Characterization and pathogenic evaluation of Bacillus thuringiensis and Bacillus sphaericus isolates from Argentinean soils

BioControl 44: 59–71, 1999.© 1999Kluwer Academic Publishers. Printed in the Netherlands.

Characterization and pathogenic evaluation ofBacillus thuringiensisand Bacillus sphaericusisolatesfrom Argentinean soils

S.C. DIAS1,∗, M.A. SAGARDOY1, S.F. SILVA2 and J.M.C.S. DIAS21Departamento de Agronomía, Universidad Nacional del Sur, 8000 Bahía Blanca,Argentina;2Centro Nacional de Pesquisa de Recursos Genéticos e Biotecnologia,CENARGEN/EMBRAPA, Caixa Postal 02372, CEP: 70.770.900, Brasília, Brasil(∗author for correspondence; e-mail: [email protected])

Received 17 March 1998; accepted in revised form 2 March 1999

Abstract. Eighty soil samples of different origin (from urban, agricultural, forested andhorticultural areas) which had not previously been treated with bioinsecticides, were collectedand examined to investigate the presence ofBacillus thuringiensisandB. sphaericus. Froma total of 1473 bacterial isolates examined by differential staining techniques and growth onnutrient agar with the addition of penicillin and streptomycin, 31 (2.1%) strains ofBacillussphaericusand 25 (1.6%) strains ofBacillus thuringiensiswere isolated. These strains weretested for their pathogenicity against Diptera (Culex quinquefasciatus) and Lepidoptera(Anticarsia gemmatalisandSpodoptera frugiperda). Seven strains ofBacillus thuringiensissubspecieskurstaki were found to be pathogenic toSpodoptera frugiperdaand twenty-twostrains showed a pathological effect againstAnticarsia gemmatalis. None of the strains ofBacillus thuringiensisnor theBacillus sphaericusinvestigated, showed pathogenic activityagainstCulex quinquefasciatus. The strains ofBacillus thuringiensiswere characterizedserologically as belonging to six serotypes (darmstadiensis, entomocidus, kurstaki, muju, sottoandxianguangiensis). One strain seemed to be a new serotype. The electrophoretic profilesof the strains ofBacillus thruringiensisshowed bands of 130 kDa similar to those foundin strains pathogenic against Lepidoptera. Some physicochemical characteristics were alsostudied in the soil samples, in order to relate them to the presence or absence of theseBacillusspecies.

Resumé.Quatre-vingt échantillons de sol de différentes origines (zones urbaines, forestièreset horticoles), qui n’avaient jamais été traités antérieurement avec des bioinsecticides, ontété collectés (prelevés) et examinés pour chercher la présence deBacillus thuringiensisetBacillus sphaericus. Sur un total de 1473 isolats bactériens examinés à l’aide de techniquesde coloration différentielle et avec croissance sur milieu agar + nutriments contenant de lapenicilline et de la streptomycine, 31 souches (2.1%) deBacillus sphaericuset 25 (1.6%) deBacillus thuringiensisont été isolées. Ces souches ont été testées pour leur pouvoir pathogènecontre les Diptères (Culex quinquefasciatus) et les Lépidoptères (Anticarsia gemmataliset Spodoptera frugiperda). Sept souches deBacillus thuringiensissousespècekurstaki sesont révélées pathogènes contreSpodoptera frugiperdaet 22 montrent un effet larvicidecontreAnticarsia gemmatalis. Aucune des souches deBacillus thuringiensisou deBacillussphaericusétudiées ne montre d’activité pathogène contreCulex quinquefasciatus. Les

60

souches deBacillus thuringiensisont été caractérisés sérologiquement comme appartenantà six sérotypes (darmstadiensis, entomocidus, kurstaki, muju, sottoet xianguangiensis).Huit souches s’autoagglutinent et une semble appartenir à un nouveau sérotype. Les profilsélectrophorètiques des protéines de cristaux des souches deBacillus thuringiensisprésententdes bandes typiques de souches actives contre les Lépidoptères. Quelques caractéristiquesphysico-chimiques ont aussi été étudiées dans les échantillons de sol, afin de les corrèler à laprésence ou à l’absence de ces espèces deBacillus.

Key words: biological control of insects, electrophoretic profiles, serotypes

Introduction

Economic losses in agriculture owing to pests are huge. It is estimated thatin the world 15% of these are due to insect attacks on crops (Bravo andQuintero, 1993). The residual effects produced by insecticides, environmentalpollution, toxicity, and induced resistance in insects, provoked the use ofmicroorganisms as an alternative and efficient means of control (Lacey andHarper, 1986).

Among the groups of organisms which are being studied because of theirpotential as biocontrol agents, are bacteria belonging to the genusBacil-lus, within which the most important, because of their entomopathogenicactivity, areBacillus thuringiensisandB. sphaericus. Bacillus thuringiensis(Bt) is a gram positive bacillus, which produces crystalline proteinic inclu-sions during the sporulating process. The spores and/or inclusion bodies ofthis species have insecticidal activity. The proteins of the crystal are calleddelta-endotoxins (Aronson, 1991), or parasporal crystals (Tanada, 1993). Theamount of crystal protein produced byB. thuringiensiscultivated under labor-atory conditions (ca. 0.5 mg protein/ml) and the size of the crystal, indicatethat each bacterial cell synthesises around 106 to 2× 106 molecules of delta-endotoxin during its resting period to form the crystal (Agaise and Lereclus,1995).

Recently, the need to develop safe pesticides has accelerated the searchfor new strains ofB. thuringiensiswith different modes of action (Bernhardand Utz, 1993). These new strains may increase the list of commercialproducts available in the battle against insect larvae belonging to the ordersLepidoptera, Diptera and Coleoptera.

Useful strains of Bt have been isolated from soils, stored grains, deadinsect larvae, silk worm farms and leaf surfaces (Smith and Couche, 1991).DeLucca et al. (1981) isolated Bt from 17% of the natural soils sampledfrom various zones of the USA. Martin and Travers (1989) showed that Bthas a worldwide distribution, being found in different kinds of soils, includ-ing those of jungles, savannahs and deserts. Dulmage and Aizawa (1982)

61

suggested that the normal habitat of Bt is soil. Smith and Couche (1991)considered Bt to be a common component of the microflora of leaves andthat the infection is a natural event since Bt belongs to the phylloplane wherethe insect larvae arrive to feed. Martin (1984), however, stated that the nor-mal environment for Bt will never be well-defined since it is a saprophyticmicroorganism which has no stringent requirements and can, therefore, adaptitself to different environmental conditions.

B. sphaericusis a heterogeneous collection ofBacillus strains, some ofwhich show different degrees of pathogenicity in mosquito larvae (Alexanderand Priest, 1990; Guerineau et al., 1991). In general,Culexsp. andAnophelessp. are more sensitive to this bacterium thanAedessp. Few of the most toxic30 strains have been isolated in Latin America, the only citations belonging toEl Salvador, Guyana and Brazil (Singer, 1977; Lysenko et al., 1985; Sckenkelet al., 1992). This bacterium has received great interest from the World HealthOrganisation through programmes which were developed to control humandiseases transmitted by hematophagic Diptera, such as malaria, dengue fever,yellow fever and filariasis (WHO, 1987).

The aim of this study was the isolation of strains ofB. thuringiensisandB. sphaericusfrom soils of different provinces of Argentina and their serolo-gical, protein electrophoretic and pathogenic characterization against Dipteraand Lepidoptera by bioassays.

Materials and methods

Soil collectionSoil samples were collected at different times during the year and from placeswith different vegetation types: uncultivated areas, urban centers, agricultural,horticultural or cattle breeding regions situated in 11 of the 22 differentprovinces in Argentina. A 1-m2 area was chosen and 10 sub-samples (ofapprox. 10 g of soil) from up to 10 cm deep were collected.

Determination of pH, moisture and organic matter of the soil samplesSamples were diluted 2.5-fold in water for pH analysis. The pH deter-mination was performed during shaking using a glass electrode. Moisturewas determined using the gravimetric method and organic matter accordingWalkley-Black (Jackson, 1964).

Isolation of strains ofBacillus thuringiensis(Bt) andB. sphaericus(Bs)The methodology which was employed is a modification of that described bythe World Health Organisation (WHO, 1985). Approximately 1 gram of soilwas diluted in 9 ml of saline solution which was then shaken for 5 minutes.

62

Successive serial dilutions were then made up to 106. From the last three dilu-tions, 1 ml aliquots were transferred to a sterile Eppendorf tube. They werethen incubated in a thermal bath at 80◦C for 12 minutes. The treated sampleswere then placed on nutrient agar. After 24 hours growth at 30◦C, colonieswith similar morphological characteristics to those ofB. sphaericusand Btwere transferred to nutrient agar containing penicillin (100 mg/l) or strep-tomycin (25 mg/l) for preliminary identification (Yousten, 1991). Isolateswhich grew in the presence of penicillin were stained with Coomassie-blueand later observed under the microscope to detect the presence of spores andcrystals. Those isolates showing both were assumed to beBacillus thuringi-ensis. Each isolate was then transferred to nutrient agar for later serologicalcharacterization, pathogenicity evaluation and electrophoresis. Those colon-ies which grew in media with streptomycin were stained with fuchsin forthe observation of spores and typical sporangia ofB. sphaericus. Strains ofB. sphaericuswere transferred to nutrient agar for later evaluation of theirpathogenic ability against dipterans.

Serological characterizationThe strains of Bt were serologically characterized using the methodologydescribed by de Barjac and Bonnefoi (1962) and adapted by Dias et al. (1993).

Evaluation of pathogenicity of the isolated strainsThe strains of Bt were investigated against two Lepidoptera:Anticarsia gem-matalisandSpodoptera frugiperda(Lepidoptera: Noctuidae). A third-stagelarva was placed in a 50 ml disposable cup with approx. 200 mg of artificialdiet (specific for each of the insects tested) to which 0.5 ml of a suspension(approximately 107 spores/ml) of the bacteria, which had been grown for48 hours in a shaker, was added. Evaluation of mortality was done everytwo days up to the 8th day. For each bioassay, two sets of 20 larvae wereutilised. To test the pathogenicity of the strains ofBacillus thuringiensisandB. sphaericusagainst Diptera, larvae ofCulex quinquefasciatuswere used.One ml of a suspension of bacteria, which had been cultured in a shaker for48 hours, was added to disposable cups which contained 100 ml of distilledwater and 25 second-stage larvae in duplicate. In both cases, the isolates wereconsidered pathogenic when they caused at least 50% of mortality. After 24and 48 hours, the survivors were counted.

Gel electrophoresis analysisExtraction of the spore-crystal complex was done according to the method-ology described by De Souza et al. (1993). Proteins were then subjected toelectrophoresis using a 13% polyacrylamide gel (Laemmli, 1970).

63

Table 1. Soil samples and characterization of 1473 spore forming bacteria from elevenprovinces of Argentina

Provinces No. of soil No. of Bacillus Bacillus Bacillus Bacillussp.

samples strainssphaericus thuringiensis cereus

Buenos Aires 37 796 15 8 528 245

Santa Fe 2 66 0 0 45 21

Chubut 3 20 0 0 7 13

Santiago del

Estero 12 36 0 1 24 11

Cordoba 1 24 3 0 14 7

Rıo Negro 10 219 4 13 139 63

Catamarca 1 21 3 0 9 9

Santa Cruz 4 78 0 0 59 19

La Pampa 3 36 2 0 21 13

Mendoza 4 114 2 1 79 32

Misiones 3 63 3 0 48 12

Total 80 1473 32 23 973 445

Results and discussion

Isolation

From the 80 soil samples, 1473 spore forming bacteria were obtained: 32B.sphaericus(2.23%), 23B. thuringiensis(1.6%), 973B. cereus(66%) and 445(30.2%)Bacillus sp.(Table 1).

The percentage recovery of Bt is higher than that found by DeLucca etal. (1981). There, authors examining 46373Bacillus isolates from soils inthe USA found that only 0.5% were Bt. The low rate of detection of Bt isprobably due to the isolation methodology used by these authors who usedthe technique described by Saleh et al. (1969). On the other hand, using theacetate technique (Travers et al., 1987), Martin and Travers identified 2.9%of Bacillusstrains as Bt from samples originating from different continents,including forested areas, savannahs and deserts (Martin and Travers, 1989).This was probably an overestimate since some of these areas had been previ-ously treated with commercial preparations of Bt. In the present study, despitehaving obtained a low percentage recovery of Bt compared to the work ofMartin and Travers (1989), the technique suggested by the WHO (1985) wassimple, fast and permitted the isolation of bothBacillus thuringiensisandB.

64

sphaericus. This does not happen using the acetate technique, which uses aconcentration of 0.5 M of sodium acetate, inhibiting the growth ofB. sphaeri-cus, as well as inhibiting the germination of some Bt (Travers et al., 1987).We consider that the recovery rate obtained by us is good for soils where nobioinsecticides have been applied.

Use of culture media with antibiotics for subcultures only (Yousten, 1991)permitted rapid and accurate identification of strains of Bt andB. sphaericus.The utilisation of Coomassie blue staining was indispensable for the differ-entiation of strains of Bt andB. cereus, where the greater number ofB. cereuscompared to Bt (39:1) should be observed. Similar data have been obtainedby investigators in Japan (Ohba and Aizawa, 1978) and Brazil (Lopes etal., 1994), showing that, as in Argentina Bt is less frequent thanB. cereus.Bt and B. sphaericuswere found together in 24 (30%) of the 80 studiedsoils samples. Bt isolates withoutB. sphaericuswere found in only 6 (7.5%)samples, whileB. sphaericusisolates without Bt were present in 16 (20.0%)samples.

Larvicidal activity and serological identification

None of the 55 strains of Bt andB. sphaericusproduced mortality in Diptera(C. quinquefasciatus).

The evaluation of the pathogenicity of the strains of Bt against Lepidopterashowed that 8 of the isolates were pathogenic againstS. frugiperdaand 21were pathogenic againstA. gemmatalis(Table 2).

The 23 Argentinean strains of Bt could be serologically characterised asbelonging to six distinct serotypes (Table 2). The strain A45 (Villa Regina,Rio Negro) and A68 (Mendoza) belonged to serotype 4a4b (subspeciessotto).Eight strains isolated from Villa Regina were characterized as subspeciesmuju (serotype H49); the strain A56 from Necochea (Buenos Aires) reactedagainst the serum H51 (subsp.xianguangiensis). Among the eight strainsbelonging to subspecieskurstaki (serotype H3a3b3c), one was from RioNegro (A27) and the others were isolated from Buenos Aires. The strainsA42 and A47, from Rio Negro were characterized as subspeciesdarmstadi-ensis(H10a10b) and the strain A26 reacted against the serum H6 (subsp.entomocidus). Finally, strain A20 (Santiago del Estero) did not react againstany of the antisera known at that time. Analysis performed at the PasteurInstitute (France) indicated that this would be a new subspecies (Lecadet, M.,Pasteur Institute, Paris, France, 1997, personal communication) (Table 2).

RegardingA. gemmatalispathogenicity, it can be observed that 16 strainscaused mortality equal or superior to 50% (Table 2). In these figures, fourstrains caused mortality of 50%, four other strains led to mortality between60 and 90% and eight strains caused total mortality onA. gemmatalislarvae.

65

Table 2. Origin, serotype and pathogenicity ofBacillus thuringiensisagainst Lepidopteraa

Strain Origin Serotype Anticarsia Spodoptera

gemmatails(%) frugiperda(%)

A20 Santiago del Estero New serotypeb 20 0

A26 Rıo Negro H6 entomocidus 40 0

A27 Rıo Negro H3a3b3c kurstaki 100 100

A36 Rıo Negro H49 muju 0 0





A42 Rıo Negro H10a10b darmstadiensis 30 0

A43 Rıo Negro H49muju 80 100


A45 Rıo Negro H4a4b sotto 50 0

A46 Rıo Negro H49muju 60 0

A47 Rıo Negro H10a10b darmstadiensis 0 0

A56 Buenos Aires H51 xianguangiensis 30 0

A59 Buenos Aires H3a3b3c kurstaki 90 0







A68 Mendoza H4a4b sotto 30 0

aData are expressed as % mortality.bLecadet, M., Pasteur Institute, Paris, France, 1997, personal communication.

Seven of the eight strains of Bt which were toxic toS. frugiperda(A27,A60, A61, A62, A63, A64 and A66) belong to subspecieskurstaki(serotypeH3a3b3c), while the strain A43 belongs to subspeciesmuju.

These results strengthen the affirmations of some authors who agree thatserological characteristics cannot be used as indicators of pathogenicity (deBarjac and Frachon, 1990; Lambert and Peferoen, 1992; Lereclus et al.,1993).

66

(A)

(B)

67

(C)

(D)Figure 1. Protein patterns ofBacillus thuringiensiscrystal analyzed by sodium dodecylsulphate polyacrylamide gel electrophoresis (acrylamide, 13%) and Coomassie blue staining.Numbers to the left indicate protein size in kDa.(A) 1 = MW, molecular weight marker proteins; 2 = strain A27; 3 = strain A61; 4 = strainA62; 5 = strain A63; 6 = train A64; 7 = strain A68; 8 = strain HD-1.(B) 1 = MW, molecular weight marker proteins; 2 = strain A36; 3 = strain A47; 4 = strainHD-1.(C) 1 = MW, molecular weight marker proteins; 2 = strain A38; 3 = strain A40; 4 = strainA43; 5 = strain A44; 6 = strain A59; 7 = strain HD-1.(D) 1 = strain A20; 2 = strain A36; 3 = strain A37; 4 = strain A39; 5 = strain A56; 6 = strainA26; 7 = strain A68; 8 = MW, molecular weight marker proteins.

68

Electrophoretic analyses of crystal proteins

Four protein patterns were found for the protein characterisation of the Btspore-crystal complex. Figure 1 presents the electrophoretic profiles of theinvestigated strains, with their respective molecular weights.

In Figure 1A, bands with molecular weights of 130 kDa and of circa65 kDa can be observed; all of these strains belong to the serotype H3a3b3c(subsp.kurstaki) and all show pathogenic activity againstAnticarsia gem-matalis (Table 2). This pattern corresponds to the finding of Lereclus et al.(1993) for the Bt strains with crystals containing proteins type CryI and CryII,which are toxic against lepidopteran larvae. The same proteinic componentis present in CryIV toxins, lethal against mosquitoes. However, none of ourstrains displaying this component were toxic againstC. quinquefasciatus.

Other strains, A40 and A44 (Figure 1C), A 37, A39, A56 and A26 (Figure1D), also showed the Cryl typical band of 130 kDa. Finally, strains A36 andA68 (Figure 1D) did not show this band and were not pathogenic againstA.gemmataliseither. These results are in agreement with those described byLereclus et al. (1993), who indicate that the proteins of the crystal producedby the strains with specific activity against Lepidoptera larvae (Cryl) are themost frequently found and are synthesised from at least seven subclasses ofgenes (CrylA to CrylG). The crystals of this toxin show bipyramidal shape.The analyses of the purified crystals solubilized through SDS polyacrylamidegel show polypeptides with molecular weights of 130–140 kDa.

The unexpected result of the strain A47, which showed a 130 kDaband (Figure 1B) but did not cause mortality in Lepidoptera larvae studied(Table 2), should be explained by the presence of different receptors in differ-ent insects. Other causes could be the variation in the number of receptors andin the efficiency of the binding toxin-receptor (Bravo and Quintero, 1993).

According to Höfle and Whiteley (1989), among the strains that have typeCryl crystals some can have a polypeptide of 71 kDa that generates cuboid-shaped crystals. These crystals can produce double pathogenicity againstLepidoptera and Diptera being classified as Cryll. In the strains investigated inthis study, that protein band was not observed, nor was pathogenicity againstDiptera (Table 2).

In none of the strains analysed electrophoretically, polypeptides werefound of molecular weight between 70 and 75 kDa, typical of Cryll strains,nor bands of 135, 128, 74 and 72 kDa, which correspond to the fragmentstoxic to Diptera larvae, characteristic of the Cry lV polypeptides. Theseresults are in agreement with the toxicity bioassays againstCulex quinquefas-ciatus (Table 2), where no pathogenicity of any of the strains was observedagainst that insect.

69

Quantification of pH, moisture, and organic matter in the soil samples

The investigated soils showed a great diversity in values of pH, moisture, andorganic matter. The pH ranges between 5.3–9.7; moisture ranges between0.09–31.7%, and organic matter between 0.2–10.8%. No significant relation-ship was found between the physicochemical parameters of the studied soilsand the presence of Bt andB. sphaericus.

Evidence presented in this paper therefore suggests that several differentserotypes of Bt are colonising Argentinean soils and also that those nativespore forming bacteria are able to control insects of economic importance.These strains could be utilised for bioinsecticide production, aiming to reducethe use of chemical insecticides in Argentina.

Acknowledgements

This work was partially supported by the National Research Council ofArgentina (CONICET). The authors thank Dr Peter Inglis for manuscriptreviewing.

References

Agaise H. and D. Lereclus, 1995. How doesBacillus thuringiensisproduce so muchinsecticidal crystal protein.J. Bacteriol.17: 6027–6032.

Alexander, B. and F.G. Priest, 1990. Numerical classification and identification ofBacillussphaericusincluding some strains pathogenic for mosquito larvae.J. Gen. Microbiol.136:367–376.

Aronson, M.J., 1991.Bacillus thuringiensisinsecticidal crystal proteins: gene structure, actionand utilization. In: K. Maramorosh (ed),Biotechnology for Biological Control of Pests andVectors. CRC Press, Florida. pp. 3–24.

Bernhard, K. and R. Utz, 1993. Production ofBacillus thuringiensisinsecticides for experi-mental and commercial uses. In: P.F. Entwistle, J.S. Cory, M.J. Bailey and S. Higgs (eds),Bacillus thuringiensis,an Environmental Biopesticide: Theory and Practice. John Wiley& Sons, Chichester, England. pp. 255–266.

Bravo, A. and R. Quintero, 1993.Importancia y potencial deBacillus thuringiensisen elcontrol de plagas. REDBIO, Santiago, Chile. 50 p.

De Barjac, H. and A. Bonnefoi, 1962. Essai de classification biochimique et serologique de24 souches deBacillusdu typethuringiensis. Entomophaga7: 5–31.

De Barjac, H. and E. Frachon, 1990. Classification ofBacillus thuringiensisstrains.Entomo-phaga35: 233–240.

DeLucca, A.J., J.G. Simonson and A.D. Larson, 1981.Bacillus thuringiensisdistribution insoils of the United States.Can. J Microbiol.27: 865–870.

70

De Souza, M.T., M.M. Lecadet and D. Lereclus, 1993. Full expression of the Cry IIIAtoxin gene ofBacillus thuringiensisrequires a distant upstream DNA sequence affectingtrancription.J. Bacteriology175: 2952–2960.

Dias, S.C., J.O. Silva-Werneck, R.G.M. Schenkel, J.B. Lopes and J.M.C.S. Dias, 1993. Téc-nicas sorológicas para caracterização deBacillus thuringiensise Bacillus sphaericus.Comunicado Tecnico. Embrapa. Cenargen15: 1–6.

Dulmage, H.T. and K. Aizawa, 1982. Distribution ofBacillus thuringiensisin nature. In: E.Kurstak (ed),Microbial and Viral Pesticides. Marcel Dekker, New York. pp. 209–237.

Guerineau, M., B. Alexander and F.G. Priest, 1991. Isolation and identification ofBacillussphaericusstrains pathogenic for mosquito larvae.J. Inverteb. Pathol.57: 1–9.

Hofte, H. and H.R. Whitelley, 1989. Insecticidal crystal proteins ofBacillus thuringiensis.Microbiol. Rev.53: 242–255.

Jackson, M.L., 1964.Analisis guimicos de suelos. Ediciones Omega, Barcelona. 662 p.Lacey, L.A. and J.D. Harper, 1986. Microbiol control and integrated pest management.J.

Entomol. Sci.21: 206–213.Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of

bacteriophage T4.Nature277: 680–685.Lambert, B. and M. Peferoen, 1992. Insecticidal promise ofBacillus thuringiensis. Facts and

mysteries about a successful biopesticide.Bioscience42: 112–123.Lereclus, D., A. Delécluse and M.M.Lecadet, 1993. Diversity ofBacillus thuringiensistox-

ins and genes. In: P.F. Entwistle, J.S. Cory, M.J. Bailey and S. Higgs (eds), Bacillusthuringiensis,an Environmental Biopesticide: Theory and Practice. John Wiley & Sons,Chichester, England. pp. 39–69.

Lopes, J.B., J.S. Giusti, S.F. Silva and J.M.C.S. Dias, 1994. Comparação entre três metodolo-gias para isolamento de bacilos entomopatogênicos em amostras de solo da região sul doBrasil. IV Simpósio de Controle Biológico.Anais. p. 145.

Lysenko, O., E.W. Davidson, L.A. Lacey and A.A. Yousten, 1985. Five new mosquito lar-vicidal strains ofBacillus sphaericusfrom non-mosquito origins.J. Am. Mosq. Control.Assoc.1: 369–371.

Martin, P.A.W., 1984. An iconoclastic view ofBacillus thuringiensisecology.Am. Entomol.40: 85–90.

Martin, P.A.W. and R.S. Travers, 1989. Worldwide abundance and distribution ofBacillusthuringiensisisolates.Appi. Environ. Microbiol.55: 2437–2442.

Ohba, M. and K. Aizawa, 1978. Serological identification ofBacillus thuringiensisand relatedbacteria isolated in Japan.J. Invert. Pathol.32: 303–309.

Sakeh, S.M., R.F. Harris and O.N. Allen, 1969. Method for determiningBacillus thuringiensisvar. thuringiensisBerliner in soil.Can. J. Microbiol.15: 1101–1104.

Schenkel, R.G.M., L. Nicolas, E. Frachon and S. Hamon, 1992. Characterisation and toxicityto mosquito larvae of fourBacillus sphaericusstrains isolated from Brazilian soils.J.Inverteb. Pathol.60: 10–14.

Singer, S., 1977. Insecticidal activity of recent bacterial pathogens of vectors. In: Briggs,T. (ed),Biological Regulation of Vectors. U.S. DHEW Publication No. (NIH) 77-1180.National Institute of Health, Washington, DC. pp. 3–18.

Smith, R.A. and G.A. Couche, 1991. The phylloplane as a source ofBacillus thuringiensisvariants.Appl. Environ. Microbiol.57: 311–315.

Tanada, Y.A., 1993 Bacterial infections: Bacillaceae. In: H.K. Kaya and Y.A. Tanada (eds),Insect Pathology. Academic Press, San Diego. pp. 83–146.

71

Travers, R.S., P.A.W. Martin and C.F. Reichelderfer, 1987. Selective process for efficientisolation of soilBacillusspp.Appl. Environ. Microbiol.53: 1263–1266.

World Health Organization, 1985. Informal consultation on the development ofBacillussphaericusas microbial larvicide. WHO Mimeograph Document 85.3.

World Health Organization, 1987. Report of an informal consultation on the detection,isolation, identification and ecology of bio control agents of disease vectors. WHOMimeograph Document 87.3.

Yousten, A., 1991. Entomopathogenic bacteria for biological control. In:Workshop Manual.Mimeography Document, Fundaçáo André Tosello, Campinas, Brazil. 50 p.

Characterization and pathogenic evaluation of Bacillus thuringiensis and Bacillus sphaericus isolates from Argentinean soils

Documents