Mafb Protein Confers Primary Resistance of Myeloma to Proteasome Inhibitors

Multiple myeloma (MM) is a malignancy of terminally differentiated clonal plasma cells displaying significant molecular heterogeneity with 7 subgroups defined by gene expression profiling (GEP). Our previous work showed that both the MS and MF subgroups associated with inferior survival (Zhan et al, Blood 2006). Furthermore, clinical studies have demonstrated that the addition of the proteasome inhibitor (PI) bortezomib (Bzb) to high dose melphalan based regimens provided a major advantage to patients in MS subgroup while patients in the MF subgroups (including C-MAF and MAFb) did not benefit from Bzb (Nair, Blood 2010). We have previously demonstrated that Bzb prevents c-MAF protein degradation leading to primary drug resistance (Abstract # 281, ASH 2013). In the present study, we assessed the ability of MAFb, another MAF family member, to influence the innate resistance to proteasome inhibitors (PI) and identify the molecular mechanism underlying the resistance of proteasome inhibitors in high MAFb-expressing patients.

To investigate the association of the limited therapeutic effect of proteasome inhibitors in the different molecular subgroup of myeloma, we compared the IC50 of Bzb and carfilzomib (CFZ) in 29 myeloma cell lines (MMCL) belonging to different GEP-based molecular subgroups. IC50 concentrations of Bzb were higher (>25 nM) in all 4 MAFb MMCL and >60 nM in one MAFb MMCL, which expressed the highest level of MAFb protein, as determined by immunoblot analysis. In contrast, Bzb IC50 levels were lower (7.5-20 nM) for the MMCL belonging to the other molecular subgroups. For CFZ, IC50 concentrations were higher (>30nM) in all 11 c-MAF cell lines, while the IC50 levels were lower (2-20 nM) for the MMCL belonging to other molecular subgroup. One MMCL harboring t (14; 20) with high IC50 of Bzb (35 nM) showed low IC50 (10 nM) for CFZ. These results indicate that high MAFb expression in myeloma cells may contribute to primary resistance to Bzb and MM cells that express high MAFb protein although resistant to Bzb are sensitive to CFZ. Mechanistically, immunoblotting analysis demonstrated that exposure to Bzb resulted in increased MAFb protein levels in a dose-dependent manner, suggesting that Bzb prevents the degradation of MAFb protein in myeloma cells.

To further confirm that drug-induced stabilization of MAFb protein confers resistance to Bzb and partially to CFZ, we generated loss of functional MAFb cells by silencing MAFb expression in a t (14;20) positive myeloma cell line using lentiviral shRNA specific to MAFb mRNA (shMAFb). shMAFb infected myeloma cells had 85% lower levels of MAFb mRNA and protein level compared with the cells infected with scrambled shRNA. Additionally, significantly decreased ITGB (9.1 fold), E-cadherin (2.5-fold), CCND2 (5.5-fold), and CCR1 (25-fold) levels were observed in these cells, compared with the cells infected with control viral vector. Silencing MAFb expression significantly decreased proliferation of myeloma cells (65.1% decrease, p=2.1E-6). Moreover, Bzb treatment of the cells infected with shMAFb led to 50.2%, inhibition (P=2.9E-8) of proliferation compared with control cells. Similarly, CFZ treatment of cells with silenced MAFb resulted in 54.3% (P=3.5E-8) inhibition of proliferation compared with control cells. Taken together, our results indicate that high expression of MAFb protein, similar to c-MAF, confers primary resistance to Bzb as well as to CFZ. In addition Bzb induces stabilization of MAFb protein, further increasing resistance to Bzb. These data provide the molecular rational for adopting an alternative therapeutic strategy for high-MAFb expressing myeloma patients.