Structure and activity of diphtheria toxin. II. Attack by trypsin at a specific site within the intact toxin molecule.

Most preparations of diphtheria toxin that we have examined consist predominantly of mixtures in various proportions of two proteins, each of molecular weight about 63,000. As judged by their electrophoretic behavior on polyacrylqide gels run in the presence of sodium dodecyl sulfate, one of these consists of intact 63,000-dalton polypeptide chains, while the other is composed of two fragments of molecular weight 24,000 and 39,000 (Fragments A and B, respectively) linked by at least one disulfide bridge. Here we report that when these preparations are treated at 25” with low concentrations of trypsin (0.25 to 1.0 pg per ml), a single scission is introduced at a specific site in the former (intact toxin) converting it into a protein apparently identical with the latter (nicked toxin). Thus nicked toxin in the original preparations must be a derivative of intact toxin which arose by attack of proteases from the bacterial culture in which the toxin was produced. Fragment A contains 1 half-cystine residue, and is therefore linked to Fragment B in nicked toxin by one disulfide bridge. Fragment A is stable to heating at loo’, and exposure to pH values from 2 through 12. Fragment B is relatively unstable, and precipitates even at neutral pH. We have shown previously that toxin is almost devoid of enzymic activity unless treated with thiols. The activity is probably entirely due to Fragment A released upon dissociation of nicked toxin in the presence of thiols. When toxin is treated with trypsin the intact chains are converted into a form which can be dissociated by thiols into fragments apparently identical with A and B. Hence the specific activity of the preparation is enhanced by trypsin treatment because of the greater amount of Fragment A released by thiols. The toxicity of toxin is not significantly changed by treatment with trypsin, which indicates that both intact and nicked toxin are toxic. Fragment A alone is not toxic. Linkage to Fragment B is apparently required for toxicity, perhaps to facilitate entry of Fragment A into cells. Treat-