1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Parasporal inclusion proteins from a total of 1744 Bacillus thuringiensis strains, consisting of 1700 Japanese isolates and 44 reference type strains of existing H serovars, were screened for cytocidal activity against human leukaemia T cells and haemolytic activity against sheep erythrocytes. Of 1684 B. thuringiensis strains having no haemolytic activity, 42 exhibited in vitro cytotoxicity against leukaemia T cells. These non-haemolytic but leukaemia cell-toxic strains belonged to several H-serovars including dakota, neoleonensis, shandongiensis, coreanensis and other unidentified serogroups. Purified parasporal inclusions of the three selected strains, designated 84-HS-1-11, 89-T-26-17 and 90-F-45-14, exhibited no haemolytic activity and no insecticidal activity against dipteran and lepidopteran insects, but were highly cytocidal against leukaemia T cells and other human cancer cells, showing different toxicity spectra and varied activity levels. Furthermore, the proteins from 84-HS-1-11 and 89-T-26-17 were able to discriminate between leukaemia and normal T cells, specifically killing the former cells. These findings may lead to the use of B. thuringiensis inclusion proteins for medical purposes.

Bacillus thuringiensis was first isolated in Japan as a pathogen of the sotto disease of the silkworm, Bombyx mori, by Ishiwata (1901) early in this century. The organism is a Gram-positive, spore-forming bacterium that produces crystalline parasporal inclusions during sporulation. The inclusions often exhibit strong insecticidal activity against several orders of insects, and this makes B. thuringiensis a reliable agent for microbial control of insect pests of agricultural and medical importance ( Lambert & Peferoen 1992; Cannon 1996). The insecticidal parasporal inclusions contain two families of insect-toxic molecules, Cry and Cyt proteins. The Cry protein is specifically toxic to insects and is currently classified into 22 genetically different major groups, Cry1 to Cry22 ( Crickmore et al. 1998 ). It has been well established that the high specificity of Cry proteins in killing insects is attributable to specific binding of the proteins to receptors that reside on the midgut cell membranes of susceptible insects ( Cannon 1996). Another family of the toxin, the Cyt protein, has a broad cytolytic activity against invertebrate and vertebrate cells including erythrocytes, and is divided into two genetically different groups, Cyt1 and Cyt2 (formerly CytA and CytB, respectively) ( Crickmore et al. 1998 ).

Historically, it has long been believed that B. thuringiensis as a species is characterized by insecticidal activity associated with its parasporal inclusions. Earlier studies, however, have demonstrated that in natural environments, B. thuringiensis strains producing non-insecticidal parasporal inclusions are more widely distributed than insecticidal ones ( Ohba & Aizawa 1986; Hastowo et al. 1992 ; Meadows et al. 1992 ; Ohba 1996, 1997; Roh et al. 1996 ). Thus, the question arises as to whether such non-insecticidal parasporal inclusions have any biological activity as yet undiscovered ( Ohba et al. 1988 ). Although antibacterial activity has recently been demonstrated in inclusion proteins of some non-insecticidal strains of B. thuringiensis ( Yudina & Burtseva 1997), very little, if anything, is known about the other activities. It is reported here for the first time that a unique activity, cytocidal against human cancer cells, is associated with non-insecticidal parasporal inclusions of certain B. thuringiensis strains.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Bacterial strains and culture conditions

Most Japanese Bacillus thuringiensis isolates used in this study were from the collection of the Institute of Biological Control, Kyushu University ( Saitoh et al. 1998 ). They had been serotyped by the method described previously ( Ohba & Aizawa 1978), although H serotyping of some new isolates was done in the present study. The reference type strains of B. thuringiensis were obtained from the Institut Pasteur, Paris. To prepare sporulated cultures, the strains were grown on nutrient agar, pH 7·6, at 28 °C for 4 d. Nutrient agar consisted of meat extract (10 g), polypeptone (10 g), NaCl (2 g), agar (20 g) and distilled water (1000 ml).

Cells and culture conditions

Seven vertebrate and insect cell lines were obtained from RIKEN Cell Bank (Tsukuba, Japan): MOLT-4, human leukaemia T cell; A549, human lung cancer cell; MRC-5, normal human lung fibroblast cell; HeLa, human uterus cervix cancer cell; GF-Scale, gold fish (Carassius auratus) scale cell; BM-N, silkworm (Bombyx mori) cell; NIAS-AeAl-2, mosquito (Aedes albopictus) cell. They were cultured under the conditions recommended by the manufacturer. Normal human T cells were prepared from buffy coats obtained from Fukuoka Red Cross Blood Centre (Fukuoka, Japan). T cells were separated from lymphocyte cells using a CellectTM Human T Cell kit (Biotex Laboratories Inc., Alberta, Canada), and were cultured in RPMI1640 medium supplemented with 10% foetal bovine serum and kanamycin (30 μg ml−1) at 37 °C. Sheep erythrocytes were purchased from Nippon Bio-Test Lab., Tokyo, Japan. Human erythrocytes (blood group O) were freshly prepared from a volunteer.

Inclusion protein preparation and production of antisera

Sporulated cultures were washed three times with distilled water before solubilization and used as crude parasporal inclusions. Parasporal inclusions were purified from sporulated cultures as previously described ( Saitoh et al. 1998 ). The crude or purified inclusions were suspended in 50 mmol l−1 Na2CO3 (pH 10) + 10 mmol l−1 DTT (dithiothreitol) + 1 mmol l−1 EDTA for 60 min at 37 °C. After solubilization of the inclusions, the solution was added with phenylmethylsulphonyl fluoride (Wako Pure Chemical, Tokyo, Japan) to stop the proteolytic reaction. The mixture was then centrifuged at 20 000 g for 5 min at 4 °C. Protein concentration of the supernatant fluid was determined by the method of Lowry et al. (1951) using bovine serum albumin as the standard. The supernatant fluid was then treated with proteinase K as previously described ( Ishii & Ohba 1993). After incubation, the pH of the fluid was adjusted to 7·2 and filtered through a 0·45 μm membrane filter. Antisera were raised in rabbits against solubilized whole inclusion proteins of the type strains of serovars israelensis, kyushuensis and kurstaki according to the method of Higuchi et al. (1998) .

Screening methods

A two-step method ( Saitoh et al. 1996 ) was used for mass screening of alkali-solubilized, proteinase K-treated, parasporal proteins. For primary screening 1744 strains were grouped into 349 sets, usually five isolates per set. For the sets exhibiting cytocidal activity against human leukaemia T cells (MOLT-4), the isolates were individually examined for cytotoxicity against MOLT-4 cells and haemolytic activity against sheep erythrocytes. The assay procedure in the screening involved (i) observation of the cytopathic effect (CPE) under a phase-contrast microscope and (ii) assessment of the degree of cytotoxicity by a cell proliferation assay technique.

Cytotoxicity assay and haemolysis assay

Each well of a microtest plate received 90 μl of cell suspension containing 2 × 104 cells. The plate was incubated at 37 °C for 16 h. Then, to each well was added 10 μl of the sample solution, which contained approximately 12 μg of protease-activated inclusion proteins. Each test was done in duplicate or triplicate and was repeated at least three times. The CPE was monitored under a phase-contrast microscope 1 h and 24 h post-inoculation. The degree of CPE was graded as follows on the basis of the proportion of damaged cells in about 1000 cells: –, < 5% ± , 5% to < 10% + , 10% to < 30% + + , 30% to < 60% + + + , 60% to < 90% + + + + , ≥90%.

For assessment of the level of cytotoxicity, a cell proliferation test using a MTT [3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide)] assay ( Behl et al. 1992 ; Heiss et al. 1997 ) was done 24 h post-inoculation using a Cell Titer 96TM kit (Promega Co., Madison, WI, USA) or a Premix WST-1 kit (Takara Co., Kyoto, Japan). In this assay, the average of absorbance values for a negative control (mock inoculation of parasporal protein) was used as a blank value. The degree of cytotoxicity (CT) was graded on the basis of the relative value of absorbance to the blank (1·00): + + + + , extremely high (<0·10); + + + , high (0·10 to <0·40); + + , moderate (0·40 to <0·70); + , low (0·70 to <0·90); ± , very low (0·90 to <0·95); –, non-toxic (≥ 0·95). The isolates with the graded + + to + + + + in CPE and/or CT were selected for further study as cytotoxic strains.

Haemolytic activity was tested on sheep or human erythrocytes as described previously ( Saitoh et al. 1998 ). The degree of haemolytic activity was graded as follows on the basis of their absorbence at 540 nm: + + + , high (>1·00); + + , moderate (1·00 to >0·50); + , low (0·50 to >0·20); –, non-haemolytic ( ≥0·20). The isolates with the levels + to + + + in haemolytic activity were considered as haemolytic strains.

Insecticidal activity test

For insecticidal activity tests, 11 species of five orders were used: Lepidoptera (Plutella xylostella and Bombyx mori), Diptera (Aedes aegypti, Culex pipiens molestus, Anopheles stephensi, Telmatoscopus albipunctatus and Musca domestica), Orthoptera (Gryllus bimaculatus and Atractomorpha bedeli), Dictyoptera (Blattella germanica) and Isoptera (Reticulitermes separatus separatus). The assay was done according to the method of Higuchi et al. (1998) .

SDS-PAGE and immunoblotting

SDS-PAGE of parasporal inclusion proteins was performed as described by Laemmli (1970) using 10% separating and 4% stacking gels. The gels were stained with 0·1% (w/v) Coomassie blue R250 (Sigma). The molecular masses of proteins were determined using protein standards (Sigma). For immunoblot analysis, the resolved parasporal inclusion proteins were transferred to a PVDF membrane using a Transblot SD semi-dry electrophoretic blotting cell (Bio-Rad, Hercules, CA, USA). The membrane was then probed with an appropriate dilution of antiserum. Bound antibodies were detected using alkaline phosphatase-conjugated anti-rabbit IgG (TAGO Inc. Burlinegame, CA, USA). Visualization was performed with nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl toluidinate (Wako Pure Chemical) as the colour substrate.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

Screening of leukaemia cell-toxic but non-haemolytic inclusion proteins

A total of 1744 B. thuringiensis strains, comprising 1700 Japanese isolates and 44 reference type strains of existing H serovars ( Table 1), were examined first for cytocidal activity of inclusion proteins against human leukaemia T cells and for cytolytic activity on sheep erythrocytes.

Table 1.  Cytotoxic activities of Bacillus thuringiensis parasporal inclusion proteins against human leukaemia T cells and sheep erythrocytes
 Number of isolates mono-toxic to 
H antigenH serovarNumber of isolates tested‡Leukaemia T cellSheep erythrocyteNumber of isolates toxic to both cellsNumber of non-toxic isolates
  • *

    Sub-factor, not identified.

  • Seroreactive for two reference antisera.

  • Bold figure includes the type strain.

  • UTE, Untestable due to no flagellation and/or autoagglutination.

  • UTY, Untypable by reference antisera.

  • NE, Not examined.

2finitimus 11
3adsumiyoshiensis 88
5abgalleriae 88
5accanadensis 66
6abentomocidus 11
6acoyamensis 99
8acostriniae 55
8bdnigeriensis 11
9tolworthi 44
11abtoumanoffii 11
12thompsoni 11
14israelensis 514
18abkumamotoensis 55
20abyunnanensis 88
20acpondicheriensis 11
21colmeri 44
23japonensis 44
24abneoleonensis 312
25coreanensis 211
26silo 11
28abmonterrey 11
31toguchini 22
32cameroun 11
33leesis 22
34konkukian 633
NE113 13496
Total 17444244161642

Of the strains tested, 60 showed haemolytic activity against sheep erythrocytes. Among these haemolytic strains, 16, including the type strains of serovar israelensis and serovar kyushuensis, exhibited significant cytotoxicity against leukaemia cells. However, the 44 other haemolytic strains did not have leukaemia cell-killing activity. Of the 1684 B. thuringiensis strains with no haemolytic activity, 42 showed cytocidal activity against leukaemia cells ( Table 1). These strains, non-haemolytic but leukaemia cell-killing, consisted of 41 Japanese isolates derived from soils and phylloplanes of nine geographically different localities, and the type strain of serovar neoleonensis from Mexico ( Rodriguez-Padilla et al. 1990 ) ( Table 2). The 41 Japanese strains belonged to H serovars dakota, shandongiensis, coreanensis and other unidentified serogroup(s). None of the predominant serovars in our B. thuringiensis library, such as alesti, kenyae, sotto, aizawai, kurstaki and morrisoni, were toxic to leukaemia cells, although these serovars contained many strains highly toxic to insects of the order Lepidoptera. Interestingly, the 42 leukaemia cell-toxic strains showed no significant insecticidal activity against 11 species from five orders (data not shown), except that one strain exhibited only low larvicidal activity against the mosquito, C. pipiens molestus. As shown in Table 2, the shape and size of parasporal inclusions varied markedly depending on the strains.

Table 2. Bacillus thuringiensis strains whose parasporal inclusion proteins are non-insecticidal and non-haemolytic but cytocidal against human leukaemia cells
H antigenH serovarNumber of isolates Locality (Prefecture)SourceInclusion morphology
  • *

    Seroreactive for two reference antisera.

  • †Type strain ( Rodriguez-Padilla et al. 1990 ).

  • Low toxicity to Culex pipiens molestus (Diptera).

  • UTE, Untestable due to no flagellation and/or autoagglutination.

  • UTY, Untypable by reference antisera.

15dakota8FukuokaSoilIrregularly pointed
15/21 *dakota/colmeri1FukuokaPhylloplaneIrregularly pointed
22shandongiensis3TokyoSoilIrregularly pointed
24abneoleonensis 1 Guanajuato, MexicoSoilTriangular
25coreanensis1KyotoSoilIrregularly pointed
UTE 19HiroshimaSoilSpherical
UTY 6TokyoSoilBipyramidal
   1 KumamotoSoilBipyramidal
Total 42   

Cancer cell-toxicity of purified parasporal inclusions

The purified parasporal inclusions of the three selected Japanese isolates (84-HS-1-11, 89-T-26-17 and 90-F-45-14), and type strains of the three H serovars (israelensis, kyushuensis and kurstaki), were examined for their in vitro cytotoxic activity against vertebrate and insect cells. Inclusion proteins of the three isolates showed neither insecticidal activity against lepidopteran and dipteran insects (data not shown), nor haemolytic activity against human erythrocytes. In contrast, they exhibited marked cytotoxic activities, with different spectra and varied activity levels, against human, fish and insect cells ( Table 3, Fig. 1). The proteins of the two strains, 84-HS-1-11 and 89-T-26-17, had cytocidal activity against leukaemia T cells ( Fig. 1a,b) but not against normal T cells ( Fig. 1e,f). The strain 84-HS-1-11 also showed strong cytocidal activity against HeLa cells ( Fig. 1i). The serovar dakota strain 90-F-45-14 was highly cytotoxic to all the human and insect cells tested ( Fig. 1c,g,k) but not active on human erythrocytes and fish cells ( Table 3). The activities of the three isolates were heat-(100 °C, 10 min)labile and dose-dependent; the minimum protein concentrations causing detectable cytopathy varied from 1 to 100 μg ml−1, depending on the cell species (data not shown). In general, the cytopathy occurred in 1 h, except for strain 84-HS-1-11 in which CPE became visible after several hours.

Table 3. In vitro cytocidal and haemolytic activities of parasporal inclusion proteins in the three selected Japanese isolates and the type strains of the three H serovars of Bacillus thuringiensis
Cytocidal activity* against
 Human T cellHuman lung cell Insect cell
StrainH antigen (serovar)Haemolysis on human erythrocyte LeukaemiaNormalCancerNormalHuman uterus cervix cancer cellGoldfish (Carassius auratus) scale cell Bombyx moriAedes albopictus
84-HS-1-11UTE++ (++) − (−) + (±) ++ (+) +++ (+++) + (+) − (+) − (+)
89-T-26-17UTY++ (++) − (−) ± (±) − (−) ++ (++) − (+) − (+) + (+)
90-F-45-1415 (dakota) +++ (+++) +++ (++) +++ (+++) +++ (+++) +++ (++++) − (+) ++ (+++) +++ (++)
israelensis type strain14 (israelensis) ++++++ (+++) ++ (+) ++++ (+++) ++++ (+++) ++++ (++++) +++ (+++) +++ (+++) ++++ (+++)
kyushuensis type strain11ac (kyushuensis) +++++ (++) ± (±) ++++ (+++) ++++ (+++) + (++) NT+++ (+++) ++++ (+++)
HD-73‡3abc (kurstaki) − (−) − (−) − (−) − (−) − (+) NT− (−) − (+)
*Alkali-solubilized, protease-treated inclusion proteins (124 μg ml−1) were examined for cytotoxic activities on: MOLT-4 (human leukaemia T cell), normal human T cell (primary culture), A549 (human lung cancer cell), MRC-5 (normal human lung fibroblast cell), HeLa (human uterus cervix cancer cell), GF-Scale (Carassius auratus goldfish scale cell), BM-N (Bombyx mori silkworm cell), and NIAS-AeAl-2 (Aedes albopictus mosquito cell). The results with microscopic observations are shown in upper rows, and the levels of cytotoxicity, assessed by cell proliferation assay, are presented in parentheses. The degree of cell damage was graded as follows: extremely high (++++), high (+++), moderate (++), low (+), very low (±), and no damage (−). †The type strain of serovar kurstaki. UTE, Untestable due to no flagellation and/or autoagglutination. UTY, Untypable by reference antisera. NT, Not tested.

Figure 1. Cytopathic effect of parasporal inclusion proteins of the three non-insecticidal Bacillus thuringiensis strains. Phase-contrast microscopy at 24 h post-inoculation. Bar shown in l represents 50 μm for a–h, 100 μm for i–l

Download figure to PowerPoint

Indiscriminative cytotoxicities to human cells, fish cells and insect cells were associated with parasporal inclusions of the two mosquitocidal reference strains, israelensis and kyushuensis. The strain HD-73 (serovar kurstaki), a Lepidoptera-toxic strain, was virtually non-toxic to the vertebrate and invertebrate cells used here.

SDS-PAGE of inclusion proteins and immunoblot analysis

When analysed by SDS-PAGE ( Fig. 2), the purified inclusions of 84-HS-1-11 contained a single major protein of 81 kDa. In contrast, at least five major proteins were evident in purified inclusions of both 89-T-26-17 and 90-F-45-14. Protease treatment generated major bands of 61 and 28 kDa in 89-T-26-17 and 90-F-45-14, respectively, while the major protein of the strain 84-HS-1-11 was degraded rapidly into smaller molecules of < 20 kDa.


Figure 2. SDS-PAGE and immunoblot analysis of the cancer cell-killing parasporal proteins of the three non-insecticidal Bacillus thuringiensis strains. Immunoblotting was done with antisera raised against inclusion proteins of the type strains of (a) serovar israelensis, (b) kyushuensis, and (c) kurstaki. Each lane contained 5 μg of inclusion proteins. Molecular markers are indicated on the left

Download figure to PowerPoint

When examined by immunoblotting, three reference polyclonal antisera against the type strains of serovars israelensis, kyushuensis and kurstaki showed little cross-reactivity to inclusion proteins of 84-HS-1-11, 89-T-26-17 and 90-F-45-14 ( Fig. 2), with the exception that the 81 kDa protein of the strain 84-HS-1-11 gave a very weak reaction for kurstaki antibodies.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

It is reported here for the first time that certain non-insecticidal B. thuringiensis inclusion proteins have a novel, pronounced biological activity, a cytocidal toxicity to human cancer cells. A large-scale screening clearly showed the existence of B. thuringiensis strains whose parasporal inclusion proteins are neither insecticidal nor haemolytic but highly cytocidal in vitro for human cells, including certain cancer cells. Earlier, two interesting observations have been made by Indian ( Prasad & Shethna 1974) and Japanese ( Seki et al. 1978 ) workers who reported that Lepidoptera-toxic B. thuringiensis crystal proteins exhibited anti-tumour activities against mouse ascite sarcoma cells. In the present study, however, we could not find cytocidal activity against human cancer cells among Lepidoptera-toxic parasporal proteins. The two mosquitocidal strains, israelensis and kyushuensis, showed strong anti-cancer cell activity. It is very likely, however, that the activity in these strains is attributable to the broad-spectrum cytolysins, Cyt1 and Cyt2 proteins ( Koni & Ellar 1994). Immunologically, it is clear that the proteins of the three strains (84-HS-1-11, 89-T-26-17 and 90-F-45-14) are not allied to the Cyt proteins and the other established classes of the insecticidal proteins, Cry1 and Cry4.

It also appeared from the results that there is a marked variation among proteins in the cytotoxic spectrum as well as in the level of cell-killing activity. Of particular interest is the fact that the proteins from 84-HS-1-11 and 89-T-26-17 are able to discriminate between human leukaemia T cells and normal T cells, killing the former cells specifically. Recent studies have demonstrated the occurrence of B. thuringiensis strains that have very narrow insecticidal spectra. The examples include: (i) a strain of the serovar japonensis that produces the Cry8C protein specifically active on only a few genera of the family Scarabaeidae (Coleoptera) ( Ohba et al. 1992 ; Sato et al. 1994 ), and (ii) a strain of the serovar leesis whose parasporal proteins exhibit larvicidal activity highly specific for the moth-fly, Telmatoscopus albipunctatus (Diptera: Psychodidae) ( Higuchi et al. 1998 ). The findings, coupled with our present observations, strongly suggest the possible occurrence, in non-insecticidal, naturally-occurring B. thuringiensis populations, of strains with proteins highly selective for given types of cancer cells. This may lead to the use of B. thuringiensis for medical purposes.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References

The authors thank Dr T. Kawarabata, Kyushu University, for the critical reading of the manuscript.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. References
  • Behl, C., Davis, J., Cole, G.M., Schubert, D. 1992 Vitamin E protects nerve cells from amyloid β protein toxicity. Biochemical and Biophysical Research Communications, 186, 944 950.
  • Cannon, R.J.C. 1996 Bacillus thuringiensis use in agriculture: a molecular perspective. Biological Reviews of the Cambridge Philosophical Society, 71, 561 636.
  • Crickmore, N., Zeigler, D.R., Feitelson, J et al. 1998 Revision of the nomenclature for the Bacillus thuringiensis pesticidal crystal proteins. Microbiology and Molecular Biology Reviews, 62, 807 813.
  • Hastowo, S., Lay, B.W., Ohba, M. 1992 Naturally occurring Bacillus thuringiensis in Indonesia. Journal of Applied Bacteriology, 73, 108 113.
  • Heiss, P., Bernatz, S., Bruchelt, G., Senekowitsch-Schmidtke, R. 1997 Cytotoxic effect of immunoconjugate composed of glucose-oxidase coupled to an anti-ganglioside (GD2) antibody on spheroids. Anticancer Research, 17, 3177 3178.
  • Higuchi, K., Saitoh, H., Mizuki, E., Hwang, S.-H., Ohba, M. 1998 A novel isolate of Bacillus thuringiensis serovar leesis that specifically exhibits larvicidal activity against the moth-fly, Telmatoscopus albipunctatus. Systematic and Applied Microbiology, 21, 144 150.
  • Ishii, T. & Ohba, M. 1993 Diversity of Bacillus thuringiensis environmental isolates showing larvicidal activity specific for mosquitoes. Journal of General Microbiology, 139, 2849 2854.
  • Ishiwata, S. 1901 On a kind of severe flacherie (sotto disease). Dainihon Sanshi Kaiho, 114, 1 5.
  • Koni, P.A. & Ellar, D.J. 1994 Biochemical characterization of Bacillus thuringiensis cytolytic δ-endotoxins. Microbiology, 140, 1869 1880.
  • Laemmli, U.K. 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680 685.
  • Lambert, B. & Peferoen, M. 1992 Insecticidal promise of Bacillus thuringiensis. Bioscience, 42, 112 122.
  • Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J. 1951 Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry, 193, 265 275.
  • Meadows, M.P., Ellis, D.J., Butt, J., Jarrett, P., Burges, H.D. 1992 Distribution, frequency, and diversity of Bacillus thuringiensis in an animal feed mill. Applied and Environmental Microbiology, 58, 1344 1350.
  • Ohba, M. 1996 Bacillus thuringiensis populations naturally occurring on mulberry leaves: a possible source of the populations associated with silkworm-rearing insectaries. Journal of Applied Bacteriology, 80, 56 64.
  • Ohba, M. 1997 Three Bacillus thuringiensis flagellar serovars widely occurring in natural environments of Japan. Journal of Basic Microbiology, 37, 71 76.
  • Ohba, M. & Aizawa, K. 1978 Serological identification of Bacillus thuringiensis and related bacteria isolated in Japan. Journal of Invertebrate Pathology, 32, 303 309.
  • Ohba, M. & Aizawa, K. 1986 Insect toxicity of Bacillus thuringiensis isolated from soils of Japan. Journal of Invertebrate Pathology, 47, 12 20.
  • Ohba, M., Iwahana, H., Asano, S., Suzuki, N., Sato, R., Hori, H. 1992 A unique isolate of Bacillus thuringiensis serovar japonensis with a high larvicidal activity specific for scarabaeid beetles. Letters in Applied Microbiology, 14, 54 57.
  • Ohba, M., Yu, Y.M., Aizawa, K. 1988 Occurrence of non-insecticidal Bacillus thuringiensis flagellar serotype 14 in the soil of Japan. Systematic and Applied Microbiology, 11, 85 89.
  • Prasad, S.S.S.V. & Shethna, Y.I. 1974 Purification, crystallization and partial characterization of the antitumour and insecticidal protein subunit from the δ-endotoxin of Bacillus thuringiensis var. Thuringiensis. Biochimica et Biophysica Acta, 363, 558 566.
  • Rodriguez-Padilla, C., Galau-Wong, L., De Barjac, H., Roman-Calderon, E., Tamez-Guerra, R., Dulmage, H. 1990 Bacillus thuringiensis subspecies neoleonensis serotype H-24, a new subspecies which produces a triangular crystal. Journal of Invertebrate Pathology, 56, 280 282.
  • Roh, J.Y., Park, H.W., Jin, B.R., Kim, H.S., Yu, Y.M., Kang, S.K. 1996 Characterization of novel non-toxic Bacillus thuringiensis isolated from Korea. Letters in Applied Microbiology, 23, 249 252.
  • Saitoh, H., Higuchi, K., Mizuki, E., Ohba, M. 1996 Larvicidal activity of Bacillus thuringiensis natural isolates, indigenous to Japan, against two nematoceran insect pests occurring in urban sewage environments. Microbiological Research, 151, 263 271.
  • Saitoh, H., Higuchi, K., Mizuki, E., Ohba, M. 1998 Larvicidal toxicity of Japanese Bacillus thuringiensis against the mosquito Anopheles stephensi. Medical and Veterinary Entomology, 12, 98 102.
  • Sato, R., Takeuchi, K., Ogiwara, K et al. 1994 Cloning, heterologous expression, and localization of a novel crystal protein gene from Bacillus thuringiensis serovar japonensis strain Buibui toxic to scarabaeid insects. Current Microbiology, 28, 15 19.
  • Seki, T., Nagamatsu, M., Nagamatsu, Y et al. 1978 Injuring reaction of δ-endotoxin upon sarcoma 180 ascites cells and silkworm midgut cells in vitro. Science Bulletin of the Faculty of Agriculture Kyushu University, 33, 19 24.
  • Yudina, T.G. & Burtseva, L.I. 1997 Activity of δ-endotoxin of four Bacillus thuringiensis subspecies against prokaryotes. Mikrobiologiya, 66, 25 31.