Bluetongue virus and Culicoides variipennis: molecular epidemiology, transeasonality, and taxonomy

Abstract

Bluetongue virus (BTV) remains a serious threat to the livestock industry in the United States, causing substantial direct and indirect costs to sheep and cattle producers. As with most arthropod-borne viruses, BTV epidemiology is not well understood. In order to design cost-effective and efficient control protocols, the vector-virus-host disease cycle must be thoroughly investigated. The biting midge, Culicoides variipennis, is the primary vector of BTV in the U.S. In Colorado, the winter weather is severe enough to break the BTV transmission cycle every year by killing the insect vectors in the area. The mechanism of observed BTV persistence, or transeasonality, in these geographical areas despite this cycle interruption year after year is unknown. Studies were conducted to investigate the possibility that vertical transmission in the insect vector acts as the transeasonality or overwintering mechanism of BTV in temperate regions such as Colorado. In addition, vertebrate host preference and temporal feeding activities of Culicoides in Colorado were studied. Finally, new techniques were developed and characterized for differentiating Culicoides species as adults, pupae, and larvae and for detecting BTV analyte in specimens.Intensive field studies were conducted in Colorado to investigate the transeasonality of BTV in endemic areas. These studies included vertebrate sentinel animal and herd animal surveillance, and extensive larvae and adult Culicoides vector collections from four farm sites. Nucleic acid hybridization, antigen detection enzyme immunoassays, and virus isolation techniques were used to investigate the putative role of the arthropod vector in BTV overwintering in temperate regions. Approximately 62,050 larvae were collected over a two year period. Eighty-seven larval pools tested positive by at least one assay; thirteen pools tested positive by two assays. The ELISA was a satisfactory screening test for BTV; virus isolation, however, had a low sensitivity for detecting BTV in these larval samples. At least one of the sentinel calves seroconverted during the study period showing that BTV transmission was occurring in the study areas. Minimum field infection rates (MFIR's) in Culicoides at the Ausman site ranged from 0% to 0.8% in 1991 and 0% to 0.4% in 1992. At the Fields site, MFIR's ranged from 0% to 0.5% in 1991, and 0% to 0.3% in 1992. MFIR's at the Howard farm ranged from 0% to 0.2% in 1992. Subsequent chi-square analysis showed a significant difference in the study site monthly MFIR's during June 1991 and significant differences in the total MFIR's between the Ausman and Fields site in 1991 and 1992. A time-line analysis of the study data showed a BTV-positive larvae peak in June, a BTV-positive adult C. variipennis peak in July, and BTV seroconversion of sentinel and herd animals in September and October. The implications of these results for the overwintering hypothesis are discussed.Prospective studies were conducted to investigate host preferences and temporal and seasonal feeding activities of C. variipennis in Colorado. These are major determinants of vectoral capacity of C. variipennis for BTV. Populations of insects attracted to sentinel animals at two farm sites were sampled to investigate host attraction rates, seasonal activity periods, site differences and BTV-field infection rates. During the study, approximately 11,000 insects, including C. variipennis, were collected. Blood feeding of C. variipennis initially peaked in June at the beginning of the study and again in August at both farm sites. The Fields site was monitored an extra month and high numbers of C. variipennis were recovered in early September. The calf consistently attracted greater numbers of Culicoides species insects than the sheep at both farm sites, except for C. crepuscularis, which preferred the sheep to the calf. MFIR's of C. variipennis collected during the study were 4.3% at the Fields site and 10.5% at the Ausman site, which was a significant difference (p<0. 001). MFIR's of C. crepuscularis were 6.5% at the Fields site and 12.2% at the Ausman site, which was not a significant difference (p>0.1).Other species of Culicoides can transmit BTV world-wide and may play an as yet unknown role in BTV transmission in the U.S. The accurate identification of these vectors to the species level has proven to be difficult, relying almost exclusively on adult morphological characteristics. Accurate species identification is often needed early in the vector life cycle to ensure effective control strategies. To address these problems, PCR amplification of tRNA gene sequences taxonomic different using tool life consensus primers was investigated as a to differentiate Culicoides species in stages. This technique was capable of differentiating four species of Culicoides commonly found in Colorado. This differentiation could be observed regardless of the life stage of the midge (i.e., adult, pupal, or larval). The technique was also able to discriminate between heterogeneous pools versus homogeneous pools of Culicoides species.A new molecular approach for BTV diagnosis was developed by coupling a reverse transcription-polymerase chain reaction (RT-PCR) with a reversible target capture (RTC) hybridization assay. This RTC-RT-PCR assay was used to detect analyte in BTV-infected cell cultures and BTV-infected Culicoides variipennis. Synthetic oligonucleotide primers were used as both poly-A tailed capture probes and as RT-PCR primers to amplify the BTV analyte to easily detectable levels. By using magnetic-bead separation, the assay bypassed the need for stringent RNA purification techniques, proteinase K treatment and other procedures necessary for the release of this cell-associated virus. This assay showed promise in detecting BTV in clinical specimens such as blood and tissue. The RTC step eliminated agents or factors that inhibit PCR.