Novel field assays and the comparative repellency of BioUD, DEET and permethrin against Amblyomma americanum

Transcription

1 Medical and Veterinary Entomology (2011) 25, doi: /j x Novel field assays and the comparative repellency of BioUD, DEET and permethrin against Amblyomma americanum B. W. BISSINGER 1, C.S. APPERSON 1,D.W.WATSON 1, C. ARELLANO 2, D. E. SONENSHINE 3 and R. M. R O E 1 1 Department of Entomology, North Carolina State University (NCSU), Raleigh, NC, U.S.A., 2 Department of Statistics, NCSU, Raleigh, NC, U.S.A. and 3 Department of Biological Sciences, Old Dominion University, Norfolk, VA, U.S.A. Abstract. Two new field bioassay methods were developed to compare the repellent activity of BioUD (containing 7.75% 2-undecanone), 98.1% DEET and 0.5% permethrin against natural populations of nymphal Amblyomma americanum (L.) (Acari: Ixodidae). In a cloth sheet assay, pieces of material measuring cm, separately treated with one of the test materials or the appropriate solvent carrier, were placed at random on the ground and baited with dry ice for 1 h. Mean numbers of ticks on repellent-treated sheets were significantly lower than on control sheets. There was no significant difference in the number of ticks collected between sheets treated with BioUD and those treated with DEET. However, significantly fewer ticks were found on sheets treated with BioUD or DEET than on permethrin-treated sheets. In a sock test, over-the-calf tube socks were treated with one of the test materials or the appropriate solvent carrier. Human volunteers wore a repellent-treated and a corresponding carrier-treated sock on either leg and walked randomly over an area of approximately 4000 m 2 for 15 min. Significantly fewer ticks were collected from socks treated with BioUD or DEET than from socks treated with the carrier and there was no significant difference in repellency between these two agents. No difference in the mean number of ticks collected was found between permethrin-treated and corresponding carrier-treated socks. To examine the mechanism of repellency of BioUD, a four-choice olfactometer was used to assess spatial repellency against adult A. americanum. As expected in the absence of a repellent, when all choices were represented by water-treated filter paper, ticks were equally distributed among the choices. When one choice consisted of BioUD -treated filter paper and the remaining choices of water-treated paper, the distribution of ticks on the repellent-treated paper was significantly lower than might be expected to occur by chance, suggesting that repellency is at least partly achieved by an olfactory mechanism. Key words. Arthropod repellent, BioUD, DEET, olfactometer, permethrin, spatial repellency, ticks, undecanone. Introduction Ixodid ticks are obligatory blood-feeding ectoparasites and important vectors of pathogens that can cause disease in humans and other animals (Sonenshine, 1993). The lone star tick, Amblyomma americanum (L.), which was used in the current study, is a vector of the rickettsial pathogens Ehrlichia chaffeensis and Ehrlichia ewingii and readily bites humans during its larval, nymphal and adult life stages (Childs & Paddock, 2003). Amblyomma americanum is widespread in Correspondence: R. M. Roe, Department of Entomology, North Carolina State University, Campus Box 7647, Raleigh, NC , U.S.A. Tel.: ; Fax: ; michael_roe@ncsu.edu Medical and Veterinary Entomology 2010 The Royal Entomological Society 217

2 218 B. W. Bissinger et al. the southeastern and south-central U.S.A. (Keirans & Durden, 1998) and has expanded its range along the east coast as far north as New York state (Childs & Paddock, 2003). Arthropod repellents play an important role in personal protection against tick bites. The U.S. Centers for Disease Control and Prevention (2008) recommend using a repellent containing permethrin on clothing or 20% DEET (N,Ndiethyl-m-toluamide) on skin. Although permethrin is sold as a repellent, the protection it provides primarily results from its rapid toxicity (Lane & Anderson, 1984). DEET is used by millions of people each year and has a good record (Osimitz & Grothaus, 1995; Koren et al., 2003; Sudaken & Trevathan, 2003). However, some people doubt its safety (Aquino et al., 2004) or find its smell and oily consistency unpleasant (Frances & Debboun, 2007). It is important that user-accepted repellent products are available to provide protection against ticks for people who choose not to use DEET or other synthetic products. Plant-based repellents have gained popularity in recent years (Gerberg & Novak, 2007) and can be used as tick repellents (Jaenson et al., 2006; Thorsell et al., 2006; Tunón et al., 2006; Mkolo & Magano, 2007; Garboui et al., 2009; Bissinger & Roe, 2010). These compounds, however, generally provide repellency for a shorter length of time than DEET-based repellents (Fradin & Day, 2002; Moore et al., 2007). The repellent BioUD (Homs, LLC, Clayton, NC, U.S.A.) contains the U.S. Environmental Protection Agency (EPA)- registered active ingredient 2-undecanone, which was originally isolated from glandular trichomes on the leaves and stems of the wild tomato plant, Lycopersicon hirsutum Dunal f. glabratum C. H. Mull (Farrar & Kennedy, 1987). Previous studies have shown that BioUD (7.75% 2-undecanone) applied to the skin of human subjects is repellent against Dermacentor variabilis Say (Acari: Ixodidae) in the laboratory and against mosquitoes in the laboratory and field (Witting- Bissinger et al., 2008). BioUD is also repellent against the ticks A. americanum, D. variabilis and Ixodes scapularis Say (Acari: Ixodidae) in the laboratory in the absence of a host (Bissinger et al., 2009a, 2009b). In laboratory trials against A. americanum and D. variabilis, BioUD applied to cotton cheesecloth provided equivalent repellency to products containing 98.1% DEET, 19.6% IR3535, 30% oil of lemon eucalyptus, and greater repellency than a product containing 0.5% permethrin. BioUD was also more repellent than products containing 5% and 15% picaridin against A. americanum and equivalent in repellency to 15% picaridin against D. variabilis. In head-to-head trials, BioUD exhibited greater repellency than IR3535 and lemon eucalyptus against A. americanum and greater repellency than IR3535 against D. variabilis at the same concentrations (Bissinger et al., 2009b). Notably, field trials of the repellency of BioUD against ticks have not been conducted. Understandably, tick repellents should be evaluated under field conditions. However, practical methods for testing repellents in field conditions are lacking. Additionally, more comparative laboratory vs. field studies are needed to examine the value of the former in predicting repellent activity under field conditions (Bissinger & Roe, 2010). Laboratory tests conducted in the absence of a host or host stimuli can be completed rapidly and inexpensively, but may overestimate repellency and may not represent the true performance of the agent in the field (Matthewson et al., 1981; Dautel, 2004). By contrast, field tests using live hosts may provide a more realistic evaluation, but can be time-consuming and require approval by an institutional review board (IRB) or an institutional animal care and use committee. Field tests also expose subjects to tick bites and, potentially, to tickborne diseases. It is well known that a number of cues are attractive to ticks, including heat, carbon dioxide (CO 2 ), butyric and lactic acid, and ammonia (Sonenshine et al., 2002). The incorporation of one or more host cues into repellency field trials could potentially serve as a substitute for a live host. Most tick repellency bioassays do not differentiate between repellency caused by olfaction and that caused by tactile chemoreception (Carroll et al., 2005). Olfaction appears to be involved in repellency for at least some compounds. For example, indalone presented in an air stream caused Amblyomma variegatum Fabricius to make oriented movements away from the repellent source (McMahon et al., 2003). Similarly, nymphal A. americanum and I. scapularis were repelled by both DEET and SS220 applied to human fingers wrapped in organdie cloth to prevent direct contact of the ticks with the repellent (Carroll et al., 2005). In Y-tube olfactometer trials, Ixodes ricinus L. would approach a DEET-treated surface but would not contact it (Dautel et al., 1999). Spatial repellency was defined by Gouck et al. (1967) as repellency produced by a compound at a distance. The spatial repellency of BioUD, a relatively new commercial repellent, has not been studied. The goals of the current study were to examine the repellency of BioUD, DEET and permethrin against A. americanum using two novel field tests, one using a host cue (CO 2 from dry ice) and the other using human hosts. Results were also compared with those of field tests and laboratory choice studies conducted earlier with BioUD. We also determined for the first time whether BioUD acts as a spatial repellent. Materials and methods Field sites Two tests to evaluate tick repellency in the field were simultaneously conducted in a mixed deciduous/coniferous forest near the White Pines Nature Preserve in Lee County, NC, U.S.A. ( N, W) on 11 and 12 June 2009 during hours and in a similar woodland approximately 1.5 km from Jordan Lake in Chatham County, NC ( N, W) on 25 June 2009 during hours. Temperatures were C (11 June), C (12 June) and C (25 June). Conditions were mostly cloudy on 11 June and clear on 12 and 25 June. The ground at both sites was covered in leaf litter with some shrubby undergrowth. Test substances and treatment procedures for field studies Repellency studies were conducted with BioUD spray (7.75% 2-undecanone), 98.1% DEET (Jungle Juice; Sawyer

3 Repellency of BioUD, DEET and permethrin against Amblyomma americanum 219 Products, Inc., Safety Harbor, FL, U.S.A.) and 0.5% permethrin (Premium Clothing insect repellent; Sawyer Products, Inc.). All repellents were purchased at local retail stores. To ensure that repellents could be applied uniformly on the test substrates (cloth), they were diluted 1 : 1 with absolute ethanol (DEET and permethrin) or distilled water (BioUD ). BioUD is formulated in water by the manufacturer. Control experiments (without repellent) were conducted with the appropriate diluting solvent. Repellents were applied using identical 59.1-mL plastic fingertip spray bottles (Bottle Crew LLC, West Bloomfield, MI, U.S.A.). The number of sprays that delivered 1 ml of repellent was determined prior to repellent application (overspray was estimated to be <5%). (A) Cloth sheet assay For the cloth sheet assay, the diluted test substances were applied at a rate of 2 ml/600 cm 2 so that the equivalent of 1 ml/600 cm 2 of undiluted substance was applied to one side of white cloth sheets (60% cotton, 40% polyester) measuring cm (Mainstays, Wal-Mart Stores, Inc., Bentonville, AR, U.S.A.). Sheets were treated 3 h before the beginning of bioassays and allowed to dry at room temperature in the laboratory (by hanging from a clothesline for 15 min) before they were placed individually in sealed plastic bags for transport to the field. Prior to treatment, sheets were cleaned using a normal cycle in a home washing machine in warm tap water without detergent and dried on high heat in a home dryer. In the field, repellent- and control-treated sheets were randomly distributed on the ground. Prior to use, sheets were stored in separate labelled plastic bags in a box; during distribution, bags were blindly selected from the box. The sheets were placed on the ground over an area of approximately 4000 m 2 and were separated from one another by at least 15 m. Sheets were laid with the treated side facing up. Pelleted dry ice ( cm pellets, 0.45 kg total amount) was placed in the centre of each sheet (Fig. 1A). The CO 2 emitted is a tick attractant and dry ice has been used before as a collection method for field-sampling (Garcia, 1962; Wilson et al., 1972; Norval et al., 1987). After 1 h, any remaining dry ice was removed and the cloths were collected and placed in separate labelled plastic bags that were sealed and returned to the laboratory. The number of nymphal A. americanum on both sides of each cloth was later recorded by visual observation without magnification. Voucher specimens were deposited in the North Carolina State University (NCSU) Insect Museum. A total of 10 assays for each treatment and control were conducted. Sock assay Treated socks worn by human subjects have been used previously to assess tick repellency in the laboratory (Smith & Gouck, 1946; Brennan, 1947), but not in the field. For the sock assay, repellents and control solvents were applied 3 h before the field assay at a rate of 1 ml/cm 2 to the outer surface of (B) Fig. 1. (A) Sheet used in field repellency trials with pelleted dry ice. (B) Human volunteer wearing one repellent-treated and one controltreated sock. white, over-the-calf tube socks (81% cotton, 18% polyester, 1% spandex) (Fruit of the Loom, Inc., Bowling Green, KY, U.S.A.) as described for the sheet assays. The outer surface area of the socks was determined by measuring the socks before they were worn. In the field, the repellents and controls were evaluated using human volunteers under protocol , approved by the NCSU IRB. All human subjects provided written informed consent before participating in the study. Each subject wore shorts that did not extend below the knee, a repellent-treated sock on one leg and an identical sock

4 220 B. W. Bissinger et al. treated with the respective control solvent on the other leg in direct contact with the skin; shoes were not worn (Fig. 1B). Prior to treatment, socks were laundered in the same manner as the cloth sheets. Subjects were instructed by the study conductor to walk randomly over an area of approximately 4000 m 2 (the same area as used in the cloth sheet assay) at a slow pace (approximately 30 steps/min) for 15 min. Subjects were not to follow the same paths as one another. During the test period, subjects collected any ticks from the legs above the socks to prevent tick bites; these were placed into respective plastic bags and counted as not having been repelled because they had crawled over the sock barrier. After the 15-min test period, subjects removed the socks and placed them separately in labelled plastic bags which were returned to the laboratory so that the number of A. americanum per sock could be recorded. The experiment was repeated three or four times per day, retaining the same subject for each compound tested and keeping the time period between the treatment of socks and the start of the assay at 3 h. Repellent-treated socks were always worn on the same leg on a given day to avoid contamination of the control. Repellent-treated socks were worn on the opposite leg on alternate days. Four volunteers (three males, one female) participated in the study. Each repellent was evaluated with three subjects per day (three of the four volunteers were involved in testing on each test day). Repellents were randomly assigned to volunteers on test day one, then re-assigned each day so that all repellents were tested on three different human subjects. Volunteers who worked on consecutive days bathed between days. Olfactometer trials The spatial repellency of BioUD against A. americanum was evaluated using a four-choice (4 glass bulbs, each cm) olfactometer [Agricultural Research Services (ARS), Gainesville, FL, U.S.A.] (Fig. 2). Unfed, adult A. americanum of both sexes were collected from the field using dry ice-baited towels during hours on 11 June 2009 near the White Pines Nature Preserve in Lee County, NC ( N, W). Nymphs were not used in olfactometer trials Fig. 2. Diagram of the four-choice olfactometer used in tick olfactometer trials.

5 Repellency of BioUD, DEET and permethrin against Amblyomma americanum 221 because their small size increased their potential to escape the olfactometer. Prior to bioassays, ticks were stored for no longer than 45 days at 28 ± 1 C, 85% relative humidity (RH) and a 14 : 10 h photoperiod (which included 60-min dusk and dawn crepuscular light periods). Tests were conducted at 25.4 ± 0.1 C under ambient (fluorescent) light; ticks were acclimatized to this temperature for 1 h before beginning bioassays. Humidified (approximately 90%) charcoal-filtered air was pumped into the olfactometer using an air compressor (model SKH32EG115E; General Electric Co., Fort Wayne, IN, U.S.A.). The pressure of the air reaching each bulb was controlled with pressure manifolds (ARS 4-choice olfactometer air delivery system) to produce an airflow of 50 ml/min measured using a #11 compact shielded flowmeter (Gilmont Instruments, Barrington, IL, U.S.A.). A line connected to a vacuum pump was used to remove air from the central region of the body of the olfactometer at a rate of 200 ml/min to avoid the accumulation of vaporized repellent; this air was vented from the test area. Semi-circular pieces of filter paper (31.8 cm 2, Whatman no. 1) were treated with either 250 μl of distilled water (control) or BioUD (7.75% 2-undecanone) (treatment) and allowed to dry for 3 h under a fume hood at room temperature before bioassays were begun. Afterwards, the filter paper was folded in half lengthwise twice so that it could be inserted into the bulb without contacting the rim of the bulb to avoid any transfer of repellent to the entrance of the bulb. Prior to each repellent bioassay, control-only trials were conducted (water-treated filter paper in all four bulbs) to observe the distribution of ticks in the absence of repellent. In the repellency trials, one bulb contained filter paper treated with BioUD and the other three bulbs contained filter paper treated with water. The bulb containing the repellent treatment was randomly selected and this position was rotated clockwise for each new trial in order to avoid positional bias. All bulbs were cleaned with 70% ethanol between trials and kept at room temperature until all of the solvent had evaporated. New repellentand water-treated filter papers were used for each trial. At the beginning of each bioassay, eight ticks were placed in the central region of the olfactometer and the number of ticks in each bulb was recorded 15 min later. Bulbs were observed intermittently throughout these trials to determine whether ticks directly contacted the treated filter paper. Additional observations during which the bulbs were watched continuously confirmed that ticks never contacted the BioUD -treated filter paper. The distance from the centre of the arena to the bulb was 25 cm (Fig. 2). Ticks that did not move into the bulbs were recorded as non-responders (only 6.25% of the total tested). Control and repellent trials were each repeated eight times using ticks not previously tested. Data analysis In the sheet assays, the response to repellent-treated cloth for nymphal A. americanum were log 10 (x) transformed to achieve approximate variance homogeneity and normality prior to running analyses. Data were analysed by fitting a general linear mixed model to observed responses using the proc mixed procedure in sas Version 9.1 (SAS Institute, Inc., 2003), with date treated as a fixed factor and replicate within date and treatments (repellents and controls) as a random factor. Preplanned pairwise mean comparisons were made to determine if differences in the mean number of ticks collected from sheets treated with different repellents and between sheets treated with each repellent and its respective control were statistically significant. Prior to analyses, tick distribution on repellent-treated socks were square root (x) transformed to achieve approximate variance homogeneity and normality. Data were analysed using proc mixed with date, subject and treatment within each set (each different treatment and its respective control) treated as fixed factors, and repeated pairs of socks treated as measures with an unstructured error covariance. Preplanned pairwise comparisons were conducted to determine if differences between the mean number of ticks collected from treated and corresponding control socks for each repellent were statistically significant. Comparisons were also conducted to determine if differences in the mean number of ticks collected between socks carrying different repellent treatments were statistically significant. For olfactometer trials, a chi-squared test for randomness across repetitions in the absence of repellent was conducted using the proc freq procedure in sas Version 9.1 (SAS Institute, Inc., 2003) to determine if the distribution of ticks in the control bulbs differed significantly (P = 0.05) from the null hypothesis that the expected proportion (p) of ticks per bulb in the absence of a repellent is 0.25 (H o : p 1 = p 2 = p 3 = p 4 ). The same procedure was conducted for treatment assays in which one bulb contained BioUD -treated paper and the other three bulbs contained water-treated paper; however, in this case the null hypothesis for the expected proportion of ticks per control bulb (n = 3) was 0.33 (H o : p 1 = p 2 = p 3 ). The average proportion for each bulb was then compared between control trials (all bulbs containing water-treated paper) and treatment trials (three bulbs containing water-treated paper, one bulb containing BioUD -treated paper) by fitting a general linear mixed model to observed responses using the proc glimmix procedure. A log link and a Poisson distribution for the count number of individuals within each chamber at the end of the experiment was used, with a correction for unequal numbers of individuals at each trial (offset variable), which roughly corresponds to fitting the rate of occupancy for each treatment/ control chamber. Results Treated sheet assays The assay format is shown in Fig. 1A. For repellent-treated sheets, a significant difference was found between all repellents and their respective controls (Fig. 3). The number of nymphs of A. americanum was significantly lower on sheets treated with BioUD than on those treated with water (F = 15.74, d.f. = 1,29, P = ). Significantly fewer ticks were also found on DEET-treated (F = 33.50, d.f. = 1,29, P<0.0001) and permethrin-treated (F = 4.66, d.f. = 1,29, P = 0.039) sheets compared with their respective ethanol controls. No difference

6 222 B. W. Bissinger et al. Fig. 3. Mean number of ticks [(±1 standard error of the mean (SEM)] collected from sheets treated with BioUD and its respective control (water) and sheets treated with DEET, permethrin or their respective controls (ethanol). Different lower case letters above bars indicate a statistically significant difference in the mean number of ticks collected from repellent-treated sheets and their respective controls. Different upper case letters above bars indicate a significant difference in the mean number of ticks collected from sheets across all treatments. was found between sheets treated with BioUD and those treated with DEET in the mean number of nymphs collected (F = 1.68, d.f. = 1,29, P = 0.21) (Fig. 3), and significantly fewer ticks were collected from sheets treated with BioUD or DEET than from permethrin-treated sheets (F = 6.33, d.f. = 1,29, P = and F = 13.79, d.f. = 1,29, P = , respectively). The number of adults collected in these experiments was minimal (82.5% nymphs, 17.5% adults; n = 451 in total). However, combining the adults with the nymphs did not alter the conclusions (analysis not shown). Treated sock assays The format for this assay is shown in Fig. 1B. Figure 4 shows the total number of ticks collected from each subject on the repellent- and solvent-treated socks in each treatment and control pair. No difference was found in the number of ticks collected from each subject (F = 1.10, d.f. = 3,29, P = 0.36). When comparing repellent-treated with untreated socks on the same human subject, the number of ticks was significantly lower on socks treated with BioUD than on those treated with water (F = 10.45, d.f. = 1,10, P = ). In addition, fewer ticks were found on DEET-treated than ethanol-treated control socks (F = 22.14, d.f. = 1,10, P = ). No significant difference in the number of ticks collected was found between the permethrin treatment and its ethanol control (F = 3.41, d.f. = 1,10, P = 0.095). Comparisons of differences between repellents in the number of ticks collected from treated and untreated socks revealed no differences between BioUD and DEET (F = 0.18, d.f. = 1,15.2, P = 0.68), BioUD and permethrin (F = 0.98, d.f. = 1,16.4, P = 0.34) or DEET and permethrin (F = 0.56, d.f. = 1,18.7, P = 0.47). All of the ticks collected from the socks were A. americanum and most were nymphs (90.6% nymphs, 9.3% adults; n = 615 in total); therefore, analysis of the results was limited to nymphal A. americanum. Because the number of adults collected per sock was minimal, the separate analysis of these results was not possible; however, combining the nymphal and adult data produced the same conclusions (analysis not shown). Olfactometer trials The format for this assay is shown in Fig. 2. The average proportion of adult A. americanum ticks in each of four bulbs in the control trials in which the filter paper was treated with water only did not differ significantly from 0.25 (χ 2 = 4.67, d.f. = 3, P = 0.20). In these studies, the average proportion of ticks in each bulb did not differ significantly (t = 1.95, d.f. = 21, P = 0.07). The mean number of ticks in each

7 Repellency of BioUD, DEET and permethrin against Amblyomma americanum 223 Ticks per sock, n DEET Fig. 4. Number of ticks collected from solvent-treated (control) and repellent-treated socks. Letters below the x-axis indicate specific human subjects. bulb was: bulb 1, 2.75 ± 0.53; bulb 2, 1.75 ± 0.56; bulb 3, 1.63 ± 0.60, and bulb 4, 1.38 ± The mean number of ticks that did not leave the central region of the olfactometer and were thus counted as non-responders was 0.5 ± When one of the bulbs contained filter paper treated with BioUD, the proportion of ticks in the three control (watertreated) bulbs did not differ significantly from 0.33 (χ 2 = 0.06, d.f. = 2, P = 0.97) and the average proportion of ticks was significantly lower in the bulb containing BioUD treated filter paper than in the bulbs containing water-treated filter paper (t = 2.76, d.f. = 21, P = 0.01), indicating spatial repellency. Over the eight replicates, the mean number of ticks in the bulb containing BioUD -treated filter paper was 0.13 ± 0.13 (only one of the 64 ticks tested entered this bulb), whereas the combined mean number of ticks in the remaining three bulbs that contained water-treated filter paper was 7.25 ± The mean number of ticks that did not move from the central body of the olfactometer (non-responders) was 0.63 ± Discussion Comparative tick repellency of BioUD, DEET and permethrin under field conditions The present experiments showed that in the assays using dry ice-baited sheets and those with human subjects, no significant differences were found in repellency between BioUD (with 7.75% 2-undecanone) and 98.1% DEET (Figs 3 and 4, respectively). Taken together, the results of these studies indicate that, when applied to cloth, the two products have equivalent tick repellency under field conditions. In the permethrin experiments, the two assay approaches produced different results. Significantly fewer ticks were collected from permethrintreated sheets compared with controls (Fig. 3), which suggests that the 0.5% permethrin treatment, like those of BioUD and DEET, was repellent. The mean number of ticks on the permethrin sheets, however, was higher than that on sheets treated with BioUD or DEET, which suggests that permethrin was less effective as a repellent for clothing treatments. By contrast, in the sock assay, no difference was found in numbers of ticks between the permethrin-treated and ethanol-treated socks, which suggests that permethrin was not a tick repellent. One problem with evaluating the repellency of permethrin in any assay method is its acute toxicity. The mechanism of toxicity of permethrin and, for that matter, of both natural and synthetic pyrethroids in arthropods and other animals is well established (Casida et al., 1983; Vijverberg & van den Bercken, 1990). Moreover, the weak repellency of permethrin to ticks has been reported in the literature. Lane & Anderson (1984) found that permethrin was initially repellent to Dermacentor occidentalis Marx, but that this repellency declined completely within 8 15 min. Similarly, these authors found that the repellency of permethrin against Ornithodoros coriaceus Koch (Acari: Argasidae) completely diminished within 4 8 min. In our field studies, we did not evaluate the time-course of intoxication, but only the number of ticks collected per cloth at one time-point. Certainly morbidity and mortality affect behaviour and repellency. Given the differences in the surface properties of the two different types of cloth used, the assay geometries (horizontal sheets vs. socks worn in both horizontal and vertical positions) and movement (static in the sheet assay vs. walking movement in the sock assay), as well as the likely tick intoxication during the course of the bioassay period, it is difficult to determine the relative impact of toxicity vs. assay method on the evaluation of the repellency of permethrin. In view of the known toxic mechanism of permethrin in insects, the compound mode of action in arthropod protection is likely to involve not only repellency, and this is also relevant to our bioassay methods. The endpoint of our assays, viewed on its own, appears to indicate that 0.5% permethrin is not consistently as effective as BioUD or DEET. Previous research from our laboratory showed that BioUD applied to filter paper provided significantly greater repellency against adult A. americanum than 98.1% DEET (Bissinger et al., 2009a). However, a second study (Bissinger et al., 2009b) showed that BioUD and 98.1% DEET provided equivalent repellency against adult A. americanum on cotton cheesecloth. In the same study, both BioUD and DEET were more repellent than a product containing 0.5% permethrin. Similar results were found in the present study in that no difference was observed in the number of ticks collected from sheets baited with dry ice and treated with either BioUD or 98.1% DEET, but significantly more ticks were collected from

8 224 B. W. Bissinger et al. sheets treated with 0.5% permethrin. Likewise, significantly fewer ticks were collected from socks treated with either BioUD or DEET compared with their respective controls. It is apparent from these findings that BioUD, at minimum, compares favourably with the current reference standards for cloth treatments for the repellency of ticks (i.e. 98.1% DEET and 0.5% permethrin). Mode of action of BioUD Most tick repellency bioassays do not differentiate between repellency caused by olfaction and that caused by tactile chemoreception (Carroll et al., 2005); the sheet and sock assays discussed earlier are no exception. However, there is some evidence that repellency of ticks in some chemicals is the result of olfaction. For example, Carroll et al. (2005), using treated human fingertips wrapped in two layers of organdie cloth, found that DEET exhibited spatial repellent properties at a short distance against nymphal A. americanum and blacklegged ticks, I. scapularis. Using a vertically oriented Y-tube, Dautel et al. (1999) made similar findings with reference to the sheep tick, I. ricinus. McMahon et al. (2003) showed that the repellent indalone presented in an air stream caused adult tropical bont ticks, A. variegatum, to move away from the repellent source. The spatial repellency of BioUD has not previously been examined. Using a four-choice olfactometer, we showed that BioUD applied to filter paper was repellent via olfaction to adult A. americanum for h after repellent application to filter paper. Although these studies clearly showed BioUD to be a spatial repellent under the conditions of the assay for A. americanum, we do not know the relative importance of tactile chemoreception vs. spatial repellency in ticks to the practical application of BioUD in host protection, or whether spatial repellency occurs for other life stages or for other tick species. Because of the design of the olfactometer, we were not able to test larvae or nymphs because they were able to escape from the device. Practical field-relevant assays to evaluate tick repellents Laboratory trials often serve as a starting point to evaluate repellent compounds. They can be conducted rapidly and at low cost and can be carried out at any time of year, irrespective of weather conditions. Most often, however, practical field validation using human volunteers of the effectiveness of a repellent is required, whereby the compound is applied to skin or clothing and the subjects are exposed to wild populations of ticks. Such tests are difficult to perform because they require multiple volunteers or testing days to achieve sufficient replication, human volunteers are at risk of acquiring ticktransmitted diseases and IRB approval is required. One impediment to the discovery of new tick repellents is the lack of standardized testing methods that demonstrate the true performance of the material as it is most likely to be used in real life. Another barrier refers to the absence of studies that correlate outcomes of laboratory assay approaches to those of field tests and are conducted by the same researchers using the same repellents. As Matthewson et al. (1981) and Dautel (2004) explained, laboratory tests in the absence of a host or host cues may result in the overestimation of repellency and incorrect conclusions about the practical value of a compound in host protection. In this regard, the present investigation allows us to compare results of field assay approaches with results of our previous laboratory studies of the repellent action to ticks of BioUD, DEETandpermethrinonfilter paper and cloth (Witting-Bissinger et al., 2008; Bissinger et al., 2009a, 2009b). The laboratory experiments involved twochoice studies between repellent-treated or untreated surfaces and head-to-head two-choice tests in which the choices were two different repellents. The studies were also conducted on different substrates [i.e. filter paper and cotton cloth (the same substrate used in each experiment)] and with three different tick species (A. americanum, D. variabilis, I. scapularis). Although there was some variability in responses between species and substrates tested in the laboratory, the more practical field sheet and sock tests mostly validated the results of the laboratory experiments. BioUD was an effective tick repellent and was essentially as effective or more effective than 98.1% DEET and 0.5% permethrin. These results suggest that the simple choice tests described by Bissinger et al. (2009a, 2009b) in the absence of host cues, and comparisons with other repellents, represent a reasonable and rapid alternative method for the evaluation of tick repellency under field conditions. Additional studies are needed to determine if this holds true for repellents other than those tested here. The advantage of the field sheet assay described in the current study is that trials can be conducted rapidly, do not place human volunteers at risk for tick bites and do not require approval by an IRB. Field bioassays using human volunteers wearing treated and untreated socks are advantageous in that they allow tests to be conducted using live hosts under conditions that represent the realistic use of the repellent on clothing and where the cloth barrier protects volunteers against direct contact with ticks. Ticks are also clearly visible on white socks and can therefore easily be removed and collected before they are able to reach skin. Trials using socks do not take into account the absorption of the repellent into the skin or the interaction of the repellent with chemicals present in the skin (Carroll et al., 2010) and therefore may not be completely representative of the efficacy of the repellent when directly applied to skin. In addition, socks may not be able to protect the subject from biting by larval ticks. In summary, previous studies from our laboratory have shown that BioUD exhibits a level of repellency against ticks that is at least similar to those of 98.1% DEET and 0.5% permethrin under laboratory conditions (Witting-Bissinger et al., 2008; Bissinger et al., 2009a, 2009b). The results of the current study demonstrate for the first time that BioUD under more realistic field conditions and in the presence of host cues and human subjects was as effective as 98.1% DEET and more effective than 0.5% permethrin against nymphal A. americanum. We also showed that the simple two-choice assays previously described by this laboratory for BioUD, DEET and permethrin produced similar conclusions about the

9 Repellency of BioUD, DEET and permethrin against Amblyomma americanum 225 relative repellencies of these compounds as the assays conducted under field conditions and can be used as a rapid and inexpensive screening tool for tick repellents. Furthermore, the two field bioassay techniques developed (i.e. the sheet and sock assays) provide new field testing approaches: the former can be conducted without any risk to human subjects, without need for IRB approval and in the presence of a host cue, whereas the latter permits the optimum realistic use of a tick repellent on clothing, which minimizes the risk of transmission of disease agents to the human subject. We have also provided the first evidence that BioUD includes spatial repellency as one component of its activity. Acknowledgements The authors are grateful to Luma Abu Ayyash and Nicholas Kimps for technical assistance. References Aquino, M., Fyfe, M., MacDougall, L. & Remple, V. (2004) West Nile virus in British Columbia. Emerging Infectious Diseases, 10, Bissinger, B.W. & Roe, R.M. (2010) Tick repellents: past, present, and future. Pesticide Biochemistry and Physiology, 96, Bissinger, B.W., Apperson, C.S., Sonenshine, D.E., Watson, D.W. & Roe, R.M. (2009a) Efficacy of the new repellent BioUD against three species of ixodid ticks. Experimental and Applied Acarology, 48, Bissinger, B.W., Zhu, J., Apperson, C.S., Sonenshine, D.E., Watson, D.W. & Roe, R.M. (2009b) Comparative efficacy of BioUD to other commercially available arthropod repellents against the ticks, Amblyomma americanum and Dermacentor variabilis on cotton cloth. American Journal of Tropical Medicine and Hygiene, 81, Brennan, J.M. (1947) Preliminary report on some organic materials as tick repellents and toxic agents. Public Health Reports, 62, Carroll, J.F., Klun, J.A. & Debboun, M. (2005) Repellency of deet and SS220 applied to skin involves olfactory sensing by two species of ticks. Medical and Veterinary Entomology, 19, Carroll, J.F., Paluch, G., Coats, J. & Kramer, M. (2010) Elemol and amyris oil repel the ticks Ixodes scapularis and Amblyomma americanum (Acari: Ixodidae) in laboratory bioassays. Experimental and Applied Acarology, 51, Casida, J.E., Gammon, D.W., Glickman, A.H. & Lawrence, L.J. (1983) Mechanisms of selected action of pyrethroid insecticides. Annual Review of Pharmacology and Toxicology, 23, Childs, J.E. & Paddock, C.D. (2003) The ascendancy of Amblyomma americanum as a vector of pathogens affecting humans in the United States. Annual Review of Entomology, 48, Dautel, H. (2004) Test systems for tick repellents. International Journal of Medical Microbiology, 293 (Suppl.), Dautel, H., Kahl, O., Siems, K., Oppenrieder, M., Müller-Kuhrt, L. & Hilker, M. (1999) A novel test system for detection of tick repellents. Entomologia Experimentalis et Applicata, 91, Farrar, R.R. & Kennedy, G.G. (1987) 2-Undecanone, a constituent of the glandular trichomes of Lycopersicon hirsutum f. glabratum: effects on Heliothis zea and Manduca sexta growth and survival. Entomologia Experimentalis et Applicata, 43, Fradin, M.S. & Day, J.F. (2002) Comparative efficacy of insect repellents against mosquito bites. New England Journal of Medicine, 347, Frances, S.P. & Debboun, M. (2007) User acceptability: public perceptions of insect repellents. Insect Repellents: Principles, Methods, and Uses (ed. by M. Debboun, S. Frances & D. Strickman), pp CRC Press, Boca Raton, FL. Garboui, S.S., Borg-Karlson, A.-K. & Pållson, K. (2009) Tick repellent properties of three Libyan plants. Journal of Medical Entomology, 46, Garcia, R. (1962) Carbon dioxide as an attractant for certain ticks (Acarina: Argasidae and Ixodidae). Annals of the Entomological Society of America, 55, Gerberg, E.J. & Novak, R.J. (2007) Considerations on the use of botanically derived repellent products. Insect Repellents: Principles, Methods, and Uses (ed. by M. Debboun, S. Frances & D. Strickman), pp CRC Press, Boca Raton, FL. Gouck, H.K., McGovern, T.P. & Beroza, M. (1967) Chemicals tested as space repellents against yellow-fever mosquitoes, I. Esters. Journal of Economic Entomology, 60, Jaenson, T.G.T., Garboui, S. & Pålsson, K. (2006) Repellency of oils of lemon eucalyptus, geranium, and lavender and the mosquito repellent MyggA Natural to Ixodes ricinus (Acari: Ixodidae) in the laboratory and field. Journal of Medical Entomology, 43, Keirans, J.E. & Durden, L.A. (1998) Illustrated key to nymphs of the tick genus Amblyomma (Acari: Ixodidae) found in the United States. Journal of Medical Entomology, 35, Koren, G., Matsui, D. & Bailey, B. (2003) DEET-based insect repellents: safety implications for children and lactating women. Canadian Medical Association, 169, Lane, R.S. & Anderson, J.R. (1984) Efficacy of permethrin as a repellent and toxicant for personal protection against the Pacific Coast tick and the pajaroello tick (Acari: Ixodidae and Argasidae). Journal of Medical Entomology, 21, Matthewson, M.D., Hughes, G., Macpherson, I.S. & Bernard, C.P. (1981) Screening techniques for the evaluation of chemicals with activity as tick repellents. Pesticide Science, 12, McMahon, C., Kröber, T. & Guerin, P.M. (2003) In vitro assays for repellents and deterrents for ticks: differing effects of products when tested with attractant or arrestment stimuli. Medical and Veterinary Entomology, 17, Mkolo, M.N. & Magano, M.R. (2007) Repellent effects of the essential oil of Lavendula angustifolia against adults of Hyalomma marginatum rufipes. Journal of the South African Veterinary Association, 78, Moore, S.J., Lenglet, A. & Hill, N. (2007) Plant-based insect repellents. Insect Repellents: Principles, Methods, and Uses (ed. by M. Debboun, S. Frances & D. Strickman), pp CRC Press, Boca Raton, FL. Norval, R.A.I., Yunker, C.E. & Butler, J.F. (1987) Field sampling of unfed adults of Amblyomma hebraeum Koch. Experimental and Applied Acarology, 22, Osimitz, T.G. & Grothaus, R.H. (1995) The present safety assessment of deet. Journal of the American Mosquito Control Association, 11, SAS Institute, Inc. (2003) User s Guide for SAS Software Navigator. SAS Institute, Cary, NC. Smith, C.N. & Gouck, H.K. (1946) Observations on tick repellents. Journal of Economic Entomology, 39,

10 226 B. W. Bissinger et al. Sonenshine, D.E. (1993) Biology of Ticks, Vol. 2. Oxford University Press, Oxford. Sonenshine, D.E., Lane, R.S. & Nicholson, W.L. (2002) Ticks (Ixodida). Medical and Veterinary Entomology (ed. by G. Mullen & L. Durden), pp Academic Press, San Diego, CA. Sudaken, D.L. & Trevathan, W.R. (2003) DEET: a review and update of safety and risk in the general population. Clinical Toxicology, 41, Thorsell, W., Mikiver, A. & Tunón, H. (2006) Repelling properties of some plant materials on the tick Ixodes ricinus L. Phytomedicine, 13, Tunón, H., Thorsell, W., Mikiver, A. & Malander, I. (2006) Arthropod repellency, especially tick (Ixodes ricinus), exerted by extract from Artemesia abrotanum and essential oil from flowers of Dianthus caryophyllum. Fitoterapia, 77, U.S. Centers for Disease Control and Prevention (2008) Stop ticks. [Accessed 2 July 2009]. Vijverberg, H.P.M. & van den Bercken, J. (1990) Neurotoxicological effects and the mode of action of pyrethroid insecticides. Critical Reviews in Toxicology, 21, Wilson, J.G., Kinzer, D.R., Sauer, J.R. & Hair, J.A. (1972) Chemoattraction in the lone star tick (Acarina: Ixodidae): I. Response of different developmental stages to carbon dioxide administered via traps. Journal of Medical Entomology, 9, Witting-Bissinger, B.E., Stumpf, C.F., Donohue, K.V., Apperson, C.S. & Roe, R.M. (2008) Novel arthropod repellent, BioUD, is an efficacious alternative to deet. Journal of Medical Entomology, 45, Accepted 14 September 2010 First published online 1 December 2010