P692: the keap1-nrf2 redox pathway is of central importance during 5-azacytidine therapy and the development of resistance via a specific set of protein oxidative modifications

Background: Hypomethylation therapy with 5-azacytidine (AZA) represents the treatment of choice for patients with high-risk myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) with dysplasia unsuitable for transplantation and also represents consolidation in AML with oral AZA. AZA inhibits tumor cell proliferation and induces cell death through inhibition of DNA methylation. While AZA often leads to a hematologic response under well-tolerated therapy, the main concern is the emergence of resistance leading to loss of response. In this work, we reveal the relationship between AZA resistance and cellular oxidation/reduction (redox) homeostasis. We previously noted that AZA regulates cell death/survival pathways through modulation of redox homeostasis and oxidative modification of proteins and activates key factors of the redox signaling pathway involving Kelch like associated protein 1 (KEAP1) and nuclear factor erythroid 2-related factor 2 (NRF2). Aims: Our aim was to identify the AZA-specific oxidized protein cysteines that would be related to AZA resistance and to investigate how specifically the KEAP1-NRF2 pathway is involved in this process. Methods: To identify AZA-associated key oxidative modifications of protein cysteine residues we used a mass spectrometry-based quantitative proteomic approach and corroborated these data with functional and transcriptomic analyses. The study was performed on MDS/AML clones of the OCI-M2 cell line that were prepared using AZA selection (AZA-R). Total cellular redox state and glutathione (GSH) levels were measured by flow cytometry. Results: AZA resistance is marked by a significantly higher oxidation state and higher GSH compared to the parental state. Thus, AZA resistant cells are less sensitive to oxidizing agents such as diamide. While in parental cells AZA has the ability to rapidly increase the oxidative state of the cells, this does not occur in AZA-R. Upon GSH depletion, we achieve a significant increase in sensitivity to AZA. We next performed a proteomic analysis to identify specific protein targets of AZA mediated oxidation and found a redox change in 15% of proteins (403 out of 2’643), many of them involved in biological processes such as AZA metabolism and cell survival/death pathways. Among proteins with the significantly altered redox state was the Sequestosome-1 (SQSTM1)-KEAP1-NRF2 antioxidant pathway. NRF2 is a transcription factor that responds to redox stress by inducing transcription of antioxidant enzymes. However, in contrast to the parental cells AZA does not stimulate the NRF2 downstream target program in AZA-R cells. Therefore, we inhibited KEAP1, which acts as a ubiquitin ligase for NRF2, and this treatment led to a change in the redox balance in AZA-R cells and restoration of sensitivity to AZA. To validate the identified mechanism under in vivo conditions, we transplanted AZA-R cells into immunodeficient CDX mice and used different KEAP1 inhibitors to demonstrate prolonged event-free survival and overall survival of MDS/AML mice after AZA treatment. Next, we studied the KEAP1-NRF2 pathway and, in particular, the redox modifications identified in model systems in myeloblasts from MDS/AML patients. Despite heterogeneous results, we found encouraging agreements with in vitro systems that are subject to validation. Summary/Conclusion: We revealed that the mechanism of AZA resistance involves redox reset and modulation of the KEAP1-NRF2 cellular antioxidant response pathway. Complete sensitization of AZA-resistant cells represents a previously unconsidered and tempting way to enhance the efficacy of AZA therapy. This opens the way for further studies to modulate the KEAP1-NRF2 pathway to block AZA resistance in patients.Keywords: Proteomics, Azacitidine, Myelodysplastic syndrome, Drug resistance