The PI3K/Akt pathway regulates BBS-stimulated COX-2 promoter activity and AP-1 activation. PC-3 cells were co-transfected with plasmids containing the 1.4 kb human COX-2 promoter 5' of a luciferase reporter gene and β-galactosidase. A) Twenty-four hours after transfection, cells were treated either with vehicle or 10 nM BBS for 2, 4, and 6 h. Changes in luciferase activity (mean ± S.D, n = 3) are expressed relative to vehicle treated at each time point after normalizing for transfection efficiency using β-galactosidase activity [* BBS vs. time-point matched vehicle control, p < 0.05]. B) Pre-treatment with LY294002 (25 μM) partially inhibited BBS-stimulated COX-2 promoter activity [* LY294002 vs. BBS alone, p < 0.05, n = 3], whereas SB203580 (10 μM) had no effect. C) Autoradiogram showing binding of nuclear proteins to 32P-labeled oligonucleotide containing the AP-1 consensus sequence [lane 1, radiolabeled probe only; lane 2, nuclear proteins + radio-labeled probe and excess unlabeled probe; lane 3, nuclear protein from vehicle-treated cells; lane 4 nuclear proteins from cells treated with BBS (10 nM) for 30 min; lanes 5 and 6, nuclear proteins from cells pretreated with LY294002 (25 μM) for 30 min followed by vehicle or BBS for 30 min]. D) PC-3 cells were treated with BBS (10 nM) or TNF-α (10 ng/ml) alone or in combination with inhibitors for 30 min, fixed, and immunostained with an antibody to the p65 subunit of NF-κB. Quantification of the effects of curcumin (Cur) (20 μM) and LY294002 (LY) (25 μM) on BBS-stimulated NF-κB nuclear translocation. Data are expressed as the ratio of NF-κB-positive stained nuclei divided by the total number of cells per high power field (200X). Four fields were counted for each condition from 3 independent experiment [† BBS or BBS + LY294002 vs. vehicle or LY294002 alone, p ≤ 0.001; * BBS + curcumin vs. BBS alone or TNFα + curcumin vs. TNFα alone, p ≤ 0.001].