Therapeutic window for melanoma treatment provided by selective effects of the proteasome on Bcl-2 proteins.

Melanoma cells depend on sustained proteasomal function for survival. However, bortezomib, the first proteasome inhibitor in clinical use, is not sufficient to improve the poor prognosis of metastatic melanoma patients. Since the proteasome is also expressed in all normal cell compartments, it is unclear how to enhance the efficacy of bortezomib without exacerbating secondary toxicities. Here, we present pharmacological and genetic analyses of mechanisms of resistance to proteasome inhibition. We focused on Bcl-2, Bcl-x(L) and Mcl-1 as main antiapoptotic factors associated with melanoma progression. Despite an efficient blockage of the proteasome, bortezomib could not counteract the intrinsically high levels of Bcl-2 and Bcl-x(L) in melanoma cells. Moreover, Mcl-1 was only downregulated at late time points after treatment. Based on these results, a combination treatment including (-)-gossypol, an inhibitor of Mcl-1/Bcl-2/Bcl-x(L), was designed and proven effective in vivo. Using a specific RNA interference approach, the survival of bortezomib-treated melanoma cells was found to rely primarily on Mcl-1, and to a lesser extent on Bcl-x(L) (but not on Bcl-2). Importantly, neither Mcl-1 nor Bcl-x(L) inactivation affected the viability of normal melanocytes. This hierarchical requirement of Bcl-2 family members for the maintenance of normal and malignant cells offers a therapeutic window to overcome melanoma chemoresistance in a tumor cell-selective manner.