Host Landing And Diel Activity of Potent Vectors of Bluetongue Disease, Culicoides Oxystoma and Culicoides Peregrinus


 Background The spread of bluetongue virus depends on the vectorial ability of Culicoides affecting the susceptible host. Animal farms in West Bengal have reported prevalence of potent vectors of BTV (C. oxystoma, C. peregrinus and C. fulvus). Besides, high seroprevalence of BTV was also reported from this cattle dense region. Henceforth host-seeking activity of two important potent vectors, C. oxystoma and C. peregrinus on cattle were studied in two farm sites of West Bengal, India.Methods The study was done in 2018-19 comprising of total 297 hours of collection over 27 nights. A comparison was made between the catches obtained by mouth aspirator and light trap. Hourly collections of Culicoides were done directly from cattle (oral aspirator) as well as light trap was operated in close vicinity of cattle at a different shed.Results A total of 11,462 Culicoides belonging to C. oxystoma, C. peregrinus and C. fulvus were collected in light trap and aspirator. In aspirator 4764 midges were collected whereas 6698 individuals were collected in light trap. The following species were aspirated: C. peregrinus and C. oxystoma; however the light trap catches consisted of C. fulvus, C. oxystoma and C. peregrinus. Light trap collection exhibited crepuscular activity whereas aspirator collection was maximum between 4.00 am and 5.00 am. Likewise maximum landing of midges was observed in neck and hump region of cattle.Conclusion It was observed that the preferred time of feeding of C. peregrinus and C. oxystoma on cattle were early morning hours though midges were ubiquitous from dusk to dawn. Surprisingly the preferential landing of the two vectors were mostly restricted to the neck and hump region of the cattle. The results obtained during the study warrants further insight into the factors influencing the landing site by the vectors which may be useful biological data in disease management and draw effective deterrent strategies.

This investigation was carried out in two rural villages of West Bengal 51 km apart i) Dharan (DH; 23°02'57.7"N, 87°51'47.5''E), ii) Sahibganj-Tantipara (ST; 23°44'29.14"N, 87°82'76.56"E). The biting midges were trapped throughout the seasons in DH during May to October, 2018 and in April to June, 2019; and in ST from May to October, 2018. In India, Indian Meteorological Department (IMD) recognizes the occurrences of four seasons: winter, December to February; summer or pre-monsoon season, lasting from March to May; monsoon or rainy season, lasting from June to September; Post-monsoon or autumn season, lasting from October to November. At ST village site, 12 night collections were made i.e., two catches/month, whereas in DH village site, monthly three collections were done (excepting for the months of October, April and June where single collections/every month). A total of 27 night catch data (297 hrs of collection) were considered for both the sites. In both the sites, the sheds housing the cattle were made of mud-brick adjoining to household. Geographically the villages are situated amidst agricultural elds in which extensive rice cultivation is practiced throughout the year, water logging even during dry seasons predominantly rice growing areas of West Bengal. Most households maintain livestock animals for their economic sustenance. It experiences tropical climate with annual rainfall of 1496 mm. The hygiene conditions are compromised within and outside the perimeter of the animal sheds as dung heaps and paddy straws were garbaged. In ST, the cattle sheds were located on the embankment of a pond. Moreover, the drainage system of the village was found to empty into the pond. For this study one cattle shed at DH and two cattle sheds at ST were chosen.

Description of collection
The adults that landed on the host body surface was retrieved by aspirators (oral & mechanical) and ying adults were trapped by LED based light traps installed at close vicinity of cattle within the shed. Adults were aspirated covering various parts of the body surface of a white coloured adult cow. For our convenience the entire body surface of the cattle was subdivided [23] following the studies of [20] and [21]. Catches made were labeled as follows: head (H1), neck (H2), hump (H3), back (H4), leg (H5), belly (H6), hip (H7) (Figure 1). Each catch duration was restricted to 10 minutes/hour the entire process of aspiration initiated on 18.00hrs in the evening that continued up to 6.00hr in the morning (12 hours per diem). The catch period (designated as T1 hr to T11 hr) was followed: T1: 18.00-19.00, T2: 19.00-20.00, T3: 20.00-21.00, T4: 21.00-22.00, T5: 23.00-00.00, T6: 00.00-01.00, T7: 01.00-02.00, T8: 02.00-03.00, T9: 03.00-04.00, T10: 04.00-05.00, T11: 05.00-06.00. During the process of aspiration all other cattle stationed within the sheds were evacuated at least half an hour prior to collection. 4W dim white light was used during the process of collection of the Culicoides from cattle. The LED light trap was operated within the cattle shed (ST), 200 m apart from the shed where aspiration based collections were done. Although a mechanical aspirator was also used for the purpose; however, the mouth aspirator was tedious but proved handy in aspirating the adults tucked within the fur.

Statistical Analysis
A logistic regression was done on the proportion of engorged and non engorged females to justify the effects of time and site on host landing. The logistic regression was carried following the binomial GLM with logit link, using time and site as the explanatory variables on the assumptions of generalized linear model (GLM). The logistic regression equation form: (y) = 1/(1+exp(-(a+b 1 x 1 +b 2 x 2 +b 3 x 3 )); where the explanatory variables, x 1 , x 2 and x 3, represented the time of landing of Culicoides on host, site of landing on host and interaction between the two components respectively and y was the response variable. The regression analysis was performed on the assumptions that the landing of non-engorged and engorged Culicoides follow binomial distribution (n, p) with n replicates for each set of independent variables (time of landing, site of landing on host and interaction between host landing and time). Maximum likelihood method has been used as a measure to estimate the logit linked parameters through statistical software. Using the value of Wald's Chi square, the parameter of the models were tested for the signi cance at P= 0.05 level.

Results
In this investigation, the trapped number of C. oxystoma females caught were 1924 individuals further categorized as 1315 non-engorged, 609 engorged and C. peregrinus females consisted of 2070 individuals of which 1192 non-engorged and 878 engorged (Table 1a). In ST, a total catch of C. oxystoma were 2119 individuals (899 engorged, 1220 non-engorged) and C. peregrinus were 2645 individuals (1630 engorged and 1015 non-engorged) respectively (Table 1b). In ST collections, the following species were aspirated: C. peregrinus and C. oxystoma; however the light trap catches consisted of C. fulvus, C. oxystoma and C. peregrinus. The total Culicoides caught through aspiration were 4764 compared to light trap catch of 6698 individuals (Figure 2a; 2b). Signi cant numbers of Culicoides individuals landed on the upper portion of the cow, while very few individuals landed on the belly and legs. Landing of Culicoides was limited to hip due to continuous tail whipping, licking and kicking re exes of the cattle. It appeared that a period of 30-40 min i.e. between 4 am and 5 am (dawn) was the actual feeding window of the C. oxystoma and C. peregrinus attacking the cattle (Table 2). During the feeding interval usually the Culicoides females swarm on to the host althoughboth the species attacked the cattle with equal intensity, the proportion of engorged C. peregrinus aspirated were more (42.42%) compared to C. oxystoma (31.65%) ( Figure 3). The results highlighted variation in abundance of the Culicoides analyzed through ANOVA with post hoc Tukey test considering the time of host-seeking activity and different body parts of cattle as the source of variations (Table 3a; Table 3b).

Discussion
The Culicoides species investigated in this study was reported to be vectors of important diseases associated with farm animals' worldwide [17]. Shielding of animals from attack of female Culicoides may be adopted as a measure to interrupt disease transmission hence information on peak activity of Culicoides from this region of world will be useful in disease management strategies [17]. The landing time and site of females belonging to C. oxystoma and C. peregrinus on the cattle in the early morning for the purpose of obtaining blood meal was observed. However, C. fulvus has been reported only in a light trap from one of the study sites thereby raising doubts on the host preference of the species. The aspirator based study validated the diel activity of the C. oxystoma and C. peregrinus. Signi cant proportion of engorged females in the study justi ed the usage of aspirator to intercept the host-seeking females. Moreover the present study ascertained that cattle not only attracted Culicoides but constituted one of the signi cant hosts. C. oxystoma and C. peregrinus were known to be one of the most prevalent species across India, from which the BTV serotypes have been isolated [24, 25] and also were enlisted as potent vectors of BTV from the subcontinent [18]. Moreover both the species were reported to be most abundant in cattle sheds of West Bengal [17] thereby contributing to the signi cance of the study. High proportion of engorged C. peregrinus and C. oxystoma feeding on the cattle suggests active blood-seeking behaviour of the species, which agrees with the opinion regarding erceness of C. peregrinus [26]. C. oxystoma has been recognized as potential vectors of BTV from India The proportion of engorged to non-engorged females aspirated ascertains landing of Culicoides on host but does not warrant feeding. It was noted that 30-40% of the total midges landing actually fed on the cattle; the nding substantiated the conclusion drawn on activity of Culicoides midges from Ireland [32].
A bimodal distribution of C. oxystoma, C. peregrinus and C. fulvus was observed in light trap based collections within the animal shed. The plausible explanation was that the resting adults within the cattle shed after completion of blood meal were also attracted towards the light traps. However, Culicoides were observed to be prevalent in resting conditions in the cracks, crevices, walls, and roof within the shed housing cattle throughout the 24 hour period. Mere prevalence of Culicoides within cattle sheds does not warrant their blood feeding activity.
Amongst the different portions of cattle body, the desired landing site of the C. oxystoma and C. peregrinus were neck, hump followed by the head of the cattle (mostly in and around the ear). Although various studies suggested temperature to be a signi cant factor in uencing the landing of Culicoides [33], our observation suggests that the thickness of epidermis and degree of vascularization might be two of the most important criteria in uencing the landing of female Culicoides. Likewise [32] reported on the preference of C. obsoletus, C. dewul , C. pulicaris, C. punctatus and C. nebeculosus for mane and lower legs of horse.
[33] remarked that 72% of the total collections were from the belly whereas 28% from the dorsal surface. C. puncticollis and C. schultzei preferred the belly region whereas C. imicola preferred the dorsal surface. A comparative preferential landing study of Culicoides on dairy cows, Shetland pony and sheep has been done [21]. It was observed that C. chiopterus favoured legs, C. punctatus, C. achrayi landed on the belly. C. obsoletus, C. dewul and C. pulicaris landed on head, back and anks respectively. Moreover C. chiopterus, C. punctatus, C. obsoletus/scoticus favoured the belly region of horses [35]. Anatomically, vascularization has been complex in the neck and hump region [23], henceforth correlating to the preferred landing of the Culicoides at these sites. Likewise the thickness of epidermis was found to be less in the aforesaid body parts. The difference in preference could not be attributed to body surface temperature as there was not much difference in temperature except belly and hip, where the difference was 1°C. However in the present study Culicoides on belly was less compared to that of neck and hump.
This investigation recorded diel activity and host-seeking activity of the potent vector species associated with cattle in the state of West Bengal. Most of the cattle sheds in West Bengal were open type and cattle were either herded in this shed or in open yard at night. In such a setup there exists a high risk of cattle being exposed to Culicoides bite during the early morning. The present study argues that in order to minimize the contact between Culicoides and cattle, the cattle should be housed in a closed shed for at least 1 hour (i.e., between 4.00 am and 05.00 am). In closed sheds the activity of Culicoides has been observed to be reduced by 14 fold (unpublished). Owing to absence of proper closed sheds in most of the rural areas, an alternative approach could also be adopted by farmers. Blankets, badges or jute bags could be placed over the hump and neck region of the cattle during the peak activity time (between 4 am and 5 am) of midges thereby interfering with the preferred landing and subsequent feeding of the Culicoides midges.      Figure 1 The body of cow was divided into head (H1), neck (H2), hump (H3), back (H4), leg (H5), belly (H6), hip (H7) Figure 2