Pyrethroid Cross-Resistance Spectrum Among Population of An. Gambiae

Home , Etofenprox

Journal of the American Mosquito Control Association, 15(1):53-59, 1999 Copyright @ 1999 by the American Mosquito Control Association, Inc.

PYRETHROID CROSS RESISTANCE SPECTRUM AMONG POPULATIONS OF ANOPHELES GAMBIAE S.S. FROM COTB D'IVOIRE

FABRICE CHANDRE,I FREDERIC DARRIEI' SYLVIE MANGUIN,I CECILE BRENGUES,I PIERRE CARNEVALE, INo PIERRE GUILLET'

ABSTRACT. Field samples of Anopheles gambiae s.s. from C6te d'Ivoire were tested with 5 pyrethroids (cyfluthrin, )t-cyhalothrin, o-cypermethrin, deltamethrin, permethrin), I pseudo-pyrethroid (etofenprox), and an organochlorine (DDT). With the use of World Health Organization diagnostic tests, 5 out of 6 samples were found cross-resistant to these insecticides. A strong decrease in knockdown effect and mortality was also observed when testing deltamethrin-impregnated nettings. With a polymerase chain reaction amplification of specific alleles diagnostic test, resistance was found associated with the presence of a kdr mutation. The strong correlation between kdr allelic frequency and resistance to DDT or etofenprox indicated that kdr was the main resistance factor for these 2 insecticides. On the contrary, a lower correlation was observed between kdr frequency and resistance to 4 of the 5 pyretbroids tested, suggesting that another mechanism was also involved, Iikely a metabolic detoxification. These results point out the necessity to monitor pyrethroid resistance and the presence of kdr before implementation of any impregnated bed-net programs for malaria control.

KEY WORDS Anopheles gambiae, pyrethroid, resistance, kdr, malalria control, C6te d'Ivoire, West Africa

INTRODUCTION Cdte d'Ivoire, where resistance was 1st pointed out, is a country with intensive agriculture where Vector control is an important component of the huge amounts of insecticides have World Health Organization (WHO) global strategy been used for many years. Therefore, we decided to carry out for malaria prevention. The main objective is to a survey in several localities from break the transmission of the parasite with indoor this country to verify whether resistance was residual spraying or pyrethroid impregnated mate- widespread. Because the kdr mutation is known to induce rials (bed-nets or curtains). Pyrethroids are prefer- a broad spectrum of cross-resistance to pyrethroids (Farn- ably used for impregnation because they are highly DDT and han L973, 1977), effective and fast acting insecticides with a strong we also tested other available pyrethroids excito-repellent effect on mosquitoes. usually regarded as possible alternatives in case of permethrin Anopheles gambiae s.l. Giles is the main malaria resistance. In addition to bioassays made with f,eld-collected mosquitoes, vector in Africa, where 9OVo of t}te world cases we also evaluated the frequency occur. Resistance of An. gambiae s.s. to pyrethroids of the kdr allele and compared it with the level was first observed in Cdte d'Ivoire (West Africa) of resistance to the different insecticides. by Elissa et al. (1993), who reported a significantly decreased mortality for permethrin and a lower susceptibility to the knockdown effect of deltamethrin MATERIALS AND METHODS and tr-cyhalothrin. More recently, knockdown re- Samples sistance (kdr) to pyrethroids and DDT was ob- and strains of mosquitoes served in several countries from West Africa Larvae of An. gambiae s.s. were collected in nat- (Chandre et al. 1999). Investigations on the rarget ural breeding sites from 6 localities of COte d'Ivoire site for pyrethoids and DDT have shown that this (Fig. 1). The samples from Abidjan were collected resistance was associated with a single point mu- within the urban area of the capital of COte tation on the gene coding for the sodium channel, d'Ivoire. Samples from Korhogo and Yaokoffikro resulting in the change of one amino acid (Marti- were collected in periurban areas of these cities. nez-Torres et al. 1998). A polymerase chain reac- The 3 other localities were sampled in rural areas, tion (PCR) amplification of specific alleles (PASA) including Kafine and M'be from the southern sa- diagnostic test has been developed to identify the vanna area and Zarpobly from the forested area mutation on a single mosquito and to score for the (Fig. 1). genotype. All tests were carried out either on nonbloodfed The current pyretlroid resistance observed in 2-5-day-old females emerged from field-collected West Africa has been suspected to result mainly larvae or on females emerged from the lst progeny from the intensive use of DDT and, later on, py- brood. Susceptibility of natural populations was rethroids for crop protection, especially cotton. compared with that of 2 laboratory strains of An. gambiae s.s. with the same lots of impregnated pa- I ORSTOM, Laboratoire de Lutte contre les Insectes pers: Kisumu, a susceptible reference strain origi- Nuisibles, BP 5045, 34032 Montpellier Cedex 1, France. nated from Kenya and maintained for many years ' OCCGE, Institut Pierre Richet, BP 1500, Bouake 01, in the laboratory; and Kou, a selected resistant C6te d'Ivoire. strain, originating from a natural pyrethroid- and JounNaL on rHB AwenlcAN Moseurro CoNrnol AssoclnttoN VoL. 15,No. 1

, 200 lan [ : Savamuea

f] : Foresteduea (f,) : Urbmuea

C0te d'Ivoire

Fig. 1. Map of Cdte d'Ivoire showing the localities of the 6 collecting sites of Anopheles gambiae s.s. situated either in the savanna area (white), forested area (shaded), or urban area (star).

DDT-resistant population collected in Burkina Faso tests were made at higher concentrations for per- in 1995. This strain was laboratory selected with methrin (l%o). The 3 other insecticides, or-c)4)er- permethrin and made homozygous for tdr. Colo- methrin, cyfluthrin, and etofenprox, were used at nies of both Kisumu and Kou strains were main- O.OO25Vo,O.O5Vo, and O.25Voconcentrations respec- tained at Bouake (Institut Pierre Richet, OCCGE) tively, providing systematically 95-100Vo mortality where all bioassays were made. for the susceptible reference strain. Then, for most insecticides, we considered the threshold for resistance below 95Vo morlality. Resistance tests Laboratory tests were made with lots of 25 adult Bioassays were carried out with WHO test kits females in WHO test tubes with at least 4 replicates for adult mosquitoes (World Health Organization per bioassay. Exposure time was 60 min with tubes 1970) with 7 insecticides of technical grade quality. maintained in the normal vertical position. After Five were pyrethroids: cyfluthrin (92Vo) (Bayer,Le- exposure, mosquitoes were kept under observation verkusen, Germany), L-cyhalothrin (84.lVo) (Ze- for 24 h, supplied with lQVo honey solution, and neca, Bracknell, United Kingdom), cr-cypermethrin maintained at a temperature around 26'C with (99.5Vo) (Phytagri, Geneva, Switzerland), delta- about SOVo RH. Mortalitv was read after this 24-h methrin (99.8Vo) (Agrevo, Berkhamsted, United period. Kingdom), and permethrin 25175 (93.8Vo) (Agre- vo). One insecticide was a nonester pyrethroid, eto- Allelic frequency of kdr in field samples fenprox (997o) (Mitsui Toatsu, Tokyo, Japan), and 1 was an organochlorine, DDT (lOOEo) (Sigma, St. PCR amplification for kdr mutation' Genotypes Quentin Fallavier, France). for kdr mutation were determined according to Impregnated papers were prepared in our labo- Martinez-Tones et al. (1998). For individual mos- ratory with acetonic solutions and silicone oil (Dow quitoes, DNA was extracted from2-3 legs with5Vo Corning 556) as a carrier. Impregnations were made Chelex@ in water solution. The PASA tests were on the basis of 3.6 mg of oil per cm2. Whatman made in buffer containing deoxynucleotides tri- filter paper sheets (12 X 15 cm) were impregnated phosphate, Taq DNA polymerase, the 2 primers for with a mix of 0.7 ml of silicone oil and 1.3 ml of the control band, and the 2 specific primers for sus- insecticide acetonic solution. Papers were stored at ceptible allele and kdr mutation. The lengths of am- 4"C and were not used more than 3 times. The con- plified sequences were 293 bp for the control band, centrations used were those recommended by WHO 195 bp for the kdr allele, and 137 bp for the sus- for deltamethrin (O.O25Vo), )r-cyhalothrin (O.lVo), ceptible allele, allowing an easy definition of ge- permetbrin (O.25Vo), and DDT (47o). Additional notypes. M.qncs 1999 Pvnenrnon Cnoss-ResrsreNcsop Av. GAMBTAE 55

PCR for identification of species: Although the Allelic frequency of kdr in natural populations &dr mutation has been found so far only in An. gambiae s.s., specimensfrom the field (except 1 All specimensexcept those from Korhogo were locality, Korhogo) were identified with specific identified by PCR and were found to be An. gam- primers according to Scott et al. (1993). biae s.s. The l

Insecticide Kisumu (susceptible straln) Insecticide Kou(resistantstrain) permethrin0. r00(r05) permethrin0.25% 0 (60) permetlrin I r00(244) permethrin1% 3.9(r02) deltamethrin roo(256) deltamethrin0.025o/o r3.8(s8) l"-cyhalothrin0.1% r00(262) l"-cyhalothrin0.1% 67.3(98) o-cypenn.0.0025% r00(378) ct-cyperm.O.OO25% 13.7 (I 08) cyfluthrin e7.e(238) cyfluthrin 0.05%15.l (98) (233) (97) etofenprox0.25% 99.6 etofenprox0.25% | | DDT 100(181) 311,1ao7"ll.7 (58) 0 20 40 60 80 100 0 20 40 60 80 100

Talpobly M'be permethrin sr.3(1e3) permetlrin 0.250% 87.2 (94) permethrin (200) deltamethrin0.02 4 (r88) l,-cyhalothrin 0. I r00(103) deltamethrin0.025Vo e8(r02) cl-cyperm. | (113) cyfluthrin .4 (95) ?r-cyhalothrin0.1% r00(100) etofenprox (100) DDT Q02) 0 20 40 60 80 100 20 40 60 80 100

Abidjan Kafine permethrin 1 78 (r00) permethrin0.25% s (100)

deltamethrin 87.4(es) deltamethdn0.025% 6r.9(10s)

l.-cyhalothrin0. l% 87.2(e4) l.-cyhalothrin0.1% 90.1(269) cyfluthrin 5r (100) c,-cypenn.0.0025% s.3(es)

etofenprox 0.25olo 64.6(e6) cyfluthrin 0.05% 68.4(9s) DDr4%II75.8(e5) etofenprox0.25olo s (e4) 0 20 40 60 80 100 0 20 40 60 80 100

Yaokoflikro Korhogo permethrin0.25% Ls.9(176) permethrin 3.6 (84) permethrin1% permethrin t6.8(r0t) deltamethrin0.025% deltamethrin0.02 36.7 (98) )r-cyhalothrin0.1% },-cvhalothrin 0.1 o-cypenn.0.0025% 2.9(r02) cr,-cyperm. t.2 (e4) cyfluthrin 0.05% cyflutlrin 0. r7.7(e6) etofenprox0.25% etofenprox0, (ee) DDT 4% r7.6(r02) DDT 7.2(e7) 0 20 40 60 80 r00 0 20 40 60 80 100 7" Mortality o/"Mortality

Fig. 2. Percentages of mortality to WHO tests with different insecticides for 6 field samples of Anopheles gambiae s.s., a susceptible reference strain (Kisumu), and a resistant one (Kou). Number of mosquitoes assayed in brackets. Mancu 1999 PvnBrunon Cnoss-RrstsreNcp or Arv. SAMBTAE 57

Table 1. Genotypes of Anopheles gambiae and allelic Table 3. Knockdown (Kd) and mortality of Anopheles frequency of kdr in natural populations of C6te d'Ivoire. gambiae after 3-min exposure to deltamethrin nettings (25 rng/m') for 2 field-resistant samples (Korhogo, Genotype Yaokoffikro) and susceptible reference strain (Kisumu). Sample SS RS RR F(R)' Sample Kd at 60 min Mortality Zaipobly 2600 OVo 26 Kisumu IOOVo lNVo 60 M'be 6331 3.7Vo 67 Korhogo 3.9Vo l.3Vo 77 Abidjan 1013438.9Vo 2'7 Yaokoffikro 21.3%o lO.lVo 89 Kafine 12610 46.4Vo 28 Yaokoffikro 5322 78.3Vo 30 Korhogo 112895.0Vo 30 'Allelic frequency of Mr in Vo. diagnostic concentrations for monitoring mosquito ? Number of mosquitoes assayed. resistance in the field. A broad spectrum of resistance to various pyrethroids and DDT was conferred by the kdr mutation, as shown from the results obtained with the ple was tested, a low but detectable resistance lev- Kou strain, homozygous fot kdr. The cross-resis- el to most of the pyrethroid insecticides was tance pattern observed in 5 populations from COte found. However, this sample would have been re- d'Ivoire was also clearly associated with the pres- garded as fully susceptible if tested only with \- ence of tlre lcdr mutation. The strong correlation cyhalothrin. observed between kdr allelic frequency and surviv- Furthermore, a collaborative study was initiated al of mosquitoes with DDI )t-cyhalothrin, and eto- by WHO after this study to determine the proper fenprox indicated that lcdr was the main resistance diagnostic concentrations of pyrethroids for the factor for these 3 insecticides. most important mosquito vector species. In the On the contrary, the correlation for other tested framework of this study, our laboratory (WHO Col- insecticides was not as signiflcant, although an in- laborating Centre for Vector Control) has found that crease of kdr frequency was clearly associated the mean mortality rate with tr-cyhalothrin 0.017o with a decreased mortality to pyrethroids. Al- was 98.7Vo with susceptible An. gambiae s.s. (4 though this poor correlation could be attributed to replicates, 40O individuals tested, impregnated pa- the small number of samples considered, it sug- pers provided by WHO). The concentrations pro- gested, above all, that kdr was not the only resis- viding 1007o mortality ranged from 0.017o to tance mechanism involved. More interestingly, al- O.OZSVo.Therefore, the concentration of 0,170 men- though t}re kdr frequency in the M'be sample was tioned above is obviously too high, and the correct low (4Vo) and mortality with DDT high as ex- diagnostic concentration for )r,-cyhalothrin should pected (98.5Vo), a high survival rate was observed be O.OSVo(twice our highest concenhation value). with cr-cypermetbin (69Vo). Considering that for Similarly, the correct diagnostic concentration for several years DDT has not been used in C6te permethrin 25175 would, be l7o instead of the d'Ivoire, the high resistance observed indicates O.25Vo proposed by WHO using permethrin with a that the An. gambiae s.s. populations were sub- different isomeric ratio. In accordance with both mitted to a strong pyrethroid selection pressure. Hemingway (1995) and our present study, the con- Possibly, this pressure selected more specific py- centration used for etofenprox (O.25Vo) was shown rethroid resistance mechanisms, such as metabolic to be discriminant for anophelines. These results detoxification. Oxidase-based metabolism through underline the necessity to standardize the case of P450 cytochromes and, to a lesser extent, esterases

Table 2. Correlation coefficients between kdr allelic frequency (Table 1) and corresponding percentage of population survival rate (Fig. 2) for different insecticides.

Correlation Insecticide coefficient t-value P(t)' n2 Cyfluthrin 0.057o 0.883 3.257 o.0236 ) \-Cyhalothrin 0.17o 0.977* 9.O98 0.0004 6 c-Cypermethri n O.OO25Vo 0.898 2.880 0.0512 4 Deltamethrin O.O25Ea 0.853 3.263 0.0155 6 Permethnn O.25Vo 0.899 J.))4 0.0190 5 Permethrin 17o 0.959 4.780 0.0206 Etofenprox 0.25Vo 0.953* 5.445 0.0061 5 DDT 4Vo 0.986* 8.379 0.0070 4 'Probability of r-value with n - 2 df. ' Number of samples. " Conelation coefficient significantly different from 0 at 95Vo level after sequential Bonfenoni procedure 58 JoURNAL oF THE AMERICIN Moseurro CoNrnol AssocrlrroN Vol. 15. No. I is commonly involved in pyrethroid resistance. ACKNOWLEDGMENTS Pyrethroid detoxification mechanisms may not confer resistance to DDT and etofenprox (nonester We thank the following companies: Agrevo, Bayer, Mitsui pyrethroid), which have different chemical struc- Toatsu,Phytagri, andZenecafor providing technical insecticides. tures and metabolic pathways. This may explain We are also grateful to R. N'guessan for technical why resistance to these 2 insecticides was strongly assistance.This study was supportedby OCCGE, ORSTOM, and MESR correlated to kdr frequencies and resistance to py- (grant 92-058). rethroids was not (except for \-cyhalothrin because the concentration used was too high). On- going studies are aimed to identify the other REFERENCESCITED resistance mechanism(s) involved. Chandre, F, E Darriet, L. Manga, M. Akogbeto, O. Faye, As previously suggested, it is likely that the kdr J. Mouchet and P Guillet. 1999. Status of pyrethroid mutation in West Africa was inherited from DDT resistance in Anopheles gambiae s.1. Bull. WHO 77: (in resistance, itself induced by the agricultural use of press). organochlorine insecticides (Chandre et al. 1999). Cheng, H., W. Yang, W. Kang and C. Liu. 1995. Large- scale spraying This resistance has probably been further enhanced of bednets to control mosquito vectors and malaria in Sichuan, China. Bull. WHO 73:321- by the intensive use of pyrethroids for crop protec- 328. tion and domestic hygiene. Although a-cypermeth- Choi, H. W., J. G. Breman, S. M. Teutsch, S. Liu, A. W. rin has never been used in COte d'Ivoire for mos- Hightower and J. D. Sexton. 1995. The effectiveness quito control, resistance was observed even in the of insecticide-impregnated bed nets in reducing cases M'be sample, where resistance to other pyrethroids of malaria infection: a meta-analysis of published re- was much lower. However, this is not surprising sults. Am. J. Trop. Med. Hyg. 52:377-382. Doannio, J., J. Dossou, M. C. Henry, considering that cypermethrin accounts for 80-907o I. Nzeyimana, F Darriet, P Guillet and P Carnevale. 1997. Pyrethroid of the overall quantity of pyrethroids used for sev- resistance in An. gambiae s.s. and efficacy of impreg- eral years for cotton, with about 1 million liters nated bednets. First results of a field trial in West Af- applied annually in this country (Vassal, personal rica. Proc. Int. Conf. Bednets and Other Insecticide communication). Treated-Materials. (October, 1997). Pyrethroid resistance in COte d'Ivoire should def- Elissa, N., J. Mouchet, E Rivibre, J. Y. Meunier and K. initely be regarded as a threat for the implementa- Yao. 1993 Resistance of Anopheles gambiae s.s. to pyrethroids in C6te d'Ivoire. Ann. tion of malaria control programs with impregnated Soc. Belge Med. Trop. 73:291-294. bed-nets. Field-collected mosquitoes from resis- Farnham, A. W. 1973. Genetics of resistance of pyre- tance af,easwere no longer killed when exposed for throid-selected houseflies, Musca domestica L. Pestic. 3 min to nettings impregnated with deltamethrin. In Sci. 4:513-520. addition, resistant individuals are much less suscep- Farnham, A. W. 1977 . Genetics of resistance of houseflies tible to the knockdown and excito-repellent effects (Musca domestica L.) to pyrethroids. Pestic. Sci. 8:631- of pyrethroids (unpublished data). Several large- 636. Hemingway, J. 1995. Efficacy of etofenprox scale control programs with pyrethroid-impregnat- against insecticide susceptible and resistant mosquito strains con- ed bed-nets, such as in the Gambia (Hemingway et taining characterized resistance mechanisms. Med. Vet. al. 1995) or in China (Cheng et al. 1995), did not Entomol. 9:423-426. result in the selection of resistance among anoph- Hemingway, J., S. W. Lindsay, G. J. Small, M. Jawara and eline populations in these areas. However, L case E H. Collins. 1995. Insecticide susceptibility status in of permethrin resistance was observed in East Af- individual species of Anapheles gambiae complex (Dip- tera: Culicidae) in an area of The Gambia where py- rica in a permethrin-impregnated bed-net area (Vul- rethroid impregnated bednets are used extensively for ule et al. 1994). The problem quite may be different malaria control. Bull. Entomol. Res. 85 :229 -234. when resistance genes are already present in target Martinez-Torres, D., E Chandre, M. S. Williamson, E Dar- populations, as found in Cdte d'Ivoire. Interesting- riet, J. B. Berge, A. L. Devonshire, P Guillet, N. Pasteur ly, in 1 of the sampled localities (Kafine), the allelic and D. Pauron. 1998. Molecular characterization of py- frequency of kdr increased from 46.4Vo to 92.5Vo rethroid knockdown resistance (kdr) in the major main only 1 year after distribution of permethrin-im- laria vectorAnopheles gambiae s.s. Insect Mol. Biol. 7: 179-r84. pregnated bed-nets (Doannio et al. 1997). However, Rice, W. R. 1989. Analysing tables of statistical tests. in the same time, the malaria morbidity of children Evolution 43 :.223--225. was reduced by more tllLar 5OVo, which is in the Scott, J. A., W. G. Brogdon and E H. Collins. 1993. Ioen- order range of results obtained with impregnated tification of single specimens of the Anopheles gambiae bed-nets in other parts of Africa (Choi et al. 1995). complex by polymerase chain reaction. Am. J. Tiop. These results underline the necessity to monitor py- Med. Hyg. 49:520-529. Vulule, J. M., R. E Beach, E K. Atieli, J. M. Roberts, D. rethroid resistance and the presence of kdr genebe- L. Mount and R. W. Mwangi. 1994. Reduced suscep- fore planning and implementing malaria control tTbllity of Anopheles gambiae to permethrin associated programs based on the use of impregnated materi- with the use of permethrin-impregnated bednets and als. curtains in Kenya. Med. Vet. Entomol. 8:71--75. MARCH 1999 PyRETHRoD Cnoss-RgstsreNcE oF AN. GAMBTAE 59

World Health Organization. 197O. R6sistance aux insec- and conditions. WHO/MAL/95.1O73, WHO/CTD/ ticides et lutte antivectorielle. Dix-septidme rapport du VBC/95.998. Geneva, Switzerland. comit6 OMS d'experts des Insecticides. Technical Re- World Health Organization. 1996. Report of the WHO port Series 443. Geneva, Switzerland. informal consultation on the evaluation and testing of World Health Organization. 1995. Supplies for monitor- insecticides. CTD/WHOPES/ICI96. 1 Geneva, Switzer- ing insecticide resistance in disease vectors. Procedure Iand.