Many inflammatory mediators retard granulocyte apoptosis. Most natural PGs studied herein (eg, PGE 2, PGA 2, PGA 1, PGF 2α) either delayed apoptosis or had no effect, whereas PGD 2 and its metabolite PGJ 2 selectively induced eosinophil, but not neutrophil apoptosis. This novel proapoptotic effect does not appear to be mediated via classical PG receptor ligation or by elevation of intracellular cAMP or Ca 2+. Intriguingly, the sequential metabolites Δ 12 PGJ 2 and 15-deoxy-Δ 12, Δ 14-PGJ 2 (15dPGJ 2) induced caspase-dependent apoptosis in both granulocytes, an effect that did not involve de novo protein synthesis. Despite the fact that Δ 12 PGJ 2 and 15dPGJ 2 are peroxisome proliferator-activated receptor-γ (PPAR-γ) activators, apoptosis was not mimicked by synthetic PPAR-γ and PPAR-α ligands or blocked by an irreversible PPAR-γ antagonist. Furthermore, Δ 12 PGJ 2 and 15dPGJ 2 inhibited LPS-induced IκBα degradation and subsequent inhibition of neutrophil apoptosis, suggesting that apoptosis is mediated via PPAR-γ-independent inhibition of NF-κB activation. In addition, we show that TNF-α-mediated loss of cytoplasmic IκBα in eosinophils is inhibited by 15dPGJ 2 in a concentration-dependent manner. The selective induction of eosinophil apoptosis by PGD 2 and PGJ 2 may help define novel therapeutic pathways in diseases in which it would be desirable to specifically remove eosinophils but retain neutrophils for antibacterial host defense. The powerful proapoptotic effects of Δ 12 PGJ 2 and 15dPGJ 2 in both granulocyte types suggest that these natural products control the longevity of key inflammatory cells and may be relevant to understanding the control and resolution of inflammation.