4-O-methylhonokiol, a PPARγ agonist, inhibits prostate tumour growth: p21-mediated suppression of NF-κB activity

Abstract

Background and purpose: The effects of 4-O-methylhonokiol (MH), a constituent of Magnolia officinalis, were investigated on human prostate cancer cells and its mechanism of action elucidated.

Experimental approach: The anti-cancer effects of MH were examined in prostate cancer and normal cells. The effects were validated in vivo using a mouse xenograft model.

Key results: MH increased the expression of PPARγ in prostate PC-3 and LNCap cells. The pull-down assay and molecular docking study indicated that MH directly binds to PPARγ. MH also increased transcriptional activity of PPARγ but decreased NF-κB activity. MH inhibited the growth of human prostate cancer cells, an effect attenuated by the PPARγ antagonist GW9662. MH induced apoptotic cell death and this was related to G(0) -G(1) phase cell cycle arrest. MH increased the expression of the cell cycle regulator p21, and apoptotic proteins, whereas it decreased phosphorylation of Rb and anti-apoptotic proteins. Transfection of PC3 cells with p21 siRNA or a p21 mutant plasmid on the cyclin D1/ cycline-dependent kinase 4 binding site abolished the effects of MH on cell growth, cell viability and related protein expression. In the animal studies, MH inhibited tumour growth, NF-κB activity and expression of anti-apoptotic proteins, whereas it increased the transcriptional activity and expression of PPARγ, and the expression of apoptotic proteins and p21 in tumour tissues.

Conclusions and implication: MH inhibits growth of human prostate cancer cells through activation of PPARγ, suppression of NF-κB and arrest of the cell cycle. Thus, MH might be a useful tool for treatment of prostate cancer.

Figure 1

Expression of PPARγ in prostate PC-3 and LNCap cells. (A) Cells were treated with 20 μM MH for 24 h. Whole cell extracts were prepared and PPARγ expression was determined by Western blot analysis as described in Methods. The value above the band indicates fold increase of PPARγ expression. (B) Transcriptional activation determined by luciferase activity. Luciferase activity was determined in the cells transfected with PPARγ plasmid construct after treatment with MH for 24 h as described in Methods. Values are mean ± SD of three experiments, with triplicate results for each experiment. (C) Immunoreactivity of PPARγ in PC-3 cells and in LNCaP cells treated with 20 μM MH. Untreated cells showed that PPARγ was localized predominantly in the perinuclear region and cytoplasm. Cells treated with MH demonstrated nuclear translocation of PPARγ. D-(a), Representations of the interaction between PPARγ and MH. MH binds to a tight-binding cavity within the agonist binding pocket in the ligand-binding domain of PPARγ shown in surface model. D-(b), The surrounding peptide residues that interact with MH are shown in the second picture. (E) Hydrogen-bonding interaction between hydroxyl group of MH and hydroxyl side chain of Ser²⁸⁹ of PPARγ. (F) MH-Sepharose 4B was used to pull down with PPARγ protein (Abcam Ltd.). WB, Western blot.

Figure 2

Effect of MH and WS9662 on cancer cell growth and apoptotic cell death. (A and C) Cells were treated with 0–30 μM MH for different times (0–72 h) (A) or treated with SW9662 for 72 h (C). Cell counting was performed as described in Methods. (B and D) Cells were treated with 0–30 μM MH for 24 h or the cells were pretreated with SW9662 for 24 h. DAPI and TUNEL staining were performed as described in Methods. Columns, mean of triplicate; bars, SD. *P < 0.05 indicates statistically significant differences from the untreated group. ^#P < 0.05 indicates statistically significant differences from the MH treated group.

Figure 3

Effect of MH on cell cycle, and expression of proteins regulating cell growth, cell cycle and apoptotic cell death, and NF-κB DNA binding, transcriptional activity and translocation of p50 and p65. (A) Cells were treated with 0–30 μM MH for 72 h. Cell cycle was performed as described in Methods. (B and C) Cells were treated with 0–30 μM MH for 24 h, and used for Western blotting as described in Methods. The blots were representative of three experiments. (D and F) Cells were treated with 0–30 μM MH for 1 h. Electrophoretic mobility shift assay (D) and immunofluorescence (F) were performed as described in Methods. Numbers, fold activation in relation to the control (D). (E) Cells were transfected with pNF-κB-Luc plasmid (5× NF-κB) for 6 h and then incubated with complete media containing 0–30 μM MH with TNF-α (10 ng·mL⁻¹) for 8 h. Luciferase activity was assessed as described in Methods. Columns, mean of triplicate; bars, SD. ^#P < 0.05 indicates statistically significant differences from the untreated group. *P < 0.05 indicates statistically significant differences from the TNF-α-treated group.

Figure 4

Transfection of PC3 cells with p21 siRNA or p21 mutant plasmid abolished the effect of MH on cell viability, apoptosis and related protein expression. (A–C) The cells were transfected with p21 siRNA for 6 h, grown for 24 h in complete media, and treated with 20 μM MH for 24 h. (D–F) The cells were treated with p21 mutant plasmid which has mutated cyclin D1/cdk4 complex binding site (Cy1 site, 17–24) for 6 h, grown for 24 h in complete media, and treated with 20 μM MH for 24 h. Cell counting (A,D), DAPI and TUNEL staining (B, E), and Western blotting (C, F) were performed as described in Methods. Columns, mean of triplicate; bars, SD. *P < 0.05 indicates statistically significant differences from the untreated group.

Figure 5

Effect of 4-O-methylhonokiol (MH) on the tumour growth in PC3 xenografts, in vivo model. Mice with SW620 or PC3 xenografts were treated with MH (40 or 80 mg·kg⁻¹ everyday) or cisplatin (10 mg·kg⁻¹ once a week) for 4 weeks. (A) Tumour volume and weight were measured as described in Methods. Data respresent mean of ten animals; bars show SD. *P < 0.05. (B) Immunohistochemistry; (C) electrophoretic mobility shift assay; and D, Western blotting were performed as described in Methods. Columns, mean of three animals; bars, SD. *P < 0.05 indicates statistically significant differences from the untreated group.