Changes in Mcl-1 and Bim Expression with Bortezomib and Melphalan Therapy for Multiple Myeloma.

Background: Proteasome inhibition has been shown to be effective against a variety of tumours. In multiple myeloma the response rates to bortezomib (B) in relapsed patients is 46% (APEX study ≥ MR), but are likely to be up to 75% when combined with melphalan (M). Mechanisms underlying this effect are yet to be fully determined. Aims: To investigate the effect of B, M and the combination of the two on myeloma cell lines and primary patient cells, focusing on the anti-apoptotic molecule Mcl-1 and the pro-apoptotic molecule Bim in the mediation of drug activity. Cell cycle analysis using propidium iodide and flow cytometry was also performed in parallel. Methods: The human multiple myeloma cell lines (HMCLs) RPMI 8226/S, U266 and purified primary patient malignant plasma cells were used for cell culture and viability assays using an ATP bioluminescence method. Cells were incubated for 24 or 48 hours with differing concentrations of B, M and combinations in varying schedules. EC50 values were calculated using a sigmoidal Emax model and the observed cell viability of the combination of the two drugs was compared with the additive effect expected. Mcl-1, Bim, caspase-3, and PARP were probed for by Western Blotting of HMCLs. Results: Both U266 and RPMI 8226/S cells showed dramatic reductions in cell viability to B, with EC50 values of 4.7nM and 5.3nM respectively, and responded to high concentrations of M with EC50 values of 95.6 uM and 91.5 uM respectively after 48 hour incubations. Synergistic responses were seen when M was added 24 hours prior to B, but not with B pre-treatment. This was also observed with primary patient cells. Mcl-1 levels increased after 6 hours of B exposure, likely due to proteasome inhibition, but decreased by 24 hours with associated cleavage. This effect was concentration-dependent with partial cleavage observed at 4nM (approximately EC30) and full cleavage at 50nM. Bim was present in untreated cells, unchanged after 6 hours of B exposure, but decreased at 24 hours at both concentrations. All of these changes were associated with cleavage of caspase-3 and the appearance of cleaved PARP, and persisted out to 48 hours exposure. Six hours following M exposure, there was an increase in Mcl-1 at the sub-toxic 10uM concentration (possibly a cell survival response), but a reduction at 100uM. Following 24/48 hour exposures changes were no different to control cells with 10uM M, however at 100uM (EC50 concentration) cleavage of Mcl-1 and a decrease in Bim were observed, similar to changes seen with 50nM B. On combining the two drugs simultaneously in a 48 hour exposure, B 4nM and M 10uM failed to induce any changes in U266 cells, but resulted in partial cleavage of Mcl-1 in RPMI 8226 cells. When the M concentration was increased to 100uM there was a decrease of both Mcl-1 and Bim and the associated cleavage of caspase-3 and PARP. There were no differences whether B preceded or followed M. Cell cycle analysis demonstrated G2 arrest following B therapy at 24 hours and in combination with M. Conclusions: This work demonstrates that in multiple myeloma, both Mcl-1 and Bim are closely involved in proteasome mediated cellular apoptosis and in M mediated cytotoxicity. In keeping with work in Jurkat cells (Nencioni et al., Blood 2005), Mcl-1 was found to transiently increase following proteasome inhibition, but then decreased at 24 hours as apoptosis occurred. An early rise in anti-apoptotic proteins such as Mcl-1 may explain why synergistic responses with B and M were seen only with M pretreatment.