A series of cyclohexanone curcumin analogues were synthesized by coupling the appropriate substituted benzaldehyde with cyclohexanone as previously described (30). Characterization of the compounds, 2,6-bis(4-hydroxybenzylidene)-cyclohexanone (A₁), 2,6-bis(3,4-dihydroxybenzylidene)-cyclohexanone (A₂), 2,6-bis(4-hydroxy-3-methoxybenzylidene)-cyclohexanone (A₃), 2,6-bis(3,5-di-tert-butyl-4-hydroxylbenzylidene)-cyclohexanone (A₄), 2,6-bis(3,4-dimethoxybenzylidene)-cyclohexanone (A₅) and 2,6-bis(4-hydroxy-3,5-dimethoxybenzylidene)-cyclohexanone (A₆), was previously described in detail (30).

PC-3 cells were obtained from the American Type Culture Collection (ATCC; Rockville, MD, USA). Curcumin was obtained from Sigma-Aldrich (St. Louis, MO, USA). The RPMI-1640 tissue culture medium, penicillin-streptomycin, L-glutamine and fetal bovine serum (FBS) were obtained from Gibco (Grand Island, NY, USA). The PC-3 cells were maintained in RPMI-1640 culture medium containing 10% FBS supplemented with penicillin (100 U/ml)-streptomycin (100 μg/ml) and L-glutamine (300 μg/ml). Cultured cells were grown in a humidified atmosphere of 5% CO₂ at 37°C, and were passaged twice a week. Curcumin and its analogues were dissolved in DMSO and the final concentration of DMSO in all experiments was 0.1%.

PC-3 cells were seeded at a density of 0.2×10⁵ cells/ml in medium in 96-well plates (0.2 ml/well) and incubated for 24 h. The cells were then treated with various concentrations (0.5–10 μM) of the different curcumin analogues for 72 h. Following treatment, 200 μl 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazoliumbromide (0.5 mg/ml in PBS) was added to each well of the plate and incubated for 2 h. The plate was then centrifuged at 1,000 rpm for 5 min at 4°C. Following removal of the medium, 0.1 ml DMSO was added to each well. The absorbance was recorded on a microplate reader at 540 nm. The effect of different curcumin analogues on cell growth was assessed as the percentage cell growth compared to DMSO-treated cells.

The number of viable cells following each treatment was determined using the trypan blue exclusion assay (31). In brief, 80 μl of cell suspension was mixed with 20 μl of 0.4% trypan blue solution and incubated for 2 min. The cells were then examined under a light microscope (Nikon Optiphot, Japan). Blue cells were counted as dead cells and cells that did not absorb dye were counted as live cells.

Apoptotic cells were determined by morphological assessment in cells stained with propidium iodide (32,33). Cytospin slides were prepared following each experiment and cells were fixed with acetone/methanol (1:1) at room temperature for 10 min, followed by 10 min of propidium iodide staining (1 μg/ml in PBS), and were then analyzed using a fluorescence microscope (Nikon Eclipse TE200, Japan). Apoptotic cells were identified by classical morphological features, including nuclear condensation, cell shrinkage and formation of apoptotic bodies (32,33).

Caspase-3 activation was measured using an EnzoLyte AMC Caspase-3 Assay Fluorimetric kit (AnaSpec, Fremont, CA, USA) according to the manufacturer's instructions (34). A total of 1×10⁵ cells were plated in triplicate in a flat-bottomed 96-well plate. Cells were treated with different curcumin analogues for 72 h. Following treatment, caspase-3 substrate was added to each well. Plates were incubated at room temperature for 30 min. Fluorescence intensity was measured in a Tecan Inifinite M200 plate reader (Tecan US Inc., Durham, NC, USA).

NF-κB transcriptional activity was measured using the NF-κB-luciferase reporter gene expression assay (35). An NF-κB luciferase construct was stably transfected into PC-3 cells and a single stable clone, PC-3 C4 (35), was used. PC-3 C4 cells were treated with different curcumin analogues for 24 h, and the NF-κB-luciferase activities were measured using luciferase assay kits from Promega (Madison, WI, USA). Following treatment, the cells were washed with ice-cold phosphate-buffered saline (PBS), and harvested in 1× reporter lysis buffer. Following centrifugation, 10 μl aliquots of the supernatants were measured for luciferase activity using a Luminometer from Turner Designs Inc., (Sunnyvale, CA, USA). The luciferase activity was normalized against known protein concentrations, and expressed as the percentage of luciferase activity in the control cells, which were treated with DMSO solvent. The protein level was determined using a Bio-Rad protein assay kit (Bio-Rad, Hercules, CA, USA) according to the manufacturer's instructions.

The analysis of variance (ANOVA) with the Tukey-Kramer multiple comparison test was used for the comparison of growth inhibition as determined by the trypan blue assay and determination of the NF-κB-luciferase activities in cultured PC-3 cells that were treated with different curcumin analogues.

The inhibitory effects of curcumin and its analogues A₁-A₆ on the growth of cultured PC-3 cells were determined using the MTT assay. For each experiment, curcumin was evaluated as the positive control. The inhibitory effects of curcumin did not significantly vary between different experiments. Data from the curcumin incubations were averaged (Fig. 1). Curcumin and its analogues A₁-A₆ inhibited the growth of PC-3 cells in a concentration-dependent manner (Fig. 1). A₄ was the strongest curcumin analogue for inhibiting the growth of PC-3 cells, as determined by the MTT assay, followed by A₂, A₆, A₅, A₃ and A₁ (Fig. 1A). In additional experiments, the effects of different curcumin analogues on cell growth were determined by the trypan blue exclusion assay. Compounds A₂-A₆ were more potent for decreasing the number of viable PC-3 cells as compared to curcumin (Fig. 1B). Statistical analysis using ANOVA with the Tukey-Kramer test demonstrated that the differences in the number of viable cells between the curcumin-treated group and any curcumin analogue-treated group (except the A₁-treated group) were statistically significant (P<0.001). The number of viable cells was significantly lower in the A₄-treated group than in the curcumin-treated or any other curcumin analogue-treated group (P<0.05 compared to the A₂-treated group; P<0.001 compared to other curcumin analogue-treated groups).

Effects of the curcumin analogues A₁-A₆ on apoptosis in PC-3 cells were determined by morphological assessment of apoptotic cells. Apoptotic cells were identified by classical morphological features, including nuclear condensation, cell shrinkage and formation of apoptotic bodies. Morphologically distinct apoptotic cells from representative samples are shown in Fig. 2B. Treatment of PC-3 cells with curcumin resulted in a small increase in apoptotic cells (Fig. 2C). Treatment with compounds A₁-A₆ stimulated apoptosis in PC-3 cells in a concentration-dependent manner (Fig. 2C). Compounds A₂ and A₄ demonstrated stronger stimulatory effects on apoptosis in PC-3 cells compared to the other compounds. The effect of the two strongest compounds A₂ and A₄ on activation of caspase-3 in comparison to curcumin was determined. Treatment of PC-3 cells with curcumin caused only a small increase in caspase-3 activity, while treatment with A₂ and A₄ caused an 8.2- and 9.3-fold increase in caspase-3 activity, respectively (Fig. 2D). Our results identified A₂ and A₄ as the two curcumin analogues that had the greatest effect for stimulating apoptosis in PC-3 cells.

To investigate the effect of A₁-A₆ on activation of NF-κB activity, we used an NF-κB-luciferase reporter gene expression assay in PC-3 C4 cells. PC-3 C4 is a cell line derived from the stable transfection of PC-3 cells with an NF-κB luciferase construct (35). In these experiments, PC-3 C4 cells were treated with different concentrations of curcumin and its analogues A₁-A₆ for 24 h. Treatment of PC-3 C4 cells with curcumin or A₁ (both 5 μM) caused only modest decreases in the activity of NF-κB (Fig. 3). Treatment with A₂-A₆ (all at 5 μM) caused a further decrease in NF-κB transcriptional activity. Statistical analysis using ANOVA with the Tukey-Kramer test demonstrated that NF-κB activity was significantly lower in the A₄-treated group than in any other treated group (P<0.01 compared to the A₂-treated group; P<0.001 compared to other curcumin analogue-treated groups). There were good correlations between inhibition of NF-κB activity and cell growth inhibition (r=0.97), and between inhibition of NF-κB activity and apoptosis stimulation (r=0.96) in the PC-3 cells treated with all compounds at 5 μM.

Six curcumin analogues (A₁-A₆) that contain a five-carbon linker with a mono-carbonyl group (cyclohexanone linker) were evaluated for anticancer activities in human prostate cancer PC-3 cells. All of the curcumin analogues, with the exception of A₁, had stronger inhibitory effects on cell growth and stronger stimulatory effects on the apoptosis of PC-3 cells compared to curcumin. Although the structures of A₃ and curcumin are the same, with the exception of their middle linker (Fig. 4), the anticancer activity of A₃ was stronger than that of curcumin (Fig. 1) suggesting that a cyclohexanone linker increases anticancer activity. A comparison of the six curcumin analogues (all with the same mono-carbonyl linker) revealed that anticancer activity was significantly influenced by substituents on the benzene rings. The presence of a methoxy group on both sides of the p-phenol group markedly increased activity compared to a compound with a methoxy group on only one side (A₃ vs. A₆). Tert-butyl substituents on both sides of the p-phenol group (A₄) had the strongest anticancer effect among all of the studied compounds. Comparison of A₁ and A₂ suggested that o-dihydroxyl substituents on both benzene rings (A₂) had stronger activity than an analogue with a single hydroxyl group on each side (A₁).