Salinomycin triggers endoplasmic reticulum stress through ATP2A3 upregulation in PC-3 cells

Human prostate cancer PC-3 and DU145 cells (ATCC, Manassas, VA, USA) were cultured as previously described [5]. Salinomycin (Sigma-Aldrich, St Louis, MO, USA), BAPTA-AM (Selleckchem, Houston, TX, USA) were dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich, St Louis, MO, USA). Sodium phenylbutyrate (4-PBA) was dissolved in water.

For tumorigenic studies, PC-3 cells were subcutaneously inoculated into the flanks of NOD/SCID male mice (5 weeks of age; Beijing HFK BioScience Co., Ltd. Beijing China). Mice were housed in a standard laboratory environment (temperature: 24 ± 2 °C; humidity: 50 ± 5%; 12 h day-night cycle) and treated intraperitoneally (i.p.) daily with either DMSO or salinomycin at a dose of 10 mg/kg/day/200 μL (each group was 5). After 3 weeks, the mice were euthanized by carbon dioxide inhalation followed by cervical dislocation. The xenografts were excised and pulverized in liquid nitrogen. Animal studies have approved by the animal ethics committee from South China university.

Cultured PC-3 cells were treated with 1.0 μM salinomycin or DMSO control for 24 h. Then, total RNA from the abovementioned cells or tumors was extracted with TRIzol reagent (Life Technologies, Inc., Carlsbad, CA, USA) according to the manufacturer’s protocol. Double-stranded cDNA (ds-cDNA) was synthesized from 5 μg of total RNA using a SuperScript ds-cDNA synthesis kit (Life Technologies, Inc., Carlsbad, CA, USA). Human 12 × 135 K Gene Expression Arrays (Roche NimbleGen) were hybridized at 42 °C for 16 to 20 h with 4 μg of Cy3 labeled ds-cDNA in NimbleGen hybridization buffer/hybridization component in a hybridization chamber (Hybridization System-NimbleGen Systems, Inc., Madison, WI, USA).

Slides were scanned at 5 μm/pixel resolution using an Axon GenePix 4000B scanner (Molecular Devices Corporation) piloted by GenePix Pro 6.0 software (Axon). Scanned images (TIFF format) were then imported into NimbleScan software (version 2.5) for grid alignment and expression data analysis. The expression data were normalized through quantile normalization and the Robust Multichip Average (RMA) algorithm included in the NimbleScan software. All gene level files were imported into Agilent GeneSpring GX software (version 11.5.1) for further analysis. Differentially expressed genes were identified through fold Change filtering.

Quantitative RT-PCR (reverse transcription-polymerase chain reaction) was performed with FastStart Essential DNA Green Master (Roche, Mannheim, Germany). All reactions were performed on the Roche LightCycler 96 Real-Time PCR system (Roche Diagnostics GmbH, Mannheim, Germany). Individual values were normalized to the GAPDH loading control. Sequences of gene primers are listed in Table 1. The mRNA expression levels were analyzed using delta cycle threshold (ΔCt) values.

Human prostate cancer tissues and para-carcinoma tissues were obtained from the Second Affiliated Hospital, University of South China with institutional review board approval. Some sections were stained with HE. Immunohistochemistry staining was performed as described previously [24, 25] using ATP2A3 (1:500, GeneTex, San Antonio, TX, USA), BIP and ATF4 antibodies (all from CST, Danvers, MA, USA). Images were captured by a CCD camera (Olympus, Center Valley, PA).

Protein extraction and Western blotting were performed as previously described [26]. The primary antibodies BIP, PERK, ATF4, eIF2a, CHOP, Caspase 12, p-PERK, p-eIF2a, p-CaMK-II (all from CST, Danvers, MA, USA), CaMK-II (Bioworld Technology, Inc., Minneapolis, MN, USA) and ATP2A3 (GeneTex, San Antonio, TX, USA) were used to detect the ER stress. PARP, cleaved PARP antibodies (all from CST, Danvers, MA, USA) were used to detect apoptosis. The β-actin antibody (CST, Danvers, MA, USA) was used as an internal control.

For electron microscopy, PC-3 cells were plated at a density of 10⁶ cells in 100 mm cell culture dishes. The cells were treated with salinomycin for 24 h. Subsequently, the cells were harvested, washed in PBS, and fixed in 2.5% glutaraldehyde/0.2 M phosphate buffer solution (pH 7.4) at 4 °C. Finally, cells were detected by transmission electron microscopy (Servicebio Co., Ltd. Wuhan, China).

Immunofluorescence staining was performed after salinomycin treatment, as previously described [27]. BIP (1:200, Abcam, Cambridge, MA, USA) and CHOP (1:200, CST, Danvers, MA, USA) antibodies were used in this experiment. Secondary antibodies conjugated with Alexa Fluor-647 (Abcam, Cambridge, MA, USA) and Alexa Fluor-488 (Abcam, Cambridge, MA, USA) were used. The fluorophore-labeled cells were examined and analyzed by laser scanning confocal microscopy (Olympus, Tokyo, Japan).

Intracellular Ca²⁺ release was detected using the fluorescent probe Fluo-3/AM (Beyotime Biotechnology Co., Haimen, China). For salinomycin treatment, PC-3 and DU145 cells were incubated with Fluo-3/AM stock solution (0.5–5 μM) at 37 °C in the dark for 10 min following three washes with PBS. Samples were analyzed by a FACS-Calibur (BD Biosciences, San Jose, CA, USA), and fluo-3 was detected using a 530/30 nm filter. The arithmetic mean of fluo-3 fluorescence intensity was expressed as [Ca²⁺]_i. Images of these cells were also captured using an inverted fluorescence microscope (Carl Zeiss, Jena, Germany) to observe the green fluorescence. The results were averaged from three independent experiments.

Small interfering RNA (siRNA) was used to knockdown ATP2A3 expression in PC-3 cells. ATP2A3-siRNAs and negative control sequences were synthesized by RiboBio Co., Ltd. (Guangzhou, China). The siRNAs were transfected by Lipofectamin 2000 (Invitrogen, Life Technologies, Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions.

PC-3 cells were seeded in six-well plates and then transfected with pCDH cDNA-ATP2A3 or empty vector for 24–48 h. For cell cycle analysis, cells were fixed and stained as previously described [26]. For apoptosis analysis, apoptosis and necrosis were evaluated by annexin V-FITC/PI staining as previously described [26]. The samples were analyzed by a FACS-Calibur (BD Biosciences, San Jose, CA, USA).

Previously, we found that salinomycin inhibited PC-3 cell proliferation and decreased xenograft tumor size [5]. To investigate the mechanism of salinomycin, a microarray analysis was used to identify differentially expressed genes (DEGs) in vitro (PC-3 cells) and in vivo (NOD/SCID mice xenograft model generated from implanted PC-3 cells) (Fig. 1a). Based on the threshold of P-value ≤0.001 and fold change ≥2.0, Venn diagrams showed that 150 DEGs were obtained (Fig. 1b). All target values were log-2 base transformed, and then a heatmap was generated for the 150 DEGs (Fig. 1c). The heatmap results showed distinguishable gene expression profiling between samples.

For the identified DEGs, we performed enrichment analysis in GO categories. The results showed that secondary metabolic process (GO:0019748, false discovery rate (FDR) =1E0) and plasma membrane organization (GO:0007009, FDR =1E0) were significantly enriched in the biological processes (Fig. 1d), and G-protein coupled receptor activity (GO:0004930, FDR =1.77E-1) was significantly enriched in molecular functions (Fig. 1e).

The P-value ≤0.001 and fold change ≥3.0 thresholds were set to select genes from the 150 DEGs. This search identified 10 genes that fit these criteria (Table. 2). Then, the 10 identified genes were validated by quantitative RT-PCR. Seven gene expression profiles were similar to those revealed by the microarray data. A significantly upregulated gene, ATP2A3, which is known to be involved in Ca²⁺ transport [20], was successfully identified (Fig. 1f). Thus, we hypothesized that ATP2A3 may be a potential target for salinomycin in PC-3 cells.

We validated the microarray and quantitative PCR results by Western blotting in prostate cancer cells. PC-3 or DU145 cells were treated with 0 (vehicle, DMSO), 0.5, 1.0 and 2.0 μM salinomycin for 24 h, and the expression of ATP2A3 was measured by WB. The results showed that ATP2A3 was upregulated by salinomycin treatment, and higher concentrations resulted in higher expression levels. To investigate the time-course of salinomycin-induced ATP2A3 expression, PC-3 or DU145 cells were also treated with 1.0 μM salinomycin for 0, 6, 12 or 24 h. The WB showed that ATP2A3 was progressively induced by salinomycin (Fig. 2a). The data therefore showed that salinomycin upregulates ATP2A3 expression in dose- and time-dependent manners.

To further confirm the role of ATP2A3, we overexpressed ATP2A3 in prostate cancer cells. The data showed that ATP2A3 overexpression arrested the progression of the cell cycle (Fig. 2b) and decreased the protein levels of cyclin D1 and cyclin E1 (Fig. 2c). Furthermore, ATP2A3 overexpression induced apoptosis (Fig. 2d), and triggered PARP cleavage (Fig. 2e). These results suggest that ATP2A3 overexpression can cause cell cycle arrest and induce apoptosis in prostate cancer cells.

It is known that ATP2A3 controls Ca²⁺ transport and that its aberrant expression can cause ER stress [28, 29]. Thus we analyzed the effect of ATP2A3 overexpression on ER stress in prostate cancer. As expected, ATP2A3 overexpression caused the upregulation of ER stress biomarkers (BIP, CHOP, Caspase 12 and ATF4). Furthermore, the phosphorylation levels of PERK and eIF2a also increased significantly (Fig. 2f). These results suggest that ATP2A3 overexpression can trigger ER stress.

Immunohistochemistry staining was used to evaluate the expression of ATP2A3 and ER stress biomarkers (BIP and ATF4) in human prostate cancer and para-carcinoma tissues. The data showed weak ATP2A3 staining, but the ER stress biomarkers were strongly stained in cancer tissues (Fig. 2g). In conclusion, our results suggest that salinomycin can upregulate ATP2A3 expression, which triggers ER stress to exert anti-cancer effects in prostate cancer cells.

We analyzed the effect of salinomycin on ER stress in prostate cancer PC-3 cells. As expected, salinomycin upregulated the expression of BIP, ATF4, CHOP and caspase-12. The phosphorylation levels of PERK and eIF2a also increased significantly (Fig. 3a and b). Furthermore, we measured the protein levels of BIP and CHOP by immunofluorescence staining. Consistent with the Western blotting data, the BIP and CHOP antibody staining signals were stronger after salinomycin treatment than before treatment (Fig. 3c).

To further confirm that salinomycin-triggers ER stress, we transfected PC-3 cells with the pDsRed2-ER plasmid, which was designed for fluorescent labeling of the endoplasmic reticulum and then treated cells with salinomycin. The results showed that the fluorescent labeling of the endoplasmic reticulum became dim (Fig. 3d). Next, we observed the endoplasmic reticulum by transmission electron microscopy. The results showed that the ER membrane of salinomycin-treated PC-3 cells was dilated (Fig. 3e). Our data suggest that salinomycin can trigger ER stress in PC-3 cells.

Because ATP2A3 re-sequesters cytoplasmic Ca²⁺ to the sarcoplasmatic/ endoplasmatic reticulum store, we determined whether salinomycin influences Ca²⁺ release. PC-3 or DU145 cells were treated with the indicated concentration of salinomycin for 24 h, stained with the calcium indicator Fluo-3 AM and analyzed with flow cytometry. As illustrated in Fig. 4a and b, salinomycin significantly inhibited Ca²⁺ release in prostate cancer cells. Next, we transfected PC-3 cells with siRNA to inhibit ATP2A3 expression. Of the three synthetic siRNAs, ATP2A3#2 and ATP2A3#3 greatly knocked down ATP2A3 expression (Fig. 4c). PC-3 cells transfected with siRNA ATP2A3#2 or ATP2A3#3 for 48 h were further treated with 1.0 μM salinomycin for 24 h; afterward, WB was performed to detect ER stress markers (Fig. 4d), or staining with Fluo-3 AM was performed to detect Ca²⁺ concentration changes (Fig. 4e). The results showed that ATP2A3 silencing weakened salinomycin-triggered ER stress, as salinomycin induced upregulation of BIP and CHOP was constrained in cells transfected with siRNA ATP2A3#2 and ATP2A3#3. Similarly, reduced salinomycin-inhibited Ca²⁺ release was also observed. Our results suggest that salinomycin inhibition of Ca²⁺ release might be related to ATP2A3 in PC-3 cells.

The calcium/calmodulin kinases (CaMKs) are important mediators of intracellular Ca^{2 +} [30]. Thus, we measured the CaMK-II protein expression level and phosphorylation level (Thr 286) to indirectly ascertain CaMK activity changes [31, 32]. The results showed that salinomycin treatment reduced the protein level of CaMK-II, but the phosphorylation level did not change significantly (Fig. 4f). These data suggest that salinomycin-mediated inhibition of Ca²⁺ release might be related to CaMKs via ubiquitin mediated protein degradation [33], which will be one focus of our future studies.

It has been reported that alterations in Ca²⁺ homeostasis are implicated in ER stress. Thus, we pretreated PC-3 cells with BAPTA-AM (10 μM, an intracellular calcium chelator) to assess whether salinomycin triggers ER stress by releasing Ca²⁺. As expected, the data showed that BAPTA-AM could weaken salinomycin-triggered ER stress (Fig. 4g). Our results suggest that salinomycin-triggered ER stress might be related to ATP2A3-mediated Ca²⁺ release in PC-3 cells.

Since ER stress has been shown to play a crucial role in apoptosis [34], we co-treated PC-3 cells with salinomycin and 4-phenylbutyrate (4-PBA, 1 mM, an ER stress antagonist) for 24 h. WB was performed to detect the expression of BIP, CHOP, cleaved caspase3 and cleaved PARP. The data showed that 4-PBA significantly reduced salinomycin-triggered ER stress (Fig. 5a) and hampered caspase3 and PARP cleavage (Fig. 5b). The treated cells were also stained with Annexin V-FITC to detect apoptosis. Flow cytometry showed that 4-PBA also significantly decreased salinomycin-induced apoptosis (Fig. 5c). These results suggest that ER stress can promote salinomycin-triggered apoptosis in PC-3 cells.

We also transfected PC-3 cells with siRNA to inhibit ATP2A3 expression. Annexin V-FITC staining illustrated that ATP2A3 silencing could inhibit salinomycin-induced apoptosis (Fig. 5d). In conclusion, our data suggest that ER stress promotes salinomycin-induced apoptosis, which might be related to ATP2A3 in PC-3 cells.