Demonstrated by Electron Microscopy

Transcription

1 JOURNAL OF BACTERIOLOGY, Feb. 1969, p American Society for Microbiology Vol. 97, No. 2 Printed in U.S.A. Toxin Production in Clostridium botulinum as Demonstrated by Electron Microscopy JOHN J. DUDA AND JOHN M. SLACK Department of Microbiology, Medical Center, West Virginia University, Morgantown, West Virginia Received for publication 2 November 1968 Spheroplasts of Clostridium botulinum 62A were prepared with the use of lysozyme. These spheroplasts were then exposed to ferritin-labeled type A antitoxin. Ultrathin sections of these specimens revealed the ferritin-labeled antibody symmetrically arranged around the outer spore coats but not within the spore cortex. The ferritin-labeled antibody was also observed in the bacterial cytoplasm. Here it was arranged in aggregates and strands, although it was not associated with any identifiable cell structure. Controls included sections of C. botulinum spheroplasts treated with a 1.5% solution of ferritin as well as spheroplasts of C. roseum and Bacillus subtilis treated with conjugated type A antitoxin or a 1.5% solution of ferritin. No intracellular or extracellular ferritin was demonstrable in these specimens. The manner of exotoxin formation in clostridia is largely unknown. Bonventre (1) suggested that it is a unique process because the greater part of the toxin is found during the catabolic stage of cell growth. Additional experiments by Bonventre indicated that the toxin of Clostridium botulinum is synthesized as a comparatively nontoxic molecule during the period of cellular proliferation, and that the period of cellular degeneration which follows is characterized,< by autolysis of the cell with a subsequent activation of a toxin precursor by the normal proteolytic enzymes of the bacterial cell. Yamagishi et al. (1) showed that an intimate correlation exists between sporulation and toxigenicity in C. perfringens. The same relationship was demonstrated in C. sordellii by Nishida et al. (8) and in C. tetani by Sanada and Nishida (9). Nishida and Nakagawara (7), who observed a relationship between sporulation and toxicity in C. novyi, suggested that this relationship commonly exists in Clostridium. It was considered possible that toxin production in spheroplasts of C. botulinum could be established by use of ferritin-conjugated antibodies and electron microscopy. The results indicate a possible correlation between sporulation and toxigenesis. MATERIALS AND METHODS Organism and culture medium. C. botulinum type 62A (ATCC 798) was grown in a Trypticase and soy peptone medium (15) modified by substituting Edamin R (Sheffield Chemical, Norwich, N.Y.) for the soy peptone. This modification resulted in increased production of toxin. The modified Trypticase and soy peptone medium (MTSP) contained, per liter of distilled water: trypticase, 15 g; Edamin R, 10 g; NaCl, 2.5 g; K2HPO, 2.5 g; and glucose, 2 g. The medium was adjusted to ph 7.0 with 1 N NaOH and was autoclaved immediately before use. Ferritin-antibody conjugation. Specific type A antitoxin was obtained from Burroughs-Wellcome (Tuchahoe, N.Y.). The globulin was precipitated from a 50% ammonium sulfate solution and dialyzed. The protein concentration was estimated by the biuret method (3). Ferritin was conjugated to the specific globulin, with the use of toluene 2,-diisocyanate as the coupling agent, according to the method of Singer and Schick (11). Anaerobiosis. Anaerobiosis was obtained by bubbling illuminating gas through a sterile cotton-stoppered pipette into the medium for several minutes. The flasks were then sealed with Parafilm (American Can Co., Neenah, Wis.). Preparation of stock spore suspension. A 30-ml amount of MTSP medium was inoculated with C. botulinum and incubated for 2 hr at 37 C. This entire culture was then transferred to 300 ml of fresh MTSP medium. After incubation for 2 hr, this entire culture was then inoculated into 3 liters of MTSP medium and again incubated. When this final culture had attained 90% sporulation, as determined by the Gram stain, it was stored at C for 7 days. The spores were sedimented by centrifugation, washed 10 times with sterile distilled water, suspended in 100 ml of sterile distilled water, and stored as a watery paste at C. Samples of this suspension served as inoculum in subsequent experiments. 900

2 VOL. 97, 1969 TOXIN PRODUCTION IN CLOSTRIDIUM BOTULINUM 901 Determination of optimal temperature for heatshocking spores. To obtain cultures with a homogeneous population, determined by light microscopy and a simple methylene blue stain to consist of one type of cell whic"h would germinate uniformly, it was necessary to heat-shock the spores. The optimal heatshocking temperature was determined by subjecting samples of a diluted spore suspension to temperatures ranging from 60 to 80 C as described by Stewart (Ph.D. Thesis, West Virginia Univ., Morgantown, 1963). The temperature was increased in increments of 5 C, and samples were removed from each treatment series of 0, 5, 10, 15, and 20 min. Then each sample was suspended in 10 ml of melted yeast extract-agar. The suspensions were poured into sterile oval tubes and the agar was allowed to harden. Anaerobiosis was obtained by overlaying the inoculated medium with 3 ml of melted non-nutrient agar. Colony counts were made on a Quebec colony counter after incubation for 2 hr at 37 C. Optimal time and heat-shocking temperature were selected on the basis of the sample giving the highest colony count. Toxin assay. Male Swiss-Webster mice weighing 20 to 25 g were used routinely in the assay for toxin in culture supernatant fluids and supernatant fluids of ultrasonically treated cells. These fluids were diluted in a series of 10-fold dilutions in phosphate-buffered gelatin (ph 6.0). Then a 0.5-ml amount was injected intraperitoneally into each mouse; five mice were used for each dilution. The death rate over a -day period was recorded. Preparation of spheroplasts. Samples were taken hourly over a 20-hr period from a growing culture. The cells were washed three times with cold distilled water, and after the last washing the sedimented cells were suspended in 5 ml of MTSP medium containing 0.3 M sucrose and 0.01 M magnesium chloride. A 0.5- mg amount of lysozyme (Nutritional Biochemicals Corp., Cleveland, Ohio) and 5 drops of ferritin-conjugated antibody were added to each 1-ml amount of cell suspension (13). The samples were incubated at 37 C, and the formation of spheroplasts was followed by phase microscopy. The time required for the formation of spheroplasts varied from 2 to 5 hr depending upon the age of the cells; the younger cells required the longer period of time. Fixation of spheroplasts. As spheroplasts readily rupture with changes in osmotic pressure, it was necessary to subject them to a series of fixation processes. This began with a prefixation accomplished by adding 0.02 ml of 25% glutaraldehyde to each tube, giving a final concentration of 0.1% glutaraldehyde. These samples were incubated at 37 C for 2 hr and then an additional 0. ml of 25% glutaraldehyde was added, giving a final concentration of 2% glutaraldehyde per sample. The samples were then placed at C for 2 hr to allow complete fixation of the spheroplasts. The fixed spheroplasts were then washed three times with tris(hydroxymethyl)aminomethane (Tris) buffer (ph 7.2) containing 1.5% sucrose and 0.01 M magnesium chloride. The pellet remaining after the last washing was processed for ultrathin sectioning and electron microscopy. Preparation of fixed specimens for electron microscopy. The pellet of the glutaraldehyde-fixed spheroplasts was suspended in a few drops of melted agar and placed on a glass slide. After the agar hardened, it was cut into 1-mm squares (5). These blocks were then stained for 2 hr in 0.5% uranyl acetate in Tris buffer (ph 7.2). The blocks were then dehydrated through graded alcohols, two changes of propylene oxide, and increasing concentrations of Epon in propylene oxide. They were then embedded in Epon 812 with DMP-30 serving as the accelerator. Polymerization followed at 60 C for 72 hr. The embedded specimens were sectioned on an MT- 2 Porter-Blum ultramicrotome (Ivan Sorvall, Inc., Norwalk, Conn.), with a diamond knife (Dupont). The resulting sections were collected on 200-mesh copper grids which had been coated with a layer of carbon and were subsequently examined in JEM-T6S and RCA EMU-3 electron microscopes. RESULTS Heat-shocking. The highest rate of germination of spores of C. botulinum type 62A, as indicated by colony counts, was obtained from spores heated at 75 C for 5 min. This temperature and time was subsequently used to provide a homogeneous culture. Production of spheroplasts. Through the combined use of a high concentration (0.5 mg/ml of cell suspension) of lysozyme and a prolonged incubation (2 to 5 hr), spheroplasts of C. botulinum 62 A were produced. A spheroplast with a centrally located spore is shown in Fig. 1. Ferritin penetration of spheroplasts. The ferritin molecule has a molecular weight of approximately 500,000 (2), and when it is conjugated to the antitoxin the combined molecular weight is approximately 650,000. Such a large molecule will not penetrate the intact cell wall but will penetrate a spheroplast, as shown in Fig. 2. The ferritin appears as electron-dense material (black dots) in the photographs. Growth curve and mouse assay of toxin. Type A toxin was detectable by mouse bioassay as early as the 12th hr of growth in the culture supematant fluid of specimens removed from an actively growing culture at 1-hr intervals. However, ferritin-labeled antitoxin was detected in the cell cytoplasm as early as the 6th hr of the same growth curve. Toxin in the cytoplasm of the protoplasts. The retention of the conjugated ferritin within the spheroplast results from the toxin-antitoxin reaction, although no identifiable structure could be associated with the areas of reaction. The ferritin was observed to be deposited as strands or aggregates of material, but the basis for sucn an arrangement could not be determined. Toxin in relation to the spore. In sections of spheroplasts which contained spores, the ferritin

3 :: -, 'SJW. 2>f '"... ' "; *,.':: :+.. '... w1 ;.,. :R :M"..* ',,..... y,: :.::::. 902 DUDA AND SLACK J. BACrRIOL I III -u A,ii :s..,... ts'' ',w's S'.'.'.s ;.t:"''3's..::.-'... ;t6vs 2- S.$jt t i.': sl ::, SC w ::..CW Al' I FIG. 1. Ultrathin section ofone unstabilized spheroplast of C. botulinum 62A. The spheroplast was prepared in MTSP medium containing lysozyme (0.5 mg/ml). The spore is centrally located. Remnants ofthe cell wall (cw) are also seen. Spore coats (sc) are readily identifiable. Specimen was not stained with lead citrate. Glutaraldehydefixation was used to eliminate artifacts; however, this resulted in loss ofcytological detail. X 36,000. In all bars, I mm is equivalent to I pum. was deposited at the outer spore coat and exosporium (Fig. 2). In Fig. 3, the outer spore coat is separated from the exosporium, and the ferritin is deposited on both surfaces of the exosporium as well as on the outer spore coat. Controls. Spheroplasts treated with a 1.5% nonconjugated solution of ferritin did not retain the ferritin, nor was any ferritin deposited in or around the spore. This rules out the possibility that ferritin (Fig. 2 and 3) was deposited as the result of a nonspecific trapping of the particles. Spheroplasts were prepared from C. roseum and Bacillus subtilis and were stained with both conjugated and nonconjugated ferritin. Sections stained in this manner showed no ferritin deposits. This provided additional evidence that the ferritin observed with C. botulinum spheroplasts resulted from a specific antigen-antibody reaction and not from a nonspecific trapping effect. DISCUSSION The physiology of toxin production in C. botulinum type A has been thoroughly investigated, but there are still unanswered questions. We attempted to supply an answer to one of these questions through the determination of a specific site of toxin production by use of the ferritinconjugated antibody technique developed by Singer (10). As the ferritin-conjugated antibody is a large molecule which does not penetrate the rigid bacterial cell wall, the first step was to produce spheroplasts (6) to facilitate the entrance of this molecule. Bonventre (1) had been unsuccessful in his attempts to produce these structures; however, using lysozyme (0.5 mg/ml), we were able to produce spheroplasts of C. botulinum 62A in 2 to 5 hr. Ultrathin sections of spheroplasts prepared from samples of cells removed at 1-hr intervals over a 2-hr period and then treated with ferritinlabeled antitoxin revealed no ferritin prior to the 6th hr of growth, which, according to this method, would indicate that toxin was not demonstrable prior to 6 hr. After this time, ferritin staining was observed in the forms of aggregates, strands, and

4 ONA 2 s FIG. 2. Ultrathin section ofa 20-hr cell of C. botulinum 62A (unstabilized). This cell was treated with ferritinlabeled antibody. Ferritin (f) is symmetrically arranged around the spore coats and exosporium. An impermeable spore cortex (cx) is the limit ofthe ferritin penetration. The ferritin is not present on the outside of the exosporium. Fixed in 2% glutaraldehyde. Specimen was not stained with lead citrate. X 52,000..~~~~~~~~~~~~~~~~~~~~~.-. AdSMF_Rfi Sf *~~~~~ Stained with lead citrate.'.x.30,000 F;bJ.;~~~~~~~~~~~~i s.::.wev FIG. 3. Ultrathin section ofan unstabilized spheroplast prepared from a 20-hr culture ofc. botutinum 62A. The ferritin (f) is located on both surfaces ofthe exosporium (ex).- Ferritin is also present on the outer spore coats (sc). Stained with lead citrate. X 30,

5 90 DUDA AND SLACK J. BAcrERIOL. spiral arrangements, but no specific cell structure could be identified. It was not until the 20th hr of growth that a location within the cell could be identified with the ferritin-labeled antibody. The indication that the spore in itself may play a role in toxin formation should not be surprising, since a considerable amount of protein synthesis occurs during sporulation. Halvorson () estimated that 80% of de novo protein synthesis occurred during spore formation, and Walker et al. (12), using ferritin-conjugated antibody in a study involving B. subtilis, demonstrated that the proteins of the vegetative cell and its spore are different. However, it should be noted that it is not necessary for a strain of C. botulinum type A to sporulate in order to be toxigenic. Bonventre (1), working with strain JTD-IV of C. botulinum type A, reported that this strain did not sporulate under any conditions; however, toxin was produced. ACKNOWLEDGMENTS We are grateful to H. G. Voelz for his valuable suggestions and help in preparation of the electron micrographs and for his review and criticism of the manuscript. This investigation was supported by Public Health Service grant AI from the National Institute of Allergy and Infectious Diseases and by Public Health Service Predoctoral Fellowship 1-Fl-GM-2, 523-OlAl from the National Institutes of Health during the years of LITERATURE CITED 1. Bonventre, P. F., and L. L. Kempe Physiology of toxin production by Clostridium botulinum types A and B. I. Growth, autolysis, and toxin production. J. Bacteriol. 79: Farrant, J. L An electron microscopic study of ferritin. Biochim. Biophys. Acta 13: Gornall, A. G., C. J. Bardawill, and M. M. David Determination of serum proteins by means of the biuret reaction. J. Biol. Chem. 177: Halvorson, H Function and structure in microorganisms. Symp. Soc. Gen. Microbiol. 15: Kellenberger, E., A. Ryter, and J. Sechaud Electron microscopic study of DNA containing plasms. J. Biophys. Biochem. Cytol. : Lee, S Ferritin labeled antibody as an electron dense stain for phage protein. Exptl. Cell Res. 21: Nishida, S., and G. Nakagawara Relationship between toxigenicity and sporulating potency of Clostridium novyi. J. Bacteriol. 89: Nishida, S., K. Tamai, and T. Yamagishi Taxonomy of Clostridium bifermentans and Clostridium sordellii. I. Their toxigenicity, urease activity, and sporulating potency. J. Bacteriol. 88: Sanada, I., and S. Nishida Isolation of Clostridium tetani from soil. J. Bacteriol. 89: Singer, S. J Preparation of an electron dense antibody conjugate. Nature 183: Singer, S. J., and A. F. Schick The properties of specific stains for electron microscopy prepared by the conjugation of antibody molecules with ferritin. J. Biophys. Biochem. Cytol. 9: Walker, P. D., R. 0. Thompson, and A. Baillie Use of ferritin labeled antibodies in the location of spore and vegetative antigens of Bacillus subtilis. J. Appl. Bacteriol. 30: Weibull, C The isolation of protoplasts from Bacillus megaterium by controlled treatment with lysozyme. J. Bacteriol. 66: Yamagishi, T., S. Ishida, and S. Nishida Isolation of toxigenic strains of Clostridium perfringens from soil. J. Bacteriol. 88: Zoha, S. M. S., and H. L. Sadoff Production of spores. by a putrefactive anaerobe. J. Bacteriol. 76: