193:265-275. 15. Pfannenstiel, M. A., E. J. Ross, V. C. Kramer, and K. W.. Nickerson. 1984. ... 24. Tyrell, D. J., L. 1. Davidson, L. A. Bulla, Jr., and W. A.. Ramoska.
Vol. 48. No. 3
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 1984, p. 665-667 0099-2240/84/090665-03$02.O0/0 Copyright © 1984, American Society for Microbiology
NOTES Oral Toxicity of Bacillus thuringiensis subsp. israelensis to Adult Mosquitoest MARC J. KLOWDEN1* AND LEE A. BULLA, JR.2
Department
of Plant,
Soil and Entomological Sciences' and Department of Bacteriology anid Biochemistry,2 University of Idaho, Moscow, Idaho 83843 Received 7 May 1984/Accepted 19 June 1984
The solubilized entomotoxin of Bacillus thuringiensis subsp. israelensis killed adult male and female mosquitoes of several genera and of various physiological states when it was administered orally. Adult mosquito mortality was further influenced when the preparation was contained in sucrose solution. The potential implication for the control of adult mosquitoes is discussed.
mosquitoes in the field. Although it is unlikely that female adults could acquire a dose of entomotoxin while bloodfeeding, advances in genetic engineering might make it possible either to endow plants with the ability to make these toxins available to nectar-feeding male and female mosquitoes or to attract the insects to bait stations that contain toxin, and thus supplement existing larval control operations. Some investigators have reported that solubilized B. thuiringiensis subsp. israelensis crystals possess a hemolytic activity or a nonspecific cytotoxicity (15, 22) thought to limit the practical use of B. thiuringiensis subsp. israelensis toxin in the field. However, there is no evidence that the hemolytic and entomocidal activities reside in the same component and no reason to believe that these activities could not ultimately be separated. Indeed, B. thloringiensis subsp. kuirstaki possesses its entomocidal activity against lepidopterans in the absence of any hemolytic component (22). In this report, we demonstrate that adult mosquitoes of three genera and of various physiological states can be killed by solubilized crystals of B. thiuringiensis subsp. israelensis acquired orally. B. thluringiensis subsp. israelensis LB2 was derived from an experimental preparation of serotype H-14 obtained from Abbott Laboratories (North Chicago, Ill.). Parasporal crystals were separated from spores and cellular debris by the method of Sharpe et al. (20) and were alkaline solubilized according to Tyrell et al. (23). The suspension was then adjusted to pH 8.0 with 0.5 M Tris-hydrochloride and centrifuged at 10,000 x g for 10 min. The supernatant containing solubilized crystal was dialyzed against insect saline (5) buffered to pH 8.0 with 50 mM Tris-hydrochloride and filter sterilized. Carbonates, which have been shown to influence the toxicity of B. thiuringiensis subsp. israelensis (18), were not present in the saline. Protein concentration was determined by the method of Lowry et al. (14). Larvae of the mosquitoes Aedes aegypti (Linnaeus), Anopheles freeborni Aitken, and Culex quinquefasciatius Say were reared at 27°C under a 14:10 (light:dark) photoperiod on a diet consisting of lactalbumin hydrolysate. brewers' yeast, and finely ground lab chow (1:1:1). Adults were maintained at 27°C and 70% relative humidity, with access to 10% sucrose solution for the first 2 to 3 days after emergence. They were starved for 2 days (1 day with only water and 1 day with neither sucrose nor water) before ingesting solubilized crystals of B. tliiriingiensis subsp. israelensis.
Health and environmental considerations have limited the of some of our most effective chemicals for insect control. As a result, attention has been directed towards biological agents, including bacterial, viral, and fungal pathogens. Among these pathogens, the crystalliferous spore-forming bacterium Bacilllus thuringiensis subsp. israelensis is particularly effective against larval mosquitoes and black flies, yet does not affect the many nontarget species against which it has been tested (reviewed in reference 2). Its toxicity to mosquitoes resides in a toxin contained within the parasporal crystal that is produced during sporulation (17, 23). The conventional method for screening isolates of mosquito entomopathogens suspends larvae in water containing a known concentration of the agent. However, this assay relies on the ability of the insect to ingest the pathogen, and its outcome often depends upon the instar of larvae that are used (3, 13, 25, 28), whether the larvae feed along the surface of the water or at the bottom (3, 7, 21), and upon the formulation of the preparation (26). Assays performed in the field are influenced by the amount of particulate matter in the water that may compete with the pathogen for uptake (16, 26). Furthermore, we have found that the larval bioassay is much less sensitive with solubilized toxins (11), perhaps because of the difficulty that filter-feeding larval mosquitoes have in ingesting solubilized substances. Consequently, we looked for a bioassay that could better evaluate solubilized entomotoxins and developed a method of administering material to adult female mosquitoes via an anal route (11). The procedure, commonly referred to as an "enema" (1), allowed us to expose target cells in the midgut of the insect to measured amounts of agent and to derive a 50% lethal dose rather than the 50% lethal concentration that is only possible with the conventional larval bioassay. This method of bioassay, using adults, is independent of the feeding behavior of the insect and also of the formulation of the preparation. Our earlier experiments (11) provided the first indication that adult mosquitoes, as well as larvae, were susceptible to B. thuringiensis subsp. israelenisis. This susceptibility of adults may have a potential application for the control of use
*
Corresponding author.
t Contribution Station.
no.
8464 from the Idaho Agricultural Experiment
665
666
NOTES
APPL. ENVIRON. MICROBIOL.
Unanesthetized mosquitoes were individually manipulated, using a vacuum probe placed against the thorax; they were observed under a dissecting microscope
as
TABLE 1. Solubilized B. thuringiensis subsp. isru-elensis crystal (1.36 ,ug of protein) fed to mosquitoes in saline solution
they ingested 1 p.l
of a 1.36-[.g/[tl solution of solubilized B. thiuringiensis subsp. isruaelensis crystal protein from a microliter syringe (Fig. 1). When the entire drop was ingested, the mosquitoes were released from the probe and maintained at 271C for 48 h. Solubilized crystal preparations of B. tlhlur ingiensis subsp. israelensis killed both male and female adult mosquitoes when administered per os, but when the same preparations were heated for 5 min, they lost their activity (Table 1). Anoplieles Jieeborni adults were significantly more susceptible than the other two species, although they have generally been reported to be less susceptible to B. thitr-ingiensis subsp. isrulelentsis formulations in larval bioassays (3, 7, 24). This apparent discrepancy in the susceptibility of larvae and adults may be a result of the feeding behavior of the larvae; unlike Aedes and Clilex species, Anopheles freeborni larvae generally feed at the water surface and therefore may ingest less of a formulation in the conventional bioassay if it sinks to the bottom of the water. We believe that the susceptibility of A noplieles .iteeborni larvae to B. thluinigienIsis subsp. is)aileinsis might be comparable to the other mosquito species if all larvae ingested the same amount of preparation. Older mosquitoes are more important epidemiologically because of their increased opportunity to acquire and transmit pathogens. Considerable evidence indicates that these older females are also physiologically different from their younger counterparts (12, 12a). For these reasons, we also fed solubilized B. tliitrinsgiensis subsp. israelensis crystal to 21-day-old females of the three mosquito species. Although there was also higher mortality in control Aedes (legvpti and Anopheles fieebor-ni apparently as a result of aging alone, the older females were significantly more susceptible than
Mosquito species
% Mortality (n) at 48 h
Controls'
Experimentals
Aedes aegypti Males 4-day-old females 21-day-old females
0 (60) 0 (66) 27.5 (40)
55.5 (146) 37.5 (80) 70.8 (120)
Anopheles freeborni 4-day-old females 21-day-old females
2 (50) 25.0 (20)
63.6 (140) 88.0 (60)
Culex qiuinquefasciatius 41.3 (80) 6.6 (30) 4-day-old females 0 (8) 85.0 (20) 21-day-old females "Controls were fed identical preparations that were inactivated by heating at 100°C for 5 min.
the 4-day-old insects (Table 1). Thus, the most important target group within a mosquito population also has the greatest susceptibility to B. thiuringiensis subsp. israuelensis toxin. A blood meal is directed into the midgut of the female mosquito, but carbohydrate meals are stored in a chitinized ventral diverticulum and are slowly passed into the midgut for digestion (4, 9). To simulate the ingestion of toxin in floral nectars that may contain 3 to 81% carbohydrates (8, 10). we next fed mosquitoes an identical dose of B. thuiringiensis subsp. isru-elensis as before, but in 10% sucrose solution. This caused less mortality in Aedes alegypti females than when saline was the solvent (Tables 1 and 2) and suggested that the toxic meal was being sequestered in the diverticulum. Indeed, feeding solubilized crystals in saline to Aedes (cegvpti females caused less mortality than when the same dose was administered directly to midgut cells as an enema (11), indicating that even some of the saline meal may be sequestered. However, in contrast to our results with Aedes aiegypti, there was significantly greater mortality in C. quinqlufes iatus and Anopheles Jreebor-ni when the toxin was ingested in sucrose than when it was contained in saline (Tables 1 and 2). In blow flies, crop emptying is regulated by hemolymph osmotic pressure (6); starvation results in fewer dissolved nutrients in the hemolymph and, hence, a lowered osmotic pressure that hastens crop emptying. If a similar mechanism operates in mosquitoes, the reduced mortality in Aedes aegvpti may be a result of sugar and toxin remaining in the diverticulae longer than they do in C. quinquefjasc iatis or Anoplieles fi-eeborni. Alternatvely, the foreguts of these latter two species may be more permeable to toxin than that of Aedes aegvpti, allowing it to escape into the hemocele. We had previously suggested that adult mosquitoes be used to assay solubilized entomotoxins administered by enema because the results are less likely to be influenced by TABLE 2. Solubilized B. thuringiensis subsp. israelensis crystal (1.36 ,ug of protein) fed to 4-day-old female mosquitoes in 10% sucrose solution
s% Mortality (n) at 48 h
Mosquito species
FIG. offered
Adult
1. a
mosquito
measured
solubilized
crystal.
volume
held
with
of B.
a
vacuum
tlIiituigienisis
probe while it is subsp. isi-aelensis
Aedes aegypti Anopheles freeborni Culex quinquefasciatus
Controls'
Experimentals
3 (87) 0 (40) 0 (36)
9 (87) 79 (53) 72 (58)
aControls were fed identical preparations that were inactivated by heating at 100°C for 5 min.
VOL. 48. 1984
the feeding behavior of the insect, as they are in the conventional larval bioassay (11). Although in this report we have demonstrated that oral ingestion is also effective, the solvent the toxin is contained in apparently can influence its effectiveness. Therefore, when screening entomotoxins, we continue to recommend that the material be introduced as an enema, so that the target cells in the midgut are exposed to a known dose. Extreme care should be exercised to prevent the accidental injection of B. thlulringiensis subsp. israelensis material into humans (27). Service (19) pointed out that, even when larval mortalities are high, the numbers of emerging adult mosquitoes may often be large enough to constitute a problem. Although the focus of biological control efforts has been on the larval mosquito stage, the oral susceptibility of adult mosquitoes to B. thiuringiensis subsp. israelensis and the potential of genetic engineering for making natural entomocidal products available to adult mosquitoes in novel ways may offer new strategies for the control of these insect pests. We thank B. E. Wilton and J. M. Hurley for excellent technical assistance. This investigation was supported in part by Public Health Service grant Al-19009 from the National Institutes of Health to M.J.K. and grant PCM 7907591 from the National Science Foundation to L.A.B. LITERATURE CITED 1. Briegel, H., and A. 0. Lea. 1975. Relationship between protein and proteolytic activity in the midgut of mosquitoes. J. Insect Physiol. 21:1597-1604. 2. Burges, H. D. 1982. Control of insects by bacteria. Parasitology 84:79-117. 3. Dame, D. A., K. E. Savage, M. V. Meisch, and S. L. Oldacre. 1981. Assessment of industrial formulations of Bacillus tlihuringietisis subsp. isrueletisis. Mosq. News 41:540-546. 4. Day, M. F. 1954. The mechanism of food distribution to midgut or diverticula in the mosquito. Aust. J. Biol. Sci. 7:515-524. 5. Ephrussi, B., and G. W. Beadle. 1936. A technique of transplantation for Drosophila. Am. Nat. 70:218-225. 6. Gelperin, A. 1966. Control of crop emptying in the blowfly. J. Insect Physiol. 12:331-345. 7. Goldberg, L. J., and ,I. Margalit. 1977. A bacterial spore demonstrating rapid larvicidal activity against Aniopleles ser8.
9. 10. 11.
12.
gentii, Uranotaenia unguicldata. Cidle.x uniitattius. Aedes (aegvpti. and Ciulex pipienis. Mosq. News 37:355-358. Hocking, B. 1968. Insect flower associations in the higher arctic with special reference to nectar. Oikos 19:359-387. Hosoi, T. 1954. Mechanism enabling the mosquito to ingest blood into the stomach and sugary fluids into the oesophageal diverticula. Annot. Zool. Jpn. 27:82-90. Kleinschmidt, M. G., A. K. Dobrenz, and V. A. McMahon. 1968. Gas chromatography of carbohydrates in alfalfa nectar. Plant Physiol. 43:665-667. Klowden, M. J., G. A. Held, and L. A. Bulla, Jr. 1983. Toxicitv of Bacillus tluuiiin i,ieisis subsp. israclensis to adult Aedes aegypti mosquitoes. AppI. Environ. Microbiol. 46:312-315. Klowden, M. J., and A. 0. Lea. 1980. "Physiologically old" mosquitoes are not necessarily old physiologically. Am. J. Trop. Med. Hyg. 29:1460-1464.
NOTES
667
12a.Klowden, M. J., and A. 0. Lea. 1984. Blood-feeding affects agerelated changes in the host-seeking behavior of Aedes aegypti (Diptera: Culicidae) during oocyte maturation. J. Med. Entomol. 21:274-277. 13. Lacey, L. A., and S. L. Oldacre. 1983. The effect of temperature. larval age. and species of mosquito on the activity of an isolate of Baceillius thluringiensis var. da-rm .sstadien.sis toxic for mosquito larvae. Mosq. News 43:176-180. 14. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. 15. Pfannenstiel, M. A., E. J. Ross, V. C. Kramer, and K. W. Nickerson. 1984. Toxicity and composition of protease-inhibited Bacillus tliiuringienisis var. isr-aele,isis crystals. FEMS Microbiol. Lett. 21:39-42. 16. Ramoska, W. A., S. Watts, and R. E. Rodriguez. 1982. Influence of suspended particulates on the activity of Bacillus thluringiensis serotype H-14 against mosquito larvae. J. Econ. Entomol. 75:1-4. 17. Samasanti, W., S. Pantuwatana, and A. Bhumiratana. 1982. Role of the parasporal body in causing toxicity of Bacillus tlhuiringiensis toward Aedes aegvpti larvae. J. Invert. Pathol.
39:41-48. 18. Schnell, D. J., and K. W. Nickerson. 1983. Toxicity of Bacillus tlhluringiensis var. israelensis crystals to Aedes aegypti larvae: carbonate reversal. Appl. Environ. Microbiol. 45:1691-1693. 19. Service, M. W. 1983. Biological control of mosquitoes-has it a future'? Mosq. News 43:113-120. 20. Sharpe, E. S., K. W. Nickerson, L. A. Bulla, Jr., and J. N. Aronson. 1975. Separation of spores and parasporal crystals of Bacillus tlhiuringiensis in gradients of certain X-ray contrasting agents. AppI. Microbiol. 30:1052-1053. 21. Sun, C.-N., G. P. Georghiou, and K. Weiss. 1980. Toxicity of Bacillus thllitngien.sis subsp. israelenisis to mosquito larvae variously resistant to conventional insecticides. Mosq. News 40:614-618. 22. Thomas, W. E., and D. J. Ellar. 1983. Bacillus thlrin,£itgientsis var. is-aelen.sis crystatl 8-endotoxin: effects on insect and mammalian cells in litro and in ivao. J. Cell Sci. 60:181-197. 23. Tyrell, D. J., L. A. Bulla, Jr., R. E. Andrews, Jr., K. J. Kramer, L. I. Davidson, and P. Nordin. 1981. Comparative biochemistry of entomocidal parasporal crystals of selected Bacillus tluhuinigicu.sis strains. J. Bacteriol. 145:1052-1062. 24. Tyrell, D. J., L. 1. Davidson, L. A. Bulla, Jr., and W. A. Ramoska. 1979. Toxicity of parasporal crystals of Bacillus tlhrigtlgietnsis subsp. isrulelensis to mosquitoes. Appl. Environ. Microbiol. 38:656-658. 25. Van Essen, F. W., and S. C. Hembree. 1980. Laboratory bioassay of Bacillu.s tluhirin-igiensis subsp. israclenisis against all instars of Aedes aegvpti and Aedes taeniorhvnchus larvae. Mosq. News 40:424-431. 26. Van Essen, F. W., and S. C. Hembree. 1982. Simulated field studies with four formulations of Bacillus tlhiin-itgiensis var. israilen.sis against mosquitoes: residual activity and effect of soil constituents. Mosq. News 42:66-72. 27. Warren, R. E., D. Rubenstein, D. J. Ellar, J. M. Kramer, and R. I. Gilbert. 1984. Bacillus tlhiuriitngienisis var. israclensis: protoxin activation and safety. Lancet i:678-679. 28. Wraight, S. P., D. Molloy, H. Jamnback, and P. McCoy. 1981. Effects of temperature and instar on the efficacy of Bacillus thuringiensis var. israelensis and Bacillus sphaericius strain 1593 against Aedces stimulans larvae. J. Invert. Pathol. 38:78-87.
