Vigorous lymphoproliferative responses to synthetic polypeptides poly(L-phenylalanine, L-glutamic acid)-poly(DL-alanine)--poly(L-lysine) [(Phe,G)-A--L], and L-glutamic acid, L-alanine, L-tyrosine (60:30:10) (GAT) were observed in cells from 92 unrelated subjects. Thirty-three percent responded to (Phe,G)-A--L and 77% to GAT. No HLA association was observed with responses to these two antigens. Family studies indicated that two complementary immune response (Ir) genes are required for response to each antigen. Eleven matings were informative for linkage analysis between HLA and these Ir genes. Families in which the complementary genes are in coupling gave maximal lod scores (log of the odds) of 4.50 for (Phe,G)-A--L and 7.57 for GAT for 0 = 0. In a HLA-BID recombinant family, the Ir-PheGAL genes are mapped towards the HLA-D region.
MATERIALS AND METHODSSubjects. Cell donors were 92 healthy, unrelated volunteers selected to include all known HLA-A, -B, -C and DR specificities and members of 35 nuclear families with 3 or more children. Eight of the families had an offspring who carried an intra-HLA recombinant haplotype.Antigens. Synthetic polymers (Phe,G)-A--L (lot MP1, Mr 260,000) and GAT (lot GAT 12, Mr 46,200) were obtained from Miles-Yeda (Elkhart, IN). Both antigens were readily soluble in RPMI 1640 medium. Before use, the pH of each antigen solution was adjusted to 7.2 and the solution was filter-sterilized (Millipore, 0.45 ,um). As described previously (3), each antigen preparation was tested for contamination with mitogenic material by coculturing with human cord blood lymphocytes and with spleen cells from unimmunized female mice of nonresponder and high-responder strains.In Vitro Lymphocyte Stimulation. Human peripheral blood lymphocytes were isolated by the standard Ficoll-Hypaque gradient technique. Varying concentrations of each antigen were cultured in triplicate with 8 x 104 peripheral blood lymphocytes in 0.2 ml of RPMI 1640 medium with 10% (vol/vol) human group AB serum, in round-bottom microtiter plates (Linbro). On day 7, 1 ,uCi of [3H]thymidine (specific activity, 6.7 Ci/mmol; 1 Ci = 37 GBq) was added to each well. The cultures were harvested 16-18 hr after labeling and samples were assayed for radioactivity in a Beckman model LS 7000 liquid scintillation counter. Results were expressed as the stimulation index (SI): SI = mean cpm of culture with antigen/mean cpm of culture without antigen.Each subject was tested at least twice with seven serial dilutions of each antigen. As previously reported, the antigen concentration eliciting maximal response varied from individual to individual (3). For the majority of subjects tested, the optimal antigen concentration was 80 gg per well for both (Phe,G)-A--L and GAT. The mean maximal SI obtained for each antigen was used to classify the subjects into responder and nonresponder groups. Subjects with SI >3 were considered as responders, in accordance with accepted procedures (4-6). There was a small number of subjects [12 for (Phe,G)-A--L and 9 ...