Mem Inst Oswaldo Cruz, Rio de Janeiro, 96(7) October 2001
Identification of Entomopathogenic Bacillus Isolated from Simulium (Diptera, Simuliidae) Larvae and Adults
Departamento de Bacteriologia
*Departamento de Entomologia, Instituto Oswaldo Cruz-Fiocruz, Av. Brasil 4365, 21045-900 Rio de Janeiro, RJ, Brasil
Entomopathogenic bacteria isolated fromu00a0Simuliumu00a0larvae and adults from breeding sites in the states of São Paulo and Rio de Janeiro, Brazil, were identified as 18 strains ofu00a0Bacillus thuringiensisu00a0and one ofu00a0B. sphaericus. Most of these strains were serotyped according to their flagellar antigens. However, nine of theu00a0B. thuringiensisu00a0samples, could not be serotyped and were designated as "autoagglutinating"; they were also shown to be toxic in preliminary tests against Aedes aegyptiu00a0larvae. Additionally,u00a0B. sphaericusu00a0was also shown to be toxic towardsu00a0Culex quinquefasciatusu00a0larvae.
Simuliids are hematophagus insects naturally occurring in waterways throughout Brazil. Although some species have been involved in the transmission of human onchocerciasis in other regions (Gerais & Ribeiro 1986, Maia-Herzog et al. 1999) the main importance of these insects in the southeast of Brazil is socio-economic. The frequent and intensive attacks by simuliids on the transitory population in this area can reduce tourism in summer months, a fact that could negatively impact the economies of affected municipalities. To decrease the presence of these insects in these areas, a full-scale programme, based on Bacillus thuringiensis serovar israelensis, has been carried out in the north littoral zone of the State of São Paulo since 1990 (Araújo-Coutinho 1995).
Recent findings of B. thuringiensis serovar oswaldocruzi and B. thuringiensis serovar braziliensis from unusual habitats (Rabinovitch et al. 1995) encourage the continued search for other endemic entomopathogens. For that purpose,Simulium larvae and adults from the north littoral zone of São Paulo and some other rivers in the State of Rio de Janeiro were evaluated for the presence of entomopathogenic bacteria.
MATERIALS AND METHODS
Collection of simuliids - Black fly larvae and adults were collected from breeding sites in Pau d'Alho river, State of São Paulo. This river was divided into two regions, one downstream of B. thuringiensis serovar israelensis applications and another, upstream. Another collection point was the Soberbo river, Guapimirim, State of Rio de Janeiro (Fonseca et al. 1998). This river has never been exposed to commercial formulations based on B. thuringiensis serovar israelensis. Only larvae were collected in this waterway. The third point where larvae were collected was in Rio das Pedras, Mangaratiba, State of Rio de Janeiro.
Isolation of Bacillus - To eliminate external contamination, the insects were sterilized following the methodology described by Alves (1986) with a slight modification; "Superbonder" glue (cyanoacrilate ester) was used to close both oral and anal cavities. This modification was adopted because Simulium larvae are very small and the dental floss used in the original technique was not applicable. After this step, the original method was performed passing the larvae through three solutions, first in 70% alcohol for 2 sec, second in 5% sodium hypoclorite for 3 min and finally in sterile 10% sodium thiossulfate for 5 min. The specimens were then washed three times in sterile distilled water.
Two different methods were used to isolate B. thuringiensis samples. In both, Simulium larvae were transferred aseptically into a sterile mortar and macerated with a sterile pestle. In the first method, the macerate was placed in tubes containing distilled water (10 ml). The suspension was heated (65ºC) for 12 min, and immediately diluted (1x10-2 to 1x10-4). Dilutions were plated on nutrient agar and plates were incubated (33ºC ± 2ºC) for 24 h. The second isolation procedure was that described by Travers et al. (1987) using the same recommended temperature for heating the larvae. The main difference between the two methods is that the second can select the spores of entomopathogenic bacteria such as B. sphaericus and B. thuringiensis. This selection occurs because sodium acetate added into the medium (0.5 M to give a final concentration of 0.25 M) is able to inhibit spore germination of entomopathogenic bacteria. On the other hand, the first method is more applicable when the amount of material is not large.
Identification of the samples - Bacteria were scraped off from the agar surface and saline water suspensions (NaCl 0.85%) were observed under light microscope (1,000x) to confirm the presence of parasporal crystals, a typical characteristic of B. thuringiensis. Biochemical, cytological and physiological characteristics according to Bergey's Manual of Determinative Bacteriology (Sneath 1986), de Barjac and Frachon (1990) and Thiery and Frachon (1997) were also assayed for these samples to confirm the classification of B. thuringiensis (data not shown). A strain of B. sphaericus was also identified using the same methodology described for B. thuringiensis. B. thuringiensis and B. sphaericus isolates were sent to Unité des Bactéries Entomopathogènes, Institute Pasteur, Paris, where they were submitted to serovar identification based on the flagellar antigens.
Protein profiles of the protoxins - The protein profiles of crystal components from the B. thu-ringiensis strains were determined by SDS-PAGE analysis, as described by Laemmli (1970), using 12% acrylamide separating gels. Samples (5-15 µg) of washed spore-crystal mixtures, prepared as described by Thomas and Ellar (1983), were placed in 4X concentrated sample buffer and heated at 100ºC for 10 min, as previously described by Lecadet et al. (1992) and loaded onto the gel immediately before electrophoresis.
Bioassays for toxicity - To evaluate toxicity against Aedes aegypti or Culex quinquefasciatus larvae, Bacillus strains were grown on Nutrient Agar plates for 48 h (33ºC ± 2ºC). Bacteria were scraped off from the surface of the culture medium and suspended in 0.85% NaCl solution at an optical density of 0.1 at 600 nm. One ml aliquots of the suspensions were added to 50 ml of water (free of chlorine) containing 15 larvae (L3-L4) and kept at 25ºC for 24 h for Ae. aegypti and 48 h for Cx. quinquefasciatus. Strains 2362 (B. sphaericus) and IPS-82 (B. thuringiensis serovar israelensis) were used as reference strains. All strains were tested in duplicate and two containers without added bacteria were maintained as a negative control. The average number of dead larvae was recorded and strains characterized as very toxic (100% mortality) or non-toxic (0% mortality) (Oliveira et al. 1998).
Isolation - Eighteen strains of B. thuringiensis were obtained from Simulium sp. collected in different rivers of the states of Rio de Janeiro and São Paulo. Seventeen were isolated from larvae of Simulium sp. and a single strain from an adult fly. Only one B. sphaericus strain was isolated, also from larvae collected in the Soberbo river (Table).
Serotype - Among the strains obtained, only two were shown to belong to serotype H-14, B. thuringiensis serovarisraelensis. Nine "autoag-glutinating" strains and seven B. thuringiensis serovar oswaldocruzi, H-39, were also identified.
Protein profiles of the protoxins - The protein profiles of crystal components were determined by SDS-PAGE analysis, useful for characterizing d-endotoxin families. The results of SDS-PAGE analysis referring to crystal proteins of the "autoagglu-tinating" strains are shown in the Figure. Two of the strains (LFB-Fiocruz 1038 and 1043) showed a different protein profile when compared to B. thuringiensis serovar israelensis strain IPS-82. These profiles had multiple components (lanes 4 and 9) ranging between 55 kDa and 100 kDa. Strains LFB-Fiocruz 1035, 1036, 1037, 1039, 1041, 1042 and 1044 presented the same protein profiles of the reference strain IPS-82.
Toxicity - The two B. thuringiensis serovar israelensis, the "autoagglutinating" strains and the strain of B. sphaericuswere shown to have toxicity (100%) in preliminary bioassays against Ae. aegypti and Cx. quinquefasciatus larvae, respectively. The strains belonging to serotype H-39 did not exhibit any toxicity (Table). It must be emphasized that these tests were qualitative.
The procedure of sealing with "Superbonder" glue was very useful for the purpose of isolation and also highly applicable to Simulium larvae.
The easiest method for bacterial isolation described in this study, which employed serial decimal dilutions in water could be useful in cases where the amount of material is limited. These findings were in accordance with Chilcott and Wigley (1993) who established that the most important factor in programme for the isolation of entomopathogenic bacteria was not the nature of the culture medium employed but rather the source of the material. Previously Asimeng and Mutinga (1992) isolated B. thuringiensis from mosquito breeding-sites in Kenya, but they did not sample Simulium larvae.
It was not possible to establish any relation between the locality of isolation and the serotype determined. Bacteria pertaining to B. thuringiensis serovar israelensis and B. thuringiensis serovar oswaldocruzi groups were found insideSimulium from different rivers.
Interestingly, the number of "autoagglu-tinating" strains of B. thuringiensis found in black flies from Caraguatatuba, in areas that were not treated and those areas treated with B. thuringiensis serovar israelensis, was higher than the number of isolated H-14 strains. Cavados et al. (1998) identified one "autoagglutinating" Brazilian strain which was as toxic as B. thuringiensis serovar israelensis applied in the region against Ae. aegypti larvae, but which was more toxic than this strain when the bioassays were performed against Chrysomya megacephala (Diptera, Calliphoridae) and it is necessary to say that this strain does not produce b-exotoxin. Rabinovitch et al. (1999) tested one Brazilian strain, classified as a
B. thuringiensis serovar israelensis, against S. pertinax, Cx. quinquefasciatus and Ae. aegypti. The Brazilian sample had the same toxicity against Ae. aegypti and Cx. quinquefasciatus, but was more toxic to S. pertinax than the B. thuringiensis serovar israelensis (IPS-82) strain. The "autoagglutinating" strains have eluded H-classification. For unknown reasons, suspensions of these strains agglutinated spontaneously in the absence of specific antiserum (Lecadet et al. 1999). It is interesting that seven of the "autoag-glutinating" strains isolated had the same protein profile of the B. thuringiensis serovar israelensis strain IPS-82, the same results described by Ragni et al. (1996). All of them were very toxic to Ae. aegypti larvae and they will be submitted to quantitative tests in order to determine the LC50. Strains LFB-FIOCRUZ 1038 and 1043 presented a different protein profile when compared with IPS82, despite the fact that both were very toxic for Ae. aegypti larvae. It will be necessary to run the LC50 bioassays in order to determine whether or not they are more toxic than the other ones.
The isolation of serotype H-39 from Simulium larvae is also a novel finding, because since the original isolation from commercial peppers, this serotype has appeared only in soil samples. The range of target insects of this serotype has not yet been determined. The strain LFB-FIOCRUZ 1040 may actually be the B. thuringiensis serovar israelensis, which has been applied in the Caraguatatuba region, where these larvae were collected. The isolation of B. sphaericus fromSimulium sp. has been reported previously. Weiser (1984) isolated the strain 2362, from an adult of S. damnosum. Strain 2362 is the strain used widely for mosquito control, however, it is extremely interesting that B. sphaericus does not demonstrate any toxicity towards Simulium sp. larvae.
This was the first report of the isolation ofentomopathogenic Bacillus from Simulium sp. larvae collected in Brazil.
To the Unité des Bactéries Entomopathogènes of the Institute Pasteur, Paris, for serotyping the bacterial strains. To the Pibic fellowship program of Fiocruz, to Superintedência de Controle de Endemias da Secretaria de Estado da Saúde de São Paulo for collecting and sterilizing the Simulium sp.larvae and adults, and to Dr Lawrence A Lacey for reviewing the manuscript.
Alves SB 1986. Isolamento de patógenos. In SB Alves, Controle Microbiano de Insetos, Manole, São Paulo, p. 253.
Araújo-Coutinho CJPC1995. Biological control against simuliids in the State of São Paulo, Brazil. Mem Inst Oswaldo Cruz 90: 131-133.
Asimeng EJ, Mutinga MJ 1992. Isolation of mosquito-toxic bacteria from mosquito breeding sites in Kenya. J Am Mosq Control Assoc 8: 86-88.
Cavados CFG, Chaves JQ, Queiroz MMC, Serra-Freire NM, Rabinovitch L 1998. An assessment of the biological activity of Bacillus thuringiensis LFB-FIOCRUZ 907 in Chrysomya megacephala (Diptera: Calliphoridae). Israel J Entomol 32: 117-123.
Chilcott CN, Wigley PJ 1993. Isolation and toxicity of Bacillus thuringiensis from soil and insect habitats in New Zealand. J Invertebr Pathol 61: 244-247.
De Barjac H, Frachon E 1990. Classification of Bacillus thuringiensis strains. Entomophaga 35: 233-240.
Fonseca RN, Cavados CFG, Chaves JQ, Araújo-Coutinho CJPC, Rabinovitch L 1998. Determinação da atividade tóxica de duas linhagens auto-aglutinantes de Bacillus thuringiensis isoladas do interior de larvas de simulídeos. Anais do VI Simpósio de Controle Biológico, Rio de Janeiro, p. 160.
Gerais BB, Ribeiro TC 1986. Onchocerca volvulus - 1º caso autóctone da região Centro-Oeste. Rev Bras Med Trop 19: 68.
Laemmli UK 1970. Cleavage of structural proteins during assembly of head of bacteriophage - T4. Nature 227: 680-685.
Lecadet MM, Chaufaux J, Ribier J, Lereclus D 1992. Construction of novel Bacillus thuringiensis strains with different insecticidal activities by transduction and transformation. Appl and Environ Microbiol 58: 840-849.
Lecadet MM, Frachon E, Cosmao Dumanoir V, Ripouteau H, Hamon S, Laurent P, Thiéry I 1999. Updating the H-antigen classification of Bacillus thuringiensis. J Appl Microbiol 86: 660-672.
Maia-Herzog M, Shelley AJ, Bradley JE, Luna Dias APA, Calvão RHS, Lowry CA, Camargo M, Rubio JM, Post RJ, Coelho GE 1999. Discovery of a new focus of human onchocerciasis in central Brazil.Trans R Soc Trop Med Hyg 93: 1-5.
Oliveira JO, da Silva SEA , Rabinovitch L 1998. A standardized protocol for the rapid detection of gelatin hydrolysis by Bacillus sphaericus. Israel J Entomol 32: 141-146.
Rabinovitch L, Cavados CFG, Chaves JQ, Araújo-Coutinho CJPC, Zanher V, Silva KRA, Seldin L 1999. A new strain of Bacillus thuringiensis serovar israelensis very active against black fly larvae. Mem Inst Oswaldo Cruz 94: 683-685.
Rabinovitch L, de Jesus FF, Cavados CFG, Zanher V, Momen H , da Silva, MH, Cosmao-Dumanoir V, Frachon E, Lecadet MM 1995. Bacillus thuringiensis susbp. oswaldocruzi and Bacillus thuringiensis subsp. braziliensis, two novel brazilian strains which determine new serotype H38 and H39, respectively. Mem Inst Oswaldo Cruz 90: 41-42.
Ragni A, Thiéry I, Delécluse A 1996. Characterization of six highly mosquitocidal Bacillus thuringiensis strains that do not belong to H-14 serotype. Current Microbiol 32: 48-54.
Sneath PHA 1986. Endospore-forming Gram-positive rods and cocci. In PHA Sneath, FG Priest, M Goodfellow, G Todd (eds), Bergey's Manual of Determinative Bacteriology, Vol. 2, Williams & Wilkins, Baltimore, p. 5-7.
Thiery I, Frachon E 1997. Bacteria: identification, isolation, culture and preservation of entomopathogenic bacteria. In Lawrence A Lacey, Manual of Techniques in Insect Pathology, Cap. III-1, Biological Techniques Series, Academic Press, London, p. 55-75.
Thomas WE, Ellar DJ 1983. Bacillus thuringiensis var. israelensis crystal d-endotoxin: effects on insect and mammalian cells in vitro and in vivo. J Cell Sci 60: 181-197.
Travers RS, Martin W, Reichelderfer CF 1987. Selective process for efficient isolation of Bacillus spp. Appl Envirom Microbiol 53: 1263-1266.
Weiser J 1984. A mosquito virulent Bacillus sphaericus in adult Simulium damnosum from Northern Nigeria. Zbl Mikrobiol 139: 57-60.