An improved sporulation medium has been developed in which all five strains of Clostridium perfringens tested exhibited a 100to 10,000-fold increase in numbers of spores when compared with spore yields in SEC medium under comparable conditions. In addition, three of five strains produced a 100to 1,000-fold increase, with the remaining two strains yielding approximately the same numbers of spores, when compared with strains cultured in Ellner medium. At the 40-hr sampling time, 18 of 27 strains produced a 10to 100-fold increase in numbers of spores in our medium, when compared to spore production obtained in a medium recently reported by Kim et al. The new medium contained yeast extract, 0.4%; proteose peptone, 1.5%; soluble starch, 0.4%; sodium thioglycolate, 0.1%; and Na2HPO4. 7H20, 1.0%. In some cases, the spore yield could be increased by the addition of activated carbon to the new medium. The inclusion of activated carbon in the me

The ability of Clostridium perfringens type A to produce an enterotoxin active in human food poisoning has been shown to be directly related to the ability of the organism to sporulate. Enterotoxin was produced only in a sporulation medium and not in a growth medium in which sporulation was repressed. Mutants with an altered ability to sporulate were isolated from an sp + ent + strain either as spontaneous mutants or after mutagenesis with acridine orange or nitrosoguanidine. All sp 0 − mutants were ent − . Except for one isolate, these mutants were not disturbed in other toxic functions characteristic of the wild type and unrelated to sporulation. A total of four of seven osp 0 mutants retained the ability to produce detectable levels of enterotoxin. None of the ent − mutants produced gene products serologically homologous to enterotoxin. A total of three sp − mutants, blocked at intermediate stages of sporulation, produced enterotoxin. Of these mutants, one was blocked at stage III, one probably at late stage IV, and one probably at stage V. A total of three sp + revertants isolated from an sp − ent − mutant regained not only the ability to sporulate but also the ability to produce enterotoxin. The enterotoxin appears to be a sporulation-specific gene product; however, the function of the enterotoxin in sporulation is unknown.

The ability of cell extracts and culture ifitrates of various strains of C. perfringens to produce ileal loop fluid accumulation and overt diarrhea in rabbits was tested. Good correlation was obtained in the ability of whole cells and a toxic factor (present in cell extracts and concentrated culture filtrates) to produce both fluid accumulation in ileal loops and diarrhea when injected into the normal ileum of the rabbit. The toxic factor was present in cell-free preparations when cells were grown in a sporulation medium, but not when they were grown in an asporogenic medium. The factor was shown to be heat labile, nondialyzable, and was inactivated by Pronase but not by trypsin, lipase, or amylase. Loss of activity occurred at pH 1.0, 3.0, 5.0, and 12.0.

The ligated loop of the rabbit intestine was investigated as a possible experimental model for the study of Clostridium perfringens food poisoning. The method of preparation of the challenge inoculum was important in determining whether a given strain would provoke a response. When cultures were grown for 4 hr at 37 C in Skim Milk (Difco), 14 of 29 type A strains isolated from food-poisoning outbreaks consistently produced exudation of fluid and consequent dilation of the ileal segments. In contrast, 15 of the 18 strains derived from other sources failed to elicit a response. By use of different inoculum preparations, nearly all strains could be made to give at least an occasional positive loop reaction. Diarrhea was not obtained in rabbits by intraluminal injection into the normal ileum or by per os administration of the cultures. Lecithinase, purified and in concentrated culture supernatant fractions, failed to produce a response in the isolated ileal loops. MATERIALS AND METHODS Cultures. Of the 46 C. perfrinigens type A strains used, 29 were isolates from feces of food-poisoning cases or from incriminated foods. Both the "classical" and "food-poisoning" strains (6, 7) were represented. The remaining 17 were isolates from random sources, such as raw vegetables, beef liver, ground beef, spiced luncheon meats, and flies. One type D strain was also tested. The known association of these strains to food poisoning is given in Table 3. Stock cultures maintained frozen in Cooked Meat Medium (Difco) were subcultured in Fluid Thioglycollate Medium (BBL) for 16 to 20 hr at 37 C. Media for preparation of the animal challenge materials were inoculated with these "activated" cultures (1%/a by volume) and were always incubated at 37 C. Vigorously growing cultures were usually obtained after 4 hr by this procedure. Spores were developed in a sporulation medium (5), harvested and washed three times with distilled water by centrifugation, and stored in water at 4 C. Surgical procedure. New Zealand white rabbits of both sexes, acclimatized to the laboratory for at least 1 week, were 7 to 10 weeks old and weighed 1.4 to 2.2 kg at the time of testing. The animals were main-1560