Search

Evidence is presented which confirms that cholera toxoids obtained by reaction of purified toxin with Formalin possess the ability to partially reactivate both in vivo and in vitro. At the same time, conditions are presented for the preparation of stable, antigenic cholera toxoids by reaction of purified toxin with glutaraldehyde. Treatment of purified cholera toxin with approximately 200 mol of glutaraldehyde per mol of toxin at pH 7.8 reproducibly resulted in the preparation of toxoids which: (i) possessed less than 20 bluing doses per 100 μg; (ii) did not reactivate in vivo or in vitro; (iii) precipitated with, and neutralized antitoxin; (iv) elevated prolonged serum antitoxin in immunized rabbits; (v) protected immunized guinea pigs against toxin skin challenge; and (vi) lent themselves to enhanced antigenicity by means of an in situ adjuvant system which may be suitable for man. Acrylamide gel electrophoresis and molecular sieve chromatography of a series of glutaraldehyde-derived toxoids suggested that the reaction products consisted of monomeric and polymeric species and that the proportion of higher-molecular-weight species was determined by the relative concentrations of toxin and glutaraldehyde. The results suggested a relationship between complete and irreversible elimination of toxicity and the formation of higher-molecular-weight toxoids.

The addition of an ultrafiltration step to the purification procedures previously described for cholera toxin (Rappaport et al., (1974) permitted the preparation of highly purified antigenic toxoids essentially free of somatic antigen(s). The purity of such toxoids is established: (i) by the absence of more than about one part limulus amebocyte lysate (LAL)-positive endotoxin per 10(5) parts toxoid and (ii) by the inability of the toxoids to elicit a significant rise in rabbit vibriocidal antibody. The antigenicity of the toxoids is demonstrated by their ability to produce the same high levels of rabbit serum antitoxin as are produced by comparable toxoids containing small amounts of somatic antigen. The results also indicate that amounts of somatic antigen of the order of less than or equal to 1 mug/100 mug of toxoid do not exert an adjuvant effect on the toxoid, at least with respect to circulating antitoxin. Other data show that, where present, the ability of somatic antigen to elicit vibriocidal antibody is influenced by the immunization schedule employed and that a correlation exists between the LAL-determined endotoxin content of the toxoids and their ability to stimulate vibriocidal antibody. Somatic antigen-free toxoids, purified and tested by the refinements herein described, were prepared for use in the National Institutes of Health sponsored field trials, and data pertaining to their purity and antigenic properties are presented.

The enterotoxin from Vibrio cholerae is selectively concentrated from cell-free culture supernatant by co-precipitation with hexametaphosphate and is further purified by adsorption on aluminum hydroxide powder. The bulk of residual somatic antigen becomes insoluble upon lyophilization of the toxin preparation and is removed by centrifugation of the rehydrated material. Other contaminants are eliminated by treatment with activated carbon. Preparations of toxin, purified by this method, have been characterized by: (i) a single immunoprecipitin line against polyvalent antisera; (ii) homogeneity on acrylamide gels; (iii) specific activities on the order of 22 limit-of-bluing doses/μg; (iv) ultraviolet spectra characteristic of pure protein; and (v) overall yields on the order of 50%, irrespective of purification scale. Such preparations, however, have been shown to contain trace amounts of somatic antigen when they are intensively tested either for their ability to elevate serum vibriocidal antibody titers in immunized rabbits or for their ability to increase resistance of immunized mice to live vibrio challenge. In the latter test system, the level of residual somatic antigen per 50 μg of toxin (toxoid) antigen generally did not exceed 0.025% of the Division of Biological Standards reference vaccine, V. cholerae Inaba IN-12. Methods for elimination of this small amount of somatic antigen have been investigated and are discussed. The particular combination of purification steps which are presently described have been easily and reproducibly applied on a production scale to prepare gram amounts of toxin with a high degree of purity, even under a variety of initial conditions.