Induction of apoptosis coupled to endoplasmic reticulum stress in human prostate cancer cells by n-butylidenephthalide

Abstract

Background: N-butylidenephthalide (BP) exhibits antitumor effect in a variety of cancer cell lines. The objective of this study was to obtain additional insights into the mechanisms involved in BP induced cell death in human prostate cancer cells.

Methods/principal findings: Two human prostate cancer cell lines, PC-3 and LNCaP, were treated with BP, and subsequently evaluated for their viability and cell cycle profiles. BP caused cell cycle arrest and cell death in both cell lines. The G0/G1 phase arrest was correlated with increase levels of CDK inhibitors (p16, p21 and p27) and decrease of the checkpoint proteins. To determine the mechanisms of BP-induced growth arrest and cell death in prostate cancer cell lines, we performed a microarray study to identify alterations in gene expression induced by BP in the LNCaP cells. Several BP-induced genes, including the GADD153/CHOP, an endoplasmic reticulum stress (ER stress)-regulated gene, were identified. BP-induced ER stress was evidenced by increased expression of the downstream molecules GRP78/BiP, IRE1-α and GADD153/CHOP in both cell lines. Blockage of IRE1-α or GADD153/CHOP expression by siRNA significantly reduced BP-induced cell death in LNCaP cells. Furthermore, blockage of JNK1/2 signaling by JNK siRNA resulted in decreased expression of IRE1-α and GADD153/CHOP genes, implicating that BP-induced ER stress may be elicited via JNK1/2 signaling in prostate cancer cells. BP also suppressed LNCaP xenograft tumor growth in NOD-SCID mice. It caused 68% reduction in tumor volume after 18 days of treatment.

Conclusions: Our results suggest that BP can cause G0/G1 phase arrest in prostate cancer cells and its cytotoxicity is mediated by ER stress induction. Thus, BP may serve as an anticancer agent by inducing ER stress in prostate cancer.

Figure 1. Effects of BP on the viability and morphological changes of human prostate cancer cells.

DU-145 (black), PC-3 (grey), LNCaP (white) cells were treated with increasing concentration of BP (25 to 100 µg/ml) for 24 (A) and 48 h (B) and analyzed with MTT assay. LNCaP and PC-3 cells were treated with 0.2% DMSO (C and E, respectively) or 70 µg/ml BP (D and F, respectively) for 24 h, were shown.

Figure 2. BP induces G0/G1 arrest and changes the expression profiles of G0/G1 regulatory proteins.

BP induced cell cycle G0/G1 arrest in (A) LNCaP and (B) PC-3 cells. Cells were seeded at 8×10⁵ (LNCaP) and 6×10⁵ (PC-3) per 6-cm plate in triplicates and treated with 70 µg/ml BP for 12–24 h. Data are presented as means ± S.D. from three different experiments. *, P<0.05; **, P<0.01. Western blot analysis of cyclin D1, CDK2, p16, p21, p27 and phospho-Rb (Ser807/811) was performed in LNCaP cells (C) and PC-3 cells (D). β-actin was used as an internal control. Western blot analysis of phospho-Akt (Ser473), Akt, phospho-GSK-3β (Ser9) and GSK-3β was performed in LNCaP cells (E) and PC-3 cells (F). β-actin was used as an internal control.

Figure 3. BP induces apoptosis in prostate cancer cells.

Western blot analysis of active caspase 8, active caspase 3 and bax was performed in LNCaP cells (A) and PC-3 cells (B). β-actin was used as an internal control. LNCaP cells and PC-3 cells were incubated in the absence (C and E, respectively) or presence (D and F, respectively) of 70 µg/ml BP for 48 h and then subjected to TUNEL assay.

Figure 4. BP induces ER stress-related genes expression in prostate cancer cells.

Western blot analysis of BiP, calnexin, PDI, ATF6, phospho-eIF2α (Ser51), IRE1-α, GADD153/CHOP, phosphor-ASK1 (Thr845), ASK1 and Fas were performed in LNCaP cells (A) and PC-3 cells (B). β-actin was used as an internal control. Western blot analysis of BiP, IRE1-α, GADD153/CHOP and Ero1-Lα were performed in LNCaP cells (C) and PC-3 cells (D). β-actin was used as an internal control.

Figure 5. Nuclear translocation of GADD153/CHOP after BP treatment in human prostate cancer cells.

(A) LNCaP cells and (B) PC-3 cells were treated with 0.2% DMSO (control) or 70 µg/ml BP for 12 h, fixed and stained with anti-GADD153/CHOP. FITC-labeled secondary antibody was used (green fluorescence) Nuclei were stained with DAPI (blue fluorescence). Images were captured by a confocal laser microscope.

Figure 6. BP induces GADD153 and IRE1-α dependent cell death.

(A) LNCaP cells were transfected with scramble siRNA (^#), 20, or 40 nM GADD153 siRNA, respectively, for 48 h using the RNAifect transfection reagent. After 24 h treatment with 70 µg/ml BP, western blot analysis was performed for GADD153. (B) BP-induced anti-proliferative activity was measured with MTT assay in LNCaP cells transfected with scramble siRNA (^#) or 20 nM GADD153 siRNA for 48 h and then treated with 70 µg/ml BP for 24 h. Data are expressed as means ± S.D. from three independent experiments. ***, P<0.001. (C) LNCaP cells were transfected with scramble siRNA (^#), 20, or 50 nM IRE1-α siRNA, respectively, for 48 h using the RNAifect transfection reagent. After 24 h of treatment with 70 µg/ml BP, western blot analysis was performed for IRE1-α and GADD153, respectively. β-actin was used as an internal control. (D) BP-induced anti-proliferative activity was measured with MTT assay in LNCaP cells transfected with scramble siRNA (^#), 20, or 50 nM IRE1-α siRNA, respectively, for 48 h and then treated with 70 µg/ml BP for 24 h. Data are expressed as means ± S.D. from three independent experiments. ***, P<0.001 versus vehicle.

Figure 7. BP-induced ER stress and cell death are JNK1/2 dependent.

(A) LNCaP cells and (B) PC-3 cells were treated with 70 µg/ml BP for the indicated times. Phospho-ERK1/2, total ERK1/2, phospho-JNK1/2, total JNK1/2, phospho-p38, and total p38 were detected by western blotting respectively. (C) LNCaP cells were transfected with scramble (^#) or 20 nM JNK1/2 siRNA for 48 h using the RNAifect transfection reagent. After treatment with 70 µg/ml BP for 24 h, western blot analysis was performed for phospho-JNK1/2, total JNK1/2, IRE1-α and GADD153. β-actin was used as an internal control. (D) BP-induced anti-proliferative activity was measured with MTT assay in LNCaP cells transfected with scramble (^#) or JNK1/2 siRNA for 48 h and then treated with 70 µg/ml BP for 24 h. Data was expressed as means ± S.D. from three independent experiments. ***, P<0.001 versus vehicle.

Figure 8. BP inhibits xenographic growth of LNCaP cells in NOD-SCID mice.

(A) NOD-SCID mice were injected with approximately 5×10⁵ LNCaP cells into the dorsal subcutaneous tissue. When the tumor reached 100–250 mm³, LNCaP tumor-bearing mice were administrated s.c. with vehicle control (▪) or 500 mg/kg BP (•) on days 0–4 for 5 days. The relative tumor volumes of control and the BP-treated groups were shown as means ± S.D. of tumor volume at each time point. (B) Tumors of control and therapeutic groups were removed and weighted on day 18. Average tumor weight from the BP-treated group was 56% smaller than control group. Data was expressed as means ± S.D. of tumor weight of the control and the BP-treated groups. **, P<0.01. (C, D) Tumor tissue sections with HE staining of control (C) and therapeutic groups (D).GADD153 expressions were immunohistochemically identified in the control (E) and the BP-treated group (F). The GADD153-positive cells were stained brown. Expression of GADD153 and active form caspase-3 in the LNCaP xenograft tumor tissue were upregulated after BP administration as compared to control group by western blotting analysis (G).

Figure 9. Schematic model of BP-induced apoptosis in human prostate cancer cells.

BP induces apoptosis in human prostate cancer cells through multiple apoptotic pathways, including G0/G1 phase cell cycle arrest, death receptor pathway and ER stress-dependent pathway.