We observe that PBMC from most adults (16 of 18 subjects tested) show a small but significant in vitro proliferative response to a 30-amino acid-long peptide (peptide 2, amino acids 34-63) derived from toxic shock syndrome toxin. By contrast, PBMC from newborn blood and thymocytes do not proliferate to this peptide, and furthermore, peptide 2 did not displace the binding of radiolabeled TSST-1 to MHC class II positive cells, nor did it induce IL-1 beta mRNA in monocytes, indicating that this peptide does not behave as a superantigen. Proliferation of PBMC to peptide 2 could be blocked by anti-HLA-DR, but not by anti-HLA-DP or DQ mAb, suggesting that HLA-DR molecules are the restriction elements for the recognition of this peptide by T cells. This premise was further confirmed by demonstrating that mouse L cells transfected with human HLA-DR, but not HLA-DP or DQ molecules, supported the proliferation of purified T cells to peptide 2. Studies with subjects of known HLA-DR types showed that all types tested are capable of responding to this peptide, PBMC from adults exposed to mycobacterial Ag showed significantly better proliferative response to peptide 2 than unexposed adults. Studies with truncations of this peptide suggest that a "core" region of eight amino acids that is conserved between low m.w. heat shock proteins and peptide 2 may be critical to T cell recognition of this peptide. The universal presentation of peptide 2 by HLA-DR molecules may contribute to the widespread natural immunity observed against toxic shock syndrome toxin.