Laboratory and Field-Based Tests of Deltamethrin Insecticides Against Adult Culicoides Biting Midges

Abstract

Bluetongue virus (BTV) is an economically important arbovirus of ruminants transmitted by Culicoides biting midges. Vector control using residual spraying or application to livestock is recommended by many authorities to reduce BTV transmission; however, the impact of these measures in terms of both inflicting mortality on Culicoides and subsequently upon BTV transmission is unclear. This study consisted of a standardized World Health Organization laboratory assay to determine the susceptibility of European Culicoides species to deltamethrin and a field trial based upon allowing individuals of a laboratory strain of Culicoides nubeculosus Meigen to feed upon sheep treated with Butox 7.5 pour-on (a deltamethrin-based topical formulation). Susceptibility in the laboratory trial was higher in colony C. nubeculosus (24-h LC₉₀ = 0.00106%), than in field populations of Culicoides obsoletus Meigen (24-h LC₉₀ = 0.00203%) or Culicoides imicola Kieffer (24-h LC₉₀ = 0.00773%). In the field, the pour-on formulation was tested with a total of 816 C. nubeculosus specimens fed upon on the thigh of treated sheep. The study revealed a maximum mortality rate of 49% at 4 d postapplication, and duration of lethal effect was predicted to be as short as 10 d, despite testing being carried out with a highly susceptible strain. The reasons for this low efficacy are discussed with reference both to the potential for lack of spread of the active ingredient on the host and feeding patterns of the major potential vector species on the sheep host. Practical implications for vector control strategies during BTV incursions are also detailed.

Culicoides, bluetongue, deltamethrin, Butox 7.5 pour-on, vector control

Culicoides biting midges (Diptera: Ceratopogonidae) are responsible for the transmission of several livestock arboviruses of international importance, including bluetongue virus (BTV) and African horse sickness virus. In recent years, BTV has inflicted huge economic losses upon Europe (Callistri et al. 2004, Velthuis et al. 2010) because of clinical disease in infected ruminants and animal movement restrictions imposed during attempts to limit the spread and impact of outbreaks. By far the most spectacular example has been the recent epizootic caused by a BTV serotype 8 strain in Northern Europe. This unprecedented event led to transmission occurring across a large number of countries that had no prior history of BTV incursions and demonstrated the previously held theory that northern European Culicoides were capable of transmitting BTV (see Carpenter et al. 2009 for review). The primary vectors involved in transmission in this region are thought to belong to the subgenus Avaritia, although the vector status of these species remains unclear.

During the period before vaccination (which has recently restricted the spread of BTV-8 and subsequently eradicated it from several countries; Zientara et al. 2010), the only method to limit virus transmission lay in restricting animal movements (economically damaging in itself) and in employing vector control. As a result of the scant data at that time concerning larval breeding sites and resting and endophagic behavior of adult BTV-8 vectors, little information was available on which to base a rational strategy for controlling midges (Carpenter et al. 2008). In this context, the use of insecticides in stables, in trucks for animal transportation, or directly on animals was suggested (EFSA 2008), and was employed on a compulsory basis by some countries (e.g., France). The susceptibility of Culicoides populations to pyrethroids, the primary class of insecticides authorized active for residual spraying and products applied directly to livestock (EFSA 2008), has rarely been assessed in a standardized fashion (with Braverman et al. 1995, 2004 being notable exceptions). The most common way currently used to protect cattle and sheep against Culicoides is the application of pour-on formulations along the backline, but the spread of these products is known to be limited on the host (Stendel et al. 1992), and active ingredient may not reach the commonly used biting sites on belly, legs, and face (Nielsen et al. 1988). Whereas studies have examined the impact of exposure of Culicoides to deltamethrin-treated hair, sampled from treated ruminants at known times postapplication (e.g., Mehlhorn et al. 2008; Schmahl et al. 2008, 2009; Papadopoulos et al. 2009), these have tended to involve a passive laboratory exposure to the agent, rather than one caused by a true feeding response. Perhaps even more worrisome, products that have been shown to reduce Culicoides biting activity may still fail to protect animals from BTV infection (Mullens et al. 2001, Melville et al. 2004).

Given the clear lack of information about Culicoides species susceptibility to insecticides, this study aimed to do the following: 1) provide a standardized test of the susceptibility of Culicoides to deltamethrin (the most used pyrethroid) using a World Health Organization (WHO) test kit against three different species of interest, Culicoides nubeculosus Meigen (IAH reference strain), Culicoides obsoletus sensu stricto Meigen, and Culicoides imicola Kieffer field-collected populations (France); and 2) assess the efficiency of Butox 7.5 pour-on (a commonly used deltamethrin-based topical formulation) against the reference strain of C. nubeculosus via direct feeding on sheep.

Materials and Methods

Intrinsic Susceptibility of Culicoides to Deltamethrin (Phase I).

Susceptibility to deltamethrin exposure was assessed in three different species of veterinary importance, as follows: C. nubeculosus, C. obsoletus s.s., and C. imicola, using a standardized WHO assay test. C. nubeculosus specimens were provided from the colony maintained by the Institute of Animal Health (Pirbright, United Kingdom). C. obsoletus s.s. and C. imicola specimens were collected from the field in Saint-Martin-de-Londres (Mediterranean region, France, April 2008) and in Figari (Corsica, France, August 2008) using an ultraviolet light trap (Agricultural Research Council - Onderstepoort Veterinary Institute [ARC-OVI] model, South Africa) and replacing the collection jar with a fine mesh cage. To prevent desiccation, cages were covered with wet papers and an outer layer of aluminum foil and retrieved at dawn.

As sensitivity to insecticides can be age specific (Chareonviriyaphap et al. 2006), 2- to 3-d-old nulliparous females of colony C. nubeculosus and field-collected, nulliparous individuals of C. obsoletus and C. imicola were used with parity determined using abdominal pigmentation (Dyce 1969). For each species, trials were conducted using a standardized assay for assessing insecticide resistance (WHO 1981). For C. nubeculosus, in each replicate, ≈15 nulliparous Culicoides were exposed in the essay system to eight deltamethrin-impregnated papers (0.0035, 0.0025, 0.0015, 0.001, 0.00075, 0.0005, and 0.0003%) and one control paper for 1 h using WHO test kit tubes (WHO/VBC/81.805). Test papers (Whatman n°1 filter paper, 90 g/m², 12 × 15 cm) were impregnated in a WHO collaborative center (Laboratoire de Lutte contre les Insectes Nuisibles/Institut de recherche pour le développement [(LIN)/(IRD)], France) using an acetone-silicon mix as solvent (2 ml per paper, 67% acetone, and 33% silicone). Control papers were impregnated with 2 ml of acetone-silicone mix only (WHO 1981). After exposure, live midges were transferred using an aspirator to observation cages. The record of dead midges gave an assessment of the 1-h mortality, evaluating immediate effects. Then dead and live midges were counted at 24 h after exposure to assess the delayed mortality. For field-collected strains, ≈100 unsorted individuals were used in each tube to be certain of obtaining at least 15 nulliparous females of the target species. As a result of the difficulty in manipulating field-collected Culicoides, only four concentrations were tested (0.005, 0.001, 0.0005, and 0.0001%), and mortality was recorded solely at 24 h after exposure, even if midges were transferred to observation cages after 1-h exposure. During the 24-h observation period, tubes were kept horizontally in an isolated room (24°C) with a 10% sugar solution provided on cotton wool pads.

After exposure during the insecticide trials, field-collected individuals were morphologically identified to species for C. imicola or to complex for specimens of the Obsoletus complex (Delécolle 1985, Delécolle and La Rocque 2002), with specimens of the Obsoletus complex further identified to species level using a diagnostic multiplex polymerase chain reaction (modified from Nolan et al. 2007). Thus, specimens from the Obsoletus complex were initially dissected under a binocular microscope using sterile forceps, and the head and thorax were individually separated into 1.5-ml Eppendorf tubes. Head and thorax were then ground in 100 μl of 5% chelex solution (Biorad, Marne-La-Coquette, France), and the tubes were maintained at 56°C for 1 h, followed by 30-min incubation at 95°C. Tubes were then centrifuged at 13,000 × g for 3 min and then held at −20°C until amplification. Amplifications were performed in a 25 μl reaction volume following the procedure of Nolan et al. (2007) using only the reverse primers for C. obsoletus s.s. and Culicoides scoticus. Products were visualized after electrophoresis on an ethidium bromide-stained 2% agarose gel. Two positive controls, one C. obsoletus s.s. and one C. scoticus females, were identified by an expert and confirmed with different molecular tools (Mathieu et al. 2007, Nolan et al. 2007).

As recommended by the WHO, only those replicates in which control mortality was <20% were considered during analyses, and if control mortality was >5%, mortality data were corrected using Abbott's method, as follows: corrected mortality = 100 × (observed mortality − control mortality)/(100 − control mortality) (WHO 1981). Mortality data were then analyzed by probit regression using CalcuSyn version 2.1 (Biosoft Copyright 1997–2010) to obtain a LC₅₀ and LC₉₀ value for each population.

Efficiency of a Topical Formulation of Deltamethrin: Butox 7.5 Pour-On.

Fifteen Arles Merino sheep (Fig. 1A) of ≈25 mo old and 40 kg weight were separated into five batches of three animals. All animals had access to identical food during trials, were checked for parasitic infections by a veterinary worker before the experiment, and were sheared 15 d before the trial. In four batches of three treated animals, 10 ml of a deltamethrin pour-on formulation (Butox 7.5; 0.75 g of deltamethrin/100 ml) was applied along the backline of 12 individuals from the head to the tail, as recommended by the manufacturer. Insecticide application was staggered between batches of animals to allow different times posttreatment to be assessed on each experimental day. One batch remained entirely untreated as a negative control. During the study, sheep were kept inside the sheepfold and were not exposed to rainfall (which can wash away insecticide) or to direct sunlight (avoiding photo-degradation of insecticide). Each batch was restricted to a dedicated pen so that contact with other groups was not possible. The impact of treatment was assessed at 1, 4, 6, and 13 d posttreatment by allowing 2- to 3-d-old nulliparous unfed females of C. nubeculosus to feed on the thighs of treated sheep. An exposure cage (Fig. 1, d and e) was placed on the sheep thigh, and 10 nulliparous females were transferred with a mouth aspirator into this cage, allowing direct contact between Culicoides and the sheep skin (Fig. 1B). During this exposure period, midges were able to feed, and the number of engorged females was recorded immediately posttrial. Three batches of C. nubeculosus were exposed to each sheep and, after an exposure period of 3 min, midges were transferred to observation cages using an aspirator (Fig. 1C). A cotton wool soaked with 10% sucrose solution was placed on top of the cages, which were then placed in an insecticide-free environment under regulated conditions (≈21°C). Finally, mortalities were recorded 1 and 24 h after exposure. If control mortality was >5%, mortality data were corrected using the Abbott's method described above. After correction, mortality data were modeled using a generalized linear model with a β-binomial distribution to account for observed overdispersion of data (Bouyer et al. 2007). Then upper and lower confidence intervals of prediction were computed for α = 0.05. Effects of mortality were considered statistically different from 0 as long as the lower value of the confidence interval was >0. The R software (version 2.9.1, R foundation for statistical computing) was used for statistical computing and graphs, and the package “aod” for R (Lesnoff and Lancelot 2006) was used to fit β-binomial models.

Fig. 1.

(a) Arles Merino sheep; (b) contact between biting midges and the tight of the sheep; (c) observation cages; (d) exposure cage top view; and (e) exposure cage bottom view (mesh only on the top of the lid).

Results

Intrinsic Susceptibility of Culicoides to Deltamethrin (Phase I).

The susceptibility of Culicoides to deltamethrin was assessed using 348 nulliparous C. nubeculosus in three replicates, 176 nulliparous C. obsoletus s.s. in one replicate, and 442 C. imicola in three replicates. Among the 182 individuals of the Obsoletus complex, 176 were molecularly identified as C. obsoletus s.s., five as C. scoticus, and one specimen did not give any polymerase chain reaction amplification and was not able to be identified; hence, the analysis was only performed with the 176 C. obsoletus s.s. At 24 h postexposure, the laboratory strain of C. nubeculosus was more susceptible to deltamethrin (LC₅₀ = 0.00039% = 0.14 mg/m², and LC₉₀ = 0.00106% = 0.39 mg/m²) than field populations of C. imicola (LC₅₀ = 0.00183% = 0.67 mg/m², and LC₉₀ = 0.00773% = 2.83 mg/m²) and C. obsoletus s.s. (LC₅₀ = 0.00077% = 0.28 mg/m², and LC₉₀ = 0.00203% = 0.74 mg/m²) (Table 1). Differences were marked between the laboratory strain of C. nubeculosus and field populations, but not between C. imicola and C. obsoletus s.s.

Table 1.

Susceptibility of field populations of Culicoides imicola (Corsica, France) and Culicoides obsoletus s.s. (Mediterranean region, France) and of a laboratory strain of Culicoides nubeculosus (IAH colony) to deltamethrin: delayed mortality 24 h after 1-h exposure to different concentrations

Complete mortality was observed at 24 h postexposure for C. nubeculosus at deltamethrin concentrations from 0.0010 to 0.0035% (from 0.37 mg/m² to 1.28 mg/m²) (Fig. 2). Mortality at the end of the 1-h exposure was observed in C. nubeculosus for concentrations >0.0005% (0.18 mg/m²), and 100% lethal effect at the end of the 1-h exposure period was observed at concentrations >0.0025% (0.92 mg/m²).

Fig. 2.

Mortality of C. nubeculosus (mean ± SD, three replicates, n = 348) at the end of 1-h exposure to deltamethrin concentrations (%) and after 24 h postexposure.

Efficiency of a Topical Formulation of Deltamethrin: Butox 7.5 Pour-On.

In 54 exposures, 816 nulliparous C. nubeculosus were exposed to the thighs of untreated and treated sheep. The maximum corrected mortality in treated sheep (45%) was observed on day 4 postexposure (Fig. 3), although mortality rates were highly variable throughout the trial. On day 4 postexposure, the engorgement rate was similar or lower for the treated compared with the untreated sheep; however, this effect was dependent upon the batch of animals considered (e.g., 41 versus 40% for one batch [P = 0.99, χ² test], and 11 versus 28% for another batch [P < 0.001, χ² test]). As the maximum mortality was observed on day 4, data from day 1 were not used for regression, which was used to model the mortality decrease and to estimate the insecticide persistence. The persistence of the lethal effect for this pour-on formulation is estimated to be shorter than 10 d (when the lower value of the predicted confidence interval reached 0).

Fig. 3.

Boxplot of the 24 h postexposure corrected mortality of C. nubeculosus (54 contacts, n = 816) after 1, 4, 6, and 13 d after the application of deltamethrin pour-on formulation (Butox 7.5 pour-on).

Discussion

All Culicoides populations tested were highly susceptible to deltamethrin when exposed in a standardized WHO laboratory assay; however, field trials to assess a formulation based upon this active ingredient led to only limited increases in mortality rates after feeding on treated hosts. Of the species investigated, the most susceptible was C. nubeculosus, whereas field populations of C. imicola and C. obsoletus s.s. exhibited a higher tolerance (in part because of the limited testing carried out upon the latter). This could be because of a variety of factors, including the inbred colony origin, lack of prior exposure to insecticides, and significantly greater size of C. nubeculosus. The use of this assay, however, provides a baseline upon which to base further studies of both efficacy of active ingredients for use against Culicoides and the potential development of resistance. The LC₉₀ values produced for C. imicola during this study were substantially lower than those produced previously using the same species and standardized assay in Israel with cyhalothrin (0.46%) and λ-cyhalothrin (0.0564%) (Braverman et al. 1995, 2004) (Table 2), indicating deltamethrin's superiority as an active ingredient for use against Culicoides, for example.

Table 2.

Susceptibility of different populations of Culicoides midges to pyrethroid insecticides in laboratory trials

In the field trials, Butox 7.5 pour-on was selected for use, as it was widely used during the BTV-8 epizootic in Europe, and particularly in Northern France in part because of its ease of application. Application according to the manufacturer's recommendations should have theoretically resulted in a concentration of 71 mg/m² active ingredient if spread evenly over the individual [estimation of the skin surface of a 42-kg sheep from Fischer et al. (1999)]. Although this theoretical concentration is ≈180, 25, and 100 times greater than the estimated LC₉₀ for the laboratory strain of C. nubeculosus, and the field strains of C. imicola and C. obsoletus s.s., respectively, only limited and short-term effects on mortality were observed in all species and a variable reduction in engorgement noted only in C. nubeculosus placed on the thigh of treated sheep. In other studies carried out in the European Union, high mortalities were observed by exposing field-collected Culicoides sp. to hair clipped from sheep and cattle on which Butox 7.5 pour-on was applied in Germany (Mehlhorn et al. 2008; Schmahl et al. 2008, 2009). Additionally, high and persistent mortality of C. nubeculosus exposed to belly hair was observed when α-cypermethrin was applied as a pour-on to sheep at the concentration of 600 mg/m² (almost 100% mortality after 3 exposure min up to 28 d and 50% at 42 d) and on cattle at 64 mg/m² (almost 100% up to 21 d) (Papadopoulos et al. 2009). The differences between these studies and the current study could be the result of a number of factors, including the use of a different exposure method (mixing adult Culicoides with hair taken from hosts in bags, in the majority of cases), and, additionally, the breeds selected for trials are in some cases not listed and the areas of the body sampled also vary significantly.

Given the application along the backline in the current study, it is unlikely that the active ingredient diffused entirely to the thigh of the sheep when applied in this current formulation, despite using recently shorn individuals. This effect has been demonstrated for a flumethrin-based product used on cattle that was found to vary from 0.2 to 126 mg/m² at day 1 and from 0.1 to 8 mg/m² at day 10 after pour-on application, depending on distance from the site of application (Stendel et al. 1992). This effect has also been documented for the United States BTV vector, Culicoides sonorensis Wirth and Jones, feeding on a membrane system through hair of both goats (Mullens 1993) and calves (Mullens et al. 2000), which had been treated with permethrin. In this study, mortality effects were recorded across several body regions, and it was notable that these were greatest at the point of application. The poor diffusion of active ingredient when applied as pour-on on sheep was also suggested by Carpenter et al. (2008) to explain the absence of toxicity to C. nubeculosus of belly and leg hair from sheep treated with 1.25% cypermethrin (at 450 mg/m²) or 1% deltamethrin (at 60 mg/m²) pour-on in the United Kingdom.

A recent study by Mullens et al. (2010) demonstrated that application of 10 ml of 7.5% deltamethrin (Butox 7.5) to shorn sheep eliminated engorgement by Culicoides. However, these authors applied the insecticide to the entire animal body rather than as a pour-on applied to the backline. Hence, whereas dipping of sheep in insecticidal products is currently in decline in most European countries because of user safety concerns and the risk of groundwater contamination, the use of spray races that reduce runoff may prove to be more effective in treating large flocks. It would be desirable, however, that this effect is quantified by chemical analyses of insecticide residues on host hair, which would allow a better understanding of spread of active ingredient and Culicoides mortality in the field.

Although it is likely that ruminants with hair that allows more rapid and active dispersion of product formulations will inflict higher mortality rates on feeding Culicoides, it is still unclear how and whether this ultimately translates to reduction in BTV transmission. To date, small-scale studies of protection provided in areas of endemic BTV transmission have produced equivocal results. In Australia, weekly application of deltamethrin pour-on on cattle (48 mg/m²) reduced incidence of BTV transmission, whereas flumethrin (216 mg/m²) and fenvalerate (59 mg/m²) pour-on applications had no discernible effect. Neither of these trials, however, eliminated BTV transmission despite the use of relatively high concentration of insecticide (Melville et al. 2004). Similarly, Mullens et al. (2001) found that applying 0.2% wt:vol permethrin to the bellies of a herd of 200 cattle had no significant effect on BTV transmission. In the case of both these studies, it is unclear whether the absence of any effect reflected the poor performance of the insecticide and/or whether the intervention itself was applied on too small a scale to have any significant impact on the local midge population in term of reduction of density and longevity. Larger scale studies, using a standardized approach across several countries and examining different primary vector species, would enable pour-on products to be both compared and optimized in formulation and application as a primary tool in control of arboviral outbreaks during the period before implementing vaccination campaigns.

Acknowledgments

We thank the Direction générale de l'alimentation from the French Agricultural Ministry, which partially funded this study; Stéphane Duchon from LIN/IRD for his help in treating the OMS paper; Laëtitia Gardès from Cirad and Sandie Bellec from EID-Med for technical help; Renaud Lancelot from Cirad for the statistical analyses; Eric Denison and James Barber from Institute of Animal Health for their valuable help; and Glenn Bellis from the Northern Australia Quarantine Strategy and Lorna Melville from the OIC Berrimah Veterinary Laboratories for their information.