Activation of PERK Contributes to Apoptosis and G2/M Arrest by Microtubule Disruptors in Human Colorectal Carcinoma Cells ‡

Abstract

Microtubule-targeting agents (MTAs) are widely used in cancer chemotherapy, but the therapeutic responses significantly vary among different tumor types. Protein kinase RNA-like endoplasmic reticular (ER) kinase (PERK) is an ER stress kinase, and the role of PERK in the anticancer effects of MTAs is still undefined. In the present study, taxol (TAX) and nocodazole (NOC) significantly induced apoptosis with increased expression of phosphorylated PERK (pPERK; Tyr980) in four human colon cancer cell lines, including HCT-15, COLO205, HT-20, and LOVO cells. Induction of G₂/M arrest by TAX and NOC with increases in phosphorylated Cdc25C and cyclin B1 protein were observed in human colon cancer cells. Application of the c-Jun N-terminal kinase (JNK) inhibitors SP600125 (SP) and JNK inhibitor V (JNKI) significantly reduced TAX- and NOC-induced apoptosis and G₂/M arrest of human colon cancer cells. Interestingly, TAX- and NOC-induced pPERK (Tyr980) protein expression was inhibited by adding the JNK inhibitors, SP and JNKI, and application of the PERK inhibitor GSK2606414 (GSK) significantly reduced apoptosis and G₂/M arrest by TAX and NOC, with decreased pPERK (Tyr980) and pJNK, phosphorylated Cdc25C, and Cyc B1 protein expressions in human colon cancer cells. Decreased viability by TAX and NOC was inhibited by knockdown of PERK using PERK siRNA in COLO205 and HCT-15 cells. Disruption of the mitochondrial membrane potential and an increase in B-cell lymphoma-2 (Bcl-2) protein phosphorylation (pBcl-2; Ser70) by TAX and NOC were prevented by adding the PERK inhibitor GSK and JNK inhibitor SP and JNKI. A cross-activation of JNK and PERK by TAX and NOC leading to anti-CRC actions including apoptosis and G₂/M arrest was first demonstrated herein.

Keywords: human colon carcinoma cells; microtubule disrupters; nocodazole; protein kinase RNA-like endoplasmic reticular kinase; taxol.

Figure 1

Taxol (TAX) and nocodazole (NOC) induced DNA ladders with increased protein kinase RNA-like endoplasmic reticular kinase (PERK) protein phosphorylation in human COLO205, HCT-15, LOVO, and HT-29 colorectal carcinoma (CRC) cells. (A) Increased DNA ladders by agarose electrophoresis and caspase-3 (Casp-3) protein cleavage by Western blotting in TAX- and NOC-treated CRC cells. Four CRC cells were treated with different concentrations (5, 10, and 20 μM) of TAX or NOC for 24 h, and DNA integrity and Casp-3 protein of cells was detected by agarose electrophoresis and Western blotting, respectively. (B) TAX and NOC reduced the viability of CRC cells according to an MTT assay. CRC cells were treated with TAX or NOC (10 μM) for 24 h, and the viability of CRC cells was examined by an MTT assay. Data are presented as a percentage of the control (% of CON) from three independent experiments. (C) TAX and NOC induced PERK protein phosphorylation at Thr980 in human CRC cells. As described in (B), total proteins from CRC cells under various treatments were applied for Western blotting, and levels of indicated proteins, including PERK, phosphorylated (p)PERK (Thr 980), and α-tubulin (α-TUB), were detected using specific antibodies. Each data point was calculated from three triplicate groups, and data are displayed as the mean ± SD. ** p < 0.01 denotes a significant difference between indicated groups.

Figure 2

TAX and NOC induced cell cycle arrest at the G₂/M phase with increased Cdc25C protein phosphorylation and cyclin B1 (cycB1) protein in human CRC cells. (A) Induction of the G₂/M phase by TAX and NOC in human CRC cells. Cells were treated with TAX or NOC (10 μM) for 24 h, and cell cycle progression of human CRC cells under different treatments was examined by a flow cytometric analysis using propidium iodide (PI) staining. (B) TAX time-dependently induced phosphorylated Cdc25C and cyclin B1 protein expressions in COLO 205, HT-29, and LOVO cells. Cells were treated with TAX (10 μM) for different times (4, 8, and 12 h), and expressions of the indicated proteins were examined by Western blotting. (C) TAX and NOC induced phosphorylation of Cdc25C and cyclin B1, but not cdc2, in COLO 205, HT-29, and LOVO cells. Cells were treated with TAX or NOC (10 μM) for 12 h, followed by Western blotting to detect expressions of indicated proteins. Data from three independent experiments were obtained and are displayed as the mean ± SD.

Figure 3

c-Jun N-terminal kinase (JNK) activation participated in TAX- and NOC-induced apoptosis of human CRC cells. (A) The JNK inhibitor SP600125 (SP) protected cells from TAX- and NOC-induced cytotoxicity in human CRC cells. Human CRC cells were treated with SP (10 μM) for 30 min followed by TAX or NOC (10 µM) treatment for 24 h. The viability of cells under different treatments was examined by an MTT assay, and results are expressed as a percentage of the control (% of CON). (B) SP inhibition of cleavage of Casp-3 protein in human CRC cells. As described in (A), expressions of the Casp-3 and α-TUB proteins were examined by Western blotting using specific antibodies. (C) Addition of JNK inhibitor V (JNKI) reversed the cytotoxicity of TAX and NOC in HCT-15, HT-29, and COLO205 cells. Cells were treated with JNKI (10 μM) for 30 min followed by TAX and NOC (10 μM) treatment for 24 h. The viability of the cells was examined by an MTT assay. (D) SP and JNKI inhibited TAX- and NOC-induced DNA ladders in human CRC cells. As described above, DNA integrity was examined by agarose electrophoresis. Data from three independent experiments were obtained and are displayed as the mean ± SD. ** p < 0.01 denotes a significant difference between indicated groups.

Figure 4

JNK activation is involved in G₂/M arrest by TAX and NOC in human CRC cells. (A) The JNK inhibitor SP inhibited the increase in cyclin B1 and phosphorylated Cdc25C protein expressions by TAX and NOC in human CRC cells. As described in Figure 3A, expressions of cyclin B1, Cdc25C, and α-TUB proteins were examined by Western blotting using specific antibodies. (B) SP decreased TAX- and NOC-induced G₂/M arrest in human CRC cells. As described above, the G₂/M ratio of indicated cells by TAX and NOC with or without SP (10 μM) treatment was detected by a flow cytometric analysis using PI staining. (C) JNKI inhibited TAX- and NOC-induced phosphorylated Cdc25C and cyclin B1 protein expressions in the indicated human CRC cells. (D) As described in (B), JNKI reduced the G₂/M ratio induced by TAX and NOC. Each data point was calculated from three triplicate groups, and data are displayed as the mean ± SD. ** p < 0.01 denotes a significant difference between indicated groups.

Figure 5

Inhibition of JNK by SP (A) or JNKI (B) inhibited PERK and JNK protein phosphorylation by TAX and NOC in human CRC cells. Cells were treated SP or JNKI (10 µM) for 30 min followed by TAX or NOC (10 μM) stimulation for 12 h. Expressions of pPERK (Thr980), pJNK, total JNK, and α-TUB proteins were examined by Western blotting using specific antibodies. (Right panel) Western blotting. (Left panel) The relative levels of pPERK (Thr980) and pJNK were normalized to α-TUB and tJNK, respectively, and quantified using Image J software [29].

Figure 6

The PERK inhibitor GSK2606414 (GSK) inhibited cell death, PERK, and JNK protein phosphorylation in human CRC cells by TAX and NOC. (A) Both COLO205 and LOVO cells were treated with different concentrations (5, 10, and 20 μM) of GSK for 30 min, followed by TAX and NOC (10 μM) treatment for 24 h. The viability of cells under different treatments was examined by an MTT assay, and results are expressed as a percentage of the control (% of CON). (B) GSK inhibition of TAX- and NOC-induced cyclin B1, phosphorylated PERK, JNK, and Cdc25C protein expressions in COLO205 and LOVO cells. Both cell lines were treated with GSK (10 µM) for 30 min, followed by TAX and NOC (10 μM) stimulation for 12 h for detecting phosphorylated PERK and JNK proteins, cyclin B1, and phosphorylated Cdc25C proteins by Western blotting. (C) GSK reversed the increase in G₂/M and the decrease in the G₁ ratio by TAX and NOC in COLO205 cells. (D) Knockdown of PERK protein by transfection of COLO205 and HT-29 cells with PERK siRNA (Si; siPERK) inhibited TAX and NOC-induced cell death. Both cells were transfected with siPERK or scramble siRNA (SC) (1 ng), and expression of phosphorylated PERK protein and viability of both cells were examined by Western blotting (left panel) and MTT assay (right panel), respectively. Each data point was calculated from three triplicate groups, and data are displayed as the mean ± SD. ** p < 0.01 denotes a significant difference between the indicated groups.

Figure 7

The PERK inhibitor GSK and the JNK inhibitors SP and JNKI inhibited TAX- and NOC-induced loss of the mitochondrial membrane potential (MMP) and phosphorylated B-cell lymphoma-2 (Bcl-2) (Ser-70) in human CRC cells. (A) HCT-15 cells were treated with GSK, SP, or JNKI (10 μM) for 30 min, followed by TAX and NOC (10 μM) stimulation for 12 h. The MMP was examined by a flow cytometric analysis via DiOC6 staining. (B) GSK inhibition of TAX- and NOC-induced phosphorylation of the Bcl-2 protein (pBcl-2) at Ser-70 in COLO205, HCT-15, HT-29, and LOVO cells. Cells were treated with GSK (10 μM) for 30 min, followed by TAX or NOC stimulation for 12 h. (C) The JNK inhibitor SP decreased TAX- and NOC-induced phosphorylation of the Bcl-2 protein (pBcl-2) at Ser-70 in HCT-15 and LOVO cells. Both cell lines were treated with SP (10 μM) for 30 min, followed by TAX or NOC stimulation for 12 h. Levels of Bcl-2, pBcl-2, and α-TUB proteins were examined by Western blotting using specific antibodies. Each data point was calculated from three triplicate groups, and data are displayed as the mean ± SD. ** p < 0.01 denotes a significant difference from NOC- and TAX-treated groups, respectively.

Figure 8

A tentative mechanism for anti-CRC actions of the microtubule-disrupting agents TAX and NOC.