Abstract

Histone deacetylase inhibitors have been shown to modulate the cell cycle, to induce apoptosis and to sensitize cancer cells to other chemotherapeutics. Among these inhibitors, valproic acid (VPA), an antiepileptic drug, is being discussed as a promising novel anti-cancer drug. Chronic Lymphocytic Leukemia (CLL) is a clinically heterogeneous disease remaining incurable despite introducing new promising treatments. The effects of VPA and its mechanism of action were evaluated on mononuclear cells isolated from 40 CLL patients. Exposure of CLL cells to increased doses of VPA (0.5–5mM) leads to a dose-dependent cytotoxicity and apoptosis in all CLL patients. VPA treatment induced apoptotic changes in CLL cells including phosphatidylserine externalization and DNA fragmentation. The mean apoptotic rates were similar between IGHV mutated and unmutated patients, the latter presenting a more aggressive clinical course. VPA induced apoptosis via the extrinsic pathway involving engagement of the caspase-8-dependent cascade. Interestingly, VPA increased the sensitivity of leukemic cells to tumor necrosis factor-related apoptosis inducing ligand (TRAIL) even among resistant patients. Moreover, VPA at physiological concentration of 1mM can significantly increase the in vitro cytotoxic effects of fludarabine, bortezomib and the natural product honokiol allowing the reduction of effective concentration 50% (EC50). In order to understand the early mechanism of action of VPA, we investigated gene expression profiles of 14 CLL-patient samples (7 with a good prognosis and 7 with a bad prognosis regarding IGHV mutational status and Zap-70 expression) treated in vitro during 4 hours with a physiological dose (1mM) of VPA and compared with their untreated counterpart using Affymetrix technology. No difference in gene modulation was observed between poor and good prognosis patients after VPA treatment. Modulation of several pro- and anti-apoptotic mRNA expression was confirmed by a real-time reverse transcription-PCR. The molecular analysis of the apoptotic machinery involved in VPA response revealed the up-regulation of APAF1 (5.5 fold, P<0.0001), BNIP3 (2.2 fold, P=0.0006), PTEN (1.9 fold, P=0.0002), CASP6 (2.5 fold, P<0.0001) and the down-regulation of CFLAR/FLIP (2.0 fold, P<0.0001), BCL2 (1.6 fold, P=0.0222), AVEN (1.9 fold, P<0.0001), BIRC4/XIAP (1.7 fold, P<0.0001) and BIRC1/NAIP (1.6 fold, P=0.0007). In conclusion, VPA induced apoptosis of CLL cells at clinically relevant concentration by selective activation of the caspase-8 (extrinsic) pathway and by targeting several pro- and anti-apoptotic genes. Therefore, the combined application of VPA with other drugs might be considered as a potential strategy for CLL treatment.

Author notes

Disclosure: No relevant conflicts of interest to declare.