CC BY
24Russian Journal of Nematology, 2011, 19 (2), 181 - 188
Antimicrobial activity of protein inclusions from bacteria symbiotic with entomopathogenic
nematodes
Tatiana G. Yudina1, Igor A. Zalunin2, Alla P., Zarubina1, Danyang Guo1,
3 3
Nadezhda S. Shepeleva and Sergei E. Spiridonov
'Department of Microbiology, Biology Faculty, Moscow State University, 119991, Moscow, Russia,
e-mail: yudinatg@mail.ru 2The Scientific Research Institute for Genetics and Selection of Industrial Microorganisms, 1st Dorozhny proezd 1, Moscow 117545, Russia. 3Centre of Parasitology, A.N. Severtsov Institute of Ecology and Evolution, Rusian Academy of Sciences, Leninskii pr., 33, Moscow, 119071, Russia.
Accepted for publication 10 November 2011
Summary. Proteins of inclusions in bacteria symbiotic to entomopathogenic nematodes Xenorhabdus spp. and Photorhabdus spp. have a pronounced antibiotic effect suppressing growth of different bacteria on agar media. The toxicity of proteins from Photorhabdus luminescens and entomopathogenic bacteria Bacillus thuringiensis inclusions towards Escherichia coli has been compared. Inclusion proteins from Xenorhabdus bovienii inhibit the growth of two assayed subspecies of B. thuringiensis: kurstaki and israelensis. The vegetative cells of X. bovienii, isolated from Steinernema feltiae, proved to be sensitive to the B. thuringiensis ssp. israelensis CytlA crystal protein.
Key words: antibiotic effect, entomopathogenic nematodes, inclusion proteins, symbiotic bacteria,
Xenorhabdus.
Xenorhabdus and Photorhabdus bacteria colonise the intestines of the infective soil-dwelling stage of entomopathogenic nematodes (EPN) of the genera Steinernema and Heterorhabditis, respectively. Infective juveniles infect susceptible insect larvae and release the bacteria into the insect haemocoel. The bacteria kill the insect larvae and convert the cadaver into a food source suitable for nematode growth and development (Goodrich-Blair & Clarke, 2007). Two forms or phases of both Photorhabdus and Xenorhabdus are reported, with the first phase mainly found in nematode infective juveniles, and the second phase arising inside insect cadavers at the late stages of nematode development. It is the first phase which produce the main part of insecticidal toxins, antibiotics and other biological active components (Baneijee et al., 2006, Koppenhofer, 2007;), including intracellular protein inclusions, which can account for 40% of the total protein content of cells (Bintrim & Ensign, 1998; Bowen & Ensign, 2001). It was also presumed that these proteins contribute to the nutrition of the host nematode. Photorhabdus has two distinct types of
crystals formed by small (10 kDa), hydrophobic proteins encoded by cipA and cipB genes (Bintrim & Ensign, 1998). Xenorhabdus nematophila also produced two crystal proteins, IP1 (26 kDa), and IP2, (22 kDa), but only the gene encoding IP1, pixA, has been identified (Couche & Gregson, 1987; Goetsch et al, 2006). Consistent with a potential role for crystal proteins in nutrition, when expressed in Escherichia coli, Photorhabdus cipA or cipB can promote nematode development (Joyce et al, 2006; You et al, 2006).
Unlike S-endotoxins from inclusion proteins of the entomopathogenic bacterium Bacillus thuringiensis, the inclusion proteins of EPN symbiotic bacteria are not insecticidal (Koppenhofer, 2007). Some S-endotoxins (Cry proteins of classes 5, 6, 12, 13) are active against nematodes (Soberon et al., 2010). The S-endotoxins are divided into two families: the invertebrate-specific Cry toxins and cytolytic Cyt toxins. Their cytolytic effect does not require binding to a membrane receptor, as in case of Cry proteins, but rather is mediated by direct binding to the lipid (Bravo et al, 2007; Li et al, 2009). Also, Cyt proteins can be the receptors for Cry proteins
Table. 1. Antibacterial activities of inclusion proteins from bacterial symbionts of entomopathogenic nematodes.
Test microorganism Specific antibacterial activities of inclusion proteins, U
(mm mkg- ):
Xenorhabdus bovienii Xenorhabdus bovienii Photorhabdus
str. T 319 from Steinernema intermedium. luminescens str.ZMl
Micrococcus luteus strain 140 67.5 ± 7.3 39.8 ± 5.0 47.5 ± 4.9
M. luteus 137 77.1 ± 8.7 42.5 ± 4.7 52.7 ± 5.5
M. aurantiacus 131 64.1 ± 6.1 45.8 ± 4.7 57.1 ± 6.1
Rhodococcus erytropolis 119 12.9 ± 1.5 11.7 ± 1.3 21.0 ± 2.7
Rh. rubroperctinctus 117 - 7.8 ± 0.9 19.7 ± 2.1
Brevibacterium citreum 213 5.2 ± 0.5 7.1 ± 0.8 15.3 ± 1.4
Lactococcus lactis 163 21.1 ± 2.5 19.5 ± 1.8 -
Nocardia calcarea 215 16.5 ± 1.9 - 15.0 ± 1.7
Streptomyces rimosus 267 19.2 ± 2.0 13.3 ± 1.4 17.0 ± 1.8
Streptomyces chrysomallus 257 11.7 ± 1.3 10.5 ± 1.3 21.7 ± 2.3
Bacillus subtilis 9 9.5 ± 0.9 3.7 ± 0.5 Trace activity
B. thuringiensis ssp. kurstaki Z-52 4.5 ± 0.6 2.7 ± 0.4 7.0 ± 0.8
B. thuringiensis ssp. israelensis B-2395 3,5 ± 0,3 - -
Bacillus megaterium 11 Trace activity 0 -
Zoogloea ramigera 1 - - 6.1 ±0.5
Pseudomonas fluorescens 70 18.9 ± 2.1 16.8 ± 1.3 11.9 ± 1.3
Pseudomonas aeruginosa 47 10.7 ± 1.3 8.5 ± 1.1 3.9 ± 0.5
Escherichia coli 52 11.8 ± 1.5 13.7 ± 1.5 12.5 ± 1.5
Erwinia carotovora 35 Trace activity - 0
Note: Trace activity - if the bacteria-free zone around the well < 1-2 mm. '-' - no values.
(Canton et al, 2011). S-Endotoxins from parasporal crystals of B. thuringiensis also display an antibiotic effect on some microorganisms (Yudina & Egorov, 1996; Yudina et al, 2003, 2007; Cahan et al, 2008).
Previously we reported the antimicrobial activity of inclusion proteins in EPN symbiotic bacteria (Yudina & Ivanova, 1996; Yudina & Egorov, 1996; Yudina & Spiridonov, 1997). The comparison of antibacterial effects of inclusion proteins produced by Xenorhabdus and Photorhabdus in comparison with those of the B. thuringiensis is presented here.
MATERIAL AND METHODS
Inclusion proteins were isolated from the X. bovienii (T319) donated by Dr Ray Akhurst. Xenorhabdus bovienii was also isolated from infective juveniles of Steinernema intermedium and Steinernema feltiae. Photorhabdus luminescens strain ZM1 was isolated from Moldavian Heterorhabditis bacteriophora. For isolation, symbiotic bacteria of nematodes were grown on the
NBTA medium (Koppenhofer, 2007). Isolated symbiotic bacteria were cultures on NBTA or in liquid YS media (Koppenhoefer, 2007). The microorganisms studied as test cultures (Table 1) were obtained from the collection of the Microbiology Department of Lomonosov Moscow State University (MDMSU).
To determine the antibiotic activity of inclusion proteins of Xenorhabdus and Photorhabdus bacteria, the test microorganisms were grown on solid media, as described previously (Yudina et al., 2003). The methods of B. thuringiensis cultivation, S-endotoxins isolation and antibacterial activity have also been described (Yudina et al., 2003). Protein inclusions were separated after cell lysis by centrifugation. The washed inclusions were dissolved at pH of 11.5 for 1 h at 35°C, and the crystal proteins were precipitated by glacial acetic acid at pH values close to pI of the proteins and then separated from the supernatant by centrifugation and redissolved in 0.05 M tris-HCl buffer pH 8.0 with 0.2M NaCl or in 0.05 M phosphate buffer, pH 7.8. The obtained dilutions were then immediately subjected
Fig. 1. TEM microphotographs of Xenorhabdus bovienii cells sections undergoing lysis after treatment with Cyt1A protein from Bacillus thuringiensis ssp. israelensis. A: control, formation of protein crystals inside cell; B : slightly swollen cell, cell coating mildly damaged; C, D: vacuolisation, cell swelling, cytoplasm clarification due to disruption of membrane permeability and wall destruction, formation of ring membrane structures from residues of cytoplasmic membrane; E, F: vacuolisation, cell lysis, cell walls rupture and emergence of protein crystals. Outer membrane ripple, and rupture and residues of membranes from lysed cell seen.
to analysis of the antimicrobial activity using an agar diffusion method. The antibiotic activity of CytlA against cells ofX.bovienii was determined in liquid YS medium (Couche & Gregson, 1987; Lancini & Parenti, 1982; Yudina et al, 2003).
The ratio of the growth inhibition zone (width mm) to the introduced protein amount (mkg) obtained in the proportional region of the dose-response curve was taken as the specific antibiotic activity (U). The minimal concentration of the protein solution resulting in the growth inhibition was taken as the minimal inhibiting concentration (MIC), as it previously reported (Lancini & Parenti, 1982; Yudina et al, 2003, 2007).
The dilutions of the inclusion proteins from EPN symbiotic bacteria and those from B. thuringiensis subsp. monterrey were immediately subjected to analysis of the toxicity. The experiments were carried out with genetically engineered bioluminescent Escherichia coli TG1 (pXen7) strain (the principal component of the Ecolum test system, see Strakhovskaya et al., 2002).
Ultrathin sections of EPN symbiotic bacteria were studied using a transmission electron microscope JEM 1011, Jeol Ltd (Japan) according to standard techniques. Xenorhabdus bovienii cells from S. feltiae were contrasted with 2% aqueous solution of uranyl acetate (Gerhardt, 1981; Yudina et al, 2003).
RESULTS
Inclusion proteins (in 0.05 M tris-HCl buffer pH 8.0 with 0.2M NaCl) from three strains of EPN symbionts suppressed the growth of the different test-bacteria (Table1).
Antibacterial action of inclusion proteins from X. bovienii T319 on micrococci (Micrococcus luteus, M. aurantiacus) was characterised by values of the specific antibiotic activity (U) at 40 - 80 mm mkg-1 (Table 1). The MIC of the inclusion protein solution from X. bovienii when used against M. luteus 137 was 5 - 7 mkg ml-1.
Nocardia calcarea, Streptomyces rimosus, S. chrysomallus, Brevibacterium citreum and Lactococcus lactis were significantly affected by EPN inclusion proteins (see Table 1). An effect of P. luminescens inclusion proteins on rhodococci, S. chrysomallus and B. citreum was at least twice as stronger as that of Xenorhabdus inclusion proteins. The bacteriostatic action of P. luminescens on Zoogloea ramigera (U = 6) was observed. The growth of the vegetative cells B. thuringiensis subspp. kurstaki and israelensis was suppressed by Photorhabdus and Xenorhabdus inclusion proteins (U = 2.7 - 7). Weak action or only traces of it were
reported for Bacillus megaterium (Table 1). Among Gram-negative bacteria, both species of Pseudomonas were found to be the most susceptible to antimicrobial activity of EPN inclusion proteins. The effect of Xenorhabdus inclusion proteins on Pseudomonas spp. was 2-5 times higher than that of P. luminescens. Erwinia carotovora cells were not affected by any inclusion proteins in the concentrations used. Escherichia coli susceptibility to X. bovienii inclusion proteins was two times lower than P. fluorescens, whereas P. luminescens effect on both Pseudomonas spp. was nearly identical (Table 1). Inclusion proteins from B. thuringiensis were also found to be bactericidal for M. luteus 140, but U values were lower than those of EPN-symbiotic bacteria: 7.5, 28.8 and 42 for B. thuringiensis subspp. kurstaki, israelensis, monterrey, respectively, and 67.5, 39.8 and 47.5 U for X. bovienii 319, X. bovienii from S. intermedium, and P. luminescens ZM1, respectively.
A bacteriostatic effect of inclusion proteins on E. coli bacteria was observed after 24 - 48 h of bacteria growth on solid media at 30°C or 37°C. After an additional 24 - 48 h growth at 30°C or 37°C, the bacteriostatic zones usually disappeared.
The susceptibility of cells E. coli in liquid media containing an inserted lux operon and a gene for ampicillin resistance for toxicological action of inclusion proteins P. luminescens and B. thuringiensis ssp. monterrey was examined using an Ecolum test system.
The primary control of the P. luminescens inclusion proteins toxicity for E. coli showed that these proteins rapidly associate with E. coli cells significantly reducing their bioluminescence (as the result of disruption in the cell membrane permeability and subsequent cell death). After 5 min of incubation, P. luminescens inclusion proteins (240 mkg ml-1) were already bound to the cells in the suspension and inhibited its bioluminescence by about 40% of the control value. The proteins of B. thuringiensis ssp. monterrey crystals (260 mkg /ml-
inhibited its luminescence by about 80% of the control value only after 15 min of incubation and up to about 50% at the end of treatment, i.e., after 60 min of incubation. About a quarter of all the E. coli cells were viable after the exposure to P. luminescens inclusion proteins and demonstrated growth on LB agar medium, whereas there were only about 40% of viable E. coli cells after exposure to S-endotoxins of B. thuringiensis.
The MIC value of the activity of Cyt1A of B. thuringiensis ssp. israelensis for the cells of X. bovienii isolated from S. feltiae and cultured on
Fig. 2. TEM microphotographs of contrasted Xenorhabdus bovienii cells. A: control, logarithmic phase cell with flagella, fimbriae; B, C: swollen cells with damaged membrane and clarified cytoplasm, inclusions present; D: cell partially destroyed; inclusion seen on intact part.
liquid YS medium was estimated as 12.5 mkg ml-1. The TEM studies on morphology of the phase 1 X. bovienii cells after the treatment with Cyt1A protein (50 mkg ml-1, 1 h treatment under 20°C and then incubated in liquid YS medium 1 h under 30°C, 200 rpm) demonstrated damaged cell surface and cytoplasmic membrane, malformations and cell lysis (Fig. 1). In controls, late logarithmic phase cells did not undergo lysis, proving that cell lysis was caused by the Cyt1A protein treatment. Numerous fimbriae were discernible on the surface of Xenorhabdus cells after the contrasting treatment with uranylacetate. It means that cells are on the exponential phase of their growth and their lysis is a result of the action of Cyt1A protein (Fig.2).
DISCUSSION
Using the agar diffusion method, we have shown that inclusion proteins of EPN bacterial symbionts demonstrate antimicrobial activity against some Gram-positive (Micrococcus, Lactococcus, Rhodococcus, Nocardia, Streptomyces,
Brevibacterium and Bacillus spp.) and Gramnegative (Pseudomonas, Escherichia spp.) aerobic bacteria. Yudina et al. (2003) reported that the manifestation of antimicrobial activity of inclusion proteins strongly depends on salt composition and the presence of reduced compounds in the growth medium for test-culture. The phenomenon that antibacterial agents are growth-rate-dependent and
nutrition-dependent was described in numerous reports (e.g., Hadas et al., 1995).
For some test organisms the activity of inclusion proteins from EPN symbionts and B. thuringiensis differs significantly. Thus, U-values of Xenorhabdus and Photorhabdus proteins against Micrococcus spp. were 2-30 times higher than that of parasporal crystals from different subspecies B. thuringiensis. No antibacterial effect on rhodococci or brevibacteria was demonstrated for B. thuringiensis prior to our study.
The determination of bioluminescence intensity was proved in our study to be a reliable, rapid and sensitive method of detection of antimicrobial activity of inclusion proteins. The disruption in the cell membrane permeability subsequently causes cell death and cessation of bioluminescence. It was shown that parameters of bioluminescence and colony-forming ability of TG1 (pXen7) E. coli were correlated (Strakhovskaya et al, 2002; Zarubina et al, 2009).
Inclusion proteins of P. luminescens were more toxic to E. coli with lux-operone than B. thuringiensis, which was demonstrated by colony count of E. coli retaining the colony-forming ability after treatment with inclusion proteins. At the same time, the Gram-negative S. feltiae bacterial symbionts of X. bovienii were found to be susceptible to the antibacterial activity of CytlA of B. thuringiensis. The TEM studies have shown that CytlA action cause malfunction of cytoplasmic membrane permeability and destruction of cell wall followed by lysis. Similar effects were observed on different microorganisms treated by crystal proteins of B. thuringiensis (Yudina et al., 2003, 2007; Revina et al, 2005). As it is seen from microphotographs (Fig. 2), the contrasted X. bovienii cells are in exponential phase of their growth (having peritrichous flagella and fimbriae) and being lysed due to CytlA protein action. Bacterial symbionts of EPN are motile only when in Phase 1, they then have peritrichous flagella, which is characteristic for the exponential phase of growth (Koppenhofer, 2007). The antibacterial effect of different inclusion proteins on Gram-negative bacteria is significant since the outer membrane usually enables the bacteria to withstand antibacterial molecules by excluding them or reducing their penetration into the cells.
It is also important to note that B. thuringiensis ssp. israelensis producing Cytla protein is susceptible to the antibacterial action of X. bovienii inclusion proteins (Table 1). The inclusion proteins of EPN symbiotic bacteria are similar to the B. thuringiensis Cyt proteins by having relatively low molecular masses of 10-28 kD. By contrast to the
Cry proteins, their cytolytic affect is less specific. The MIC values on micrococci of the proteins studied were also similar: 5-7 mkg ml"1 for X. bovienii and 6-7 mkg ml"1 for CytlA (Yudina et al, 2003).
ACKNOWLEDGEMENT
The authors are grateful to Dr Ray Akhurst, who presented Xenorhabdus bovienii strain T319. The work of S.E. Spiridonov and N. S. Shepeleva was supported by RFBR grant 11-04-00590a.
REFERENCES
Banerjee, J., Singh, J., Joshi, M.C., Ghosh, S. & Banerjee, N. 2006. The cytotoxic fimbrial structural subunit of Xenorhabdus nematophila is a pore-forming toxin. Journal of Bacteriology 188: 79577962.
Bintrim, S.B. & Ensign, J.C. 1998. Insertional inactivation of genes encoding the crystalline inclusion proteins of Photorhabdus luminescens results in mutants with pleiotropic phenotypes. Journal of Bacteriology 180: 1261-1269. Bowen, D.J. & Ensign, J.C. 2001. Isolation and characterization of intracellular protein inclusions produced by the entomopathogenic bacterium Photorhabdus luminescens. Applied and Environmental Microbiology 67: 4834-4841. Bravo, A., Gill, S.S. & Soberón, M. 2007. Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control. Toxicon 49: 423-435. Cahan, R., Friman, H. & Nitzan, Y. 2008. Antibacterial activity of Cyt1Aa from Bacillus thuringiensis subsp. israelensis. Microbiology 154: 3529-3536. Cantón, P.E., Reyes, E.Z., Escudero, I.R., Bravo, A. & Soberón, M. 2011. Binding of Bacillus thuringiensis subsp. Israelensis Cry4Ba to Cyt1Aa has an important role in synergism. Peptides 32: 596-600. Couche, G.A. & Gregson, R. P. 1987. Protein inclusions produced by the entomopathogenic bacterium Xenorhabdus nematophilus subsp. nematophilus. Journal of Bacteriology. 169: 5279-5288. GERHARDT, P. 1981. Manual of Methods for General Bacteriology. American Society for Microbiology, washington, DC, 936 pp. Goetsch, M., Owen, H., Goldman, B. & Forst, S. 2006. Analysis of the PixA inclusion body protein of Xenorhabdus nematophila. Journal of Bacteriology 188: 2706-2710. Goodrich-Blair, H. & Clarke, D.J. 2007. Mutualism and pathogenesis in Xenorhabdus and Photorhabdus: two roads to the same destination. Molecular Microbiology 64: 260-268. Hadas, H., Einav, M., Fishov, I. & Zaritsky, A. 1995. Division-inhibition capacity of penicillin in
Escherichia coli is growth-rate dependent. Microbiology 141: 1081-1083. Joyce, S.A., Watson, R.J. & Clarke, D.J. 2006. The regulation of pathogenicity and mutualism in
Photorhabdus. Current Opinion in Microbiology 9: 127-132.
Koppenhofer, H. S. 2007. Bacterial symbionts of Steinernema and Heterorhabtitis. In: hematology Monographs and Perspectives. (K.B. Khuong & D.J. Hunt. Eds.), vol. 5, pp. 735-808, Brill, Leiden-Boston. Lancini, G. & Parenti, F. 1982. Antibiotics - an
integrated view. New York; Springer, 253 p. Li, X., Nevels, K.J., Gryczynski, Z., Gryczynski, I., Pusztai-Carey, M., Xie, D. & Butko, P. A. 2009. Molecular dynamics model of the Bt toxin Cyt1A and its validation by resonance energy transfer. Biophysical Chemistry 144: 53-61. Soberon, M., Pardo, L., Munoz-Garay, C., Sanchez, J., Gomez, I., Porta, H. & Bravo, A. 2010. Pore formation by cry toxins. in: Proteins; Membrane binding and pore formation. (G. Anderluh & J. Lakely, Eds.), pp. 127-142, Landes and Springer Science+Business Media. Revina, L. P., Kostina, L. I., Dronina, M. A., Zalunin, I. A., Chestukhina, G. G., Yudina, T. G., Konukhova, A. V. & Izumrudova, A. V. 2005. Novel antibacterial proteins from entomocidal crystals of Bacillus thuringiensis ssp. israelensis. Canadian Journal of Microbiology 51: 141-148. Strakhovskaya, M.G., Parkhomenko, I.M., Rumbal, Ya.V., Zarubina, A.P., Danilov, V.S. & Stradanko, E.F. 2002. Photoquenching of the bioluminescence of the genetically engineered
Escherichia coli TG1 (pXen7) strain in the presence of photodithazine. Microbiology (Moscow) 71: 294-297. You, J., Liang, S., Cao, L., Liu, X. & Han, R. 2006. Nutritive significance of crystalline inclusion proteins of Photorhabdus luminescens in Steinernema nematodes. FEMS Microbiology Ecology 55: 178-185. Yudina, T.G. & Egorov, N.S. 1996. Antibacterial activity of protein inclusions of different bacteria. Doklady Biochemistry and Biophysics 349: 369-372. Yudina, T.G. & Ivanova, E.S. 1996. Antibiotic activity of protein inclusions of Xenorhabdus nematophilus and X bovienii from entomogenous nematodes. Russian Journal of hematology 4: 102. Yudina, T.G. & Spiridonov, S.E. 1997. Antimicrobial activity of protein inclusions from bacteria symbiotic with entomopathogenic nematodes. Russian Journal of hematology 6: 41. Yudina, T.G., Konukhova, A.V., Revina, L.P., Kostina, L.I., Zalunin, I.A. & Chestukhina, G.G. 2003. Antibacterial activity of Cry- and Cyt-proteins from Bacillus thuringiensis subsp. israelensis. Canadian Journal of Microbiology 49: 37-44. Yudina, T.G, Bryukhanov, A.L., Zalunin, I.A., Revina, L.P., Shestakov, A.I., Voyushina, N.E., Chestukhina, G.G. & Netrusov, A.I. 2007. Antimicrobial activity of the different proteins and their fragments from Bacillus thuringiensis parasporal crystals against clostridia and archaea. Anaerobe 13: 6-13.
Zarubina, A., P., Lukashev, E.P., Deev, L.I., Parkhomenko, I.M. & Rubin, A.B. 2009. Biotesting the biological effects of single-wall carbon nanotubes using bioluminescent bacteria test-system.
Nanotechnologies in Russia 4: 871-875.
Юдина, Т. Г., И. A. Залунин, A. П., Зарубина, Даньян Го, Н. С. Шепелева, С. Э. Спиридонов.
Антимикробное действие протеиновых включений симбиотических бактерий, ассоциированных с энтомопатогенными нематодами.
Резюме. Протеины из внутриклеточных включений ассоциированных с энтомопатогенными нематодами бактерий Xenorhabdus spp. и Photorhabdus spp., проявляют выраженное антибиотическое действие, подавляя рост различных бактерий на агаризованных средах. Проведено сравнение токсичности для Escherichia coli протеинов включений из Photorhabdus luminescens с таковой из энтомопатогенных бактерий Bacillus thuringiensis. Протеиновые включения из Xenorhabdus bovienii также подавляют рост двух исследованных подвидов B. thuringiensis: kurstaki и israelensis. В то же время вегетативные клетки X. bovienii, изолированные из Steinernema feltiae, оказались чувствительными к действию кристаллических протеинов CytlA, полученных из кристаллов B. thuringiensis ssp. israelensis.
