We have investigated the potential use of peroxisome proliferator-activated receptor gamma (PPARgamma) agonists as anti-inflammatory agents in cell-based assays and in a mouse model of endotoxemia. Human peripheral blood monocytes were treated with LPS or PMA and a variety of PPARgamma agonists. Although 15-deoxy-Delta12,14-prostaglandin J2 (15d-PGJ2) at micromolar concentrations significantly inhibited the production of TNF-alpha and IL-6, four other high affinity PPARgamma ligands failed to affect cytokine production. Similar results were obtained when the monocytes were allowed to differentiate in culture into macrophages that expressed significantly higher levels of PPARgamma or when the murine macrophage cell line RAW 264.7 was used. Furthermore, saturating concentrations of a potent PPARgamma ligand not only failed to block cytokine production, but also were unable to block the inhibitory activity of 15d-PGJ2. Thus, activation of PPARgamma does not appear to inhibit the production of cytokines by either monocytes or macrophages, and the inhibitory effect observed with 15d-PGJ2 is most likely mediated by a PPARgamma-independent mechanism. To examine the anti-inflammatory activity of PPARgamma agonists in vivo, db/db mice were treated with a potent thiazolidinedione that lowered their elevated blood glucose and triglyceride levels as expected. When thiazolidinedione-treated mice were challenged with LPS, they displayed no suppression of cytokine production. Rather, their blood levels of TNF-alpha and IL-6 were elevated beyond the levels observed in control db/db mice challenged with LPS. Comparable results were obtained with the corresponding lean mice. Our data suggest that compounds capable of activating PPARgamma in leukocytes will not be useful for the treatment of acute inflammation.