HEXIM1: a new player in myocardial hypertrophy?

This editorial refers to ‘Inducible re-expression of HEXIM1 causes physiological cardiac hypertrophy in the adult mouse’ by M.M. Montano et al., pp. 74–82, this issue. Cardiac hypertrophy is a consequential response to stress; however, not all types of hypertrophy are the same. In fact, the molecular signatures underlying different forms of hypertrophy depend on the type of stress and of the subsequent activated signalling cascade. Preferential activation of specific signal transduction cascades, such as the calcineurin and MAPK pathways, is characteristic for pathological cardiac hypertrophy (e.g. in response to ventricular pressure overload), while the PI3 K/Akt/mTOR signalling cascade seems to be preferentially activated in exercise- or IGF1-dependent physiological hypertrophy.1–3 Remarkably, the activation of each signalling pathway is not specific but rather preferential for one type of hypertrophy. For instance, the PI3 K/Akt pathway is activated in both pressure-overload and exercise-induced cardiac hypertrophy.2,4 Akt induction indirectly activates mTOR, which, in turn, promotes protein synthesis through mRNA translation, while at the same time activating vasculogenesis.5 This increase in capillary density is a fundamental component of the cardiac response to stress since it counterbalances the augmented demand for oxygen by increasing its supply. On the other hand, in the absence of Akt, angiogenesis is impaired and the pressure-overloaded heart goes into failure more quickly.6,7 A different situation is found in exercise-induced hypertrophy, where the match between cardiomyocyte hypertrophy and angiogenesis is more balanced compared with the models of continuous pressure overload.8 A typical hallmark of cardiac hypertrophy, and in particular the maladaptive condition, is an overall increase in RNA synthesis.9 Generally, this is characterized by a reactivation of genes that are typically expressed at the foetal stage, such as NPPA and NPPB.4 The pro-angiogenic vascular endothelial growth factor-A (VEGF) is among the genes whose synthesis is increased during hypertrophy and its activation is regulated by Akt signalling10 and more directly by the hypoxia-inducible factor 1-alpha (HIF-1α).11 Interestingly, VEGF is produced by cardiomyocytes and enhances cell growth through its autocrine effect.12 In line with all this, accumulating evidence has demonstrated an important role of the molecular elements that control the general RNA processing in the progression to cardiac hypertrophy with particular emphasis on the RNA positive transcription elongation factor-b (P-TEFb) and its modulators, including Hexamethylene bis-acetamide inducible protein 1 (HEXIM1). In the paper by Montano et al.13 the authors studied the effects of HEXIM1 re-expression in the mouse adult heart. This gene (the human homologue of the mouse cardiac lineage protein CLP-1), was the first mammalian RNA kinase protein to be identified and has been shown to be involved in the regulation of many physiological and pathological processes, including cancer, inflammation, and neuronal degeneration as well as cardiac development and hypertrophy.14 At the structural level, HEXIM1 is composed of several domains of which (i) the N-terminal domain is a self-inhibitory domain, (ii) the central region acts as a nuclear localization signal as well as a binding domain for the 7SK small nuclear RNA, (iii) an adjacent region binds to P-TEFb, and (iv) the C-terminus binds to Cyclin T. At the molecular level, HEXIM1 acts as a transcriptional repressor by interfering with the effect of P-TEFb15 on mRNA transcription. In the catalytically inactive state, P-TEFb exists as a complex between HEXIM1, cyclin-dependent kinase 9 (Cdk9), 7SK small nuclear RNA, cyclin T, and other proteins.16 In response to various physiological and pathological stimuli, the silent P-TEFb complex is converted into a catalytically active form through dissociation of HEXIM1 from cyclin. This allows Bdr4 to bind to P-TEFb, activating Cdk9 and through interaction with and phosphorylation of RNA polymerase II leading to increased elongation activity and transcription of nascent mRNAs. Interestingly, the affinity of the bromodomain protein Brd4 to acetylated histones directs the active P-TEFb to specific regions of the genome that are epigenetically marked as transcriptionally active.17 Among the triggering stimuli are those associated with pathological hypertrophic signals. Indeed, in transgenic mice for calcineurin,18 the hypertrophic response observed in the heart is accompanied with reduced expression of HEXIM1 with a consequential activation of P-TEFb.19 Similar results have been obtained in cultured cardiac myocytes where stimulation with hypertrophic agonists caused the switch from the inactive to the active state of P-TEFb. Thus, HEXIM1 possibly represents a key modulator of the dynamic equilibrium of P-TEFb between the inactive and the active state and in doing so contributes to the global transcriptional regulation of a wide range of mRNAs associated with physiological and pathological cellular processes. This might represent a fine tuning modulation system for the control of cellular homoeostasis. Furthermore, several pieces of evidence have shown that the HEXIM1/P-TEFb pathway can promote the expression of specific genes recruited by specific transcription factors.20 For example, it has been shown that HEXIM1 can directly bind to and modulate transcription factors, such as the oestrogen receptor alpha, CCAAT/enhancer-binding protein alpha, and nuclear factor-kappa β.21 Taken together these data highlight a mechanism whereby the increase in P-TEFb activity is involved in increasing the global rate of RNA synthesis, thus contributing to the hypertrophic response in cardiomyocytes. In line with this, a lethal foetal phenotype together with a clear evidence of myocardial hypertrophy has been shown in null mice for HEXIM1.22 A similar phenotype has been reported for a mutant mouse expressing a truncated form of HEXIM1 in which its C-terminal domain is absent.23 More specifically, this transgenic mouse showed prenatal lethality with significant evidence of several heart defects such as abnormal coronary distribution, decreased vascularization, and thinner ventricular walls. Moreover, it is worth mentioning that overexpression of cyclin T or Cdk9 was found to provoke specific activation of P-TEFb that was associated with hypertrophic growth and acceleration of the progression to heart failure.24 Although the expression level of HEXIM1 is high during embryonic development and the early postnatal period, it is gradually reduced at the adult stage.25,26 Montano et al.13 determined the effects of acute reactivation of HEXIM1 in the adult heart. The authors generated an inducible and cardiac-specific transgenic mouse for HEXIM1 and showed that it up-regulates the protein level and thus activity of HIF-1α through direct interaction, leading to up-regulation of VEGF, suggesting that HEXIM1 is fundamental for the regulation of VEGF gene expression. Interestingly, the authors also found down-regulation of genes related to glucose metabolism and a concomitant up-regulation of genes associated with fatty acid utilization. The absence of activation of typical stress genes, such as NPPA and NPPB, led the authors to speculate that HEXIM1 overexpression induces a hypertrophy that more resembles exercise-induced rather than pressure-overload induced hypertrophy. In that respect, it is intriguing that HEXIM1 re-expression induces bradycardia without any compromise of other cardiomechanical parameters. It should be pointed out that the activation state of other major signalling pathway molecules involved in pressure-overload or exercise-induced hypertrophy (such as MAPK, mTOR, calcineurin, etc.) was not determined. Care should be taken when data are extrapolated from transgenic mouse models in which non-physiological levels of the transgene are reached during activation. The fact that the majority of the endogenous HEXIM1 is not associated with P-TEFb suggests the possibility that P-TEFb itself may play a role in hypertrophy. Moreover, the equilibrium between P-TEFb complexes is further controlled by additional regulators and several phosphatases, such as PP2B and PP1A, which have been involved in the dephosphorylation and thus deactivation of Cdk9.27 Montano et al. suggest a significant role of HEXIM1 on cardiac angiogenesis and its possible role in promoting adaptive cardiac hypertrophy (Figure 1). Schematic model of HEXIM1activity. HEXIM1 controls mRNA elongation and gene expression by modulating pTEF-b activity and through a direct effect on specific transcription factors. Conflict of interest: none declared. G.C. is supported by grants from Fundation LeDucq, Fondazione CARIPLO (grant number 12-4-5157157-3) and ERC ‘Advanced’ ‘CardioEpigen’.