FUNCTIONAL GENOMIC APPROACHES TO SENSITIZE HEMATOLOGICAL MALIGNANCIES TO PROTEASOME INHIBITORS

Proteasome inhibitors (PIs) are extensively used for the therapy of multiple myeloma (MM) and mantle-cell lymphoma (MCL). Nevertheless, patients continuously relapse or are intrinsically resistant to PIs.Here, to identify druggable targets that synergize with PIs, we carried out a functional screening in MM cell lines using a shRNA library targeting cancer driver genes. The Isocitrate Dehydrogenase 2 (IDH2) and Lysin Specific-Demethylases 1 (LSD1) genes were identified as top candidates, showing a synthetic lethal activity with the PI Carfilzomib (CFZ). We first validated the role of IDH2 in mediating PIs sensitivity. Combinations of the pharmacological IDH2 inhibitor AGI-6780 with FDA approved PIs significantly increased apoptotic cell death in ten MM cell lines, both sensitive and resistant to PIs. Combined treatments triggered synergistic cytotoxicity also in others hematological malignancies, such as Burkitt’s lymphoma, mantle cell lymphoma and diffuse large B-cell lymphoma. Importantly, CFZ/AGI-6780 treatment increased death of primary MM cells from nine patients and exhibited a favorable cytotoxicity profile towards normal human cells. Mechanistically, CFZ/AGI- 6780 combination decreased TCA cycle activity and ATP levels, due to enhanced IDH2 enzymatic inhibition. Specifically, CFZ treatment decreased nicotinamide phosphoribosyltransferase (NAMPT) expression, a rate-limiting enzyme required for IDH2 activation, through the NAD+-dependent deacetylase SIRT3. Consistently, combination of CFZ with either NAMPT or SIRT3 inhibitors impaired IDH2 activity and increased MM cells death, thus phenocopying CFZ/AGI-6780 effects and putting the proteasome in a direct link with IDH2 inhibition. Moreover, inducible IDH2 knock-down enhanced the therapeutic efficacy of CFZ in xenograft mouse models of MM, resulting in inhibition of tumor progression and extended survival. Finally, preliminary results also showed that LSD1 may represent a potential therapeutic target to combine with PIs. Indeed, genetic inhibition of LSD1 increased sensitivity to CFZ in MM cell lines, both sensitive and resistant to PIs. Remarkably, treatment with the non-competitive LSD1 inhibitor SP2509 significantly increased CFZ efficacy in eighth out of ten MM cell lines, both sensitive and resistant to PIs. However, GSK2879552 and GSK-LSD1, two LSD1 enzymatic inhibitors, did not display synergistic activity with CFZ, thus establishing the basis for future research of non-canonical functions of LSD1 or alternative SP2509 targets. In conclusion, our data demonstrate that IDH2 inhibition increases the therapeutic efficacy of PIs, thus providing compelling evidence for treatments with lower and less toxic doses and broadening the application of PIs to other malignancies. Moreover, preliminary results suggest LSD1 targeting as an alternative promising strategy to enhance PIs sensitivity in multiple myeloma.