Histone deacetylase inhibitors: Keeping momentum for neuromuscular and cardiovascular diseases treatment

Abstract

Histone deacetylases (HDACs) are enzymes with a pleiotropic range of intracellular localizations and actions. They are principally involved in the withdrawal of acetyl-groups from a large number of nuclear and cytoplasmic proteins including nuclear core histones as well as cytoskeletal proteins and metabolically relevant enzymes. Initial findings indicated that HDAC inhibitors (DIs) could be successfully applied in a variety of cancer treatment protocols as a consequence of their anti-proliferative and pro-apoptotic properties. Recent observations, however, enlightened the important therapeutic effects of DIs in experimental animal models for arthritis, neurodegenerative and neuromuscular disorders, heart ischemia, cardiac hypertrophy, heart failure and arrhythmias. A small number of clinical trials are now open or planned for the near future to verify the therapeutic properties of DIs in non-cancer-related diseases. This review summarizes some of the most important observations and concepts aroused by the most recent experimental application of DIs to neuromuscular and cardiac diseases.

Introduction

Histone deacetylases (HDACs) are a family of 18 molecules divided in four sub-classes (I–II–III–IV) defined according to structural similarities [1]. Their functional role is variegate ranging from regulation of chromatin structure, repression or activation of gene expression, histone and non-histone protein modification and regulation of cell metabolism [1]. Their prototypes were first identified in yeast where they play important roles in DNA replication and repair, gene expression and protein function although their number is much smaller than in mammalian [1]. The presence of such a cohort of molecules with apparent similar enzymatic activity may be incorrectly interpreted as redundancy while it probably reflects very specialized functions, the necessity of narrow actions on selected substrates and the possibility to obtain a wide variety of combinatorial complexes with different functional activities [2], [3], [4].

The best characterized HDACs are the members of classes I, II and III. The latter are also known as sirtuins which are important regulator of metabolism and aging [5]. Their importance has been emphasized in a series of recent articles and will not be matter of further discussion in this review [6], [7]. Class I HDACs encompasses HDAC1, 2, 3 and 8 with conserved structure, function similarities and a prevalent intra-nuclear distribution among the first three while HDAC8 is tissue specific and present either in the nuclear or cytoplasmic compartments [8]. HDACs belonging to class II are more structurally complex and their functions are not well characterized yet. Class II is further divided in classes IIa and IIb reflecting internal similarities among different molecules. Members of class IIa are HDAC4, 5, 7 and 9 while HDAC6 and 10 are grouped in class IIb. HDAC11 is probably the less characterized enzyme of this family and is the only component of class IV [1]. At present it does not seem implicated in neuromuscular or cardiovascular disorders and will not be further discussed.

In recent years, prompted by their prominent anticancer properties, pressure has been put on pharmaceutic companies and academic research laboratories to identify class- or isoform-selective HDAC inhibitors (DIs) starting from few prototypes mainly represented by hydroxamic acids or benzamide derivatives [9]. A number of DIs are currently undergoing clinical trials as anticancer drugs, the majority of which have been characterized as pan-DIs, while some of them are class-selective. In 2006, vorinostat (suberoylanilide hydroxamic acid, SAHA) (Fig. 1) was approved by FDA (Zolinza, Merck) for the treatment of refractory cutaneous T-cell lymphoma [10]. In addition to vorinostat, other hydroxamates such as belinostat, panobinostat, trichostatin A (TSA), ITF2357, as well as some short-chain fatty acids (SCFAs) such as sodium valproate (VPA), or cyclic peptides such as apicidin and romidepsin (FK-228), or some benzamide derivatives such as MS-275 and MGCD0103 (Fig. 1) are currently in phase I/II or phase III clinical trials as anticancer agents [11], [12]. In some cases, however, their potential for clinical drug development could be limited by low potency and lack of selectivity, cytotoxicity, low solubility in the aqueous vehicle and/or low stability in cell culture and in animal models. Regarding HDAC selectivity, the hydroxamates (such as vorinostat, TSA, belinostat, panobinostat, ITF2357) are typically pan-HDAC inhibitors active at submicromolar/nanomolar level, with a drop of inhibiting potency against HDAC8, whereas the SCFAs such as VPA showed millimolar activity against class I HDACs (Fig. 1) [13]. Among the cyclic peptides, apicidin was highly active against HDAC2 and 3, less potent against HDAC8 and no active against HDAC1 and class II HDACs, while romidepsin showed high HDAC1 and 2 inhibiting activity being near 10- and 300-fold less potent against HDAC4 and 6 respectively (Fig. 1) [13], [14]. The two benzamide derivatives MS-275 and MGCD0103 displayed similar but not superimposable class I selectivity. Specifically MS-275 was highly active against HDAC1 and 6, about 13-fold less potent against HDAC2 and 3 and inactive against HDAC8. MGCD0103 showed high inhibitory activity against HDAC1 and 2, 30-fold lower potency against HDAC3 and no activity against HDAC8 (Fig. 1) [13].

Some other class- or isoform-specific DIs have been developed, either belonging to hydroxamates/benzamides or bearing different enzyme inhibiting groups (thiolate analogues, trifluoromethylketones) (Fig. 2). Among the hydroxamates, tubacin was described as an in-cell HDAC6-selective inhibitor [15], and MC1568 and MC1575 were reported as pan-class II (IIa + IIb) selective DIs [16], [17], [18], [19], [20]. In addition, the indole derivative PCI-34051 has been recently identified as a HDAC8-selective inhibitor, specifically active against T-cell lymphoma through PLCγ1 activation and calcium-induced apoptosis [21]. Some 2-amino-5-arylbenzamides were independently disclosed by MethylGene [22] and Merck [23] as HDAC1/2-specific inhibitors (Fig. 2). A number of sulphur-containing compounds were described as HDAC6-selective inhibitors (mercaptoacetamides and thiolate analogues) [24], [25] or HDAC8-specific inhibitor (SB-379278A) [26], and some trifluoromethylketones [27] were reported with different degrees of selectivity against the various class II HDAC isoforms (Fig. 2).

The quest for more specific compounds is still ongoing, however selective molecules targeting a restricted number of class I or II molecules have been successfully generated and are currently under evaluation alone or in combination with other compounds in several experimental and clinical studies.

Very recently a series of reports described the successful application of DIs in experimental animal models of neuromuscular and other degenerative diseases as well as in cardiac hypertrophy, heart failure and some types of arrhythmic disorders [28], [29], [30], [31], [32]. Therefore non-selective or target-restricted DIs represent a new and growing field of potential application for these drugs in neuromuscular and cardiovascular diseases which hopes and pitfalls will be discussed in this review.