Targeting nicotinamide salvage pathway is a unique metabolic vulnerability of high-risk MDS stem cells

Abstract Background: High-risk myelodysplastic syndrome (MDS) is an incurable age-related clonal disorder of the hematopoietic stem cell (HSPC). The current standard of care for MDS patients is hypomethylating agents, like azacitidine; however, only ~30% of patients respond, usually with remission of less than two years. Thus, to improve MDS patient outcomes, an unmet clinical need, it is essential to identify better therapeutic targets by exploring vulnerabilities of MDS HSPCs, the source of the disease. To this end, we have previously shown that inhibiting protein synthesis is a vulnerability for MDS. So, we further pursued studying the unique proteome of MDS to identify targetable vulnerabilities. Methods: We used CD34+ HSPCs isolated from 12 high-risk MDS patients and 15 age-matched normal bone marrow (BM) for most of our studies. For cell-intensive experiments, we used CD34+ MDS-L cell line derived from a high-risk MDS patient. Results: To study the unique proteome of MDS compared to healthy, we performed global proteomics on MDS and normal HSPCs. Protein enrichment analysis identified a significant upregulation of metabolic proteins. Indeed, Seahorse analysis of MDS and normal HSPCs validated significantly increased maximal respiration and spare respiratory capacity in MDS compared to normal HSPCs. Most of the proteins enriched in MDS, such as G6PD, MDH1, ME2, IDH2, NADK2 and SIRT2, either maintain NAD(H) redox balance or use NAD as a cofactor for their function. To establish the role of NAD metabolism in MDS, we inhibited NAMPT, the rate-limiting enzyme in the nicotinamide salvage pathway of NAD anabolism, using small molecule inhibitors (KPT-9274 and OT82) and genetic approaches. NAMPT inhibition significantly decreased NAD levels in both normal and MDS HSPCs. Subsequently, to determine the importance of NAD in MDS, we performed global proteomic analysis on MDS-Ls post 24h of OT82 treatment. Interestingly, the most significantly downregulated proteins were related to the ribosomal machinery (RRM2), cytoplasmic and mitochondrial translation (EIF4H, MRPL15). Indeed, we found that translation, as measured by puromycin incorporation post 24h of OT82 treatment, was significantly reduced in MDS HSPCs indicating a differential reliance of MDS HSPCs on NAD for protein synthesis. Furthermore, we found that NAMPT inhibition positively enriched metabolic proteins associated with amino acid metabolism, lipid metabolism, and the citric acid cycle. Thus, to further determine how NAMPT inhibition, and its effect on translation ultimately impact metabolism in MDS, we carried out global metabolic analysis and heavy carbon labeled glucose and glutamine tracing in MDS-Ls post 24h of OT82 treatment. We found that NAMPT inhibition, reduced overall carbon flux, suggesting metabolic dysregulation, while increasing carbon flux into the aspartate-malate shuttle and the pentose phosphate pathway, both of which maintain NAD(P)H balance independently of nicotinamide metabolism. Consistently, 24h of OT82 treatment significantly reduced maximal respiration and spare respiratory capacity in MDS HSPCs compared to normal. These data indicate that NAD anabolism via the nicotinamide salvage pathway is an essential regulator of metabolic activity, exclusive to MDS HSPCs. Finally, we addressed whether NAMPT inhibition represents a therapeutic target in MDS by performing functional assays using patient samples. NAMPT inhibition significantly impaired self-renewal and colony forming potential of MDS HSPCs compared to normal. Additionally, we observed increased cell death of MDS HSPCs compared to normal post 48h of OT82 treatment. Moreover, NAMPT-induced apoptosis in MDS was revealed to be synergistic with azacitidine, the current standard of care. Importantly, to establish pre-clinical relevance of our findings, we performed xenograft studies by transplanting MDS patient BM cells and MDS-Ls into NSG-S mice. Strikingly, after 2 weeks of OT82 treatment, we observed a significant reduction in disease burden, indicating NAMPT as a tractable target for therapeutic inhibition in high-risk MDS patients. Conclusion: Our data suggest that NAMPT is uniquely required for the function and survival of MDS HSPCs compared to normal and thus can be exploited as a promising therapeutic vulnerability. The results of this pre-clinical study provide strong support for initiating NAMPT inhibitor phase I/II clinical trials to improve outcomes for high-risk MDS patients.