Selective Inhibition of CDK4 and CDK6 Primes Chemoresistant Myeloma Cell for Cytotoxic Killing through Induction of Cell Cycle-Coupled Mitochondrial Dysfunction and Apoptosis

Abstract

Dysregulation of cyclin-dependent kinase (CDK)4 or CDK6 activity by gain of function or loss of inhibition is one of the most frequent aberrations in cancer. This includes multiple myeloma (MM), where overexpression of CDK4 (CDK6) precedes unrestrained proliferation of CD138+ bone marrow myeloma cells in vivo, in particular during aggressive tumor growth and relapse. In complex with the D-type cyclin, CDK4 and CDK6 promote cell cycle entry and progression through G1 by inactivating the retinoblastoma protein Rb and antagonizing the INK4 family of CDK inhibitors, suggesting that inhibition of CDK4/6 is a promising approach for cell cycle control in MM. We have now developed a novel approach to both halt cell proliferation and enhance cytotoxic killing of MM cells by selective inhibition of CDK4/6 in combination with cytotoxic agents. We show that knocking down CDK4 and CDK6 expression by shRNA interference or inhibiting CDK4/6 activity with PD 0332991, the only known CDK4/6-specific small molecule inhibitor, leads to sustained G1 arrest and induction of synchronous cell cycle progression upon removal of PD 0332991. Induction of sustained early G1 arrest is not accompanied by apoptosis. However, it primes MM cells for synergistic killing by low dose cytotoxic agents of diverse modes of action, which is further augmented during synchronous S phase entry. Most importantly, induction of sustained G1 arrest with PD 0332991 primes freshly isolated chemoresistant CD138+ bone marrow myeloma cells for killing by low dose proteasome inhibitors in the presence of bone marrow stromal cells. Synergistic killing by PD 0332991 combined with low dose bortezomib (2–6 nM) in early G1 (referred to as PD-B) is mediated by increased neutralization of Mcl-1 and Bcl-2 in the absence of Noxa, as PD-B augments bortezomib activation of Bim and Mcl-1 transcription while silencing Noxa in early G1. This leads to aggregation of Bak, but not Bax, on the mitochondria, mitochondrial membrane depolarization, preferential release of Smac/DIABLO, but not cytochrome c, from mitochondria, reduction of c-IAP and caspase-9 activation. Apoptosis is further amplified through activation of caspase-8 without inducing TRAIL, FASL and TNF-α, the major ligands that trigger the extrinsic apoptosis pathway. Cytotoxic killing by PD-B is recapitulated in synergistic tumor suppression in animal models. Collectively, our ex vivo and in vivo data demonstrate that PD-B induces synergistic killing of MM cells through cell cycle-coupled regulation of Bcl-2 family genes and induction of mitochondrial dysfunction. As PD 0332991 is orally bio-available, potent and low in toxicity, our approaches have formed the basis for an ongoing, first-inclass Phase I/II clinical trial to selectively target CDK4/6 with PD 0332991 in combination with bortezomib and dexamethasone in multiple myeloma. Selective targeting CDK4 and CDK6 in combination with cytotoxic killing, therefore, provides a new and promising mechanism-based therapeutic strategy for multiple myeloma and potentially other cancers.