Acetylome and phosphoproteome modifications in imatinib resistant chronic myeloid leukaemia cells treated with valproic acid

Abstract

Chronic myeloid leukaemia has a specific therapy: BCR/ABL inhibitor imatinib. Resistance due to BCR/ABL dependent and independent mechanisms is partially reversible by histone deacetylase inhibitors. We analysed by 2D-electrophoresis and anti-pan-acetylated and anti-phosphotyrosine immunoblots, followed by spot-matching and MALDI-TOF mass spectrometry, which proteome modifications would parallel restoration of sensitivity to imatinib by valproic acid (VPA). VPA plus imatinib significantly increased acetylation of HSP90 and hnRNP L and decreased phosphorylation of HSPs and hnRNPs in imatinib resistant cells. VPA was able to modify profoundly acetylome and phosphoproteome of CML cells, while reverting resistance to imatinib.

Introduction

Imatinib is an extremely effective therapy for chronic phase CML BCR/ABL positive, but patients may develop resistance, caused most frequently by point mutations within the kinase domain of BCR–ABL, or less often by amplification of the BCR/ABL genomic locus [1]. Mechanisms independent of BCR/ABL activity are also supposed to cause resistance. Novel ABL kinase inhibitors or dual SRC/ABL kinase inhibitors with higher potency against native and imatinib-resistant mutants of BCR/ABL have substantial clinical utility, but at least one mutation located in the p-loop of ATP domain of ABL (T315I) remains resistant to any kinase inhibitor until now in clinical use [2]. Thus, the search for alternative drugs effective in imatinib-resistant CML is still cogent [3]. Treatment of CML cells with histone deacetylase inhibitors (HDACi) of the class of hydroxamic acid analogues has been shown to promote proteasomal degradation of BCR/ABL [4], associated with apoptosis, and to result in anti leukemic effect together with imatinib [5], even in T315I mutation positive cells [6]. These findings prompted the design of new clinical trials with agents inhibiting Heat Shock Protein 90 (HSP90) [7], as the pivotal role of this stress-activated chaperone protein was quite clear. HSP90 in fact, in combination with HSP70 [8], contributes to protect BCR/ABL from proteasomal degradation and this activity is inactivated by chaperon protein acetylation [6], [9], [10], [11]. Although histones are the most prominent and abundant substrates for HDACs, they can de-acetylate non-histone substrates, situating themselves in cell cytoplasm, nucleus and in mitochondria [12]. Candidate substrates are at least two hundred “acetylable” proteins, whose functions and localization are modulated by this post-translational modification. The mechanisms by which HDAC inhibitors may exert their broadly demonstrated antineoplastic activity could therefore lay not uniquely on their epigenetic effects [12], [13].

We evaluated which protein modifications were involved in reversion of imatinib resistance by valproic acid (VPA), a feeble HDACi, belonging to the short chain fatty acid class [14], [15]. We have recently shown that CML cells resistant to imatinib express a different protein phenotype respect to sensitive cells [16], [17]. In this study, we support the suggestive observation of a substantial effect of HDACi/short chain fatty acids and imatinib in terms not only of inhibition of proliferation, but also BCR/ABL expression and induction of apoptosis [17], [18]. We then present an analysis of the modifications of the acetylation and phosphorylation of proteins expressed after exposure to the combination of imatinib and VPA. To perform this analysis, we analysed human CML cell lines, with a clone resistant and a clone sensitive to imatinib and utilized a two dimensional Western blot (WB) proteomic approach [19].