Search for a command to run...
Ovarian cancer is a rare cancer but with a poor prognosis. Although patients respond favorably to resection surgery and chemotherapy (platinum salts and taxanes), chemoresistance emerges, leading to patient death. It is therefore essential to define predictive tools for treatment response and to find alternatives to chemotherapy. During carcinogenesis, the calcium signaling of malignant cells is considerably reshaped to allow them to proliferate at the expense of normal cells. Our work led us to focus on the SERCA2 calcium pumps (Sarcoplasmic/Endoplasmic Ca2+ ATPases), which are the only pumps to bring calcium into the endoplasmic reticulum (ER); they thus play a crucial role in cellular homeostasis. We showed that inhibiting these pumps (via RNA interference or the use of thapsigargin) unbalanced the [pro-]/[anti-apoptotic] ratio in chemoresistant cell lines. Furthermore, our results revealed that ABT-737, a BH3-mimetic targeting the anti-apoptotic BCL-xL, potentiated this effect, leading to an over-induction of the pro-apoptotic NOXA. The mechanistic analysis highlighted the crucial role of ER stress and the ATF4/NOXA axis. Interestingly, these results were also obtained by combining ABT−737 with ONC201 (an imipridone class molecule, an ATF4 inducer, and used in phase III clinical trials), not only on cell lines but also on patient tumoroids, supporting the use of this combination in the clinic.The study of SERCA2 expression in different cell lines surprisingly revealed that SERCA2 was less expressed in chemoresistant cells than in their chemosensitive parental lines. This result could be explained by the fact that the decrease in SERCA2 expression leads to a reduction of calcium stores in the ER, which would impede the cell's ability to create a calcium signal of sufficient amplitude to overload the mitochondria and lead to apoptosis. We confirmed this result by immunohistochemistry on a panel of tumors and tumoroids from chemotherapy-naive patients whose clinical response we knew. SERCA2 could therefore serve as a predictive biomarker for chemotherapy response. Subsequently, microspectrofluorimetry analyses seem to show that chemoresistant cells store less calcium in their ER than their chemosensitive counterparts. Finally, the overexpression of SERCA2 resensitizes chemoresistant cells to cisplatin. We also highlighted that SERCA2 might be repressed in chemoresistant cells post-transcriptionally via miR−25−3p.To understand the mechanisms involved in the acquisition of chemoresistance and on SERCA2 expression, we aimed to create chemoresistant tumoroid lines from chemosensitive tumoroids that underwent repeated carboplatin treatments, thus mimicking the courses received by patients. We successfully generated 3 resistant organoid lines, showed that the [anti-]/[pro-apoptotic] ratio was disrupted, and that SERCA2 expression was indeed repressed in the lines rendered chemoresistant after the first treatment cycles. In conclusion, placing our results in a clinical context: 1) SERCA2 could serve as a predictive biomarker for chemotherapy response, 2) the induction of its activity and/or its expression would sensitize resistant patients to chemotherapy, and 3) in case of relapse, the combination of an ER stress inducer (SERCA2 inhibitor or ONC201) with ABT-737 would constitute a relevant second line of treatment.