T-helper cells (ThC) play an important role in the induction of both cytotoxic T-cell responses and B-cell responses against the grafted organ. Furthermore, ThC alone are capable of causing graft rejection in T cell-deprived mice and rats. In view of these observations we found it important to analyse the frequency and functions of donor-specific ThC in the allograft and in the recipient lymphoid system during the course of acute renal allograft rejection. A limiting dilution assay was developed which, due to the absence of exogenous interleukin 2 (IL-2) and the low numbers of stimulator cells used, appears to be highly selective for the proliferation of specific ThC. Kidney transplants were performed from LBN (RT1n) to congenic Lewis (RT1l) strain differing in major histocompatibility complex (MHC) only. The inflammatory (white) cells were recovered from the graft, and blood and recipient spleen and the frequency of RT1n-responding ThC were determined at different times after transplantation. In the kidney graft itself, the frequency of ThC responding to RT1n MHC antigens was 1:3000 on day 2 and increased to 1:670-1320 at the peak of inflammation. In the spleen, the frequency increased from 1:1000 on day 0 to 1:200 on day 8, and remained high even after the graft was rejected. In the blood, the frequency stayed at the 1:400-1:800 level, and increased to 1:200 only after the graft had been completely destroyed. Individual ThC clones deriving from limited dilution assays of kidney and spleen cells were recovered and expanded with irradiated donor cells without IL-2 and finally with exogenous IL-2 only. All clones showed the T-helper (W3/25) phenotype, seven out of eight tested clones showed a specific anamnestic response to RT1n alloantigens and no response to RT1l or RT1a in a secondary MLC, 12 out of 12 clones produced IL-2 and 11 out of 11 clones produced gamma interferon upon re-stimulation with relevant allogeneic cells, and eight out of ten clones collaborated with syngeneic B cells for Ig synthesis, indicating that they were indeed derived from specific ThC and/or from their precursors. Taken together, the results demonstrate that specific ThC and/or their precursors represent only a very small minority in the graft-infiltrating inflammatory population. This makes it most unlikely that the ThC themselves are responsible for graft destruction; the results indicate rather that a major role of ThC in situ may be instruction of immunologically specific and nonspecific components of inflammation.