There have been several reports that TNF-related apoptosis-inducing ligand (TRAIL) has the ability to suppress the development of experimental autoimmune diseases, including a mouse model of experimental autoimmune encephalomyelitis, a rabbit model of rheumatoid arthritis, type 1 diabetes mellitus, in mice and experimental autoimmune thyroiditis (EAT) in mice. However, the mechanism underlying TRAIL effect is not well defined. In the present study, we specifically examined TRAIL effects on CD4(+)CD25(+) regulatory T cells. CD4(+)CD25(+) T cells prepared from mouse thyroglobulin (mTg)-immunized CBA/J mice proliferate in the presence of TRAIL and dendritic cells in vitro. These CD4(+)CD25(+) T cells included both CD4(+)CD25(+)CD45RB(Low) (regulatory) and CD4(+)CD25(+)CD45RB(High) (effector) T cells. Our results demonstrated that mTg-immunized mice treated with TRAIL showed significant increases in the number of CD4(+)CD25(+)CD45RB(Low) T cells compared with mice immunized with mTg alone. CD4(+)CD25(+)CD45RB(Low) T cells expressed much higher levels of the forkhead family transcription factor, IL-10, and TGFbeta1 than CD4(+)CD25(+)CD45RB(High) T cells, and these cells can completely suppress the proliferation of the mTg-primed splenocytes in lower concentrations than the unfractionated CD4(+)CD25(+) T cells. Furthermore, transfer of these cells into CBA/J mice prior to mTg-primed splenocyte injection could markedly reduce the frequency and severity of EAT development. CD4(+)CD25(+)CD45RB(Low) T cells were more effective at suppressing histological thyroiditis than unfractionated cells. These results indicated that TRAIL can increase the number of mTg-specific CD4(+)CD25(+)CD45RB(Low) T cells, inhibiting autoimmune responses and preventing the progression of EAT. These findings reveal a novel mechanism by which TRAIL could inhibit autoimmune disease.