Induction of Apoptosis in Human Pancreatic Cancer Stem Cells by the Endoplasmic Reticulum-Targeted Alkylphospholipid Analog Edelfosine and Potentiation by Autophagy Inhibition

SIMPLE SUMMARY: Pancreatic cancer remains an incurable cancer with a gloomy prognosis and a median survival of months since the time of diagnosis. Disappointingly, no significant improvement in clinical outcomes has been achieved for the last fifty years. Cancer stem cells (CSCs) are suggested to be critical and responsible for tumor development, drug resistance and cancer relapse, therefore, an effective antitumor therapy against pancreatic cancer should deal with pancreatic CSCs. Here, we found that the alkylphospholipid analog edelfosine induces apoptosis in pancreatic CD44(+)CD24(+)EpCAM(+) CSCs through its accumulation in the endoplasmic reticulum, leading to sustained endoplasmic reticulum stress and cell death. Edelfosine was effective against primary cultures from pancreatic cancer patients and their corresponding pancreatic CSCs. Autophagy inhibition potentiated the proapoptotic action of edelfosine in pancreatic CSCs. Thus, endoplasmic reticulum targeting and its combination with autophagy inhibitors could open a new framework in pancreatic cancer chemotherapy, directly involving pancreatic CSCs. ABSTRACT: Pancreatic cancer is one of the most lethal malignancies with a poor and gloomy prognosis and the highest mortality-to-incidence ratio. Pancreatic cancer remains an incurable malignancy, and current therapies are ineffective. We isolated cancer stem cells (CSCs) from the human PANC-1 pancreatic cancer cell line as CD44(+)CD24(+)EpCAM(+) cells. These CSCs form pancreatic cancer spheres or spheroids and develop tumors in SCID mice after subcutaneous injection of as few as 100 cells per mouse. Here, we found that the alkylphospholipid analog edelfosine inhibited CSC pancreatic cancer spheroid formation and induced cell death, as assessed by an increase in the percentage of cells in the sub-G(0)/G(1) region by means of flow cytometry, indicative of DNA breakdown and apoptosis. This correlated with an increase in caspase-3 activity and PARP breakdown, as a major substrate of caspase-3, following PANC-1 CSC treatment with edelfosine. The antitumor ether lipid edelfosine colocalized with the endoplasmic reticulum in both PANC-1 cells as well as PANC-1 CSCs by using a fluorescent edelfosine analog, and induced an endoplasmic reticulum stress response in both PANC-1 cells and PANC-1 CSCs, with a potent CHOP/GADD153 upregulation. Edelfosine elicited a strong autophagy response in both PANC-1 cells and PANC-1 CSCs, and preincubation of CSCs with autophagy inhibitors, chloroquine or bafilomycin A1, enhanced edelfosine-induced apoptosis. Primary cultures from pancreatic cancer patients were sensitive to edelfosine, as well as their respective isolated CSCs. Nontumorigenic pancreatic human cell line HPNE and normal human fibroblasts were largely spared. These data suggest that pancreatic CSCs isolated from established cell lines and pancreatic cancer patients are sensitive to edelfosine through its accumulation in the endoplasmic reticulum and induction of endoplasmic reticulum stress.