Previous studies on the immunotoxicity of a complex mixture of polynuclear aromatic hydrocarbon (PAH) by-products from a manufactured gas plant indicated possible synergistic interactions which were investigated by determining the immunosuppressive effects of a reconstituted PAH mixture in female B6C3F1 mice challenged with TNP-haptenated sheep red blood cells (SRBCs) (T-cell-dependent) or trinitrophenyl-lipopolysaccharide (TNP-LPS) (T-cell-independent) antigens. The reconstituted PAH mixture contained the following 17 congeners: 2-rings (indan, naphthalene, 1- and 2-methylnaphthalene), 3-rings (acenaphthylene, acenaphthene, dibenzofuran, fluorene, phenanthrene and anthracene), and > or = 4-rings (pyrene, fluoranthene, benz[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene and benzo[a]pyrene), and resembled mixtures identified as by-products from manufactured gas plants. The reconstituted mixture and the 2-, 3- and > or = 4-ring PAH fractions all caused a dose-dependent decrease in the splenic plaque-forming cell (PFC) response to SRBCs or TNP-LPS, and their ED50 values for the four treatment groups were 86, 354, 145, and 23 or 163, 439, 637 and 31 mg/kg, respectively. The corresponding ED50 values for decreased serum anti-TNP IgM levels for these same mixtures were (TNP-haptenated SRBCs, T-cell-dependent) 144, 231, 42 and 27 units, respectively, and (TNP-LPS, T-cell-independent) 161, 406, 312 and 69 units, respectively. The suppression of anti-TNP IgM titers was similar to the suppression of the PFC response and shows that antigen-specific immunoglobulin titer can be used as a biomarker of PAH exposure. A direct comparison of the immunotoxic responses of the reconstituted PAH mixture and the corresponding dose of the > or = 4-ring PAHs indicated that the latter fraction was primarily responsible for the activity of the reconstituted mixture.