It is important to describe effective and non-toxic therapies for multiple sclerosis (MS), an autoimmune demyelinating disease. Experimental autoimmune encephalomyelitis (EAE) is an immune-mediated inflammatory disease that serves as a model for MS. Earlier we and others have shown that, gemfibrozil, a lipid-lowering drug, exhibits therapeutic efficacy in EAE. However, the underlying mechanism was poorly understood. Although gemfibrozil is a known ligand of peroxisome proliferator-activated receptor α (PPARα), here, we established that oral administration of gemfibrozil preserved the integrity of blood–brain barrier (BBB) and blood–spinal cord barrier (BSB), decreased the infiltration of mononuclear cells into the CNS and inhibited the disease process of EAE in both wild type and PPARα–/– mice. On the other hand, oral gemfibrozil was found ineffective in maintaining the integrity of BBB/BSB, suppressing inflammatory infiltration and reducing the disease process of EAE in mice lacking PPARβ (formerly PPARδ), indicating an important role of PPARβ/δ, but not PPARα, in gemfibrozil-mediated preservation of BBB/BSB and protection of EAE. Regulatory T cells (Tregs) play a critical role in the disease process of EAE/MS and we also demonstrated that oral gemfibrozil protected Tregs in WT and PPARα–/– EAE mice, but not PPARβ–/– EAE mice. Taken together, our findings suggest that gemfibrozil, a known ligand of PPARα, preserves the integrity of BBB/BSB, enriches Tregs, and inhibits the disease process of EAE via PPARβ, but not PPARα.