Systemic administration of thiazolidinediones reduces peripheral inflammation in vivo, presumablybyacting at peroxisome proliferator-activated receptor g (PPARg) in peripheral tissues. Based on a rapidly growing body of literature indicating the CNS as a functional target of PPARg actions, we postulated that brain PPARg modulates peripheral edema and the processing of inflammatory pain signals in the dorsal horn of the spinal cord. To test this in the plantar carrageenan model of inflammatory pain, we measured paw edema, heat hyperalgesia, and dorsal horn expression of the immediate-early gene c-fos after intracerebroventricular (ICV) administration of PPARg ligands or vehicle. We found that ICV rosiglitazone (0.5-50 ug) or 15d-PGJ2 (50-200 ug), but not vehicle, dose-dependently reduced paw thickness, paw volume and behavioral withdrawal responses to noxious heat. These anti-inflammatory and anti-hyperalgesia effects result from direct actions in the brain and not diffusion to other sites, because intraperitoneal and intrathecal administration of rosiglitazone (50 ug) and 15d-PGJ2 (200 ug) had no effect. PPARg agonists changed neither overt behavior nor motor coordination, indicating that non-specific behavioral effects do not contribute to PPAR ligand-induced anti-hyperalgesia. ICV administration of structurally dissimilar PPARg antagonists (either GW9662 or BADGE) reversed the anti-inflammatory and anti-hyperalgesic actions of both rosiglitazone and 15d-PGJ2. To evaluate the effects of PPARg agonists on a classic marker of noxious stimulus-evoked gene expression, we quantified Fos protein expression in the dorsal horn. The number of carrageenan-induced Fos-like immunoreactive profiles was less in rosiglitazone-treated rats as compared to vehicle controls.We conclude that pharmacological activation of PPARg in the brain rapidly inhibits local edema and the spinal transmission of noxious inflammatory signals.