altitudes or their potential to transmit BTV.
Yunnan Province is located in the southwest of China and shares a border with Laos, Myanmar and Vietnam. While the southern tropical and sub tropical districts of Yunnan are endemic for BTV [12], the north of Yunnan is mostly cold, high altitude mountains and river valleys, and there is little information on the prevalence of BTV or other arboviruses in this area.
Shangri-La district is located in the northwest of Yunnan, next to the Tibetan Plateau, with an area of 11,613 km 2 . The Shangri-La district consists of relatively warm river valleys and cold highland meadows. The river valleys are located about 1900 m above sea level along the Changjiang River and contain many small tributary streams. The highland meadows average about 3450 m above sea level and are covered by snow for almost four months of the year. There is, however, some evidence that this extreme climate is becoming milder as historical weather records indicate that the average monthly minimum temperatures between 2011 and 2018 are about 1 °C higher than that of the years 1971 to 2000 (Fig. 1a). Furthermore, the monthly minimum temperatures between 2011 and 2018 were milder than between 1971 and 2000 ( Fig. 1b) [16]. The impact of such changes is difficult to predict but could result in increased activity of Culicoides spp. and associated arboviruses in this region.
During an investigation between 2014 and 2017, an average of 5% of yaks living in a 3290 m altitude highland meadow of Shangri-La were found to have BTV antibodies [17]. In a nearby river valley at a lower altitude, a 35-65% seroprevalence of BTV and a BTV-21 strain was isolated from some goat farms [17] indicating that BTV is prevalent at these altitudes. There are scarce data on the species diversity and abundance of Culicoides spp. at high altitudes in Asia so it is unclear if any local species can survive the cold and long winters experienced in these areas. Some midge species are, however, able to survive at high altitudes in South America [18] and Europe [19] so it is likely that at least some Asian species are also able to survive in the Shangri-La area. Also unclear is whether the BTV found in yaks in the highland meadows is endemic or introduced seasonally by the immigration of infected midges from the warmer river valleys.
This paper provides a brief account of the diversity and abundance of Culicoides spp. on livestock farms in the Shangri-La district of Yunnan Province, China and the screening of the most common species for the presence of BTV RNA using real-time quantitative PCR.

Culicoides spp. collection
Culicoides were collected from farms at three different sites.

Morphological identification
Insect specimens were sorted into morphospecies based on gross morphology and wing pattern using keys and illustrations of Yu et al. [20]. Parous and gravid females free from visible blood were selected based on abdominal pigmentation [21] and contents and submitted individually for nucleic acid extraction. Following digestion, representative specimens were mounted onto microscope slides using the technique of Bellis et al. [22] and identified morphologically using the keys and descriptions of Yu et al. [20].

DNA and RNA extraction
Whole midges were digested non-destructively following the methods of Bellis et al. [22]. Briefly, insects were immersed in 300 µl of DXT tissue digestion buffer (Qiagen, Hilden, Germany) incorporating 1% DX digestion buffer additive (Qiagen) and incubated at 40 °C overnight. Following digestion, the insect cadaver was removed from the digestion lysate and preserved in 70% ethanol for subsequent morphological examination as described above. From the lysate, 220 µl was used for DNA extraction and the remaining 50 µl was used for RNA extraction. The RNA was extracted using a MagMAX ™ -96 viral RNA Isolation kit (Am1836; Ambion, Austin, TX, USA) following the manufacturer's directions and using a MagMAX ™ Express-96 machine (Ambion). The DNA was extracted using a DNeasy Blood & Tissue kit (#69560; Qiagen) following the manufacturer's directions. The final DNA elution of 120 µl and RNA of 50 µl were stored at − 20 °C.

Pan BTV serotype RT-qPCR
The primers BTVF-MH, BTVR-MH and probe BTVP-MH for real-time quantitative polymerase chain reaction (qPCR) described by Hofmann et al. [23] were used to detect the NS3 gene of BTV in insect specimens. Reverse transcription quantitative PCR (RT-qPCR) was performed on a Fast7500 Realtime PCR machine (Applied Biosystems, Carlsbad, CA, USA) using an AgPath-ID ™ one step RT-PCR Kit (Ambion). The reaction was performed as described by the manufacturer on a 2 µl RNA sample in a total volume of 20 µl, using PCR master mix with 6 mM MgCl 2 , 1 μM of each PCR primer and 0.2 μM of each probe. The RT-qPCR program consisted of the following: 10 min reverse transcription at 45 °C; 10 min denaturation at 95 °C; 45 cycles of 15 s denaturation at 95 °C, 45 s annealing and extension at 65 °C. Fluorescence was measured at the end of each annealing step. Only reactions with Cq values of 25 or lower were regarded as indicative of replication of BTV in the midge [4,24,25].

cox1 gene amplification
Two specimens morphologically referable to C. nielamensis Liu & Deng, 19 specimens morphologically referable to C. tainanus Kieffer and 9 specimens morphologically referable to C. obsoletus (Meigen) including the 4 specimens found to be infected with BTV, were processed and sequenced for cytochrome c oxidase subunit 1 (cox1). Primers BC1culicFm and JerR2m, and techniques described by Bellis et al. [22], were used to amplify the cox1 barcode region. Resultant sequences were compared with published data on Barcode of Life Data System (BOLD) and the National Center for Biotechnology Information (NCBI).

Sequence analysis
The PCR products of cox1 (692 bp) genes from Culicoides spp. were sent to a local company for sequencing using the Sanger method on an ABI3739XL machine (Kunming Shuoqing Biological Technology Company, Kunming, China). cox1 sequences of representative Culicoides specimens and the most similar Culicoides spp. from NCBI data were analyzed by alignment using MEGA-X. Pairwise Kimura 2-parameter (K2P) distances among sequences were calculated and compared as a neighbor-joining (NJ) tree (bootstrap = 1000) using MEGA-X [26].

Culicoides species
Weather conditions at the three sites were typical for the season. The temperatures in the afternoon at the collection sites were 28 °C (Jinjiang), 25 °C (Nixi) and 20 °C (Xiaozhongdian). The total number of midges collected at Jinjiang, Xiaozhongdian and Nixi was 591, 108 and 107, respectively. Specimens with morphology consistent with C. tainanus, C. nielamensis and C. obsoletus dominated collections, comprising 98% of all specimens collected ( Fig. 2 and Table 1). Other species collected included a species similar to C. punctatus Latreille, another close to C. sinanoensis Tokunaga, and five species which remain unidentified due to lack of description in reference books (Table 1).
cox1 sequences of 8 specimens morphologically identified as C. obsoletus (BOLD accession numbers SGRL234-SGRL240 and SGRL242) and 16   A neighbor-joining tree of these specimens (Fig. 3) revealed the presence of two clades of C. tainanus. One of these clades was 99.47% similar to C. tainanus (= C. maculatus Shiraki) sequences on BOLD while the other was only 90.12% similar to those specimens ( Table 2). So far, we have been unable to distinguish specimens from these two clades morphologically, so we are tentatively calling this second species Culicoides sp. close to C. tainanus.
Similarly, we found no close matches for our specimens resembling C. obsoletus (Figs. 3, 4). The closest match available on NCBI or BOLD was for C. obsoletus from Canada, suggesting that our specimens are close to, but different from, true C. obsoletus and possibly represent a cryptic species.
No reference sequences in BOLD nor NCBI were close to our cox1 sequences from C. nielamensis; the nearest sequence was 84.14% similar and belonged to C. minutissimus Zett ( Table 2).
The relative abundance of these four dominant species varied with altitude, C. tainanus and Culicoides sp. close to C. tainanus were only found at low and midaltitudes, C. nielamensis was most abundant at high altitude and Culicoides sp. close to C. obsoletus was found at all altitudes. From high to low altitude, the relative abundance of C. nielamensis was 97.22, 49.53 and 3.55%, C. tainanus and Culicoides sp. close to C. tainanus was 0, 35.51 and 91.70% and Culicoides sp. close to C. obsoletus was 2.78, 8.4% and 3.89% (Fig. 2 and Table 1).

BTV RT-qPCR of Culicoides
A total of 150 specimens of C. tainanus and Culicoides sp. close to C. tainanus, 35 Culicoides sp. close to C. obsoletus and 80 C. nielamensis were analyzed for the presence of BTV. Many specimens of all species had detectable Cq values, including 70 C. tainanus and Culicoides sp. close to C. tainanus, 32 Culicoides sp. close to C. obsoletus and 42 C. nielamensis. Cq values exhibited a bimodal distribution with 4 specimens having values of 25 or less and the remainder of specimens with values between 30 and 35 (Fig. 5).
Cq values below 25 have been shown in both colony [24] and field-collected [4,25] specimens to indicate that viral replication has occurred in the insect so the 4 specimens with Cq values less than 25 were considered infected with BTV. These 4 specimens were all collected from Nixi and were all parous females; 2 were identified as C. tainanus, 1 was Culicoides sp. close to C. tainanus and 1 was Culicoides sp. close to C. obsoletus (Fig. 3 and Table 3).

Discussion
To our knowledge, this is the first investigation of BTV vectors at high altitude areas of Yunnan and follows the detection of an average 5% BTV seroprevalence in yaks at 3290 m altitude of Jiantang and 35-65% BTV seroprevalence in goats at 2000-2800 m altitude of Nixi in Shangri-La [17]. Based on our limited data, the abundance of midges in this area is low compared to tropical or sub-tropical areas of Yunnan where over 10,000 midges can be collected after 12 hours trapping (our unpublished data). The low prevalence of BTV observed in yaks at 3300 m could be due to a variety of factors, many of which are related to the impact of the harsh environment on vector biology, for example low vector abundance, low vector competence or short season of vector activity. The data from this brief study are not sufficient to allow firm conclusions about the absolute abundance of vector populations at different altitudes but some differences in midge diversity were evident between the three sites studied. The change in dominance from C. tainanus and Culicoides sp. close to C. tainanus at low altitude to C. nielamensis at high altitude is evident and coupled with the detection of BTV in C. tainanus and Culicoides sp. close to C. tainanus but not in C. nielamensis may be a contributing factor to the lowered BTV prevalence at high altitude. Culicoides sp. close to C. obsoletus was present only in low numbers at all sites which may reduce its potential to act as a significant vector of BTV. Specimens from Shangri-La are morphologically similar to the European C. obsoletus, but significantly different in cox1 (about 9%). The Obsoletus group of Culicoides subgenus Avaritia has been shown to contain several species with very similar morphology and it is likely the species present in Shangri-La is a further cryptic species within this complex. Species from this complex are Holarctic so it is not surprising that populations are able to survive in high altitude sites such as Shangri-La. Further study on this group in Asia may reveal that the species present in Shangri-La is more widely distributed than currently known. Some species of the Obsoletus group have also been implicated in the transmission of BTV [6,27,28], so it is not surprising that a further member of this group may also be acting as a vector of BTV.
Recent work in Japan [29] found evidence that C. tainanus (= C. maculatus Shiraki) may be acting as a vector of BTV and this is supported by the detection of BTV in specimens from Shangri-La. The presence of a morphologically similar species in Asia that may also be a vector of BTV highlights the need for ongoing studies of the taxonomy of species in this region, particularly species belonging to the subgenus Avaritia of Culicoides. Integrative taxonomy using a combination of morphological and molecular analyses is currently the best method of establishing the identity of specimens [14] and should be applied to any studies of the vector status of species.
Specimens of Culicoides sp. close to C. obsoletus, C. tainanus and Culicoides sp. close to C. tainanus were found to contain BTV RNA in concentrations consistent to indicate virus had replicated in the insect thus fulfilling one of the criteria of these species being proven vectors of BTV [30,31]. Consistent with the findings of previous studies [4,24,25], many specimens of midges from Shangri-La exhibited a positive response in the RT-PCR but the high Cq values from these specimens suggest either a non-specific response or a response due to remnant virus in the midge following a viraemic blood meal. The absence of specimens with Cq values between 25 and 30 supports the conclusion of Veronesi et al. [24] and van der Saag et al. [4,25] that the virus has indeed multiplied in those midges with Cq values of less than 25.
None of the 80 specimens of C. nielamensis tested showed evidence of infection with BTV suggesting a prevalence of less than 1 in 80 (< 1.25%) compared to a prevalence of > 2.0% for C. obsoletus, C. tainanus and Culicoides sp. close to C. tainanus. More specimens would need to be tested to discount the possibility of   Neighbor-joining tree for Culicoides sp. close to C. obsoletus collected in Shangri-La based on the pairwise genetic distances among the mitochondrial cox1 sequences from 9 Culicoides sp. close to C. obsoletus from Shangri-La and 28 representative sequences for C. obsoletus from NCBI. Specimens from Shangri-La bear a number prefixed with "SGRL" (NCBI accession numbers MN513318-MN513326). Specimen containing BTV RNA is labelled by solid circle. For the specimens from NCBI, the isolates, GenBank accession numbers, country of collection and collection years are provided.  [32][33][34], so it is difficult to establish when a species is not acting as a vector of BTV. However, the lower prevalence of BTV at high altitude, despite the higher proportion of C. nielamensis at these altitudes, suggests that C. nielamensis may not be acting as a vector, or at least not an efficient vector. This study has shown that biting midge species capable of carrying BTV are present at 3300 m altitude sites in Shangri-La, which may explain the occurrence of BTV in yaks at that altitude. It is unknown if BTV moves seasonally to these highland meadows from nearby low altitude areas but as the seasonal movement of midges and viruses has been documented elsewhere [35], it is possible that this is also occurring in Yunnan or on the neighboring Tibetan Plateau. Based on the available evidence, any such movement could possibly be due to the dispersal of Culicoides sp. close to C. obsoletus which occurs at all altitudes studied here and has been shown to be infected with BTV; however, further work is needed to confirm that C. nielamensis or some other species is not also acting as a vector capable of moving virus between mid and high altitude sites.
Yaks are an important economic source for the Zang people who live in and around the Tibetan Plateau, and a recent study [36] indicated that the abortion rate of yaks in Qinghai Province, China, of 21.39% could be caused by BTV or other pathogens. Further study is needed to confirm the ability of BTV to cause clinical disease in yaks and subsequently to assist with disease control and prevention. Knowledge of the ecology of BTV in the extreme climate of Shangri-La may improve understanding of the effects of global warming on vector biology and disease epidemiology, particularly given the evidence that temperatures in the region are slowly rising.

Conclusions
This study has identified three species of Culicoides, two of which appear to be new to science, which may be acting as vectors of BTV in high altitude areas of Yunnan. Further studies including laboratory infection and transmission of BTV is required to confirm the vector status of these species. The seroprevalence of BTV at high altitudes of Yunnan and the neighboring Tibetan Plateau could be caused by the seasonal movement of BTV vectors, such as Culicoides sp. close to C. obsoletus, from lower altitude areas where BTV is endemic.