Requirement of Caspase-8 Versus Caspase-9 during Apoptosis in Multiple Myeloma Cells Induced by Bortezomib- or a Novel Proteasome Inhibitor NPI-0052.

Proteasome inhibition is an effective therapy for the treatment of relapsed/refractory multiple myeloma (MM); however, the sequence of events leading to apoptosis following proteasome inhibition is unclear. Here we defined the requirement for caspase-8 versus caspase-9 during Bortezomib (Velcade^TM)- or novel proteasome inhibitor NPI-0052-induced apoptosis in MM cells. Incubation of MM.1S cells with pan-caspase inhibitor (Z-VAD-FMK) markedly abrogates both NPI-0052- and Bortezomib-induced apoptosis. Inhibition of caspase-8 (IETD-FMK) led to a significant decrease in NPI-0052-triggered cell death, whereas inhibition of caspase-9 (LEHD-FMK) only moderately blocked NPI-0052-triggered decreased viability in MM.1S cells (P < 0.005). In contrast, Bortezomib-induced decrease in viability is equally blocked by either caspase-8 or caspase-9 inhibitors (P < 0.005). These biochemical data were further examined by genetic studies using dominant-negative (DN) strategies. Treatment of DN-caspase-8-transfected MM cells with NPI-0052 (IC₅₀: 7 nM) markedly increases survival compared to cells transfected with DN-caspase-9. In contrast, treatment of either DN-caspase-8 or DN caspase-9-tranfected MM.1S cells with Bortezomib (IC₅₀: 5 nM) increases survival to a similar extent. The proteolytic processing of pro-caspase-8 is mediated by Fas Associated Death-Domain (FADD) protein, and we next examined blockade of FADD with DN-FADD to further confirm the role of caspase-8. DN-FADD significantly attenuated NPI-0052-induced cytotoxicity compared to empty vector-transfected MM.1S cells (42 ± 2.0% viable cells in vector- versus 76 ± 5.1% viable cells in DN-FADD-transfected cells; P < 0.05). Importantly, treatment of DN-FADD-transfected MM.1S cells with Bortezomib results in only 16% increase in survival compared to vector-transfected cells (39 ± 2.4% viable cells in vector- versus 55 ± 4.1% viable cells in DN-FADD-transfected cells; P < 0.05). These data, coupled with caspase-8 or caspase-9 inhibition studies, suggest that NPI-0052 relies more on FADD-caspase-8 signaling than does Bortezomib, confirming a differential mechanism of action of NPI-0052 versus Bortezomib in MM cells. To further address this issue, we examined alterations in Bax, a proapoptotic protein which translocates from cytosol to mitochondria during apoptosis, inhibits Bcl-2, and facilitates release of cyto-c and activation of caspase-9. NPI-0052 induces little, if any, increase in Bax levels in mitochondria, whereas Bortezomib triggers a significant accumulation of Bax in mitochondria. Experiments using Bax wild type (WT) or knockout mouse embryonic fibroblast (MEFs) show that NPI-0052 decreases viability in both Bax (WT) and Bax (knock-out) MEFs, whereas deletion of Bax confers significant resistance to Bortezomib. These findings suggest a differential requirement for Bax during NPI-0052- versus Bortezomib-induced apoptosis. Collectively, our data suggest that 1) NPI-0052-induced MM cell apoptosis is predominantly mediated by caspase-8; and 2) Bortezomib-induced apoptosis requires both caspase-8 and caspase-9 activation. These data provide rationale for combining agents based on differential signaling cascades to amplify apoptosis and enhance anti-tumor activity.