A beef cattle farm in Lichuan County, Jiangxi Province, served as the collection point for the blood-feeding arthropods. Specimens were collected using UV mosquito traps (Wuhan Jixing Environmental Protection Technology Co., Ltd.). The collection time was from 6pm to 6am. In this study, entomologists separated mosquitoes and midges with the naked eye at the specimen collection site, then selected the midge one by one under a stereoscopic microscope. Since morphological identification of midge species requires complex chemical treatment of an individual midge, which cannot be carried out in the field of specimen collection, the midges are referred to only as “ Culicoides biting midges “. Collection environment and insect species data were recorded, and the specimens were stored in liquid nitrogen before being transferred to the laboratory for testing (Fu et al., 2017).

BHK-21 (golden hamster kidney) and C6/36 (Aedes albopictus) cells were maintained in our laboratory. BHK-21 cell culture conditions were as follows: 90% Eagle’s medium (laboratory preparation), 7% fetal bovine serum (FBS; Invitrogen), 1% penicillin and streptomycin (100 U/mL), 1% glutamine (30 g/L), and 1% NaHCO₃. The C6/36 cell culture conditions were as follows: 89% RPMI1640 (Invitrogen), 10% FBS (Invitrogen), and 1% penicillin and streptomycin (100 U/mL). BHK-21 and C6/36 cells were cultured in an incubator at 37°C with 5% CO₂, and at 28°C, respectively (Cao et al., 2016; Fu et al., 2017).

1.5 mL of washing solution [90% Eagle’s medium (laboratory preparation), 1% penicillin and streptomycin (100 U/mL), 8% glutamine (30 g/l) and 1% NaHCO3 (75 g/l)] was added to each pool (50–100 mosquitoes or 500 midges) of samples followed by two washes (for three minutes/each wash) for cleaning the body surface of the specimen. Then, 1.5 mL of grinding solution was added (Same with washing solution), and the samples were ground using a glass grinder (Huayi Experimental Equipment Factory) under ice bath conditions. The samples were centrifuged after grinding (20,000 × g, 4°C, 20 min), and 100 μL of the supernatant was inoculated into 80% monolayers of BHK-21 and C6/36 cells on culture plates (24-well plates, Corning Inc.). BHK-21 and C6/36 cells were continuously cultured in an incubator at 37°C with 5% CO₂, and at 28°C, respectively. Cells were observed for signs of CPEs under a microscope every 12 h. When the cells appeared to be showing CPEs, the virus solution was collected and stored at -80°C until further identification. Specimens that did not show CPEs were blindly passaged for three generations, and those without CPEs were discarded (Cao et al., 2016; Fu et al., 2017).

BHK-21 cells were transferred onto a 6-well culture plate (Corning Inc.). The next day, once cells had grown into a monolayer covering 80% of the plate, a 10-fold dilution of the virus (10^-1–10^-6) was sequentially added to a 6-well culture plate (0.1 mL/well). After adsorption for 1 h in an incubator at 37°C with 5% CO₂, 1.3% methylcellulose-MEM semi-solid medium (4 mL/well) containing 2% FBS was added to each well. After 72 hours of culture, when a plaque was obvious under the microscope, the medium was discarded, and the plaque was stained with crystal violet. Finally, the number of plaque-forming units (pfu) expressing the virus was calculated (Cao et al., 2016; Feng et al., 2019).

Suckling mice (2-day-old Kunming mice) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. All mice were raised under sterile conditions at the animal facility of the Chinese Center for Disease Control and Prevention. All animal experiments were conducted in strict compliance with the regulations set by the Animal Ethics Committee of China CDC(NO.2014112509).

Each suckling mice was inoculated with 20 µL of supernatant infected with virus on the fifth passage of BHK-21 cells intracranially. The incidence and mortality of mice were observed and recorded every day. After death, the suckling mice were sacrificed and dissected. The mouse brain was ground, and the ground liquid was centrifuged at 12,000 rpm for 30 min at 4°C, and the supernatant was used to inoculate the next generation of suckling mice. The virus was passaged three times in suckling mice. Animals inoculated with the virus without disease or death were routinely sacrificed after 14 days (Cao et al., 2016).

The Viral RNA Mini Kit (Qiagen) was used to extract total RNA from the supernatant of virus-infected cells samples according to the manufacturer’s instructions. The extracted RNA was immediately placed in a water bath at 65°C for 10 min and placed on ice for 2 min. Then, 32 μL of RNA was absorbed into the first-strand reaction tube (Ready-To-Go You-Prime First-Strand Beads), which was maintained at room temperature for 1 min before 1 μL of random primers were added (phosphorylated random hexamer primers -pd (N) 6); this was centrifuged immediately and incubated in a water bath at 37°C for 1 h. Moloney Murine Leukemia Virus (M-MuLV) reverse transcriptase was used for reverse transcription. The total volume of the cDNA library was 33 μL, which was used immediately or stored at -80°C for future use (Cao et al., 2016; Fu et al., 2017; Feng et al., 2019).

Total RNA was extracted from the supernatant on the fifth passage of BHK-21 cells infected with the JXLC1806-2 isolate using the QIAamp Viral RNA Mini Kit (QIAamp, Qiagen, Valencia, CA). A metagenomics shotgun sequencing library was constructed using the TruSeq total RNA library Prep kit (Illumina, San Diego, California) according to the manufacturer’s instructions. The libraries were sequenced on the MiSeq platform (Illumina, San Diego, California) in paired-end 150-bp mode. The sequencing data generated via mNGS were imported to the CLC Genomics Workbench (Version: 20.0.4, Qiagen) and were trimmed based on quality and ambiguity. The clean reads were applied for de-novo assembly using the CLC De Novo Assembler and contigs of less than 300 bp were discarded from all assembles. The virus-related contigs were extracted based on the taxonomic information of the BLAST hits. To avoid mis-assembly, reads were mapped back to the virus genome and inspected with CLC Genomics Workbench. The viral genome sequences obtained by NGS was used as template for overlapping PCR primers design for Sanger sequencing confirmation.

Viral gene amplification (PCR) was carried out in a total volume of 25 μL, and included the cDNA template, GoTaq^® Green Master Mix, 2× (Promega, Madison, WI, USA), and 10 μmol/L of upstream and downstream primers. The PCR reaction conditions were pre-denaturation at 95°C, and the amplification conditions of the virus isolate were 94°C 30s, 50°C 30s, 72°C 60s, and a total of 35 cycles were followed by extension at 72°C for 10min. After the PCR reaction was completed, 5 μL of the gene amplification product was detected by 1% agarose gel electrophoresis. The nucleotide sequences of the amplification products were determined by Sanger sequencing (Cao et al., 2016; Fu et al., 2017; Feng et al., 2019). JXLC1806-2 virus gene amplification primers are shown in Supplementary Material 1 .

Primers were designed based on the obtained partial sequences, and the 5’/3’ RACE system for rapid amplification of cDNA ends (Invitrogen) was used to amplify the gene sequences of the 5’ untranslated region (UTR) and 3’ UTR. For the 3’ UTR without a poly(A) tail, the Poly (A) Polymerase Tailing Kit (Epicentre) was used to add the A-tail prior to amplification. The amplified products were detected by 1% agarose gel electrophoresis, purified using a QIAquick Gel Extraction Kit (Qiagen), cloned into the pGEM-T Easy vector (Promega), and 10 clones of each product for each region were sequenced by the Sanger method (Wang et al., 2015; Cao et al., 2016).

The sequences of representative strains of Orthobunyavirus isolated from different hosts and countries for different years were downloaded from GenBank. The virus strain information used in the analysis is shown in Supplementary Material 2 . The nucleotide sequences were subjected to a Blast (NCBI) alignment. The Seqman software (DNAStar, Madison, WI, USA) was used to splice the amplified sequence of the complete set of primers, and splicing errors were found to make the sequence more accurate.; BioEdit software (Version 7.0, Thomas) was used for nucleotide multiple sequence alignment; MEGA6.0 software was used for neighbor-joining phylogenetic analysis with a bootstrap value of 1000; homology analysis of nucleotide and amino acid sequences was performed using MegAlign (DNAStar, Madison, WI, USA) (Wang et al., 2015; Cao et al., 2016; Fu et al., 2017).

In July 2018, 12,400 Culicoides biting midges, two genera, and two species of mosquitoes were collected in Lichuan County, Jiangxi Province, which contained 270 Culex tritaeniorhynchus and 230 Anopheles sinensis ( Table 1 ).

The collected blood-feeding arthropods specimens were pooled and ground in the laboratory. There were six pools of mosquitoes and 21 pools of midges. The ground supernatants were inoculated into BHK-21 and C6/36 cells, and the culture was maintained, with the CPEs status observed on a daily basis. After three consecutive generations of cultivation, only JXLC1806-2 sample caused CPEs in the second and third generation BHK-21 cells, these effects included cell shrinkage, shedding, and reduction of adherent cells ( Figure 1 ). No CPEs was observed in BHK-21 cells in the remaining 26 sample pools. The CPEs time of JXLC1806-2 on second and third generations BHK-21 was 72 h and 48 h, respectively. After 5 passages of JXLC1806-2 virus in BHK-21 cells, the time of CPEs was stable at 48h. When 27 griding supernatant sample pools, including JXLC1806-2, were continuously cultured in C6/36 cells for three generations, no CPEs was observed. And no replication was detected by RT-PCR in the supernatant of three consecutive generations of C6/36 cells for JXLC1806-2.

The JXLC1806-2 virus isolate could form virus plaques with regular plaque shapes and clear edges in BHK-21 cells. The diameter of the plaques was 1.19 mm (1.45 ± 0.30 mm, n = 10, 3d) ( Figure 1 ). The virus titer determined by the plaque formation method showed that the JXLC1806-2 virus titer of the fifth passage of BHK cells was 1 × 10^5.6 pfu/mL.

The results showed that four of ten mice inoculated in the first generation gradually developed illness (no food intake, stiff limbs, and negative righting reflex) and died within 96 h of inoculation, five mice died within 108 h of inoculation, with the remaining one mice being maintained to 14 days without illness or death; Two of five mice inoculated in the second generation died 72 h after inoculation, with the remaining three mice died 84 h after inoculation; Three of six mice inoculated in the third generation died 60 h after inoculation, with the remaining three mice died 72 h after inoculation ( Figure 2 ).

As mentioned earlier, the Orthobunyavirus contained S, M, and L segmentss, and the S segment was generally 1000 nt in length and encoded two proteins (N and Ns protein) (Shchetinin et al., 2015; Lefkowitz et al., 2017). However, recent results have shown that the S segment open reading frame sequences of all viruses in the Tete serogroup lacked the Ns protein sequence, the Tete virus S segment open reading frame of the Orthobunyavirus was 774 nt in length and encoded a 258-aa N protein (Mohamed et al., 2017). Therefore, the S gene nucleotide sequence only encoded the N gene and lacked the Ns protein nucleotide sequence, which is characteristic of Tete serogroup viruses (Mohamed et al., 2017; Vuren et al., 2017). The S segment of the JXLC1806-2 virus was 774 nt in length (encoding 258 aa), which was the same as the nucleotide and amino acid length of the S segmentof the Tete virus ( Table 2 ), suggesting that the JXLC1806-2 virus S segment exhibited characteristics of a Tete serogroup virus, both in terms of nucleotide sequence length and open reading frame sequence.

Evolutionary molecular genetic analysis based on the nucleotide sequence of the virus S segment revealed that the Tete serogroup virus was significantly different from the OYO virus, which was placed outside of the serogroup, and was divided into two large evolutionary branches; the first branch contained the Tete, Batama, and I612045 viruses (the I612045 virus was independent of the first two viruses), and the second branch contained the Matruh and Bahig viruses. Evolutionary analysis of the JXLC1806-2 virus S segment and the five Tete serogroup viruses showed that all six of these viruses were divided into two evolutionary branches. The JXLC1806-2 virus was independent of the Matruh and Bahig viruses and formed a separate evolutionary branch ( Figure 2 ). The amino acid homology of N protein between JXLC1806-2 virus and Tete virus serogroup virus was 69.8%-77.2%, less than 90%. From the evolutionary molecular genetic results and homology analysis results, it can be seen that the JXLC1806-2 virus represents a new species of Tete virus group as part of Orthobunyavirus.

The Tete serogroup which includes the Tete (Taylor, 1967; Shchetinin et al., 2015), Bahig (Converse et al., 1974), Matruh (Balducci et al., 1973), I612045 (NCBI, 2012), Batama (Karabatsos, 1978), and Weldona viruses (Calisher et al., 1990) were isolated from birds, ticks and midges, Thus, birds may be the natural hosts of Tete serogroup viruses, and ticks and midges may be the vectors of these viruses. In this study, the JXLC1806-2 virus was isolated from Culicoides biting midges, suggesting that midges were vectors for the Tete serogroup virus. A number of viruses that cause human and animal diseases have been isolated from midges collected from across China (Liang et al., 2017), including the Banna virus of the Seadornavirus genus of Reoviridae (Song et al., 2017), Tibetan circovirus of the Orbivirus genus (Lei et al., 2015; Wang et al., 2017), Getah virus of the alphavirus (Dong et al., 2017), and Japanese encephalitis virus of the flavivirus (Zheng et al., 2012). In this study, the JXLC180602 virus was the first isolate of the Orthobunyavirus Tete serogroup virus isolated from Culicoides biting midges specimens, further suggesting that Chinese midges can carry a variety of arboviruses. This makes our study of great significance for public health, as it will increase our ability to identify and isolate viruses present in midges.

Lichuan County, located in the eastern part of Jiangxi Province, has a mild climate, sufficient sunshine, abundant rainfall, and four distinct seasons, and has a mid-subtropical humid climate (People’s government of lichuan county, 2018), making it suitable for breeding various blood-feeding arthropods. Survey results of midge species in livestock pens in the northeastern region of Jiangxi Province, including Lichuan County, showed that Culicoides crakawai, Culicoides oxystoma, and Culicoides puncticollis were the main local midges in the livestock pens (Liu et al., 2019). However, there were differences in the proportions of these three midges in the different livestock pens. Among them, Culicoides crakawai in duck and hen houses accounted for 96.51% and 72.87%, respectively, while 82.98% of the total midges from the piggery were Culicoides oxystoma, and Culicoides puncticollis accounted for 59.74% in the sheep pens. Midges collected from the cattle farm were identified as Culicoides crakawai (53.31%), Culicoides oxystoma (12.16%), and Culicoides puncticollis (7.89%). Culicoides crakawai was dominant in the local duck and hen houses, Culicoides oxystoma dominated in the piggery, and Culicoides puncticollis dominated in the sheep pens. The cattle farm had various types of midges (Liu et al., 2019). The midges collected in Lichuan County, Jiangxi Province, came from a beef cattle farm that covered an area of approximately 30 hectares. The farm was divided into three areas that were separated by roads. Each area contained approximately 100 heads of beef cattle, and the farm was approximately 10 km away from the village. We collected 3,500 midges from the second breeding ground ( Table 1 ), and these were separated into eight pools for virus isolation. Among them, the JXCL1806-2 pool of midge grinding liquid contained the JXLC1806-2 virus, which was isolated in BHK cells. For morphological identification of the midges, each midge body was dissected, fixed, covered with a glass slide, and viewed under a microscope. However, following this process, the midge specimens were no longer suitable for virus isolation. Therefore, the JXLC1806-2 midge pool was identified based on morphology alone, and the midge species could not be identified. In the future, molecular biology methods will be used (via the COI gene) to identify the midge species prior to virus isolation. This will enable us to clarify the specific species of midges carrying the viruses.