Curcumin, the principle component of the spice turmeric, has been used as an anti-inflammatory medication in India and China for centuries. Recent studies, predominantly using actively dividing cell lines, have suggested that this compound could be used as a chemopreventative or therapeutic agent for epithelial tumors. As curcumin has been reported to inhibit the NIK/IKK complex, an activity that would be expected to induce apoptosis in B cell malignancies, we sought to determine whether curcumin induces apoptosis in vitro in primary chronic lymphocytic leukemia (B-CLL) cells. Primary leukemic cells were incubated with varying dosages of curcumin, followed by assessment for apoptosis. The role of PPAR gamma or NF-kappa B signaling in curcumin-induced apoptosis was examined by cotreatment with a PPAR gamma antagonist or EMSA of nuclear NF kappa B complexes. We also examined whether a clinically achievable concentration of curcumin (1 mu M) would augment the apoptotic effects of fludarabine, dexamethasone, vincristine or the PDE4 inhibitor rolipram. In B-CLL cells from 14 patients, curcumin-induced apoptosis with a mean EC50 of 5.5 mu M. In contrast, the EC50 for whole mononuclear cells from a healthy donor was 21.8 mu M. In a 48 hr wash-out time course, curcumin-induced apoptosis was time-dependent, with a substantial reduction in apoptosis observed when curcumin was removed after 5 hr. Curcumin treatment reduced basal nuclear NF-kappa B levels and 1 mu M curcumin augmented both vinca alkaloid and PDE4 inhibitor-induced apoptosis in B-CLL cells. Our studies suggest that curcumin may augment the efficacy of established or experimental therapies for B-CLL.