Nymphs of Ixodes ricinus Are More Sensitive to Deet Than Adult Females
The use of repellents is a unique measure of personal protection, which can avoid tick attachment and thus reduce the risk of tick-borne infections. In the European Union, the efficacy of the repellents coming onto the market has to be evaluated according to the guidelines published by the European Chemical Agency before registration. The United States Environmental Protection Agency has a similar role. Despite obvious differences in morphology and behavior, both these guidelines allow the use of nymph or adult female ticks for laboratory testing. Here, we provide evidence that sensitivity to diethyltoluamide (deet) (P < 0.0001) of Ixodes ricinus nymphs within the in vitro trial was significantly higher than in adult females. In the experiment, we also observed that feral ticks were less sensitive to repellent than were laboratory-reared ticks (P < 0.01) and that mobility decreased when the trial was repeated (P < 0.05). This study showed that the results of efficacy tests could vary significantly even when the protocol was conducted in accordance with the recommended methods. To refine the results of efficacy tests, we recommend a reevaluation of the guidelines, with emphasis on the developmental stage and origin of ticks.ABSTRACT
INTRODUCTION
Ticks are hematophagous arthropods and ectoparasites of mammals, birds, and reptiles (ECDC 2014). Currently, ticks are known to be vectors of a large variety of pathogens of both medical and veterinary importance. Due to climate change, ticks have recently been colonizing new areas (Jaenson et al. 2012, Ragagli et al. 2016, Tarageľová et al. 2016) and so the incidence of tick-borne diseases has increased in such areas. Beside using protective clothing and tucking trousers into socks, the use of repellent represents the only possible protection against tick attachment (Slunge and Boman 2018).
Repellent is defined as a substance of synthetic or natural origin that forces ticks to move away from a repellent's source (Dethier et al. 1960). Currently, various repellent products containing dozens of effective substances in different concentrations are available worldwide. The leading national public health institute of the USA, the Centers for Disease Control and Prevention recommends the use of insect repellents containing diethyltoluamide (deet), picaridin, IR3535, oil of lemon eucalyptus, para-methane-3,8-diol, or 2-undecanone (CDC 2018). Laboratory trials are a crucial tool to identify repellent effectiveness. Dautel (2004) classified testing of tick repellency into 3 categories: in vitro bioassays with absence of host stimuli, in vitro bioassays including host stimuli, and in vivo bioassays using living hosts. Generally, the performance of in vitro test could be a very useful tool to compare the repellency of different substances or in the development of novel repellent products (Adenubi et al. 2018); however, only the in vivo test could inform the customers about the duration of the protection against ticks. On the other hand, the results of the in vivo and in vitro with host stimuli tests are known to correlate with each other (Dautel 2004; Kulma and Bubová, unpublished data).
The tick repellents for humans marketed in the European Union (EU) need to be approved by the authorities of the member states according to the Biocidal Products Regulation, Regulation (EU) 528/2012. To provide information for the evaluation of repellent efficacy, the European Chemical Agency (ECHA) periodically publishes the Guidelines on Biocidal Products Regulation, which serves as a standard for most of the European laboratories working on testing of such products. Any deviation from the guidelines will result in the product being not registered for use within EU. The United States Environmental Protection Agency (USEPA) plays a similar role in the USA. To evaluate the efficacy against ticks, guidelines of both agencies currently recommend the use of either nymphs or adult females, when each specimen could be used only once and must be specific pathogen free (USEPA 2010, ECHA 2018). However, the behavior of nymphs and adult females is known to be different (Mejlon and Jaenson 1997, Vassallo and Perez-Eid 2002).
For this reason, the objective of this study was to carry out an in vitro bioassay to reveal the difference in tolerance of nymphs and adult ticks to repellents as well as their mobility (distance covered when seeking for the host). Moreover, we tested the effects of repetition and origin of ticks. The results should contribute to the refinement and precision of the methods of repellency tests against ticks. As a model organism, we investigated this phenomenon using the sheep tick, Ixodes ricinus (L.), which is the most important vector for tick-borne pathogens in Europe.
MATERIALS AND METHODS
Ticks
The ticks were either obtained from a commercial laboratory (Insect Services, Berlin, Germany) or captured in the field using a 1-m2 white flag. The feral ticks were collected in June 2017 from vegetation at 4 sites (2 urban areas and 2 natural habitats) in the region of Bohemia, Czech Republic. Krejcárek (250–290 m above sea level [masl]; 50.095°N, 14.477°E) and Prokopské údolí (260–280 masl; 50.039°N, 14.366°E), are urban public forest parks in Prague that are frequently used for leisure activities and dog walking. Local fauna includes mainly small mammals and birds, and vegetation consists of deciduous and coniferous trees and irregularly mowed grass. Nature reserve Dománovický les (220–230 masl; 50.114°N, 15.348°E) is a mixed forest with a majority of oaks and few meadows with no apparent management that is rarely visited by people. The fauna is varied, comprised of amphibians, reptiles, birds, and mammals including squirrels, hares, deer, and wild boars. The last site is forest edge near the area of a Summer Children's Camp in Ostrovec, South Bohemia (420 masl; 49.403°N, 14.118°E), surrounded by reeds, meadow, and coniferous forest with abundant nesting birds, reptiles, and small and large mammals.
In total, 175 I. ricinus ticks (99 nymphs and 76 adult females) were used for the test (18 females and 12 nymphs for each locality apart from Prokopské údolí, where we collected 4 females and 51 nymphs). Prior to testing, all the ticks were kept unfed in polypropylene tubes (with strips of filter paper inside) enclosed by fine insect mesh and stored in a desiccator outfitted with a glass cup with soaked cotton wool (to keep humidity >80%) at 26°C for 72 h.
Bioassay
To determine the movement capacity (distance covered by ticks when seeking for the host) and ability to overcome different deet concentrations, a modified in vitro test described by Belova et al. (2012) and Kulma et al. (2017) was used (see Fig. 1). The experimental bioassays consisted of 5 concentric circles (diam 60–180 mm) drawn on an A4-size sheet of medium-flow filter paper (Whatman™ GE Healthcare, Buckinghampshire, United Kingdom). In the experimental arena, 15-mm-wide zones between these circles were treated by a flat paintbrush with an ethanolic solution of 0.1%, 0.5%, 1.0%, or 5.0% deet (Fig. 1). The central circle (diam 60 mm) remained untreated. After evaporation of the solvent (15 min), the test tick was individually placed into the center of the concentric circles and its movement was monitored for 2 min. In parallel, the observed trajectory of the tick was drawn by hand onto another identical A4 sheet. The obtained trajectories were then measured by a curvimeter (Recta, Zurich, Switzerland). The tick had not been repelled if it overcame the outer boundary line of the area treated by a certain deet concentration. At the beginning of each trial the observer's breath was used as the host stimuli, which activated ticks to the host-seeking mode. In all the experiments, the same persons were used as the bait. We performed 3 separate tests (2 repetitions after 60 and 120 min) for each individual tick. Between the repetitions, ticks were kept individually in 5-ml Eppendorf tubes (unique code was assigned to each tested tick and marked on the tube by permanent marker), each closed by insect mesh and put into the desiccator as described above. All the bioassays were conducted in June 2017, in the National Reference Laboratory for Vector Control in the National Institute of Public Health in Prague (NIPH) under standard laboratory conditions (RH 50.4 ± 2.2%, temperature 24.6 ± 0.7°C).
Citation: Journal of the American Mosquito Control Association 35, 4; 10.2987/19-6849.1
After testing, all the ticks were put into microcentrifuge tubes, killed by freezing (−20°C), and sent to the National Reference Laboratory for Lyme Borreliosis, NIPH in order to determine the presence of Borrelia pathogens and thus exclude their potential effect on the performance of any feral ticks.
Statistical evaluation
The mobility and ability to overcome different deet concentrations were tested separately. In both analyses we used the developmental stage of tick (nymph/female), borreliosis infection (yes/no), locality (laboratory/Ostrovec/Dománovický les/Krejcárek/Prokopské údolí), and order of trial (1st/2nd/3rd) as the factors. For these calculations, all the tested ticks (laboratory-reared and feral) were pooled altogether.
Mobility was tested by Generalized Linear Mixed Model (GLMM) (lme function) using R studio software (RStudio, Boston, MA). First, we tested the random component of the models by Akaike information criterion (AIC): model without random component (AIC = −155.66), model with tick individual as only random variable (AIC = −197.62), and model with order of trial and tick individual (AIC = −208.99). The models that best fit to our data have order of trial and tick individual as the random section; therefore, subsequent model simplification was focused on fixed factors only. Due to the limited sample size and number of factors, only 2-way interactions of explanatory variables were tested during modeling. After the process of model simplification, we controlled for autocorrelation of particular individual measurements and also for heteroscedasticity in the random components of the model (corAR1 function). The model with an autocorrelation structure was better (AIC = −185.49) than the model without an autocorrelation structure (AIC = −124.15). We log-transformed (log) the data of the mobility because they did not have a normal distribution. Some ticks (3.66%) were able to get out of the testing arena before the 2 min expired. These ticks have a higher probability of lower mobility than average; for these cases we used the original mobility and recomputed it by the rule of three sum. Because the explained data from the test of overcoming of different deet concentrations have a categorical character, Cumulative Link Mixed Model (CLMM) (clmm2 function) was used for resolution of this task. First, we tested full models with (AIC = 1,248.59) and without random (AIC = 1,274.00) factor of tick individual, and thus in all the models the tick was set as the random effect. The data for ticks that died before they completed all 3 trials (1.14%) have been excluded from the final sample.
RESULTS
Mobility proved to be influenced by the developmental stage (GLMM: F = 189.82, P < 0.0001) and the order of trial (GLMM: F = 4.79, P < 0.05). Females covered longer distance than nymphs. The lowest value of mobility was in the 3rd trial (2nd repetition) regardless of the developmental stage, as expected (Fig. 2). The values were more similar between repeats for 1 individual (Phi = 0.48) and heteroscedasticity increased with trial order, when about a 20% higher value was determined for nymphs (Fig. 2).
Citation: Journal of the American Mosquito Control Association 35, 4; 10.2987/19-6849.1
Tolerance to deet was affected by developmental stage, where nymphs were less successful in overcoming the higher concentrations of deet than adult females (CLMM: estimate = −3.13, z-value = −9.28, P < 0.0001) (Fig. 3). The origin of ticks also influenced their ability to overcome deet, whereas laboratory-bred ticks had significantly lowest tolerance to deet (CLMM: estimate = −1.19, z-value = −2.81, P < 0.01). On the other hand, no statistical differences were found within the feral ticks from other localities.
Citation: Journal of the American Mosquito Control Association 35, 4; 10.2987/19-6849.1
DISCUSSION
The guidelines for testing biocidal products published by the ECHA and USEPA recommend both nymphs and adult females as models for the conduction of efficacy tests against ticks, although their behavior is obviously not the same. First of all, desiccation tolerance generally increases during development and causes differences in host-seeking behavior, whereas adults can survive in the higher parts of vegetation than nymphs (Mejlon and Jaenson 1997) and their peak level of activity also varies (Daniel et al. 2015). Therefore, nymphs are able to obtain blood meals on a large scale from animals, including reptiles, and medium-size or large mammals and birds, whereas the adult females feed almost exclusively on large mammals (Talleklint and Jaenson 1994, Gern et al. 1998, Grech-Angelini et al. 2016). In the light of this, we hypothesized that the life stage could also be an important factor influencing tick mobility and ability to overcome the repellent barrier during testing.
As expected, both nymphs and females proved to be sensitive to higher concentrations of deet as was described by Büchel et al. (2015) and Semmler et al. (2011). Therefore, the highest concentration of 5.0% used in the experiment stopped a majority of the tested ticks regardless of their stage. Nymphs were reported to be more attracted to humans than adult females (Vassallo and Perez-Eid 2002); therefore, we expected their higher motivation to attach to the observer may also cause an increase in their tolerance to the repellent. The experiment provided evidence that the developmental stage significantly influenced the ability of ticks to overcome repellent concentrations, but unexpectedly, tested adult females were able to overcome significantly higher deet concentrations than nymphs and they thus proved to be significantly more sensitive to the repellent. Moreover, only females overcame the highest concentration of deet in our experiment—5%.
As Ixodes species belong among the questing ticks, which are waiting for potential hosts on the vegetation, each tick has to make the all-or-none decision to stay away or to cling to a host within a very short time period (Dautel et al. 2013). Therefore, I. ricinus species should be highly motivated to be active when such a host approaches. In this study, both stages of the tested ticks responded well to the host stimuli and were trying to overcome the treated area to attach to the host during the whole duration of the trial, but females covered significantly longer distances than nymphs (Fig. 2). The differences found in mobility could be partly explained by the differences in behavior as well as in morphology of both stages. Nymphs are less robust, they measure 1.2 × 1.5 mm and weigh approximately 4 mg, while the average weight of engorged females is 18 mg and size 2.0 × 4.0 mm (Lees and Quastel 1946, Chrdle et al. 2016). However, both ECHA and USEPA guidelines consider a tick as nonrepelled if it crosses the boundary line more than 3 cm above the wrist (USEPA 2010, ECHA 2018). If we take into account that the distance covered by nymphs was found to be almost 2 times lower in comparison with the females, those 3 cm undoubtedly represent a more serious obstacle to overcome for nymphs and such a phenomenon may thus cause distortions in the evaluation of efficacy test results.
According to USEPA and ECHA guidelines, pathogen-free ticks must be used for the experiments. Therefore, testing on feral ticks from areas with no endemic pathogen occurrence is also possible for the purpose of repellency evaluation (USEPA 2010, ECHA 2018). As there is probably no such area in the Czech Republic, we also investigated the presence of Borrelia burgdorferi Johnson, the most common local tick-borne pathogen with prevalence up to 50% (Kybicová et al. 2009), in collected ticks to exclude their potential influence on tick performance. To our knowledge, the effect of other pathogens on tick behavior is not known. However, their prevalence within the Czech Republic is very low and therefore we expect that the robustness of our data set overcomes any possible bias caused by other rare pathogens. The tolerance of ticks to repellent is reported to be affected by the presence of encephalitis virus, which could increase their motivation to attach to the host, and so overcome the repellent barrier (Belova et al. 2012). We do not expect any influence on results by possible presence of encephalitis virus due to its low prevalence of <1% (Hönig et al. 2015). In this study, the presence of the borreliosis pathogen varied from 10% to 40% and did not affect mobility as well as the ability of ticks to overcome deet. From a methodical point of view, this conclusion cannot be taken as definitive as we did not manipulate the presence or absence of pathogens in ticks, and further research on this topic is needed.
Apart from pathogen presence, many other factors can cause deviations in behavior between feral and laboratory-reared arthropods. The difference in behavioral responses to repellents between feral and laboratory-reared populations is well known for mosquitoes, whose laboratory-reared populations become more sensitive to such substances (Frances et al. 1993, Walker et al. 1996). Therefore, we focused on this phenomenon in I. ricinus. According to the results obtained, laboratory-reared ticks were significantly more sensitive to deet than feral ticks from all the localities. On the other hand, the behavior of feral ticks did not vary by locality (urban parks or forests). Since we cannot exclude the use of pesticides and repellents surrounding the sites where we collected the ticks, the laboratory-reared ticks could have been less resistant to such products due to their isolated development in the artificially controlled environment. The increase in sensitivity to insecticides as well as repellents in laboratory-reared arthropods is documented (Becker 1970, Strong et al. 1997). A sufficient amount of resources provided to the laboratory-reared ticks can contribute to a loss of host attachment motivation, and so, it could also partly explain the high sensitivity of laboratory-reared ticks to the repellent.
Finally, the guidelines allow each tick to be used only once. For both investigated variables (mobility and tolerance to deet), an integral part of best fitting models was the random factor of tick individuals. This means that repeated measurements of the same individual will be more similar than measurements among different individuals. This situation can be explained by the existence of a tick personality, which was not described for this species yet; however, it has been proved for another invertebrate species (Kralj-Fišer and Schuett 2014), or even more simply by individual constraints ensuring consistency of behavior (e.g., different levels of energy reserves and body size, or different tolerance to starvation or injury, etc.). Surprisingly, the repetition did not have a significant effect on the sensitivity of ticks to deet. On the other hand, repeatability of individual performance expressed by the mobility in this study decreased over time, which suggests that the effect of exhaustion varies in its influence among individual ticks. A single use of each tick seems to be justified.
Based on the data presented, it is obvious that the current recommendation published by USEPA and ECHA in the guidelines for testing of biocidal products could cause imbalances in the results of repellent efficacy tests against ticks. In this study, adult females of I. ricinus overcame the repellent barrier more easily and also covered significantly longer distances. From this point of view, the repellent tested solely against nymphs would seem more effective than in the same test using adult females. Additionally, laboratory-reared ticks proved to be more sensitive to deet than those collected in the field, which can also bend the results of an efficacy test. On the other hand, laboratory tests expected to inform the customers about an approximate duration of the effect and they will thus never fully simulate real use in the field. Therefore, the guidelines should mainly lead laboratories to reach the most unified and comparable results possible. In light of this, we recommend reconsidering further discussion or verification of the guidelines, with emphasis on the developmental stage and origin of ticks.
Bioassay used for evaluation of the tolerance of sheep ticks (Ixodes ricinus) to deet and mobility.
Mobility (2-min duration test) of the individual ticks in repeated trials (open squares represent mean for females, full squares represent mean for nymphs, box represents 25–75% quartiles and brackets min. − max. values). The increased bracket range with a smaller box area in the 3rd trial points to heteroscedasticity.
Proportion of all the tested ticks according to developmental stage that successfully overcame the different deet concentrations.