Peptide constructs comprised of multideterminant Th peptides from the envelope glycoprotein of HIV previously identified to induce proliferative responses in four different haplotypes of mice and IL-2 responses in 52 to 73% of HIV positive, Ag-responsive patients, were colinearly synthesized with the peptide 18 of the V3 loop of HIV-1 gp 160, corresponding to the principal neutralizing determinant of HIV-IIIB. The segments containing clusters of overlapping Th epitopes were called cluster peptides. Cognate help for peptide 18 antibody was elicited after a single immunization in all strains of mice that had previously responded to a T cell epitope encompassed by the cluster peptides. Animals boosted with cluster peptide-peptide 18 constructs 36 to 52 wk later displayed secondary antibody responses. Cluster peptide 3-peptide 18 induced antibody that neutralized homologous virus in one strain of mice although strong peptide 18 antibody responses were detected in all four strains of mice. The most promising construct, cluster peptide 6-peptide 18, induced neutralizing antibody in all strains of mice tested, and in two strains the level of neutralizing antibody achieved was comparable to levels adequate for protection from homologous viral challenge in chimpanzees. After a single boost, antibody titers for 90% neutralization in the range of 1/1000 to 1/16,000 were achieved. These neutralizing titers against the homologous viral strain, after just two immunizations, are at least four- to eightfold higher than the highest titered other polyclonal V3-specific immune sera we have ever observed in our laboratories. We also asked why some sera neutralized and others with similar ELISA titers did not. No correlation was found between neutralization and isotype or affinity for peptide or gp 120. We could not account for neutralization by antibodies to the helper sites. Substitutions made in the central loop region of peptide 18, amino acid residues PGRAF, dramatically reduced binding of both neutralizing and non-neutralizing sera although some fine specificity differences between neutralizing and nonneutralizing sera were noted. These results have implications for the design of synthetic peptide vaccines for HIV.