Raloxifene Compromises Mitochondria, Induces ROS Stress and the Unfolded Protein Response, and Synergizes Gemcitabine's Cytotoxicity in Pancreatic Cancer Cells

Pancreatic adenocarcinoma (PCa) is a deadly disease associated with high mortality and a very poor 5‐year survival rate of only 6–8%. There is a lack of knowledge about how mutations in the human homolog of the Kirsten Rat Sarcoma viral oncogene (KRAS), which occur in 90% of the patient population, lead to its development. This knowledge gap, coupled with substantial chemoresistance, has led to a scarcity in effective therapy to combat PCa. To reduce the bench‐to‐bedside time for the development of drugs for use in PCa, we repurposed FDA‐approved drugs, screening for their cytotoxic potential in KRAS ‐mutant human PCa cell cultures. Our data show that the FDA‐approved raloxifene, a selective estrogen receptor modulator used in the treatment of osteoporosis caused apoptosis of multiple PCa cell lines (MIA PaCa‐2, PANC‐1, BxPC‐3). IC50 for raloxifene was 1.81 ± 0.38 μM (MIA PaCa‐2), 58.16 ± 5.37 μM (PANC‐1), and 4.20 ± 2.16 μM (BxPC‐3). This compared well with the IC50 for gemcitabine, the gold‐standard drug used for treating PCa since 1997, values for which were 4.75 ± 1.67 μM (MIA PaCa‐2), 58.60 ± 14.78 μM (PANC‐1), and 6.10 ± 1.70 μM (BxPC‐3) because in each PCa cell line, raloxifene had a lower IC50 than gemcitabine. Further, combination index (CI) analysis using a 1:1 raloxifene‐gemcitabine combination showed evidence for synergism with CI values <1 for a broad range of dose concentrations within the three cell lines, and specifically at 50% of cells affected. To investigate the mechanism of cytotoxicity, we examined mitochondrial function and found that in MIA PaCa‐2 cells, raloxifene treatment induces reactive oxidative species (ROS) production and decreases mitochondrial membrane potential. Pre‐treatment of MIA PaCa‐2 cells with the NADPH oxidase inhibitor diphenylene iodonium (DPI) reduced ROS production for each raloxifene concentration used, suggesting that the mitochondria were the source of the ROS. To further investigate the mechanism, we examined the effect of raloxifene treatment on the enzymes involved in the ROS pathway. Our data show that while raloxifene induced the expression of glutathione synthetase, glutathione s‐transferase, it had no effect on superoxide dismutase 1 or 2 or glutathione peroxiredoxin. However, catalase expression was decreased in MIA PaCa‐2 cells, suggesting that raloxifene induced oxidative stress in these cells, making them susceptible and responsive to further apoptotic stimuli like treatment with gemcitabine. Finally, our data also show that raloxifene induces the unfolded protein response and endoplasmic reticulum stress in PCa cells as evident by an increase in endoplasmic reticulum oxidoreductase 1 (ERO1‐α) and cleavage of Activating transcription factor 6 (ATF6), confirming that it helps overcome the chemoresistance of PCa cells. Support or Funding Information Project was supported in part by the Elsa U. Pardee grant and the School of Pharmacy Research Incentive Grant to AM. This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .