Estrogen treatment has been shown to exert a protective effect on experimental autoimmune encephalomyelitis (EAE), and is under clinical trial for multiple sclerosis. Although it is commonly assumed that estrogens exert their effect by modulating immune functions, we show in this study that 17beta-estradiol (E2) treatment can inhibit mouse EAE without affecting autoantigen-specific T cell responsiveness and type 1 cytokine production. Using mutant mice in which estrogen receptor alpha (ERalpha) has been unambiguously inactivated, we found that ERalpha was responsible for the E2-mediated inhibition of EAE. We next generated irradiation bone marrow chimeras in which ERalpha expression was selectively impaired in inflammatory T lymphocytes or was limited to the radiosensitive hemopoietic compartment. Our data show that the protective effect of E2 on clinical EAE and CNS inflammation was not dependent on ERalpha signaling in inflammatory T cells. Likewise, EAE development was not prevented by E2 treatment in chimeric mice that selectively expressed ERalpha in the systemic immune compartment. In conclusion, our data demonstrate that the beneficial effect of E2 on this autoimmune disease does not involve ERalpha signaling in blood-derived inflammatory cells, and indicate that ERalpha expressed in other tissues, such as CNS-resident microglia or endothelial cells, mediates this effect.