Recent attention to mucosal immunization strategies has been focused on the nasal route for vaccine delivery. This study was designed to determine the effectiveness of a liposome-protein vaccine compared to that of a protein-only vaccine in inducing immune responses in humans. Healthy subjects were randomly assigned to two groups and immunized intranasally with a crude antigen preparation rich in glucosyltransferase (C-GTF) from Streptococcus mutans, alone or in liposomes. Parotid saliva, nasal wash, and serum were collected prior to and at weekly intervals following immunization and were analyzed for anti-C-GTF activity by enzyme-linked immunosorbent assay. The levels of immunoglobulin A (IgA) anti-C-GTF activity in the nasal wash from both groups after immunization increased to a mean peak of fivefold over the baseline level on day 28. Salivary IgA anti-C-GTF responses were induced to a lesser extent. IgG and IgA anti-C-GTF responses in serum were detected on day 14. The IgA responses were predominantly of the IgA1 subclass. These results show that C-GTF vaccines were more effective in inducing a local secretory IgA antibody response than a salivary or serum response when they were given intranasally. The IgA1 anti-C-GTF response in nasal wash samples for liposomal antigen versus antigen only was the only response which was significantly different (P < 0.04). This suggests that the form of the antigen affects the magnitude of the local mucosal response but not that of a disseminated response. These results provide evidence for the effective use of a nasal protein vaccine in humans for the induction of mucosal and systemic responses.