Chelerythrine induces apoptosis via ROS‐mediated endoplasmic reticulum stress and STAT3 pathways in human renal cell carcinoma

Abstract Renal cell carcinoma (RCC) is a heterogeneous histological disease and it is one of the most common kidney cancer. The treatment of RCC has been improved for the past few years, but its mortality still remains high. Chelerythrine (CHE) is a natural benzo[c]phenanthridine alkaloid and a widely used broad‐range protein kinase C inhibitor which has anti‐cancer effect on various types of human cancer cells. However, its effect on RCC has not been fully elucidated. In this study, we evaluated the effect and mechanism of CHE on RCC cells. Our study showed that CHE induced colony formation inhibition and G2/M cell cycle arrest in a dose‐dependent manner in RCC cells. In addition, CHE increased cellular ROS level, leading to endoplasmic reticulum (ER) stress, inactivating STAT3 activities and inducing apoptosis in RCC cells which were suppressed by NAC, a special ROS inhibitor. We further found that both knockdown of ATF4 protein and overexpression of STAT3 protein could reduce CHE‐induced apoptosis in Caki cells. These results demonstrated that the apoptosis induced by CHE was mediated by ROS‐caused ER stress and STAT3 inactivation. Collectively, our studies provided support for CHE as a potential new therapeutic agent for the management of RCC.


| 51
HE Et al. being less expensive than conventional drugs. 6,7 Approximately 60% of clinically used anti-cancer drugs are natural products or derivatives. 8 Chelerythrine (CHE) is a natural benzo[c]phenanthridine alkaloid extracted from plant species, such as Chelidonium majus, Macleaya cordata, and Sanguinaria canadensis et al. 9,10 CHE has a wide range of biological activities and plays an important role in anti-diabetes, 11 anticancer, 12 anti-fungus, 13 and the lipopolysaccharide-induced endotoxic shock 14 etc. The anti-cancer effects of CHE have been studied both in vitro and in vivo. Regulation of Bcl-2 family protein expression and activation of the mitochondrial pathway were reported to be associated with CHE induced hepatoma cell apoptosis. 15 in addition, CHE can inhibit proliferation and promote apoptosis in prostate cancer, 16 triplenegative breast cancer, 17 non-small cell lung cancer 12 (NSCLC) etc.
Here in the present study, we aimed to investigate the effect of CHE on the cultured human RCC. The results showed that CHE inhibited the growth of human RCC, induced G2/M cell cycle arrest and apoptosis. It also demonstrated that CHE induced apoptosis in human RCC through ROS mediated endoplasmic reticulum (ER) stress pathway. Moreover, we discovered for the first time that CHE induced apoptosis in human RCC through the inactivation of signal transducer and activator of transcription-3 (STAT3) signalling pathway. In summary, our findings indicated that CHE might be a therapeutic candidate in the treatment of RCC and a novel mechanism of CHE in anti-cancer activities was reported in this study.

| Cell viability assay
To measure viability of cells, we plated 6 × 10 3 cells per well of a 96well plate. Cells were attached overnight in complete growth media and were treated with CHE (dissolved in DMSO; diluted in RPMI medium) for 24 h, after which were subjected to the MTT assay.

| Colony formation assay
Caki and 786-O cells were seeded at 600 cells per well in six-well plates and treated with 2, 4 or 8 µmol/L CHE for 10 hours. Cells were allowed to grow for 8 or 9 days and stained with crystal violet solution to assess colony growth.

| Cell cycle and apoptosis analysis
For cell cycle analysis, cells were treated with CHE (6, 9 or 12 μmol/L) for 20 hours. The cells were then stained with PI (BD Biosciences, San Jose, CA) at a final concentration of 0.05 mg/mL and incubated at 4°C for 10 min in the dark. Cell cycle analysis was performed in Accuri C6 plus flow cytometer (BD Biosciences, CA).
For apoptosis determination, cells were treated with CHE (6,9 or 12 μmol/L) for 24 hours. Cells were then harvested and washed with PBS, resuspended in binding buffer containing Annexin V and propidium iodide (PI) (BD Biosciences, San Jose, CA). Cell apoptosis analysis was performed in Accuri C6 plus flow cytometer (BD Biosciences, CA).

| Determination of intracellular ROS
Intracellular ROS levels were measured through flow cytometry using DCFH-DA. In short, 5 × 10 5 cells were seeded in six-well plates, attached overnight, and then treated with CHE (6, 9 or 12 μmol/L) for 3 hours. NAC pretreatment for 1 hours if required. Cells were stained with 10 μmol/L DCFH-DA at 37°C under dyeing 30 minutes in the dark. Analysis of DCF fluorescence in the presence of ROS using Accuri C6 plus flow cytometer.

| Western blot analysis
Lysates from cells were prepared to determine protein levels using the Bradford assay (Bio-Rad, Hercules, CA). Proteins were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride transfer membranes. The blots were blocked with freshly prepared 5% nonfat milk in TBST for 2 hours at room temperature. Then the blots were incubated with specific primary antibodies overnight at 4°C. HRP-conjugated secondary antibodies and ECL substrate (Bio-Rad) were used for detection.

| Electron microscopy
Caki cells were seeded in 60 mm plates and then were treated with 12 μmol/L CHE in the presence or absence of NAC (5 mmol/L). The collected cells were fixed in phosphate buffer (pH 7.4) including 2.5% glutaraldehyde for 12 hours at 4°C. The cells were post-fixed in 1% OsO4 for 60 min at room temperature, stained with 1% uranyl acetate, dehydrated by graded acetone solutions and embedded in Epon. Areas containing cells were block-mounted and cut into 70 nm sections and examined with the electron microscope (H-7500, Hitachi, Ibaraki, Japan).

| Cell transfections
To knockdown ATF4 expression, Caki cells were seeded in sixwell plates at a density of 6 × 10 4 and cultured for 24 hours.
siRNA against ATF4 or non-targeting control were transfected in a final concentration of 50 pmol mL −1 using lipofectamine 3000 reagent (Invitrogen, CA). After 6-8 hours, the medium was replaced with fresh medium and cells were cultured for 36 hours.
Then, cells were treated with 12 μmol/L CHE for 3 hours and After 36-48 hours of transfection, the protein of STAT3 expression was detected by Western blotting analysis.

| Statistical analysis
All experiments were repeated at least three times. Statistical analyses were performed only when a minimum of n = 3 independent samples were acquired. All data were expressed as mean ± SEM.
Statistical analysis was performed with GraphPad Prism 6.0 software (GraphPad, San Diego, CA).

| CHE reduces cell viability in human renal cancer cells
The structure of CHE is shown in Figure 1A.

| CHE induces cell apoptosis in human renal cancer cells
Reduced viability in human RCC after CHE exposure prompted us to determine whether CHE induced apoptosis. We investigated the

| CHE increases ROS levels in human renal cancer cells
Previous studies have been shown that CHE may increase ROS level in prostate cancer cells and non-small cell lung cancer cells. 12,16 Hence, we investigated the effects of CHE on treatment renal cancer cells to produce ROS. We firstly performed flow cytometry analysis using DCFH-DA fluorescent dyes. As shown in Figure 3A-D, 6-12 µmol/L CHE treatment in Caki and 786-O cancer cells, the results showed that CHE could dose-dependently increase the ROS level in renal cancer cells. However, pretreatment with NAC (the specific ROS inhibitor) for 1 hour significantly suppressed the ROS levels caused by CHE ( Figure 3E-H). We also used other 4 ROS scavengers (BHA, Vita-E, Trolox and CTH) to narrow ROS species. Interestingly, all of these data did not reverse the ROS generation induced by CHE ( Figure S3) and clearly demonstrated that CHE increased ROS levels in renal cancer cells. Besides, only NAC inhibited CHE-induced ROS, while other lipid ROS inhibitors or lipid peroxide quenchers did not have this effect.

| CHE caused ROS-dependent ER stress activation in human renal cancer cells
Previous studies have proved that CHE causes prostate cancer cell apoptosis mainly related to the activation of ER stress signalling pathway. 16 Besides, it has been report that oxidative stress modulating drugs activate the ER stress-related apoptosis. 18 Thus, we speculated that ER stress contributed to renal cancer cells apoptosis by CHE treatment. Activating transcription factor 4 (ATF4) is a key transcription factor in the ER stress pathway. Next, we determined the protein expression which is associated with ER stress, such as ATF4 and p-eIF2α in CHE-treated Caki cells. Western blot results indicated that CHE (12 μmol/L) could time dependently activate ER stress ( Figure 5A). We also found that CHE could dose-dependent increase the p-eIF2α and ATF4 protein expression in Caki cells ( Figure 5B). As shown in Figure 5C, pre-treatment with NAC could completely reverse the CHE-induced changes of ER stress-related proteins. To further test effects of CHE, we directly observed the morphology of ER in Caki cells through electronic microscopy and DMSO-treated Caki cells (×20000 amplification) showed the normal appearance of smooth ER (arrow). Treatment with CHE (12 μmol/L) for 8 hour made the ER swelling (arrow) in Caki cells, which suggested the misfolded protein was accumulated in ER ( Figure 5D). (E) STAT3 overexpressing cells and vector control transfected cells were exposed to CHE and the cell apoptosis was measured by Annexin-V/PI staining and flow cytometry. All images shown here are representative of three independent experiments with similar results. Data are shown as mean ± SEM (n = 3). **P < 0.01 and ***P < 0.001 compared with the dimethylsulfoxide group; ## P < 0.01 compared with the CHE-12 group However, pretreatment with NAC (5 mmol/L) could reverse this morphological alteration in Caki cells, while treatment with NAC (5 mmol/L) alone had no effect on the ER morphology ( Figure 5D).
To confirm the upstream role of ER-stress pathway during the renal cancer apoptosis, we assessed the effect of CHE after altering ATF4 levels in Caki cells. The ATF4 expression markedly reduced after knockdown of ATF4 by siRNA in Caki cells ( Figure 5E,F). We assumed that if ATF4 (ER stress) was involved in CHE-mediated Caki cell death, the cell apoptosis induced by CHE would be expected to decrease after knocking it down. As shown in Figure 5G, the data clearly demonstrated our hypothesis, ATF4 knockdown reduced apoptotic cell death by CHE.

| CHE inactivates STAT3 activity, which contributes to CHE lethality in human renal cancer cells
STAT3 plays a key role in cell proliferation through transcriptional activation of pro-survival genes. In addition, the STAT3 signalling inhibition may promote apoptosis in human cancers. Firstly, we tested the effect of CHE on the expression of cell proliferation markers transcriptionally regulated by STAT3. Our data showed that CHE diminished the constitutive phosphorylation at Y705 residues of STAT3 in Caki cells ( Figure 6A,B). Exposure of cells to CHE reduced the p-STAT3 levels but had no effect on total STAT3 levels.
All these effects changed in both a dose-and time-dependent manner ( Figure 6A,B). Next, we explored the upstream of STAT3 pathway. Since ER stress pathway was induced by the accumulation of ROS, we conjectured that the productive ROS could inactivate the STAT3 activity. As shown in Figure 6C, pre-treatment of NAC, the specific ROS inhibitor, for 1 hour significantly reversed the CHE-induced change in p-STAT3 level. Then, we confirmed the involvement of STAT3 in CHE-induced cytotoxic effects through overexpressing STAT3. As shown in Figure 6D, we transfected cells with STAT3 expressing plasmid to increase STAT3 and p-STAT3 levels in Caki cells.
Our results showed that overexpression of STAT3 reduced apoptosis caused by CHE in Caki cells ( Figure 6E). All these data demonstrated that the inhibitory activity of CHE in human renal cancer cells was, at least partly, mediated through the inactivation of STAT3.

| D ISCUSS I ON
CHE is considered an anti-cancer drug but its underlying mechanism has not been well defined yet. In this study, we provided sufficient evidence to demonstrate that CHE inhibited RCC cells growth through by causing cell cycle arrest apoptosis. We also discovered that CHE produced these beneficial inhibitory effects in RCC cells through ROS generation. CHE also induced activation of ER stress and suppression of STAT3 in RCC cells. Knockdown of ATF4 or overexpression of STAT3 both altered the CHE-induced apoptotic cells. CHE induced apoptosis via ROS-mediated ER stress and STAT3 pathways in human RCC cells. These salient findings are summarized in Figure 7. Collectively, all of these results suggest that CHE has the potential to be a promising candidate for RCC treatment.
In general, the proliferation of normal cells could be controlled by cell cycle progression, while many cancer cells often lack this regulation. 19 Our data showed that CHE inhibited the proliferation in RCC cells by causing cell cycle arrest in G2/M phase. The G2/M phase in cell cycle has potential therapeutic effects as various cancer cells always respond effectively to chemotherapy and/or radiation in this cell period. 20 In addition, the researchers have found that CHE can induce G1 phase arrest of human leukaemia and NSCLC cells in the F I G U R E 7 Schematic illustration of the underlying mechanism of CHE's anticancer activity recent study. 21,22 This contradiction may be due to the differences of cell types, so the effects of CHE in cell cycle progression should be further studied in more cancer cell lines.
Maintaining moderate levels of intracellular ROS under physiological conditions, for keeping redox balance as well as cell proliferation are extremely important. 23 ROS is a key regulator of ER stress which generally triggers unfolded protein reaction to protect cancer cells away from cell death. 24,25 We found that the p-eIF2a and ATF4 protein levels were increased in a dose-dependent man- A recent in vitro study on human NSCLC cells, A549 and SK-MES-1 cells showed that CHE inhibited cell proliferation when used in combination with erlotinib. 26 The combination treatment of CHE and erlotinib effectively blocked the EGFR signalling pathway through reducing phosphorylation in downstream targets such as STAT3, ERK1/2 and p38 MAPK. It is worth noting that this suppression of signalling was found in the combination treatment and it is not clear whether CHE along inhibits the pathway. STAT3 (a member of STAT transcription factors) mediates multi aspects of proliferation, immunity, apoptosis and differentiation. 27,28 Since CHE induced both STAT3 inhibition and ER stress activation, it would be interesting to study the relationship between the two effects. Furthermore, recent studies have found that the increasing ROS level is an important factor leading to apoptosis induced by protein misfolding and ER stress. It has been also reported that ROS inhibits the STAT3 signalling pathway. Thus, we speculate that ROS may be a common upstream mediator of ER stress and STAT3 signalling pathways. 29,30 The constitutive activation of STAT3 signalling pathway has high frequency detection in many human cancer cell lines or tumours, and including RCC. [31][32][33] In conclusion, we have identified the anti-tumour activity of CHE against RCC cells and its potential underlying mechanisms and found that CHE could induce ROS-dependent cell cycle arrest in G2/M phase and apoptosis. Furthermore, we demonstrate that CHE induced cells apoptosis mainly through activation of ER stress pathway and inhibition of STAT3 phosphorylation. This inhibitory activity is partly reversed by ATF4 knockdown or STAT3 overexpression.
Taken together, our findings not only show CHE is a promising candidate for RCC therapy but also indicate that targeting ER stress and STAT3 is a significant strategy for the development of novel anti-RCC drugs.

CO N FLI C T O F I NTE R E S T
All authors declare no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
HH and RZ performed the research. JD and XW designed the research study. XH contributed essential reagents or tools. HW analysed the data. DX wrote the manuscript. All authors have read and approved the final manuscript.