Rabbits immunized with the protease resistant recombinant Treponema pallidum surface-associated Ag 4D showed an altered course of experimental syphilis after intradermal challenge with virulent T. pallidum. Vaccination trials using three different protocols were examined. In one experiment, a combined i.m.-i.v. immunization protocol was compared with an i.m. injection schedule. Four of five rabbits immunized by the i.m.-i.v. protocol developed morphologically atypical lesions at each of 16 sites, whereas the i.m. injected animals showed no evidence of attenuated disease. Moreover, immunization by both protocols elicited equally high ELISA titers of 4D-specific IgG antibody suggesting that cellular immune mechanisms may have been involved. In an effort to augment cell-mediated responses to 4D, seven rabbits received i.v. immunizations with an emulsion of 4D, purified BCG cell wall skeletons, and trehalose dimycolate in an oil-microdroplet form (4D-cell wall skeleton-trehalose dimycolate). Splenic lymphocytes from three representative immunized animals had a strong in vitro proliferative response to 4D as compared to controls. Further, all animals developed high anti-4D titers in response to immunization. After challenge with virulent T. pallidum, six of the seven rabbits developed an early cutaneous response (3-8 days) at the sites of inoculation consistent with a hypersensitivity reaction followed by morphologically atypical lesions. Aspirates from each of three representative atypical lesions were devoid of treponemes by darkfield examination at day 14, whereas motile T. pallidum were observed in aspirates from four of five control sites. Between days 20 and 24, lesions in the 4D-cell wall skeleton-trehalose dimycolate immunized rabbits began to enlarge and become more typical in appearance; aspirates from six of seven representative lesions were darkfield positive. We conclude that 1) immunization with the r4D Ag alters the course of experimental syphilis in a manner consistent with previously defined parameters of partial protection and 2) immunizations protocols containing an i.v. component are more effective than protocols using the i.m. route alone.