Toxicity and Bioaccumulation of Fipronil in the Nontarget Arthropodan Fauna Associated with Subalpine Mosquito Breeding Sites P

Home , Chironomus Annularius, Midge

Ecotoxicology and Environmental Safety 52, 8}12 (2002) Environmental Research, Section B doi:10.1006/eesa.2002.2166, available online at http://www.idealibrary.com on

Toxicity and Bioaccumulation of Fipronil in the Nontarget Arthropodan Fauna Associated with Subalpine Mosquito Breeding Sites P. F. Chaton, P. Ravanel, M. Tissut, and J. C. Meyran Laboratoire Ecosyste` mes et Changements Environnementaux, Centre de Biologie Alpine, UniversiteH Joseph Fourier, BP 53, F-38041 Grenoble Cedex 9, France

Received March 13, 2001

species resistant to other insecticides. At present, "pronil In order to examine ecological impact of 5pronil use for larval appears to be one ofthe most e !ective synthetic insecticides culicine control in natural hydrosystems, toxicity and bioac- for mosquito larvae control (Chaton et al., 2001). Bioac- cumulation of this new insecticide were analyzed on aquatic cumulation studies have indicated a low direct bioavailabil- species representative of the nontarget arthropodan fauna (non- ity in the aquatic environment, in addition to a low-rate culicine larval Diptera:Chaoboridae, Chironomidae; planktonic integumental route for penetration into larvae. Moreover, Crustacea:Cladocera, Copepoda, Ostracoda) associated with importance ofthe digestive route forinsecticide accumula- target larval mosquito populations in the subalpine breeding tion into larvae has suggested that ingestion might be the sites. Standard toxicological bioassays using 5pronil aqueous best way of "pronil administration to mosquito larvae solutions from 1 to 2000 nM indicated di4erent sensitivity levels among species. Insecticide bioaccumulation analyses, using (Chaton et al., 2001). [14C]5pronil solutions in simpli5ed laboratory ecosystem, also However, before considering the operational use of "p- indicated large di4erences among species. These di4erences may ronil for reducing target populations of larval Culicidae, come from biological parameters characteristic of each species. some ecotoxicological information on this relatively new Taking into account these nontarget e4ects of 5pronil, a possible insecticide against the nontarget arthropodan fauna is re- strategy of use of this insecticide for integrated mosquito control quired. This investigation is currently being carried out on management was proposed, which is based upon selective dietary planktonic crustacean and nonculicine larval dipteran spe- absorption of the insecticide by larval Culicidae. cies. Planktonic Crustacea represent an important part of 2002 Elsevier Science (USA) the "ltering fauna in the subalpine mosquito breeding sites Key Words: 5pronil; biocidal activity; bioaccumulation; aqua- (TeH tard, 1974; Amoros, 1984). Among nonculicine larval tic environment; Diptera; Crustacea. dipteran species, Chaoboridae are predators (Durieu and Thomas, 1995), thus involved in a food chain including larval mosquitoes (Rey et al., 2000a). Chironomidae, which INTRODUCTION are sediment-dwelling organisms, are involved in recycling organic detritus (Ali, 1980) and are major items for numer- Among the new chemical insecticides likely to be widely ous aquatic predators (see references in Dickman, 2000). used in the environment, "pronil is a promising phenyl- Such a comparative analysis ofthe nontarget e !ects of pyrazole compound (Colliot et al., 1992) because ofits "pronil will allow understanding ofthe biological para- unique toxicity mechanism (Buckingam et al., 1994; Cole meters involved in the toxicity ofthis insecticide in aquatic et al., 1993). Due to the e!ect of "pronil being highly speci"c organisms, and adjustment ofa new possible selective insec- to the invertebrate -aminobutyric acid (GABA)-receptor ticide strategy usable in the aquatic medium. (Hainzl and Casida, 1996), it was used primarily in the terrestrial environment against insect pests (Gaulliard, 1996; Martin et al., 1996; Balania and De Visscher, 1997; Valles MATERIALS AND METHODS et al., 1997). In the aquatic environment, preliminary evalu- Test Organisms ations in laboratory (Ali et al., 1998) and in the "eld (Ali et al., 1999; Sulaiman et al., 1997) have suggested a possible Test species analyzed here originated from di!erent sub- use of "pronil against larval culicine and chironomid midge alpine mosquito breeding sites known for the abundance of their nontarget arthropodan populations (Pautou, 1981; To whom correspondence should be addressed. Fax: 33 4 76 51 44 63. Rey et al., 2000a). Among the Crustacea, three ecologically E-mail: [email protected]. di!erent subclasses were chosen and the most representative

0147-6513/02 $35.00 2002 Elsevier Science (USA) All rights reserved. NONTARGET EFFECT OF FIPRONIL IN MOSQUITO BREEDING SITES 9 species ofeach subclass were studied (Cladocera: Daphnia TABLE 1 pulex de Geer; Copepoda: Acanthocyclops robustus Jurine, Di4erential Sensitivity to Fipronil among Species of the Non- Diaptomus castor Jurine; Ostracoda: Eucypris virens Jurine). target Arthropodan Fauna:Log Probit Determination of LC 50 Among the larval nonculicine dipteran fauna, Chironomus (Median Lethal Concentrations) after 48 Hours of Treatment annularius Meigen (Chironomidae) and Chaoborus crystal- 48-h LC linus de Geer (Chaoboridae) were selected. Immediately Species n Slope$SE (nM) 95% CL after collecting, specimens were determined and then subjected to bioassay and bioaccumulation processing. Crustacea Cladocera Daphnia pulex 900 Insensitive Bioassay Protocol Copepoda Acanthocyclops robustus 900 4.8$0.1 194.2 143.9}262.1 Tolerance to "pronil was analyzed by standard bioassay Diaptomus castor 600 4.7$0.1 7.9 6.4}9.9 technique (WHO, 1981) adapted to planktonic Crustacea Osracoda following Rey et al. (2000b) and to the larval nonculicine Eucypris virens 900 Insensitive dipteran fauna following Rey et al. (1998). Bioassays were Diptera conducted in triplicate on samples of20 same-sized speci- Chaoboridae Chaoborus crystallinus 600 1.5$0.3 1478.5 1056.2}2096.3 mens placed in disposable plastic vials containing 20 ml Chironomidae solution without food. Controls were reared in de- Chironomus annularius 300 1.2$0.4 5.6 3.6}24.5 chlorinated tap water. Treated specimens were reared in aqueous solutions with di!erent concentrations (i.e., 1, 10, Note. n"total number ofspecimens assayed. 25, 50, 100, 250, 500, 1000, 1500, and 2000 nM) of "pronil (99.9% purity, Aventis Crop Science, La Dargoire, Lyon, France). These concentrations were obtained by serial dilution in tap water ofa 0.1 M initial ethanol stock solution. water. Fipronil samples and animal homogenates were ana- Bioassay medium for chironomid larvae included 5 g of lyzsed by liquid scintillation counting, using a 1414 Win- sterilized sand (Prolabo, granulometry from 150 to 210 m) spectral (EG & G Wallac) and Ready Safe scintillation #uid in order to prevent larval cannibalism (Ali et al., 1998). The (Beckman). Each experiment was repeated three times. absence ofinsecticide loss to the sand was checked using [C]"pronil. RESULTS The mortality was recorded 48 h after the beginning of Diwerential Toxicity of Fipronil the treatment. After correction by Abbott's formula (Ab- bott, 1925), dose}mortality data were subjected to log- Table 1 reveals large di!erences in tolerance levels to probit analysis (Raymond, 1985). "pronil among species after 48 h of treatment. Among Crus- tacea, D. pulex and E. virens are insensitive, even to concentrations close to the solubility limit of "pronil in water. Bioaccumulation Studies Strong di!erences in tolerance exist between species within Fipronil accumulation was studied on 800 specimens for the same subclass (e.g., within Copepoda: between A. robust- each species. Each assay was performed for 48 h on 20 us, less sensitive, and D. castor, more sensitive). specimens placed in plastic vials containing 20 ml ofwater Larval nonculicine dipteran species appear either sensi- solutions of[ C]"pronil at concentrations used for the tive, such as C. annularius, or strongly insensitive, such as toxicological bioassays. These radiolabeled "pronil solu- C. crystallinus, which shows a larval mortality ofmerely tions were obtained by serial dilution in tap water of 68.7% after 48 h of treatment for the maximum "pronil a 4.35 mM [C]"pronil ethanol solution (Aventis Agricul- concentration used. ture Ltd, Ongar, UK) (speci"c activity, 947.2 MBq mmol\; radiopurity, 97}99%). [C]Fipronil Accumulation After the 48 h treatments, accumulation of [C]"pronil within the treated organisms was measured and the accu- After the 48 h treatments, the AF varied according to the mulation factor (AF) was calculated. The AF corresponds to species (Fig. 1). Among Dipteran larvae, the AF was less the ratio obtained at the end ofthe experiment between the than 1 for C. crystallinus and more than 60 for C. annularius. [C] concentration per gram fresh weight of animal and Among Crustacea, the AF was close to 10 for E. virens, close the [C] concentration per milliliter ofmedium. A 1-ml to 30 for D. pulex and D. castor, and more than 40 for A. sample ofthe "pronil solution was taken and the whole robustus. No clearcut relationship was observed between treated animal sample was rinsed in 2 ml distilled water, AF and LC, except for C. crystallinus, where a very low weighed, ground, and then homogenized in 5 ml distilled AF might explain the high LC value. 10 CHATON ET AL.

FIG. 1. Comparison of "pronil toxicity (expressed as 48-h LC, nM) and bioaccumulation (expressed as AF) among nontarget arthropodan species (Crustacea: D. pulex, A. robustus, D. castor, E. virens; larval Diptera: C. crystallinus; C. annularius).

DISCUSSION Crustacean species can accumulate in vivo large amounts of "pronil, regardless oftheir sensitivity to the insecticide. Fipronil shows di!erent toxic e!ects against the nontar- This strong bioaccumulation ability may come from a mass- get arthropodan fauna of the subalpine mosquito breeding ive insecticide dietary ingestion, probably enhanced by the sites, with various patterns ofbioaccumulation. These non- water-"ltering feeding mode of these small-sized organisms. target e!ects question the possible "eld use ofthis new The high ability ofresistant species to bioaccumulate "p- phenylpyrazole insecticide against larval mosquitoes in ronil without toxicity (e.g., D. pulex) suggests that they aquatic ecosystems. possess either a resistant GABA-gated chlorine channel with a speci"c structure (Cole et al., 1993; Liegeois, 1998) or high-rate preexisting detoxifying equipment. Fipronil may Diwerential Ewects of Fipronil on the Nontarget Fauna also accumulate within this species in especially large apo- As summarized in Fig. 1, the correlation between toxicity plastic spaces (e.g., fat reserves or cuticle constituents). and bioaccumulation of "pronil varies from one species to Larval nonculicine Diptera show large di!erential resist- another, suggesting di!erent patterns ofinsecticide bio- ance to "pronil. Chironomidae were known to be sensitive availability. to this insecticide (Ali et al., 1998) as well as to other The di!erential sensitivity to "pronil observed for Crusta- chemical insecticides (reviewed in Morrill and Neal, 1990) cea resembles that observed for those nontarget species and also to bacterio-insecticides (reviewed in Dickman, treated with numerous chemical insecticides (reviewed in 2000). The high "pronil AF value obtained with C. an- Milam et al., 2000; Xue et al., 2000), and bacterio-insecti- nularius suggests that "pronil accumulation occurs even cides (reviewed in Rey et al., 1998). The harmlessness of after the death of the animal and is therefore a passive "pronil against D. pulex is congruent with that previously process. However, our results, obtained in simpli"ed labor- reported for other Daphniidae species (e.g., D. magna: atory conditions, may be poorly indicative ofthe "eld biolo- Rho( ne Poulenc, 1996). More generally, cladoceran Daph- gical reality, as observed for conventional insecticides niidae appear to be relatively resistant to numerous (Charbonneau et al., 1994). Numerous factors may interfere xenobiotics (Rey et al., 1998; Brown et al., 1999), contrary to with the insecticide bioavailability for chironomid larvae in Copepoda which are also sensitive to "pronil. the aquatic medium, e.g., low "pronil solubility in water, NONTARGET EFFECT OF FIPRONIL IN MOSQUITO BREEDING SITES 11 insecticide partition between the environmental water and means ofinsecticide bioaccumulation in mosquito larvae, substrate, and insecticide accessibility to larvae. Bioavaila- which may be enhanced by using appropriate food (Chaton bility characteristics of "pronil for chironomid larvae may et al., 2001). Under such conditions, researching on a selective be particularly important, owing to the basic role of absorption of "pronil by target larval mosquitoes through chironomid larvae in recycling organic matter (Ali, 1980). feeding together with avoiding di!usion ofthe insecticide The impact of "pronil on these species may have indirect molecules throughout the aquatic medium is ofinterest. e!ects on higher trophic levels, as also reported for other An adjustment ofa selective insecticide feedingstrategy chemical insecticides (Morrill and Neal, 1990; Harkey and against mosquito larvae may be possible as nontarget arth- Klaine, 1992). ropodan species show various diets, most ofthem di !ering Chaoboridae larvae appear to be strongly resistant to from those of larval Culicidae. However, the harmlessness of "pronil, as also observed for other conventional insecticides this dietary strategy may not be absolute, because ofthe (reviewed in Rey et al., 1998). The low toxicity of "pronil nonnegligible toxicity of "pronil on the nontarget fauna. associated with the low AF measured within C. crystallinus This problem also exists for other existing chemical or larvae may have an origin di!ering from that observed for biological insecticides. resistant Crustacea. The biomass ofChaoboridae larvae is far higher than that of Crustacea. Moreover, the integumen- ACKNOWLEDGMENTS tal route ofinsecticide penetration into these larvae may be ofvery low importance, much less than that observed in We thank Aventis Crop Science, La Dargoire, Lyon, France, for provid- larval mosquitoes (Chaton et al., 2001). This particular ing technical "pronil and Aventis Agriculture Ltd (Aventis), Ongar, United Kingdom, for providing [C]"pronil. We also thank Entente InterdeH par- performance of larval Chaoboridae against "pronil may be tementale pour la DeH moustication Ain-Ise`re-Rho( ne-Savoie for supplying reinforced by their active predatory diet (Durieu and some crustacean "eld strains, and J. Patouraux for technical assistance. Thomas, 1995), which contrasts with the passive suspension feeding habits of planktonic Crustacea and with the detri- REFERENCES tivorous diet oflarval Culicidae (Clements, 1992). Our results obtained with C. crystallinus strongly suggest that the Abbott, W. S. (1925). A method ofcomputing the e!ectiveness ofan partition equilibrium between water and larval body could insecticide. J. Econ. Entomol. 18, 265}267. not be reached after 48 h through integumental exchange. Ali, A. (1980). Nuisance chironomids and their control: A review. Bull. Insecticide penetration therefore seems to be obtained only Entomol. Soc. Am. 26, 3}16. through feeding on "pronil, contaminated prey. Ali, A., Chowdhury, M. A., Hossain M. I., Ameen, M. U., Habiba, D. B., and Aslam, A. F. M. (1999). Laboratory evaluation ofselected larvicides In summary, three types ofcomportments can be ob- and insect growth regulators against "eld-collected Culex quinquefas- served among the species studied. (1) In the case of ciatus larvae from urban Dhaka, Bangladesh. J. Am. Mosq. Contr. Assoc. A. robustus and D. castor as well as oflarval C. annularius 15, 43}47. and Aedes aegypti (Chaton et al., 2001), "pronil penetration Ali, A., Nayar, J. K., and Gu, W. D. (1998). Toxicity ofa phenylpyrazole from the water solution into the body is easy, allowing the insecticide, "pronil, to mosquito and chironomid midge larvae in the toxic e!ect to appear at a low insecticide concentration. laboratory. J. Am. Mosq. Contr. Assoc. 14, 216}218. Di!erences in the LC obtained among those species may Amoros, C. (1984). Introduction pratique a` la systeH matique des organismes des eaux continentales franiaises: 5-CrustaceH s Cladoce`res. Bull. Mens. result from speci"c biodistribution patterns, or various me- Soc. ¸inn. ¸yon 53, 72}143. tabolization rates, or even from speci"c sensitivity ofthe Balania, G., and De Visscher, M. N. (1997). Impacts on nontarget insects of target. (2) In the case of D. pulex and E. virens, "pronil a new insecticide compound used against the desert locust (Chistocerca accumulates in the body without any toxic e!ect. Such gregaria, Forskal 1775). Arch. Contam. ¹oxicol. 32, 58}62. a tolerance ofthese species may be due to target resistance, Brown, M. D., Thomas, D., Mason, P., Greenwood, J. G., and Kay, B. H. high metabolization rate, or segregation ofthe insecticide in (1999). Laboratory and "eld evaluation ofthe e$cacy offourinsecticides apoplastic spaces. (3) In the case of C. crystallinus,no"p- for Aedes vigilax (Diptera: Culicidae) and toxicity to the nontarget shrimp ¸eander tenuicornis (Decapoda: Palaemonidae). J. Econ. ronil accumulation occurs, as also observed with other Entomol. 92, 1045}1051. insecticides (Rey et al., 1998). Buckingam, S. D., Hosie, A. M., Roush, R. T., and Sattelle, D. B. (1994). Actions ofantagonists and convulsant antagonists on a Drosophila melanogaster GABA receptor (Rdl) homooligomer expressed in Xenopus Selective Dietary Strategy of Use of Fipronil in Larval oocytes. Neurosci. ¸ett. 181, 137}140. Mosquito Control Charbonneau, C. R., Drobney, D., and Rabeni, C. F. (1994). E!ects of Because ofthe nontarget e !ects of "pronil associated with Bacillus thuringiensis var. israelensis on non target benthic organisms in a lotic habitat and factors a!ecting the e$cacy ofthe larvicide. Environ. the particular behavior in the aquatic medium ofthis insec- ¹oxicol. Chem. 13, 267}279. ticide that is poorly soluble in water, the direct use of Chaton, P. F., Ravanel, P., Meyran, J. C., and Tissut, M. (2001). Toxicolo- "pronil for larval mosquito control appears to be hazard- gical e!ects and bioaccumulation of "pronil in larval mosquito Aedes ous. However, ingestion has been evidenced to be the main aegypti in aqueous medium. Pestic. Biochem. Physiol. 69, 183}188. 12 CHATON ET AL.

Clements, A. N. (1992). ¹he Biology of Mosquitoes, Vol. 1. Chapman inondeH speH riodiquement: L'exemple des CulicideH s. Rapport pour le & Hall, London. Ministe%re de l1Environnement, Comite& Faune-Flore, Vol. 2, pp. 1}16. Cole, L. M., Nicholson, R. A., and Casida, J. E. (1993). Action ofphenyl- Raymond, M. (1985). PreH sentation d'un programme Basic d'analyse log- pyrazole insecticide at the GABA-gated chloride channel. Pestic. probit pour micro- ordinateur. Cahiers ORS¹OM Se&r. Entomol. Me&dic. Biochem. Physiol. 46, 47}54. Parasitol. 23, 117}121. Colliot, F., Kukorowski, K. A., Hakins, D. W., and Roberts, D. A. (1992). Rey, D., Long, A., Pautou, M. P., and Meyran, J. C. (1998). Comparative Fipronil: A new soil and foliar broad spectrum insecticide. In Pro- histopathology ofsome Diptera and Crustacea ofaquatic alpine ecosys- ceedings, Brighton Crop Protection Conference: Pests and Diseases, tems after treatment with Bacillus thuringiensis var. israelensis. Entomol. pp. 29}34. British Crop Protection Council, Farnham, England. Exp. Applic. 88, 255}263. Dickman, M. (2000). Impact ofa mosquito selective pesticide, Bti, on the Rey, D., Martins, D., David, J. P., Pautou, M. P., Long, A., Marigo, G., and macroinvertebrates ofa subtropical stream in Hong Kong. Chemosphere Meyran, J. C. (2000a). Role ofvegetable tannins in habitat selection 41, 209}217. among mosquito communities from the Alpine hydrosystems. C.R. Durieu, M., and Thomas, A. (1995). Le reH gime alimentaire de stades Acad. Sci. Paris Sci.