Emergence of Resistance and Resistance Management in Field Populations of Tropical Culex Quinquefasciatus to the Microbial Control Agent Bacillus Sphaericus

Journal of the American Mosquito Control Association, 19(l):39-46,2OO3 Copyright @ 2OO3 by the American Mosquito Control Association, lnc.

EMERGENCE OF RESISTANCE AND RESISTANCE MANAGEMENT IN FIELD POPULATIONS OF TROPICAL CULEX QUINQUEFASCIATUS TO THE MICROBIAL CONTROL AGENT BACILLUS SPHAERICUS

MIR S. MULLA,' USAVADEE THAVARA,' APIWAT TAWATSIN,' JAKKRAWARN CHOMPOSRI'ENO TIANYUN SUI

ABSTRACT. In recent years, highly potent mosquitocidal strains of the microbial agent Bacillus sphaericus (Bsph) have been isolated and developed for the control of mosquito larvae around the world. Laboratory selection experiments with the most active strains and their use in large-scale operational mosquito control programs resulted in the emergence of resistance in larvae of the Culex pipiens complex. This generated great concern among vector control agencies around the world, who feared reduced efficacy of this highly active larvicidal agent. To address this issue, the current studies were started to find practical strategies for controlling resistant mosquitoes and more importantly to develop resistance management strategies that would prevent or delay development of resistance. We initiated field studies in 3 low-income communities in Nonthaburi Province, Thailand. In I of the communities, larvae of Culex quinquefasciatas that were highly resistant (>125,0fi)-fold) to Bsph strain 2362 were successfully controlled with applications of Bacillus thuringiensis vat israelensis (Bti) alone or in combination wilh Bsph. To prevent or delay resistance to Bsph,2 other sites were selected, I treated with Bsph 2362 alone and the other treated with a mixture of Bsph 2362 and Bri. Mosquitoes treated wit}r Bsph 2362 alone showed some resistance by the 9th treatment and almost complete failure of control occurred by the lTth treatment. After 9 treatments with the mixture over a 9-month period at another site, no noticeable change in susceptibility to Bsph was detected. During this period, the site treated with Bsph alone required 19 treatments, whereas the site treated with mixtures took only 9 treatments because of slower resurgence of larvae at the site treated with the mixture than at the site treated with Bsph alone. This is the lst field evidence for delay or prevention of resistance to microbial agents in larval Cx. quinquefasciatzs by using mixtures of Bti and Bsph. Further studies on the use of mixtures for the management of field resistance are warranted.

KEY WORDS Culex quinquefasciatus, Bacillus sphaericus, resistance, resistance management, prevention

INTRODUCTION velopment of resistance to Bsph 2362 in larvae of Culex quinquefasciatus Say in the laboratory. A In the past 25 years, several isolates and strains moderate to high level of resistance was obtained of 2 species of spore-forming bacteria have been on selection with this strain (Rodcharoen and Mulla found that produce parasporal proteins that show 1994, Wirth et al. 2000, Za}:'iri et al. 2002). Mod- high toxicity against mosquito larvae (WHO 1985, erate to high levels of resistance in the C.x' pipiens de Barjac and Sutherland 1990). Bacillus thurin- L. complex were reported in the field after 2-3 giensis var. israelensis de Barjac (Bri) was discov- years of use of strains 2362 (Sinbgre et al. 1994' ered in 1976 (Goldberg and Margalir 1977) and was Silva-Filha et al. 1995), 1593M (Adak et al. 1995' developed for use in mosquito and black fly control Rao et al. 1995), and C3-41 (Yuan et al. 2O0O).In programs (de Barjac 1990a, Mulla 1990). During the evaluation and development process, Bsp& this same period, research on isolating and devel- strains were tested once or twice in field trials for oping mosquitocidal strains of Bacillus sphaericus initial activity in several countries (Liu et al 1989; (Bsph) Neide was intensified. Of the approximately Zlnang et al. 1989; Karch et al. 1992; Mulla et al' 3OO strains isolated, 17 were highly toxic to mos- 1997, 1999; Ali et al. 2000). Once implemented in quito larvae (de Barjac and Sutherland 1990). Of field control programs, little or no monitoring of (Wickreme- these, strains 2297 from Sri Lanka susceptibility levels of mosquito species was con- from Indonesia singhe and Mendis 1980), 1593M ducted and resistance was only detected when gross (de Barjac 1990b, Singer 1990), 2362 from Nigeria failure in field control programs was noted. (Liu (Weiser 1984), and C3-41 from China et al. During the course of several years of field eval- are now avail- 1989) were studied extensively and uation of Bsph forrnulations against Cx. quinque- able commercially as mosquito larvicides. Strain fasciatus in polluted water in squatter communities used material, 2362, which is the most commonly in Thailand (Mulla et al. 1997, 1999,2OOI), almost used in Europe and elsewhere since 1989 has been IOOVo initial and 35 days of residual control of lar- and in the USA since 1996. vae was achieved with the lst few treatments of In 1990, we initiated studies on the potential de- Bsph 2362. However, with further treatrnents, the level of control declined and longevity of control also progressively decreased (Mulla et al. 2001). I Department of Entomology, University of California, Rapid failure of control occurred with the Vectolex Riverside, CA, 92521. (WDG) at 'NIH, Department of Medical Sciences, Ministry of water dispersible granules formulation Public Health, 87/7 Tiwanon Road, Nonthaburi ll0o0, Wat Pikul Community in Nonthaburi Province in Thailand. 2000. Control of larvae was reduced after the 4th

39 40 Joumll op rnn AurnrclN Mosqurro CoNrnor-AssocrlrroN Vor-. 19, No. I treatment, and no control was achieved in the 5th trol- Detailed descriptions of this site were pub- treatment at the maximum label dosage of 200 mg/ lished by Mulla et al. (1999,2001). This commu- m' (Mulla et al. 2001). A high level of resistance nity is similar in size to Wat Pikul and has similar in Cx. quinquefasciatus evidently had developed af- potential for supporting polluted-water mosquitoes. ter 4-5 repeated applications of Bsph at this site. However, monitoring of this site was terminated at This rapid development of resistance caused con- the time of the l0th treatment at Wat Pikul on May cern in vector control programs. 23,2OO1. The trends of immature mosquitoes were To respond to these concerns of rapid develop- rather stable during the lst 4 months of the exper- ment of resistance, the current studies were initiated iment and we saw no point in funher following to document the magnitude of resistance tn Cx. these populations as a control. We have noted that quinquefasciatus atWat Pikul and to develop prac- mosquitoes in these types of developmental sites tical strategies for the control of Bspy'l-resistant resurge quickly after cessation of treatment and mosquitoes. Laboratory assays provided evidence reach pretreatment populations; a comparison of for lack of cross resistance to Bti in Bspft-resistant posttreatment counts versus pretreatment counts mosquitoes (Rao et al. 1995; Rodcharoen and Mul- provides an accurate estimation of control. la 1996; Zahiri et al. 2O02; Su and Mulla, unpub- Emergence of resistance to Bsph and prevention lished data) and this information led to the hypoth- of resistance with mixtures.' These studies were esis that Bsp&-resistant populations at Wat Pikul carried out in 2 communities in the Nonthaburi could be controlled with Bri alone or with a mixture Province, Thailand. The purpose of these studies of Bti and Bsph. Bacillu.s sphaericus produces high- was to determine if repeated treatments of a small er mortality and longer residual activity than does breeding source with Bsph 2362 alone would result Brl in polluted water. The belief was held that a in the development of resistance, as was observed mixture of Bsph and Bti could yield initial and per- at Wat Pikul. At the same time, we examined sistent control. Therefore, field studies were initi- whether mixtures of Bti and Bsph could prevent or ated to ascertain control of Bspft-resistant mosqui- delay emergence of resistance to Bsph. The lst toes by mixtures of Bsph and, Bti, to investigate community, Wat Lahan, Bang Bua Thong District, prevention of resistance by use of Bsph-Bti mix- was subjected to repeated treatments of Bsp& WDG tures, and to study the emergence of resistance as formulation at 100 mglm,, a highly eff'ective dosage a result of treatments with Bsph 2362 alone. Results for susceptible larvae.. This site encompassed a to- are presented in figures. tal mosquito-producing area of 4OO m2, an area much smaller than that of Wat Pikul and Soi Jumpa. This area MATERIALS AND METHODS also had some inaccessible breeding sources of mosquitoes that could not be treated. pik- Control ofBsph-resistant mosquitoes at Wat The studies in Wat Lahan community started on u/.' To control Bspft-resistant Cx. quinquefasciatus, May 17, 2001, and terminated on March 15, 2OO2, treatments with B/i or Bsph 2362 alone or in a mix- after the last (l9th) treatment at 200 ms/m, failed ture were initiated. Wat Pikul (Bang yai District, to provide satisfactory control of larvael Nonthaburi Province) is a low-income community In previous years, Soi Jumpa community was with high-density housing and a dense human pop- treated intermittently with WDG formulations of ulation (Mulla et al. 2001). In some dwelline units. Bsph 2362, receiving 7 treatments in 1997,3 treat- as many as 9 families live rogerher.in soml cases ments in 1998, and 4 trearments in 1999 (Mulla et with I family per room. As in other low-income al. 1999, 2001). In each period, the last trearment communities in this area, wastewater and solid or- was found to be highly effective, pointing to the ganic and inorganic wastes are thrown outside. absence of resistance in this population. Since Oc- These accumulate under structures built above tober 1999, this site remained untreated. To test the ground on posts and pylons. These conditions (solid susceptibility of larvae at this site, it was treated in waste mixed with polluted water) create ideal hab- early May 2O0l at 100 mg/m, of Bsph WDG, which itats for the propagation of mosquitoes, especially yielded almost 1O07ainitial control. Because of the Cx. quinquefasciatus. The breeding area for mos- lack of apparent resistance to Bsph, this site was quitoes at this location was about 4,000 mr. a rel- selected for the mixture trials. atively large mosquito-producing area of this type. Sampling: Mosquito larval populations were After detecting resistance to Bsph in this area in sampled by 400-ml-capacity dippers. Larvae were December 2OOO, the total area was subiected to sorted into lst, 2nd, 3rd, and 4th instars and pupae treatments of WDG formulations of Bti or Bsph and counted. The sites were sampled before treat- 2362 alone or in various mixtures from January to ment and at various intervals after treatments. late October 2001. When the larval and pupal counts resurged to with- Another communiry, Soi Jumpa (Pak-Kret Dis- in 6O-8OVo of the pretreatment levels, the sites were trict, Nonthaburi Province), was used as an untreat- treated again. ed control. Mosquitoes at this site were.not resistant Larval sampling was biased toward areas where to Bsph, because a tfeatment with 100 mg/m2 of the mosquito larvae and pupae were dense because we WDG formulation yielded almost 100% initial con- wished to judge eff'ectiveness of control in areas of Me.ncn 2003 MANAGEMENToF RESISTANCETo BA..LLUS spHAERrcIlsBy Cutax ourNoutptscrATus 4l dense aggregations of larvae and pupae. The num- quito population, treatments were initiated mostly ber of dips taken varied depending on the size of with Bti-Bsph 2362 mixtures starting in January the treated area. In Wat Pikul. Wat Lahan. and Soi 2O0l and continued them until late October 2001. Jumpa, 20, lO, and 2O dip samples were taken each In total, 18 treatments with single agents or mix- time, respectively. tures of the 2 agents were made (Fig. 1). Materials and applicat ion: Commercial products To further confirm the resistance at this site, the of Bri (VectoBac WDG) and Bsph 2362 (YectoLex lst treatment of Vectolex WDG was made on Jan- WDG) formulations were used in these experi- tr;.ary22,2OO1, at 2OO mg/m2, a highly efficacious ments. Both products were provided by Valent dosage against susceptible mosquitoes in polluted BioSciences Corporation (Libertyville, IL). The water. This rate of application gave no control of dosages given in mg/m2 are for the formulations 3rd and 4th instars and pupae 48 h after treatment. used. The product containers were kept at room The larval density increased after this treatment temperature until used. The required amounts of the (Fig. 1A). In our studies at this site in 2000, before WDG formulation was mixed with water in a plas- the mosquitoes developed resistance to Bsph 2362, tic bucket, transferred to a 7-liter stainless steel we obtained almost 1007ocontrol of larvae with the spray tank, pressurized, and sprayed onto the water lst treatment at 2OO mg/m2 of Vectolex WDG for surface of the habitat. The aqueous spray was ap- over 35 days (Mulla et al. 2001). After document- plied through a cone jet nozzle that projected the ing lack of control at a high dosage of Vectolex, spray jet up to 4-5 m from the sprayer. Most, if the effectiveness of an equal dosage of VectoBac not all, of the habitat was covered by the applica- WDG was tested against this Bsph 2362-resistant tion. population. This treatment gave almost complete In Wat Pikul, the treatable area was 4,000 m' control of larvae, lasting, as expected, for 14 days. through the 9th treatment, but drained to 2,000 m' After this treatment, the efficacy of lower dosages after this time. Wat Lahan had only 400 m2 of mos- of VectoBac WDG was determined by applying this quito-breeding area. In Soi Jumpa, the control area formulation at 5O,25, and 20 mg/m2 in the next 3 for the Wat Pikul test, and subsequently an area treatments. The 50 mg/m2 treatment gave IOOVo treated with the mixture to examine prevention of control of larvae 2 days after treatment, with the resistance, had a breeding source of 3,000 m2. At control dropping to 78Vo 8 days after treatmen| the Wat Pikul, the large area was sprayed with 7 tanks population resurged to the pretreatment level 13 (7 liters each), whereas the reduced area was days after treatment. The treatments at 25 and 2O sprayed with 5 tanks. Wat Lahan was sprayed only mg/m2 yielded 98 and 9lVo tnitial control of larvae, with one 7-liter tank, whereas Soi Jumpa was which dropped to 40 and 47Vo 7 days after treat- sprayed with 5 tanks (7 liters each) each time. ment, respectively (Fig. 1A). Two more applica- Population reduction: Population reduction in tions of Bti alone (13th and 14th treatments) were percent was calculated by comparing counts of 3rd made at the 20 mg/m' dosage of VectoBac WDG. and 4th instars and pupae to the pretreatment counts These treatments gave 99-lOOVo control of larvae of the same stages. Pretreatment counts used in the 2 days after treatment. The level ofcontrol dropped calculations were those just before each treatment to 60 and 33Vo \n the 13th and l4th treatments 7 at a given site. Although Bti and Bsph do not kill days after treatment, respectively. The 17th treat- pupae, the pupal numbers quickly declined because ment in this series was made with VectoBac WDG of larval kill by the treatments and, therefore, they at 200 mg/m2. This treatment, as for the 2nd treat- were included in calculations of percent reduction. ment at 2OO mg/m2, yielded IOOVo conttol initially, Comparison of posttreatment with pretreatment which declined to 92Eo in 7 days and to 64Vo 75 populations in the treated site provide consistent days after treatment (Fig. 1B). These treatments and reliable levels of control. with Bti showed that the Bsph-resistant mosquitoes at Wat Pikul are completely susceptible to Bri. This was confirrned by laboratory studies (Su and Mulla, RESULTS AND DISCUSSION unpublished data). The 7 treatments (2nd through 5th, 13th, l4th, and 17th) made with VectoBac Control of resistant mosquitoes at Wat Pikul WDG also provided evidence that Bti treatments Culex quinquefasciatus, the dominant mosquito, have no residual activity (although this increased was treated 5 times at this site with Bsph 2362 slightly at higher dosages) beyond 7 days and that WDG formulation in 2000. After the 4th and 5th repeat treatments would be necessary every 7-10 treatments, larval control was not satisfactory days at practical dosages. This study also indicated (Mulla et al. 2001). The magnitude of resistance in that the optimum dosages for the control of this this population was further quantified by laboratory B sp h-t esistant polluted-water mo squito are between bioassays, which showed more than 125,000-fold 2O and 50 mg/m2 for initial and short-term control. resistance compared to susceptible mosquitoes from After 4 treatments (2nd through 5th) with B/l Soi Sirichai, Nonthaburi Province, Thailand, to against Bsp&-resistant mosquitoes, various mixtures Bsph 2362 technical powder (Su and Mulla, unpub- of Bti and Bsph 2362 were used for treatment. The lished data). To control this highly resistant mos- 6th and 7th treatments were made with mixtures of 42 JounNer-or rne AugntcAN Moseurro CoNrnol AssoctarroN VoL. 19,No. I

A --]- I st-2ndinstars -+- 3rd-4th instars -# Puoae "o" % Reduction

lst 210 3rd 5th 6rh 7th 8rh Bsph 200* Bti 200 Bti 50 Bti 20 Bti + Bsph Bti + Bsph Bti + Bsph 5 ea. l0 ea. 20 ea. Y^ lr 100 i': VYo le :1

:1 90

ti :i 175 t; 80 l! :i ii ,! l: ;: P r l2s li i 60^ li : ; : (n.Y i = IUU : i ! tn& 2ts bi) 30 50 20

022 7 t42 8 132 8 152 7 t42 72 14 2l 242 Daysposttreatrnent B

9th 10th I lth tzth 13rh l4th 15rh l6th 17th Bsph 30* Bti + Bsph Bti + Bsph Bti + Bsph Bti 20 Bti 20 Bti + Bsph Bti + Bsph Bti 200 20 + ll| 20 ea. 100 I I I t^ io VO Vo ji VQ 100 Yi :1.' iii ii 90 :l I 80 i- 70 .: lSth o Bsph200 z 12s 60^ + ; s0.9 F too 40d Or< hb 30

20262 9162'1 1427 1421272 12't t42127142'7 t521 282.1 t4 2l Days posttreatment

Fig' l. Control of Culex quinquefasciat rs resistant to Bacillus sphaericus (Bsph) 2362 (resistance ratio 125,00O- fold) with (A) lst through 8th treatments (January to April 2001) or (B) 9th through lSth treatments (May to October 2OOl) of Bacillus thuringiensis var. israelensis (Bti) or Bsph alone or Bti-Bsph rnixtures at Wat Pikul, Bang Yai District, Nonthaburi Province, Thailand. Dosage is given as mg/mr. MencTT 2OO3 MeNecEMENT oT RBsTsTaNcn To BACILLUS SPHAERICIJSSY CUTNX QIJINQUEFASCIATUS 43

VectoBac WDG and Vectolex WDG at 5 mglm2 high label dosage of Bsph WDG at 200 mg/m2 was of each (total l0 mg/m') and l0 mg/m' of each tested. Little or no initial control (3IVo) was (total 20 mg/m'), respectively. The treatment with achieved with this treatment (Fig. lB). This clearly 5 mg/m2 of each formulation yielded little or no indicated that the Bsplr resistance in this population control of larvae, but treatment with 10 mg/m' of was stable, with no marked shift noted after 15 each yielded 96Vo control of larvae 2 days and 527o treatments of Bti or Bti-Bsph mixtures, showing control 7 days after treatment. The populations in- that once a high level of resistance was established, creased beyond the pretreatment level 14 days after it could not be reversed with 15 treatments of Bri the 7th treatment. For the 8th treatment, the dosages or of Bti-Bsph mixtures and that we of Bsph 2362 were increased to 20 rng/m2 each (total 40 mg/m'). against Bsp&-resistant Cx. quinquefas c iatus w ould This treatment yielded similar control as the 7th not be practical. Increasing the ratio of Bsph in treatment, 98Vo at 2 days after treatment and 3l%o mixture did not result in persistence against resis- 1 week later (Fig. 1A). tant larvae. Thus, development of strategies to pre- After 7 treatments with Bti alone or with a Bti- vent or delay development of resistance to mos- Bsph mixturq a single treatment with Vectolex quitocidal Bspft strains is needed, From the WDG at 30 mg/m' was administered to determine foregoing, Bspy'r-resistant Cx. quinquefasciatus in if a shift in level of Bsph resistance occurred. This the tropics clearly can be controlled easily by the treatment yielded no control of the larvae, hence use of Bti alone at relatively low rates, with the (Fig. showing no decline in Bsph resistance lB)' drawback being that .Bti would have to be applied After noting these results, 2O mg/m2 of each agent at least weekly, which would increase treatment was applied in the 10th treatment. This treatment costs. yielded 98Va control initially and only 72Vo control 9 days after treatment (Fig. 1B). The mixture treatments thus far offered no residual activity against Development of resistance in sites treated only Bsplr-resistant mosquitoes, and the results were with BsPh similar to those with Bri alone. To determine if a higher ratio of Bsph in the mix- To provide further information on the speed of ture would generate residual activity beyond that of development of resistance as found in Wat Pikul in Bti alone, a mixture of Bti-Bsph at 2O and 80 mg/ previous studies (Mulla et al. 2001), a small pol- m2, respectively, was applied in the llth and 12th luted water source of mosquitoes (4OO m') at Wat treatments. In both treatments, initial control was Lahan was subjected to repeated treatments with high (95-99Vo), but control 7 days after treatment Vectolex WDG alone at 100 mglm2. In total, 19 was 69 and 6OVo in the llth and 12th treatments' treatments were made (Flg. 2) between May 2001 respectively. No residual control was noted in either and March 2OO2. The larval threshold for repeat treatment at 14 days after treatment. These 2 ex- treatments was set low at 30 larvae/dip. periments indicated that residual activity against The lst 6 treatments yielded approxirnately 997o Bsplr-resistant populations could not be increased control for 7 days and 79Vo control 14 days after substantially by increasing the ratio or dosage of treatment (Fig. 2). Howevel the 7th treatment Bsph rn the mixture. As evidenced in the 13th and yielded a low level of control on day 2 (857o) and rng/m2' l4th treatments with B/i WDG alone at 2O day 7 (l8vo). The 8th treatment yielded excellent the the initial control and longevity were essentially initial control (96Vo) on day 2 and 76Va control on mg/m2 and same as in the mixture with Bsph at 80 day 8. During the following treatments (9th through of Bsph to B/i, even at Bti at 20 rng/m2. Addition 16th), extent of control declined substantially both the dosage of Bri, did not increase residual '7 4 times on day 2 and day after treatment. Likewise, sub- Use of Bti and Bsph at 2Q mg/m2 each activity. stantial residual activity was not noted in any of (15th treatment) yielded 1007o initial control, sim- these 8 treatments (9th through 16th). The 17th and ilar to that with B/i alone at 2O mg/rn2. Addition of l8th treatments exhibited complete failure of con- Bsph to Bti did not improve the extent or longevity trol, because initial control was only 8Vo for the of control. In 1 further treatment (16th), where the 17th treatment. After the 18th treatment, the initial quantity of Bsph in the mixture was increased fur- population density increased 3O7o and then doubled ther (Bti 2O mg/m2, Bsph lOO mg/m'), the level of over the pretreatment level 14 days after treatment. control was similar to that with Bri alone (20 mg/ treatment (19th), the dosage of m2) or with the mixture of Bti-Bsph at l0 mg/m' For the next from lO0 mg/m2 to each. For the lTth treatment, Bti was again applied Vectolex WDG was increased dosage, only at 2OO mglrrP to test for residual activity. This treat- 2OO mgrfi. At this relatively high ment produced l0o7o initial control, but the level 60Vo control on day 2, 55Vo control on day 9, and of control was 92 and 64Va at 7 and 15 days after 297o control on day 15 was achieved. The conclu- treatment, respectively. No significant level of per- sion was made that larval Cx. quinquefasciatus at sistence was noted beyond 7 days (Fig. 1B). Wat Lahan community had developed resistance to For the final treatment (l8th) against Bsplz-resis- Bsph 2352, and therefore further treatments were tant mosquitoes at Wat Pikul, the efficacy of the stopped. JounNu- oF rns AN4pnrcaNMosgurro CoNrnol AssocrerroN Vor-. 19, No. I

lst I Y 175

150

125

100

q 50

'=o z5

o 0 z '7 g t'7 t4 212'1 142 7 t4 212'7 15 2t 282 15 2't 2 'l r2thl roo'! E l3thl l4thl lsthI l6thI o o% E Y Reduction ! .aVla. ^ia t : ;' o 19th o 80 z 2OO mg/m2 nth I lOth .9 L lj: j t8th Y..... o Y. Y 75 a 50 + 25

0 2 83 7 142 '7 72 7 14 212 1 t42 7 14 2t 282 r42 8 6 14372 I 15 Daysposttreatment

Fig' 2' Emergence of resistance in larval Culex quinquefasciatus tteared lg times (May 2001 through March 2002) with Bacillus sphaericus 2362 water dispersible granulei fbrmulation at 100 mg/m, at wat Lahan Community, Bang Bua Thong District, Nonthaburi province, Thailand.

Preventive measures against resistance to Bsph of control was quite low on day 14 after treatment (54 and 46Vo, As shown in the previous section, once Bsph re_ respectively). However, the following treatment (9th) sistance emerges, ]useof Bsph after treatments with yielded excellent control on day 2 (99%) (9lVo) Bti or a mixture of Bti and Bsph is impractical. and day 7 after treatment; this de- Establishment of proactive antiresistance measures clined to 84Vo bttt was still good on day 14 after is more desirable before the widespread tse of Bsph treatment. or other chemical and microbial agents. Based on Although the field treatments at Wat Lahan the previous study and our laborat6ry studies (Za- (Bsph alone) and Soi Jumpa (with mixture) cov- hiri and Mulla, unpublished data), a mixture of ered the same period, 19 treatments were made at VectoBac and Vectolex WDG at 50 me/m2 each the former site compared to g at the latter site. The (total 100 mg/m2) was tested in the lo-w-income reason for fewer treatments at Soi Jumpa was that community of Soi Jumpa, where the total breedins resurgence of larvae here was slower than at Wat habitat was treated repeatedly. Lahan. In Soi Jumpa community, g treatments with a From these studies, mixtures of Bti-Bsph (l:1) mixture of Bti-Bsph were made from May 23, used against -Bspft-susceptiblelarvae did not cause 2001, to January 2OO2(Fig.3). The 1st 6 treatments any noticeable shift in the susceptibility of larval provided 98-lOOVo control 2 days after treatment, Cx. quinquefasciatus at Soi Jumpa. AII treatments 98-99Vo control 7 days after treatment, and up to were equally effective, with good residual activity 86Vo control 14 days after treatment, indicatine re- up to 14 or 2l days after treatment. This is in con- sidual activity for up to 2-3 weeks after treatrient. trast to the Bsph-resistant populations at Wat Pikul, Thereafter, moderate levels of control were where mixtures readily controlled the resistant lar- achieved 21-28 days after trearment (Fig. 3). The vae but provided no residual control. One other as- 7th and 8th treatments produced 96-99Vo control pect that was observed was that over the 8-month on days 2 and'l after treatment; however, the level period of our study at Soi Jumpa, the immature Mlncn 2003 MeNecevr,Nt oF RgsrsreNce ro BA.TLLUSspHAERrcus sv Cutrx eutNeuEFASctATUS 45

"o" o%Reduction 100

aa o 0 62021t421 282 7 t4 2l a E 250- too'-'- q E -f- l st-2ndinstars 200. 80 zo -a- 3rd-4thinstars o0 --t- 150. Pupae

100

2 8 14 21 28 35 42 492 7 142 7 1'4 2l 282 7 14 21 2a r) DaYsPostfieatment

Fig. 3. Prevention of resistance development in Culex quinquefasciatas treated 9 times (May 2001 through January 2002J with mixture of water dispersible granules formulation of Bacillus sphaericus-Bacillus thuringiensis var. israe' lensis (1:l) at 100 mg/m, total in Soi Jumpa Community, Pak Kret District, Nonthaburi Province, Thailand-

populations subjected to treatments with Bti-Bsph velopment of resistance to Bsph are that only 400 mixtures occurred at lower numbers, and never re- m2 of possible mosquito-breeding source was treat- covered to the level prior to the I st treatment. On ed at Wat Lahan and that sources that could not be the basis of these few treatments, the conclusion treated were nearby. Factors such as refugia, size was made that development of resistance will be of area treated, and migration of susceptible genes materially delayed or possibly prevented by admin- into the target area can influence the speed, nature, istering a mixture of Bti-Bsph to susceptible larval and magnitude of emergence of resistance in mos- populations. Howeveq additional field tests and quitoes. Note that resistance to Bsph is a feature of possibly more treatments with mixtures are war- Cx. pipiens complex and no reports of development ranted to document prevention of resistance. At the of resistance to Bsph in other species of mosquitoes present time, adequate information is not available have surfaced to date. that will result on the optimum ratios of Bti-Bsph As a result of these studies, the conclusion is initial and persistent control of in desired level of made that Bsph-resistant mosquitoes can be con- susceptible larvae. trolled with Bti alone or mixtures of Bti and Bsph. The conclusion also is made that use of a mixture CONCLUSIONS of Bti and Bsph against susceptible Cx. quinquefas' ciatus can delay or prevent emergence of resis- In previous and present studies, evidence was although additional studies are needed. The provided that tropical Cx. quinquefasciatus can de- tance, mixture is justified, be- velop resistance rapidly to Bsph 2362. Emergence inclusion of Bsph in the of suscep- of a high level of resistance (> 125,000-fold) at Wat cause it provides longer residual control Pikul (Su and Mulla, unpublished data) was noted tible, but not resistant, mosquito larvae. Prolonging after 5 treatments. This study documents the re- the efficacy of control for 1 or 2 more weeks has emergence of resistance at Wat Lahan after 17 advantages in operational vector control programs treatments. Possible reasons that more treatments at in terms of costly surveillance, monitoring, and ap- Wat Lahan than at Wat Pikul were needed for de- plication of treatments. z+o JounNx- on rnr ANaenrceNMosquno CoNrnol AssocrATroN VoL. 19, No. I

REFERENCES CITED in low-income communities in Nonthaburi Province, Thailand. J Vector Ecol 26l.221-231. Adak T Mittal PK, Rhaguvendra K, Subbarao MA, Ansarr Rao DR, Mani TR, Rajendran R, Joseph AS, Gajanana S, MA, Sharma VP 1995. Resistance to Bacillus sphaer- Reuben R. 1995. Development of a high level of resis- icus in Culex quinquefasciatas Say 1823. Curr Sci 69: tance to Bacillus sphaericus in a field population 695-698. of Culex quinquefasciatus from Kochi, India. J Am Mosq Ali A, Chowdhury MS, Aslam AFMM, Ameen M, Hos- Control Assoc l7:l-5. sain MI, Habiba DB. 2O00. Field trials with Bacillus Rodcharoen J, Mulla MS. 1994. Resistance development sphaericus and Bacillus thuringiensis serovar israelen- in Culex quinquefascianrs {Diptera: Culcidae) to Bacil- sls commercial formulations against Culex quinquefos- lus sphaericus. J Econ Entomol 87:1133-1140. ciatus larvae in suburban Dhaka, Bangladesh. Med En- Rodcharoen J, Mulla MS. 1996. Cross-resistance to Ba- romol Zool 5l:257-264. cillus sphaericus strains in Culex quinquefasciatus. J de Barjac H. 1990a. Characterization prospective and Am Mosq Control Assoc 12;247-250. view of Bacillus thuringiensis israelensis. In: de Barjac Silva-Filha MH, Regis L, Nielsen-LeRoux C, Charles JF H, Sutherland DJ, eds. Bacterial control of mosquitoes 1995. Low-level resistance to Bacillus sphaericus in a and black New Brunswick, NJ: Rutsers /ies Univ. field-treated population of Culex quinquefasciatus (Dip- Press.p l0-15. tera: Culicidae). J Econ Entomol 88:525-530. de Barjac H. 1990b. Classification of Bacillus sohaericus Sindgre G, Babinot M, Quermel JM, Gaven B. 1994. First strains and comparative toxicity to mosquito iaruae. In: field occurrence of Culex pipiens resistance to Bacillus de Barjac H, Sutherland DJ, eds. Bacterial control of sphaericus in southern France. VIIIth European Meet- mosquitoes and blackfiies New Brunswick, NJ: Rutgers ing 1994; Barcelona, Spain. Society of Vector Ecology. Univ. Press. p 228-236. p. 17. de Barjac H, Sutherland DJ, eds. 199O. Bacterial control Singer S. 1990. Introduction to the study of Bacillus of mosquitoes and black New Brunswick, NJ: Rut- flies sphaericus as a mosquito control agent. In: gers Univ. Press. de Bar;ac H, Sutherland DJ, eds. Bacterial control of rnosquitoes Goldberg LJ, Margalit J. 1977. A bacterial spore demon_ and black /ies New Brunswick, NJ: Rutgers Univ. strating rapid larvicidal activity against Anopheles ser_ Press. p 221-227. gentii, Uranotaenia unguiculata, Culex univittatus, Ae- Weiser J. 1984. A mosquito-virulent Bacillus sphaericu.s des ercgypti and Culex pipiens. Mosq News 37:355*35g. in adult Simullium damnosum from northern Niseria. Karch S, Asidi N, Manzambi M, Salaun JJ.1992. Efficacy Zentralbl Mikrobiol I39:57-60. of Bacillus sphaericus against the malaria vector WHO [World Health Organization]. 1985. Informal con- Anopheles gambiae and other mosquitoes in swamps sultation on the development of Bacillus sphaericus as and rice field in Zaire. J Am Mosq Control Assoc g: a microbial larvicide TDR.BCV/sphaericus 376-380. 85.3. WHO/VBC. Geneva, Switzerland: World Health Orea_ Liu El ZhangYM, Cai CJ, Chen ZS. 1999. Developmenr nization. of Bacillus sphaericus C3-41 mosquito larvicidil for_ Wickremesinghe RSB, Mendis CL. 1980. Bacillus sphaer- mulation. Chin J Epidemiol lO:7-1O. icrs spore from Sri Lanka demonstrating rapid larvicid_ Mulla MS. 1990. Activity and field efficacy and use of al activity on Culex quinquefasciatus. Mosq Bacillus thuringiensis News 40: israelensis against mosquitoes. 387-389. In: de Barjac H, Sutherland DJ, eds. Bacterial iontrol Wirth MC, Georgiou GP, Malik JI, Abro GH. 2000. Lab- of mosquitoes and black New Brunswick, NJ: Rut- flies oratory selection for resistance to Bacillus sphaericus gers Univ. Press.p 134-160. in Culex quinquefasciatus tDiptera: Culicidae_; from Mulla MS, Rodcharoen J, Ngamsuk W, Tawatsin pan_ A, California, USA. "/ Med Entomol 37:534-54O. Urai P, Thavara U. 1997. Field trials with Bacillus Yuan ZM, Zhang YM, Cai QX, Liu Ey. 200O. Hish-level sphaericus formulations against polluted water mosqur_ field .fharland. resistance to Bacillus sphaericus C3-41 ii Culex toes in a suburban area of Bangkok, Am J quinquefasciatas from southern China. Biocontol Sci Mosq Control Assoc 13:297-304. Technol lO:41-49. Mulla MS, Su TY, Thavara U, Thwatsin A, Ngamsuk W, Zahiri NS, Su TX Mulla MS. 2002. Strategies for the Pan-Urai P. 1999. Efficiacy of new formulations of the management of resistance in mosquitoes to the micro- microbial larvicide Bacillus sphae ric us against polluted bial control agent Bacillus sphaericus. J Med Entomol water mosquitoes in Thailand.. J Vector Ecol 24:99_ 30:l-9. I 10. Zhang YM, Cai CJ, Liu EY Chen ZS. 1989. Affecting Mulla MS, Thavara U, Tawatsin A, Kong-ngamsuk W' factors and evaluation of Bacillus sphaericus C3-41 for. Chjompoosri J, Su T, 2001. Mosquito larval control mulation on controlling mosquito larvae in fields. Chin with Bacillus sphaericus: reduction in adult populations J Epidemiol l0:2O-25.