17-DMAG

Glucose-regulated protein 78 mediates the therapeutic efficacy of 17-DMAG in colon cancer cells

Abstract

Glucose-regulated protein 78 (GRP78) is expressed as part of the molecular response to endoplasmic reticulum (ER) stress and mediates protein folding within the cell. GRP78 is also an important biomarker of cancer progression and the therapeutic response of patients with different cancer types. However, the role of GRP78 in the cytotoxic effect of 17-DMAG in colon cancer cells remains unclear. GRP78 ex- pression was knocked down by small interfering RNA (siRNA). The anticancer effects of 17-DMAG were assessed by an 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, a flow cytometric cell-cycle analysis,and an Annexin V-propidium iodide (PI) apoptotic assay. We found that HT-29 cells expressed a lower level of GRP78 compared with DLD-1 cells. The MTT assay revealed that HT-29 cells were more sensitive to 17-DMAG treatment than DLD-1 cells. GRP78 knock down (GRP78KD) cells demon- strated an increased sensitivity to 17-DMAG treatment com- pared with the scrambled control cells. Based on the cell-cycle analysis and Annexin V-PI apoptotic assay, apoptosis dramat- ically increased in GRP78KD cells compared with scrambled control DLD-1 cells after these cells were treated with 17- DMAG. Finally, we observed a decrease in the level of Bcl- 2 and an increase in the levels of Bad and Bax in GRP78KD cells treated with 17-DMAG. These results are consistent with an increased sensitivity to 17-DMAG after knock down of GRP78. The level of GRP78 expression may determine the therapeutic efficacy of 17-DMAG against colon cancer cells.

Keywords : Colon cancer . Anti-proliferation . GRP78 . 17-DMAG . Hsp90

Introduction

Colorectal cancer is one of most commonly diagnosed malignancies in the world. More than 1 million people are diagnosed with colorectal cancer annually, and more than 600 thousand people die from colorectal cancer an- nually [1]. Surgery is the treatment of choice for nonmetastatic colorectal cancer. However, for patients with nodal involvement or distant metastases, the admin- istration of chemotherapy can improve their long-term survival. To prevent disease recurrence, adjuvant chemotherapy is recommended for patients with stage III disease and high-risk patients with stage II disease, and fluorouracil/leucovorin plus oxaliplatin (FOLFOX) has been the standard of care since 2003 [2]. Approximately half of the patients who receive this treatment, however, will develop metastasis. To treat metastatic disease, mo- lecular therapy involving VEGF neutralization or EGFR inhibition can further improve the treatment response of patients with stage IV disease. However, an improvement in the therapeutic response and a reduction in the side effects are still needed.

Heat shock protein 90 (Hsp90) is the key chaperone in mediating protein folding and the protein stability of client proteins in normal and malignant cells [3–7]. Many of the Hsp90 client proteins, such as cdk4, erbB2, and mutant p53, are key modulators of cancer progression [8–11]. Inhibitors of Hsp90, including 17-AAG and 17-DMAG, may potentially be used for cancer therapy [12–15]. A comparison of 17-AAG and 17-DMAG demonstrated that 17-DMAG has better solubility in water and better oral bioavailability [16, 17]. Some clinical trials have demon- strated that 17-DMAG is an effective drug for castration- resistant prostate cancer, melanoma, and acute myeloid leukemia, but it is associated with several common side effects, such as peripheral neuropathy, fatigue, and cardiotoxicity [18–20]. In the case of colon cancer, inhibi- tion of Hsp90 by 17-DMAG suppresses the metastatic po- tential of colon cancer [21]. Additionally, 17-DMAG may also induce reactive oxygen species in colon cancer and impart an antitumorigenic effect [22]. Furthermore, 17- DMAG was shown to improve the effects of oxaliplatin treatment of metastatic colon cancer [23]. To eliminate the limitation of chemotherapy for patients with colon can- cer and to increase the clinical application of 17-DMAG, the search for reliable and predictive biomarkers is urgent.

Glucose-regulated protein 78 (GRP78), also referred to as BiP, is a stress protein and a member of the heat shock protein 70 (HSP70) family. GRP78 is encoded by a single copy gene in the eukaryotic genome, and its induction is primarily regulated at the transcriptional level [24, 25]. Most GRP78 is located in the endoplasmic reticulum (ER), and only a small fraction of GRP78 is found on the cell surface [26]. The expression of GRP78 is highly cor- related with the progression of various cancers [27–29] and has been shown to prevent apoptosis by inhibiting BAX activation [30] and suppressing caspase 7 [31]. GRP78 has also been shown to cause drug resistance in certain cancer types, such as gastric cancer, colon cancer, breast cancer, and hepatocellular carcinoma (HCC) [32–35]. In contrast to sorafenib, our previous study found that treatment with curcumin was more effective in HCC cases in which GRP78 was highly expressed. However, the role of GRP78 in 17-DMAG treatment of colon cancer remains elusive. In this study, we demonstrate that the GRP78 ex- pression level may influence the therapeutic efficacy of 17- DMAG in patients with colon cancer.

Materials and methods

Chemicals, reagents, and cell culture

The following items were obtained from Sigma-Aldrich (St. Louis, MO, USA): 17-Dimethylaminoethylamino-17- demethoxygeldanamycin (17-DMAG), propidium iodide (PI), Tris–HCl, trypan blue, ethylenediaminetetraacetic acid (EDTA), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoli- um bromide (MTT), ribonuclease A, and dimethyl sulfoxide (DMSO). GRP78 (H-129) and GAPDH (SC-32233) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Bcl2 (1071-1) and Bax (1063-1) antibodies were purchased from Epitomics (Burlingame, CA, USA). The Bad (GTX61181) antibody was purchased from GeneTex Inc. (Irvine, CA, USA).

Human colon adenocarcinoma HT29 (HTB-38) and DLD- 1 (CCL-221) cell lines were purchased from American Type Culture Collection (ATCC, Rockville, MD, USA); these cell lines were originally isolated from human colon adenocarci- noma at Dukes’ stages B and C, respectively. The cells were cultured in RPMI 1640 medium with 10 % fetal calf serum (FCS), penicillin (100 U/ml), and streptomycin (100 μg/ml) in a humidified incubator (37 °C, 5 % CO2). Cells were either subcultured or used for experimentation before they reached 80 % confluence.

Generation of GRP78 knockdown cells using siRNA

The expression of GRP78 was ablated in DLD-1 cells using small interfering RNA (siRNA) as previously de- scribed [36]. The target sequence for the human GRP78 mRNA was 5′-AAGGTTACCCATGCAGTTGTT-3′, and 5′-AAGGTGGTTGTTTTGTTCACT-3′ was the sequence for the scrambled siRNA. The GRP78 siRNA and the scrambled siRNA were inserted into the pSUPERIOR vector and transfected into cells using the Neon® Transfection System (Life Technologies, Grand Island, NY); the stably transfected cells were selected by antibi- otics as previously described [37].

MTT assay

Cells (2 × 104) were seeded onto 24-well plates and incu- bated overnight. Various concentrations of 17-DMAG or its vehicle (dH2O) were added to the cells. The medium was aspirated at specific times, and the cells were incubat- ed in a final concentration of 0.25 mg/ml MTT for 1 h.Formazan was solubilized with DMSO, and the optical density was measured at 550 nm using a spectrophotome- ter (GE Healthcare, Piscataway, NJ, USA) [38].

Flow cytometry for cell-cycle analysis

Cells (3×105) were seeded onto 6-well plates overnight and then incubated with 17-DMAG or vehicle for 48 h. At specific time intervals, cells were harvested, washed with PBS, fixed in pure methanol, treated with RNase A (at a final concentration of 40 μg/ml), and finally stained with PI (40 μg/ml) for 30 min at room temperature. The stained cells were analyzed using a flow cytometer (BD Biosciences, San Jose, CA, USA), and the DNA content was quantified using Modfit software (Verity Software House, Topsham, ME, USA). The percentage of hypodiploid (sub-G1) cells was used to quantify the dead cells [39].

Annexin V-PI and polycaspase apoptotic detection assay

Cells (5×105) were seeded onto 6-well plates overnight and then treated with 17-DMAG or vehicle for 48 h. The cells were then harvested and washed with PBS. Next, the staining solution was prepared by diluting 2 μl of Annexin-fluorescein isothiocyanate (FITC) in 100 μl binding buffer and then adding 2 μl PI. The Annexin V-FITC kit uses FITC-conjugated Annexin V in concert with the nucleic acid dye PI. As the cell membrane becomes more permeable during the later stages of apopto- sis, PI will readily move across the cell membrane and bind to DNA. An Annexin V-FITC Apoptosis Detection Kit (Immunochemistry Technologies, USA) was used for the fluorescent detection of Annexin V-bound apoptotic cells and quantitative analysis by flow cytometry [40].

Polycaspase activities were determined by a flow cytometer (BD Bioscience, USA) using FAM-FLICA reagent (Immunochemistry Technologies, USA) according to the manufacturer’s instructions. Briefly, the harvested cells were incubated with FLICA reagent for 1 h at 37 °C and then washed twice to remove the unbound reagent. Next, the cells were resuspended in 300 μl of wash buffer that contained PI and incubated for 5 min at 37 °C. Finally, the cells were ana- lyzed by flow cytometry.

Fig. 1 The cytotoxic effect of 17-DMAG in colon cancer cells. The cytotoxic effect of 17-DMAG in colon cancer cells (HT-29 and DLD-1 cells) was determined by an MTT assay. a HT-29 and DLD-1 cells were treated with various doses of 17-DMAG (0∼400 nM) for 48 h. Cell survival was determined by an MTT assay. b The cell-cycle distribution of colon cancer cells after 17-DMAG treatment is shown. Cells were exposed to 100 nM 17-DMAG for 48 h. Cells were then harvested and stained with PI. The cell-cycle distribution of each sample was analyzed. c The percentages of cell population of each phase at different samples were presented by plot. d DNA cleavage was detected using an Annexin- PI apoptotic assay according to the manufacturer’s instructions. The positive cells were then analyzed by flow cytometry. All of the experiments were independently repeated at least three times. *p<0.05. Western blot analysis A Western blot analysis using anti-GRP78, Bcl-2, Bax, Bad, or GAPDH was performed as described previously [35, 41]. Statistical analysis All experiments were repeated at least three times. All of the data are presented as the mean±standard deviation. Statistical significance was determined using Student’s t test (two-tailed) to compare two datasets. A value of p<0.05 was considered and the sensitivity of the cell to 17-DMAG. To identify the role that GRP78 plays in determining the cytotoxic effects of 17-DMAG in colon cancer cells, we knocked down GRP78 expression in DLD-1 cells using siRNA. The GRP78 expression level in the scrambled control and GRP78 knockdown (GRP78KD) cells was deter- mined by Western blot analysis. As shown in Fig. 2b, the GRP78 level in the GRP78KD cells was 80 % less than the expression level in the scrambled control cells. GRP78 modulates the effects of 17-DMAG To further investigate the influence of GRP78 on the antican- cer effects of 17-DMAG against colon cancer, we treated the scrambled control and GRP78KD cells with different doses of Both 17-AAG and 17-DMAG exhibit antiproliferative ef- fects against several cancers. To quantify this activity in colon cancer cells, HT-29 and DLD-1 cells were incubat- ed with different concentrations of 17-DMAG, and cell viability was determined using an MTT assay. Dose- dependent cytotoxic effects of 17-DMAG were observed (Fig. 1a). DLD-1 cells were more resistant to 17-DMAG treatment than HT-29 cells. We also determined the cell- cycle distribution by flow cytometry after PI staining. As shown in Fig. 1b, c, the population of HT-29 cells in G2- M phase increased dramatically after 17-DMAG treat- ment. Additionally, the apoptotic cell population (sub- G1) was also larger in the 17-DMAG-treated HT-29 cells compared with the 17-DMAG-treated DLD-1 cells. Similar results were obtained using the Annexin-PI apo- ptotic assay (Fig. 1d). The number of apoptotic cells was higher in the treated HT-29 cells than in the treated DLD- 1 cells. These results indicate that HT-29 cells are more sensitive to 17-DMAG than DLD-1 cells. Determination and manipulation of the GRP78 expression level in colon cancer cells We determined the GRP78 expression levels and found that GRP78 was highly expressed in the DLD-1 cells compared with the HT-29 cells (Fig. 2a). There appears to be a correlation between the GRP78 expression level 17-DMAG. As shown in Fig. 2c, we found that GRP78KD DLD-1 cells demonstrated a greater sensitivity to 17-DMAG than the scrambled control cells. These results indicate that a decrease in GRP78 expression may sensitize cells to 17- DMAG treatment. Fig. 2 GRP78 expression levels were correlated with the cytotoxic effect of 17-DMAG in colon cancer cells. The endogenous GRP78 expression level was determined and manipulated in colon cancer cells. a The expression level of GRP78 in HT-29 and DLD-1 cells was determined by Western blot analysis. b GRP78-siRNA and scrambled control siRNA were introduced into DLD-1 cells as described in the “Materials and Methods” section. The GRP78 expression level in the scrambled control cells and the GRP78KD cells was determined by Western blot analysis. c The cytotoxic effects of 17-DMAG in the scrambled control cells and GRP78KD DLD-1 cells were determined by an MTT assay.Cells were treated with various doses of 17-DMAG (0∼400 nM) for 48 h. Cell survival was determined using an MTT assay. All of the experiments were independently repeated at least three times. *p<0.05. Knockdown of GRP78 reduces 17-DMAG-induced apoptosis Scrambled control and GRP78KD DLD-1 cells were treated with 17-DMAG (100 nM) or vehicle control for 24 h. The cells were then harvested, fixed, and subjected to flow cyto- metric analyses for cell-cycle distribution and apoptosis. As shown in Fig. 3, after treatment with vehicle, the cell-cycle distribution was similar in the scrambled control and GRP78KD cells. After exposure to 17-DMAG, cell-cycle ar- rest in S phase was observed in the GRP78KD control cells (27 to 44 % S phase cell population). The scrambled control cells only slightly increase the S phase cell population (23 to 28 %) after 17-DMAG treatment. After treatment with 17- DMAG for 48 h, the percentage of sub-G1 cells was higher in the GRP78KD DLD-1 cells than in the scrambled control cells. These results indicate that 17-DMAG treatment may cause cells to arrest in S phase and that GRP78 KD cells are more sensitive to 17-DMAG treatment. Detection of apoptosis by an Annexin V-PI apoptotic assay To further study the role of GRP78 expression in the cytotoxic effects of 17-DMAG against colon cancer cells, we used an Annexin V-PI apoptotic assay to visualize apoptosis induced by 17-DMAG. To further support these findings, PI and Annexin V double staining was also performed to detect apo- ptosis. As shown in Fig. 4a, after treatment with 17-DMAG, the number of Annexin V-positive cells increased in the GRP78KD DLD-1 cells compared with the scrambled control cells. The polycaspase apoptotic assay also revealed a higher number of positive cells in the GRP78KD DLD-1 cells treated with 17-DMAG (Fig. 4b). These results indicate that the GRP78KD DLD-1 cells were more sensitive to the 17- DMAG treatment than the scrambled control DLD-1 cells. Fig. 3 Cell-cycle distribution of scrambled control and GRP78KD DLD-1 cells after treatment with 17-DMAG. Scrambled control and GRP78KD DLD-1 cells were exposed to 100 nM 17-DMAG for 48 h. The cells were then harvested and stained with PI. a The cell-cycle distribution of each sample was analyzed by a flow cytometer. b The percentages of cell population of each phase at different samples were presented. All of the experiments were independently repeated at least three times. 17-DMAG treatment influences the expression patterns of Bcl-2, Bad, and Bax in GRP78KD DLD-1 cells To characterize the molecular mechanism underlying GRP78 roles in 17-DMAG-induced apoptosis in DLD-1 cells, we examined the expression of apoptosis-associated proteins in response to 17-DMAG. Scrambled control and GRP78KD DLD-1 cells were treated with 17-DMAG for 48 h, and the expression levels of anti-apoptotic protein (Bcl-2) and pro- apoptotic protein (Bax and Bad) were detected using Western blot analysis. As shown in Fig. 5, a decrease in the Bcl-2 level was observed in GRP78KD cells treated with 17- DMAG, but very little change was observed in the treated scrambled control cells. In GRP78KD cells, the levels of Bad and Bax increased significantly after exposure to 17- DMAG. In contrast, the Bax level did not change and Bad slightly increased in the scrambled control cells after treatment with 17-DMAG. These results are consistent with an increase in sensitivity to 17-DMAG after silencing GRP78 gene ex- pression. These data suggest that silencing GRP78 gene ex- pression sensitizes cells to 17-DMAG by modulating the ex- pression of apoptotic proteins. Discussion ER stress refers to a condition in which misfolded or unfolded proteins accumulate in the ER [42–44]. Increasing evidence suggests that cancer cells can adapt to ER stress and prevent the stress-induced apoptotic pathway by activating the unfolded protein response [45–47]. GRP78 was originally regarded as an ER chaperone that facilitates protein folding and assembly and is currently widely used as a marker for ER stress. Additional evidence suggests that GRP78 is not only an ER stress marker but also a biomarker for the aggressive be- havior of cancer and tumor responsiveness to therapy [48–51]. For example, inhibiting GRP78 may enhance the sensitivity to chemotherapy in breast cancer cells [52]. Furthermore, recent studies have suggested that GRP78 is a predictive marker of the chemotherapeutic response in different cancer types [50, 53]. In the case of colon cancer, targeting GRP78 resulted in the suppression of cell proliferation [54], which indicates that GRP78 may be a potential target for colon cancer therapy. Colorectal cancer remains a major health problem world- wide. Although new treatment approaches for metastatic co- lon cancer have resulted in consistent improvement in clinical outcomes [55], additional progress is still needed to achieve better treatment results. The Hsp90 family of chaperone pro- teins has been shown to interact with many different client proteins to maintain the stability of those proteins [8–11]. Many Hsp90 client proteins also play an essential role in can- cer progression and therapeutic responses [8–11]. Hsp90 in- hibitors (e.g., 17-AAG, 17-DMAG) have also demonstrated anticancer effects and therapeutic potential for treating cancer [14, 56]. In this study, we investigated whether inhibiting GRP78 improves the therapeutic effect of the Hsp90 inhibitor 17-DMAG. We found that a reduction in GRP78 expression resulted in greater 17-DMAG treatment efficacy in colon cancer-derived cells. This finding is interesting and has clini- cal implications because our results imply that lower GRP78 expression in colon cancer cells may increase their sensitivity to 17-DMAG treatment. A previous study demonstrated that treatment with 17-DMAG may increase Hsp70 expression levels and that targeting Hsp70 may enhance the therapeutic efficacy of 17-DMAG [57–59]. In our study, we did not ob- serve an increase in the GRP78 expression level after 17- DMAG treatment, and thus, GRP78 may be a unique bio- marker that can be used in colon cancer therapy. Therefore, targeting both GRP78 and Hsp90 may be a novel direction for clinical colon cancer therapy. Apoptosis is programmed cell death that is triggered through an extrinsic pathway or an intrinsic pathway. In the case of chemotherapeutic agents that induce cell death, apoptosis is usually triggered by the intrinsic pathway [60, 61]. The intrinsic pathway requires mitochondria as central integrators and coordinators of the apoptotic process and is characterized by mitochondrial membrane permeabilization. This event leads to the release of pro- apoptotic proteins from the intermembrane space and the subsequent activation of the caspase cascade [62]. The Bcl-2 family of proteins, including proteins with pro- (e.g., Bax, Bak, and Bad) and anti-apoptotic (e.g., Bcl-2 and Bcl-xL) activities, are crucial regulators of MMP. A previous study demonstrated that the up-regulation of GRP78 inhibited the activation of BAX and chemothera- peutic drug resistance in epidermoid carcinoma cells [30]. In our study, apoptosis dramatically increased after 17- DMAG treatment in GRP78KD colon cancer cells, which expressed a lower level of the anti-apoptotic protein Bcl-2 and a higher level of the pro-apoptotic protein BAX. Fig. 5 Treatment with 17-DMAG alters the expression levels of apoptosis-related signaling molecules in both the scrambled control and GRP78KD DLD-1 cells. After a 48-h treatment with 17-DMAG (100 nM), proteins were extracted from scrambled control and GRP78KD DLD-1 cells. a The expression levels of Bcl-2, Bax, Bad, and GAPDH were determined by Western blot analysis. GAPDH was used as the internal control. b The intensity of the signal was calculated by a densitometer, and the relative expression levels were plotted. All of the experiments were independently repeated at least three times. All of the experiments were independently repeated at least three times. *p<0.05. Recent studies have suggested the pivotal of store-operated Ca2+ entry (SOCE) in the progression of colon cancer. Stromal interaction molecule 1 (STIM1), an essential component in SOCE, is an overexpression in colon cancer and significantly associated with tumor size, depth of invasion, and lymph node metastasis status [63]. Furthermore, STIM1 has been shown to positively regulate the migration and invasion abilities of co- lon cancer cells through increasing cyclooxygenase-2 expres- sion and inducing prostaglandin E2 production in colon can- cer cells [63]. SOCE drives most Ca2+-dependent signaling cascades and controls cell-cycle progression by regulating the expression of several calcium-dependent signaling path- ways, such as calmodulin, CaM-kinase, and calcineurin [64–66]. Orai and STIM proteins are the canonical compo- nents of SOCE. Recent study has shown that Orai1/STIM1 pathway constituted a native SOCE functioning as a crucial regulator of proliferation in clear cell renal cell carcinoma [67]. Also, SOCE-mediated melanoma proliferation was con- trolled via CaMKII/Raf-1/ERK signaling pathway activation [68]. Furthermore, it was demonstrated that Ca2+ entry involv- ing TRPC6, together with STIM1 and Orai1, increases cyclin D1 expression [69].In conclusion, 17-DMAG is a potential treatment for met- astatic colon cancer, and GRP78 may be a strong, predictive marker of 17-DMAG efficacy.