Résumé
The unfolded protein response (UPR) is a key stress resistance pathway that has become a key potential target for improving the efficacy of cancer chemotherapy. The UPR involves the activation of three ER-resident stress sensors: PERK, IRE-1 and ATF6 with different signalling outcomes leading to cell death or survival. These cell-fate decisions are difficult to predict and are the result of the complex interaction of PERK, IRE-1 and ATF6 downstream events that have differences in their dynamics and their interplay. These characteristics of the UPR are still poorly defined due to lack of methods to monitor their activation simultaneously at single-cell level. We developed SNUPR (Single Nuclei analysis of the Unfolded Protein Response), an accessible technique that allows the profiling of the three UPR branches in nuclear suspensions by flow cytometry, and applied it to study UPR dynamics in a cancer-specific context. By performing transcriptomic analysis, we found that ER-stress sensor specific gene signatures correlate with patient survival in several blood malignancies, and by using SNUPR, we detected high heterogeneity during UPR activation in vitro in different human cancer cell lines, which could not be have been predicted by the level of expression of the sensors. Our SNUPR analyses further indicate that this heterogeneity is explained by variations in the intensity and duration of ER stress-induced protein synthesis inhibition via PERK, acting as upstream regulator of both the IRE-1/XBP1 and ATF6 dependent transcriptional programs. We extend the relevance of these observations by demonstrating that IRE-1/XBP1s pathway plays a critical role in bortezomib resistance of multiple myeloma cells and patients. We present here SNUPR, that can be used to monitor UPR dynamics with single-cell resolution and identified clinical contexts in which targeting a specific UPR branch could be detrimental or help circumventing chemotherapy resistance. One Sentence Summary SNUPR method enable single-cell UPR profiling and reveals the role of IRE-1 axis in predicting bortezomib resistance in multiple myeloma. Highlights SNUPR allows simultaneous profiling of PERK, IRE-1 and ATF6 activation with single- cell resolution. Inhibition of protein synthesis via PERK control the activation levels of the IRE-1/XBP1s and ATF6 pathway. IRE-1 activation and associated transcriptional signatures predict the outcome of patients with multiple myeloma treated with Bortezomib. IRE-1 activity, but not PERK or ATF6, is essential to acquire bortezomib resistance in multiple myeloma cell lines. Graphical abstract