Insecticidal and repellent activities of pyrethroids to the three major pyrethroid-resistant malaria vectors in western Kenya

Background The dramatic success of insecticide treated nets (ITNs) and long-lasting insecticidal nets (LLINs) in African countries has been countered by the rapid development of pyrethroid resistance in vector mosquitoes over the past decade. One advantage of the use of pyrethroids in ITNs is their excito-repellency. Use of the excito-repellency of pyrethroids might be biorational, since such repellency will not induce or delay the development of any physiological resistance. However, little is known about the relationship between the mode of insecticide resistance and excito-repellency in pyrethroid-resistant mosquitoes. Methods Differences in the reactions of 3 major malaria vectors in western Kenya to pyrethroids were compared in laboratory tests. Adult susceptibility tests were performed using World Health Organization (WHO) test tube kits for F1 progenies of field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s., and laboratory colonies of An. gambiae s.s. and An. arabiensis. The contact repellency to pyrethroids or permethrin-impregnated LLINs (Olyset® Nets) was evaluated with a simple choice test modified by WHO test tubes and with the test modified by the WHO cone bioassay test. Results Field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s. showed high resistance to both permethrin and deltamethrin. The allelic frequency of the point mutation in the voltage-gated sodium channel (L1014S) in An. gambiae s.s. was 99.3–100%, while no point mutations were detected in the other 2 species. The frequency of takeoffs from the pyrethroid-treated surface and the flying times without contacting the surface increased significantly in pyrethroid-susceptible An. gambiae s.s. and An. arabiensis colonies and wild An. arabiensis and An. funestus s.s. colonies, while there was no significant increase in the frequency of takeoffs or flying time in the An. gambiae s.s. wild colony. Conclusion A different repellent reaction was observed in the field-collected An. gambiae s.s. than in An. arabiensis and An. funestus s.s. It might be that resistant mosquitoes governed by knockdown resistance (kdr) loose repellency to pyrethroids, whereas those lacking kdr maintain high repellency irrespective of their possessing metabolic resistance factors to pyrethroids. Further genetic evaluation is required for the demonstration of the above hypothesis.


Background
Pyrethroids are the predominant insecticides and are used in various formulations for mosquito control. Globally, pyrethroids comprise 40% of the insecticides used annually for indoor residual spraying against malaria vectors [1]. Since the World Health Organization (WHO) adopted the use of long-lasting insecticidal nets (LLINs) as a principal strategy for effective malaria control in the Roll Back Malaria Partnership [2], pyrethroids have been the only class of insecticides used for LLINs [1,3]. Pyrethroids have unique modes of action such as fast knockdown and excito-repellent effects [4].
A dramatic success of insecticide treated nets (ITNs) and LLINs was recorded in African countries. However, it has been countered by the rapid development of pyrethroid resistance in vector mosquitoes over the past decade [5]. Pyrethroid resistance has developed in the African malaria vectors primarily through 2 resistance mechanisms. The first is resistance at the target site, in which only 1 point mutation at 1014 L (L1014F or L1014S) in the voltage-gated sodium channel causes insensitivity to pyrethroids, resulting in knockdown resistance (kdr). The second is metabolic resistance that relates to the elevated activity of 1 or more detoxification enzymes (cytochrome P450s, etc.). The distribution patterns of the types of point mutations in voltage-gated sodium channels in African malaria vectors are unique [5][6][7]. The point mutations in the voltage-gated sodium channel are commonly reported in Anopheles gambiae Giles (An. gambiae sensu stricto [s.s.]). L1014F mutations are widespread in western African countries, L1014S mutations are distributed in eastern Africa, and L1014F/ L1014S hybrids in An. gambiae s.s. are reported in the central region. In contrast, such kdr mutations do not seem to be common in An. arabiensis Patton [8][9][10], but the metabolic resistance seems to be most common in this species [10,11]. Fortunately, kdr mutations have not been reported in African malaria vectors except for these 2 species. Some metabolic resistance, however, has been reported in the An. funestus Giles group [11][12][13][14]. Additionally, the co-occurrence of these 2 resistance factors (metabolic and kdr) in a single mosquito population would lead to further threats to malaria control, since both factors might reinforce the resistance of each other. Darriet et al. reported that, even in areas where An. gambiae s.s. was resistant to permethrin and deltamethrin (>90% kdr), bed nets treated with these insecticides remained effective [15]. This apparent paradox was explained by behavioral changes in the resistant mosquitoes; they were less repelled by the insecticide, remained on the pyrethroid-treated material for longer periods, and thus received a higher dose of insecticide [16]. On the other hand, excito-repellency is thought to be an advantage of the use of pyrethroids, which provide personal protection from mosquito bites even when ITNs accumulate holes during the course of daily usage [17]. Use of the excito-repellency of pyrethroids might be biorational, since such repellency will not induce or delay the development of any physiological resistance since it does not kill the affected insects or reduce the chance of contact to the insecticides and causes low selection pressure on insect populations [18,19]. However, little is known about the relationship between the mode of insecticide resistance and excito-repellency in pyrethroidresistant mosquitoes.
The aim of this study was to investigate the difference in the repellent reaction of pyrethroid resistant vector mosquitoes to pyrethroids and pyrethroid-impregnated nets.

Mosquito collection site
The mosquito collection sites were located in the Mbita and Suba districts of Nyanza Province in western Kenya. The rainfall pattern in the area is bimodal, with a long rainy season occurring from March to May and a short rainy season occurring in November and December. Malaria infection rates rise steadily between September and February and peak briefly in June, following the long rains [20]. The Mbita and Suba districts have been identified as high vector transmission areas in Kenya, and more than 50% of the population is exposed to malaria at a rate of ≥40% PfPR 2-0 (Plasmodium falciparum parasite rate corrected to a standard age range of 2 to <10 y) [21].
Adult susceptibility tests with insecticides were performed using WHO test tube kits for F1 progenies of field-collected An. gambiae s.s., An. arabiensis, An. funestus s.s., and the laboratory colonies of An. gambiae s.s. and An. arabiensis reared at the International Center of Insect Physiology and Ecology (ICIPE, Mbita, Nyanza, Kenya) according to WHO instructions (WHO/CDS/CPC/MAL/ 98.12). Papers impregnated with 0.75% permethrin, 0.05% deltamethrin, 0.1% propoxur, or 1.0% fenitrothion were used for the tests. Ten 1-to 3-day-old unfed female mosquitoes were released into WHO test tubes for exposure to an insecticide-impregnated paper for 1 hour, and the time to knockdown was recorded. Insects were then transferred to a clean tube and fed via cotton soaked with a 5% glucose solution, and mortality was recorded after 1 day. The time required for 50% knockdown (KT 50 ) was obtained, and average mortality was calculated. Two to 4 replications were performed for each insecticide.

Contact repellency test using WHO test tubes
The contact repellency of F1 progenies of field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s. was evaluated with a simple choice test modified by Grieco et al. [22] using WHO test tubes. Papers impregnated with 0.75% permethrin and 0.05% deltamethrin were used. Ten 1-to 3-day-old female mosquitoes were transferred into a WHO test tube lined with untreated paper that was connected to another test tube lined with an insecticide paper. Just after releasing the shutter, mosquitoes were transferred into a tube with an insecticide paper by blowing. The test tube was placed horizontally in a dark container to avoid the influence of light for 10 min with a shutter kept open. The shutter was closed after 10 min, and the number of mosquitoes in the tube with untreated paper (number repelled) was counted. For every test a control assay was performed in which an untreated paper was used in place of the insecticide-treated paper. Three to 7 replicates were performed with each insecticide.

Contact repellency test modified by the WHO cone bioassay test
The contact repellency of adult mosquitoes was evaluated by the test modified by the WHO cone bioassay test of field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s., and the laboratory colonies of An. gambiae s.s. and An. arabiensis reared at the ICIPE were used for the test. Each 1-to 3-day-old female mosquito was exposed to Olyset® Net materials (25 × 25 cm) or a paper impregnated with 0.75% permethrin for a WHO test tube assay for 3 minutes under standard WHO cones. A control assay using an insecticide-untreated filter paper was performed concurrently using another female mosquito of the same colony. After releasing a mosquito, the number of takeoffs from the net surface and the cumulative flying time after taking off to the next touchdown over 3 minutes was recorded. A total of 21, 22, and 5 replicates were performed for An. gambiae s.s., An. arabiensis, and An. funestus s.s., respectively. Because of the short time contact (3 minutes), neither mortality nor knockdown was observed during the test.

Species identification
Adult mosquitoes were examined microscopically to distinguish An. gambiae s.l. and An. funestus s.l. from other anophelines based on identification keys developed by Gillies and Coetzee [23]. Multiplex PCR methods described by Scott et al. [24] and Koekemoer et al. [25] were used for species identification.
Detection of point mutations in the voltage-gated sodium channel PCR and direct DNA sequencing were used to identify point mutations at 1014 L in the field-collected mosquitoes according to the method of Kawada et al. [10,11]. The legs of an adult were homogenized in a mixed solution of extraction solution (20 μl) + tissue preparation solution (5 μl) (REDExtract-N-Amp™ Tissue PCR Kit; SIGMA, St. Louis, MO, USA) for extraction of DNA. Initial fragment amplification was carried out using primers AGKF1 (CATGATCTGCCAAGATGGAA) and AGKR1 (GTTGGTGCAGACAAGGATGA) for An. gambiae s.l.; and AFF1 (ACCAAGATCTGCCAAGATGG) and AFR1 (TGGTGCAGACAAGGATGAAG) for An. funestus s.s., respectively. The PCR mixture contained 4 μl of REDE xtract-N-Amp™ ReadyMix (SIGMA), 0.5 μM of each primer, and 1 μl of the DNA template in a total volume of 10 μl. PCR was performed under the following conditions: 94°C for 3 min and 35

Data analysis
A digital map in shape file format (Kenya-Boundaries, FAO Africover, http://www.africover.org/index.htm) was used for mapping the collection sites. The geographical positions of the collection sites were plotted on the map using ArcGIS 10.1 (ESRI Japan Corp, Tokyo, Japan).
The median knockdown times (KT 50 s) in the insect susceptibility test were calculated using the Bliss' probit method [26]. The mosquito repellency in the simplified contact repellency test using WHO test tubes was calculated as the mean percentage of female mosquitoes repelled into untreated tubes corrected by a control test. The square root of the percent of repellency in each test was arcsin converted, and analysis of variance (ANOVA) and the multiple comparison of the repellency by a Tukey honestly significant difference test were performed using R × 64 Ver. 2.15.1 (http://www.R-project. org). The comparison of the repellency of the Olyset® Net, permethrin-impregnated paper, and untreated paper in the contact repellency test modified by the WHO cone test was performed with a Kruskal-Wallis test using JMP 7.0 J (SAS Institute Japan Inc., Tokyo, Japan).

Ethics statement
The protocol for the study (case no. 1775) was approved by the Scientific Steering Committee and the National Ethics Review Committee of the Kenya Medical Research Institute. All necessary permits were obtained for the described field studies. No mosquito collection was done without the approval of the head of the village and the owner and occupants of the collection house.
No mortality was observed in control groups in the susceptibility tests. F1 adults of the field-collected An. gambiae s.s., An. arabiensis, and An. funestus s.s. showed high resistance to both permethrin (3.4%, 63.6%, and 36.8% mortality, respectively) and deltamethrin (34.4%, 82.1%, and 36.4% mortality, respectively). The KT 50 s of the 3 species for permethrin and deltamethrin were >60 minutes except for deltamethrin in An. arabiensis (KT 50 , 29.5 minutes), indicating low knockdown activities of the 2 pyrethroids against these mosquitoes as well as low killing activities. In contrast, lethal activities against An. gambiae s.s., An. arabiensis, and An. funestus s.s. of fenitrothion (100% mortality for 3 species) and propoxur (100%, 100%, and 95.0% mortality, respectively) were higher than those of the 2 pyrethroids. The knockdown activity of propoxur (KT 50 , 23.2, 25.0, and 35.2 minutes, respectively) was higher than that of the 2 pyrethroids. The laboratory colonies of An. gambiae s.s. and An. arabiensis reared in ICPE showed high susceptibility to all of the insecticides used ( Figure 2). The allelic frequency of the point mutation in the voltage-gated sodium channel (L1014S) in F1 An. gambiae s.s. was 100% in Ragwe (n = 214, Accession AB776709) and 99.3% in Mfangano (n = 211, Accession AB776707, AB776708), while not a single point mutation was detected in the other 2 species, indicating that the major pyrethroid resistance mechanism in this species was kdr.

Discussion
High pyrethroid resistance in 3 major malaria vectors was observed in the study area as previously reported [11,14]. Kawada et al. [11] reported that the resistance mechanisms were multimodal, including kdr, caused by the point mutation of the voltage-gated sodium channel (L1014S) in An. gambiae s.s., and mixed cytochrome P450-related metabolic factors in both An. arabiensis and An. funestus s.s. In Nyanza province, dieldrin was reported to be administered mainly through aerial spraying, especially for tsetse fly control [27], while the organized intensive spraying of DDT for mosquito control was not performed in the 1970s and 1980s, and no IRS has been administered since then. Therefore, the extensive use of ITNs and LLINs in the study area is thought to be a major factor causing such high pyrethroid resistance. Fortunately, the present results showing no cross-resistance in carbamate-and organophosphate-class insecticides might provide a specific remedy for the  Figure 3 Contact repellency of permethrin and deltamethrin against the F1 progenies of An. gambiae s.s., An. arabiensis, and An. funestus s.s. collected in the study area by a simplified contact repellency test modified by Grieco et al. [22] using WHO test tubes.
Repellency was calculated as the average percentage of mosquitoes in the untreated test tubes. Different alphabetical letters indicate significant differences by means of a Tukey honestly significant difference test. emergence control of the vectors in this area. Two ICIPE laboratory colonies of An. gambiae s.s. and An. arabiensis were collected in the same field as in this study (ca. 10 years ago in An. gambiae s.s. and several years ago in An. arabiensis). The frequency of allelic L1014S point mutations in the laboratory colony of An. gambiae s.s. was 6.3% [11], while that in field-collected An. gambiae s.s. was >99%. The historical increase in kdr mutations in An. gambiae s.s. after the spread in the use of ITNs and LLINs in 2000s, as reported by Mathias et al. [6], has also been found in our study area (unpublished data), indicating that the An. gambiae s.s. ICIPE colony has been maintaining high susceptibility to pyrethroids with their low frequency kdr mutations. On the other hand, the cytochrome P450related resistant factors [11] in the laboratory colony of An. arabiensis were thought to be absent when they were transferred into the laboratory from the field, or otherwise to have declined in the course of generations of rearing. Such a decline in resistance without selection pressure from insecticides in the latter case might sometimes be common in the laboratory colonies with metabolic resistance factors [28]. The repellency of permethrin (0.75% paper) was significantly higher than that of 1/15 the amount of deltamethrin (0.05%) in the wild colony of An. arabiensis. Siegert et al. [29] pointed out the same difference in repellency between an Olyset® Net (containing 1000 mg of permethrin per square meter) and a PermaNet® (55 mg of deltamethrin per square meter). The authors reported that the Olyset® Net reduced the landing attempts of mosquitoes and elevated their flight frequency, while there was a higher incidence of mosquito landing attempts on the PermaNet® [28]. The difference in repellency between the 2 pyrethroids might be explainable by their structural differences (permethrin belongs to the non-cyano-containing Type I pyrethroids and deltamethrin to the alpha-cyano-containing Type II pyrethroids) resulting in the different neurotoxicity [30].
It is interesting that the lack of repellency to pyrethroids was observed only in the wild colony of An. gambiae s.s. even though the 2 other wild colonies also possess high resistance to pyrethroids. It might be reasonable to consider that the point mutation in the voltage-gated sodium channel interferes with the sensitivity of the sensory nervous system to pyrethroids as well as with the central nervous system, causing less irritancy to mosquitoes, resulting in slower avoidance or reduced repellency [18]. However, the phenomena previously reported on the relationships between kdr and pyrethroid repellency were not simple. Chandre et al. [31] reported that a laboratory-selected An. gambiae s.s. colony originating from Burkina Faso with a homozygous kdr factor (RR) lost contact repellency to permethrin 1% impregnated paper as compared to an insecticide susceptible Kisumu colony (SS) and a kdr heterozygous colony (RS). RS hybrids in the above report were exactly intermediate between RR and SS individuals in repellency. Corbel et al. [32] also reported a non-linear relationship in the survivorships of RR, RS, and SS genotypes with permethrin dosage, and higher dosages of permethrin more efficiently killed the RS genotypes of An. gambiae s.s. than did lower dosages. The authors concluded that heterozygous mosquitoes (RS) were more efficiently killed than susceptible mosquitoes (SS), since kdr resistance to the irritant effect appeared to be codominant while resistance to the lethal effect was recessive [31], so the RS mosquitoes stayed longer than SS mosquitoes on the permethrin-treated surface, picking up more insecticide and being killed in higher proportions. There was no answer, however, to the question of why the irritant effect of kdr was co-dominant while the lethal effect of the same gene was recessive. The kdrgoverned absence of pyrethroid repellency was also reported by Virgona et al. [33]  role in the repellency resistance of houseflies, since the pen gene causes a reduced rate of penetration of insecticides [34] and possibly also reduces the amount of pyrethroids reaching the sensory nerves, which are closely related to repellency. Although the relationships or synergisms between kdr and pen have not been well clarified, it seems to be more plausible to assume the existence of the second factors, which synergize the repellency caused by kdr gene. Chandre et al. [31] and Corbel et al. [32] concluded that a high proportion of An. gambiae s.s. possessing homozygous kdr mutations were killed by prolonged contact with pyrethroids because of their reduced sensitivity to the excito-repellent effects of pyrethroids, and that permethrintreated nets seemed unlikely to select for pyrethroid resistance in areas where the kdr mutation is rare and present mainly in heterozygous form. The above hypothesis seems to be incorrect since the kdr mutations in An. gambiae s.s. populations seem to have increased in accordance with the extensive use of LLINs from a low allelic percentage to the maximum (>90%) [6]. Recently, Kawada et al. found that permethrin-impregnated LLIN were effective against the 3 major pyrethroid-resistant malaria vectors, An. gambiae s. s., An. arabiensis, and An. funestus s.s., in western Kenya, since the frequency of human feeding was found to be reduced to a low level during 'bedtime'. However, the large proportion of human blood feeding was shown to take place during the time people were active outside LLINs in both An. arabiensis and An. funestus s.s., whereas no such event was shown in An. gambiae s.s. [unpublished data]. One plausible explanation for such a species-dependent difference in host feeding activities under LLIN use might be the difference in the repellency to pyrethroids observed in the present report. The historical population decline of An. gambiae s.s. reported by Bayoh et al. [35] might also be explained by the above behavioral characteristics in the An. gambiae s.s. population. Anopheles gambiae s.s. might have not changed its characteristics as a "midnight feeder" because of the lack of repellency despite the existence of LLINs, and it still has to rely on the limited human blood sources, most of which are protected by LLINs, resulting in the decline in their population size.

Conclusion
A different repellent reaction was observed in the fieldcollected An. gambiae s.s. than in An. arabiensis and An. funestus s.s. It might be that resistant mosquitoes governed by knockdown resistance (kdr) loose repellency to pyrethroids, whereas those lacking kdr maintain high repellency irrespective of their possessing metabolic resistance factors to pyrethroids. The comparative studies of the repellency of mosquitoes have mainly studied kdr, and those studying enhanced metabolic factors have been unexpectedly few. Further genetic evaluation should be required to demonstrate the above hypothesis.