Vaccines of inactivated Bordetella pertussis have provided an effective means of controlling whooping cough. Questions regarding the safety of these vaccines have led to public fear of vaccine injuries, litigation for which has reached proportions that ultimately threaten the nation's vaccine supply and have chilling effect on new vaccine development. To meet these challenges, we have taken a recombinant DNA approach to the improvement of pertussis vaccine. Pertussis toxin is a major virulence factor of the organism; recent clinical evidence suggests that inactivated toxin alone may confer disease protection. We have molecularly cloned portions of the toxin operon into bacterial expression vectors and obtained very high levels of selected toxin subunits as non-fusion proteins in recombinant E. coli. Preliminary studies demonstrate certain of these subunits retain their biological activity, are highly immunogenic, and confer protection against disease challenge in experimental animals. We propose to express other toxin subunits, and perhaps other virulence factors, in various recombinant hosts, genetically modify these proteins to eliminate their reactogenicity and test them for immunoprotection in a number of formulations. Genetically-engineered vaccines are expected to be economical to produce, free of reactogenic components, and both highly safe and efficacious.