Abstract
Proteasome inhibition with bortezomib has revolutionized the treatment of multiple myeloma, but the vast majority of patients eventually develop clinical bortezomib resistance through poorly understood mechanisms. One of the most conserved cellular responses to proteasome inhibition is to up-regulate proteasome subunit expression, presumably with the goal of enhancing proteasome activity and restoring intracellular protein homeostasis. We therefore hypothesized that proteasome inhibitor resistance could be associated with enhanced proteasome assembly, and that suppression of this assembly process could help restore drug sensitivity. The current studies focused on POMP, which is involved in addition of catalytically active b subunits to the hemiproteasome ring initially formed by structural a subunits.
We studied ANBL-6, KAS-6/1, OPM-2, and RPMI 8226 multiple myeloma cell lines which had acquired bortezomib resistance through prolonged exposure to increasing drug concentrations, and compared them to their drug-naïve, vehicle-treated counterparts. In addition, we evaluated primary cells derived from patients with myeloma, and examined an in vivo murine xenograft model of human myeloma.
Bortezomib-resistant (V10R) ANBL-6, KAS-6/1, OPM-2, and RPMI 8226 cell lines showed enhanced levels of POMP mRNA by quantitative (q) PCR compared to their drug-sensitive counterparts, which was associated with higher POMP protein levels seen by immunoblotting. Exogenous over-expression of POMP in drug-naïve OPM-2 and KAS-6/1 cells was sufficient by itself to induce resistance to both bortezomib and carfilzomib. Conversely, suppression of POMP with one of two different Lentiviral small hairpin (sh) RNAs restored sensitivity in OPM-2 and KAS-6/1 V10R cells to bortezomib and carfilzomib. Since no known pharmaceuticals directly target POMP, we examined its promoter region, and found a consensus binding site for nuclear factor (erythroid-derived 2)-like (NRF) 2. Consistent with a role of NRF2 in POMP expression, NRF2 mRNA and protein were increased in V10R myeloma cells, and in drug-naïve cells treated with bortezomib. Moreover, transfection of cells with NRF2 cDNA activated a POMP promoter reporter, while chromatin immunoprecipitation with anti-NRF2 antibodies preferentially precipitated sequences near the POMP promoter. Also, NRF2 over-expression induced POMP and enhanced proteasome chymotrypsin-like activity, while its suppression had the opposite effects. All trans-retinoic acid (ATRA) blocked nuclear accumulation of NRF2 in OPM-2 and KAS-6/1 V10R cells, and reduced expression of POMP. Combinations of bortezomib with ATRA showed enhanced activity against these drug-resistant cell lines in association with greater proteasome inhibition, and were synergistic in drug-naïve cells. In primary samples, ATRA with bortezomib induced a greater reduction in viability than did either treatment alone. Finally, in a murine xenograft model with OPM-2 V10R cells, neither ATRA nor bortezomib showed substantial activity, while the combination regimen, by comparison, retained efficacy.
Taken together, our data support the hypotheses that NRF-2-influenced POMP over-expression contributes to proteasome inhibitor resistance in multiple myeloma, while approaches targeting POMP hold promise in overcoming resistance. Moreover, they provide a framework for translation of proteasome inhibitors with ATRA to the clinic to enhance activity, and to overcome resistance to this important class of anti-myeloma agents.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
