Inhibition of Hypertrophy Is a Good Therapeutic Strategy in Ventricular Pressure Overload

Introduction

Hypertrophic growth of ventricular myocytes is a hallmark feature of numerous forms of cardiovascular disease¹. The hypertrophic process is complex, involving a vast array of structural, signaling, transcriptional, electrophysiological, metabolic, and functional events within the growing cell^{2, 3}. Other cellular elements within the ventricle – fibroblasts, vascular smooth muscle cells, endothelium – also manifest intricate stress responses, resulting in fibrosis, inflammatory cell infiltration, endothelial dysfunction, and vascular stiffness. Current thinking holds that these events, the heart’s reaction to a host of pathological stresses, provide short-term benefit. However, if disease-related stress remains unchecked, these remodeling events become maladaptive and predispose to cardiovascular morbidity and mortality.

Among the risks conferred by disease-related ventricular hypertrophy are ventricular tachyarrhythmia, predisposing to sudden cardiac death, and transition to heart failure. Ultimately, these events derive from wholesale reprogramming and relative dedifferentiation of the cardiac myocyte, coupled with similar events in other cell types.

Conventional thinking holds that hypertrophic growth of the myocardium is a “compensatory” response of the heart to increases in workload demand, serving to minimize wall stress and maintain contractile function. However, several lines of evidence – preclinical and epidemiological – highlight the maladaptive features of chronic ventricular hypertrophy. Indeed, in many instances suppression of load-induced growth is well tolerated⁴. Further, left ventricular hypertrophy is among the most robust markers of increased risk for developing chronic heart failure⁵. Therefore, we submit that suppression of load-induced ventricular hypertrophy warrants careful consideration as a therapeutic strategy.

Cardiomyocyte hypertrophy: Comprehensive reprogramming of the cell

Whereas evidence suggests that a small fraction of cells within the ventricle are capable of re-entering the cell cycle^6-8, the vast majority of cardiomyocytes are post-mitotic and hence do not retain the ability to divide. Rather, they respond to stress by growing, shrinking, or dying. In the context of many disease-related stresses, cardiac myocytes undergo hypertrophic transformation which entails significant cellular growth. In the setting of uncontrolled hypertension, for example, cardiomyocytes hypertrophy. Similarly, following myocardial infarction, surviving cardiac myocytes in border and remote zones of tissue increase in size in response to increases in hemodynamic demand arising secondary to loss of ventricular tissue. In both cases, when the pressure overload state is persistent, the hypertrophic phenotype of the myocardium inexorably progresses to a state of decompensation and clinical heart failure. Mechanisms governing this transition from adaptive hypertrophy to maladaptive failure remain poorly understood.

Hypertrophic transformation of the cardiomyocyte involves much more than simple cell growth. Rather, it entails a near-comprehensive retooling of multiple aspects of cellular architecture and machinery. One element of this process is relative dedifferentiation of the cell and reactivation of numerous transcriptional, signaling, electrical, and metabolic events that characterized the cell during development. As part of this, a wide range of transcriptional and post-translational events occurs, including activation of a pattern of gene expression reminiscent of that observed during fetal development (“fetal gene program”). Indeed, the fetal gene program, which was extinguished shortly after birth, re-ignites rapidly in the setting of disease.

Based on the pattern of sarcomere reorganization, it is possible to distinguish two broad patterns of ventricular hypertrophy⁹. Pressure stress provokes concentric hypertrophy, which is characterized by recruitment of sarcomeres laid down in parallel; on the contrary, excess volume elicits eccentric hypertrophy in which cardiomyocytes respond with addition of sarcomeres in series⁹. In both cases, increases in wall thickness occur. [Concentric hypertrophy is marked by increases in wall thickness with relatively little change in ventricular volume; eccentric hypertrophy is marked by increases in both wall thickness and ventricular cavity size.] According to Grossman’s pioneering stress-adaptation hypothesis, these increases in wall thickness are an adaptive response¹⁰; based on Laplace’s law, ventricular wall stress is proportional to both ventricular pressure and cavity radius and inversely proportional to ventricular wall thickness¹¹. Thus, increases in wall thickness tend to lessen wall stress and thereby diminish oxygen demand.

Myocyte growth is dictated by a delicate balance between protein synthesis and protein degradation. In the setting of elevated afterload, protein synthesis predominates, culminating in hypertrophic growth. However, it is important to recognize that hypertrophic remodeling is not a simple process of addition of new sarcomeres. Rather, this highly dynamic cellular response involves intricate coordination of de novo protein synthesis and organelle biogenesis, sarcomere remodeling, protein degradation, organelle breakdown, transcriptional reprogramming, and metabolic shifts. In many ways, the entire cellular architecture of the myocyte – frame, chassis, drive train, and engine – is retooled.

In contrast to disease-related triggers, physiological stresses, such as endurance exercise and pregnancy, induce a hypertrophic response characterized by normal or enhanced contractile function coupled with normal architecture and organization of cardiac structure¹². Beyond differences in growth triggers, biological phenotypes, and clinical outcomes, pathological and physiological hypertrophy differ in the signaling cascades which drive the process. Some evidence suggests that the distinct phenotypes do not derive simply from differences in the duration of the stimulus, highlighting significant gaps in our understanding of these two remodeling responses¹³. Other data suggest that cell size is regulated by shared signaling pathways, but cell shape and sarcomeric organization are regulated by distinct pathways¹⁴. Current understanding does not allow us to parse the effects of the myriad genes and pathways involved, but it is thought that some confer benefit whereas others are maladaptive. Substantial evidence points to alterations in transmembrane Ca²⁺ fluxes – another central feature of pathological remodeling – as a proximal trigger contributing to the pathogenesis of hypertrophy and failure¹⁵. These alterations perturb excitation-contraction coupling, alter mitochondrial metabolism, and abnormally activate Ca²⁺-responsive signaling pathways. Recent evidence indicates that β-MHC protein is induced by pressure overload in rodents in only a minor subpopulation of smaller cardiac myocytes¹⁶. The myocytes that hypertrophied after surgical constriction of the thoracic aorta express α-MHC only. Thus, hypertrophic transformation may manifest yet another layer of complexity in that it manifests significant heterogeneity among myocytes.

Cardiac myocyte death

Cell death within the myocardium is characteristic of a number of cardiac diseases, and it can occur to some extent in cardiac hypertrophy. The major types of cardiomyocyte death are necrosis and apoptosis, with the former occurring to a greater extent. An emerging literature has demonstrated that necrosis can occur as a result of a series of programmed events, not just as simple catastrophic dismantling of the cell. Indeed, programmed necrosis and apoptosis share a number of features and may represent different manifestations of a common mechanism termed necroptosis^{17, 18}.

Both dying and hypertrophying cells often harbor signs of activated autophagy, an evolutionarily ancient process of ordered recycling of intracellular contents¹⁹. Whether activation of the autophagic cascade reflects a cellular response to stress, serving to promote cell survival, or is a process contributing to cell death and disease progression, is context-dependent²⁰.

Fibrosis

Another hallmark feature of pathological hypertrophic remodeling is accumulation and deposition of excessive extracellular matrix (ECM)²¹. This surplus ECM, which constitutes tissue “scar” or fibrosis, perturbs electrical conduction, thereby predisposing to rhythm disturbances. It also promotes dysfunction of both mechanical contraction and relaxation. As a result, cardiac fibrosis contributes importantly to morbidity and mortality in cardiac hypertrophy. Indeed, the amount of fibrotic scar in the myocardium correlates directly with the incidence of arrhythmias and sudden cardiac death^22-24.

ECM deposition and fibrosis arise through the action of cardiac fibroblasts. These cells, the most abundant cell type in the myocardium, proliferate in response to pathological stress. Further, they differentiate into myofibroblasts, thereby gaining the capacity to contract and secrete collagen I, collagen III, and fibronectin²⁵. Fibrosis can be categorized as reactive (perivascular or interstitial) or replacement, occurring at the site of an eliminated myocyte. Myofibroblasts derive from activated, resident fibroblasts, but may also originate from adult epicardial cells²⁶ and circulating, collagen-secreting, bone marrow-derived cells^{27, 28}. Both individual myofibroblasts and collagenous septa within the tissue facilitate and propagate the arrhythmic phenotype of the hypertrophied heart^29-31

Cardiac fibrosis is an independent and predictive risk factor for heart failure development in the setting of ischemic or non-ischemic cardiomyopathy^32-34. Importantly, strong evidence indicates that cardiac fibrosis, long held to be irreversible, may regress under certain conditions^{21, 35}. However, whereas several signaling pathways have been implicated in fibrogenesis, tailored therapeutic approaches targeting cardiac fibrosis remain elusive²¹.

Electrical remodeling

Patients with left ventricular hypertrophy are at increased risk of malignant arrhythmia, which contributes significantly to morbidity and mortality. Indeed, arrhythmia, especially ventricular tachyarrhythmia, is a major cause of death in patients with cardiac hypertrophy or failure. Underlying mechanisms, collectively termed “electrical remodeling”, encompass alterations in multiple electrogenic transport and signaling processes within the cardiac myocyte. Whereas numerous insights have emerged in elucidating the molecular pathogenesis of cardiac hypertrophy^{3, 36}, our understanding of mechanisms underlying the myriad facets of electrical remodeling remains limited. As a result, pharmacological treatment of hypertrophy-associated arrhythmias lacks efficacy, and device-based therapy has emerged as a widely used surrogate.

The action potential phenotype of ventricular myocyte hypertrophy is characterized by delayed repolarization leading to prolongation of action potential duration (APD). This derives, at least in part, from distorted transmembrane electrical currents^{2, 15, 37}. Indeed, a wide range of alterations in myocyte ion channels and electrogenic ion transporters contribute to APD prolongation^{2, 15, 37}. Delayed recovery of excitability, in turn, predisposes to early and late after-depolarizations. Hypertrophy is also associated with myocardial fibrosis (vide supra), altered electrotonic coupling among cells, slowed conduction, and dispersion of refractoriness, which together promote re-entrant arrhythmias.

This electrical remodeling response is heterogeneous within the ventricle. In the setting of excessive afterload, such as in severe transverse aortic constriction-induced heart failure, APD is prolonged more in subepicardial ventricular myocytes than in subendocardial myocytes³⁸. Further evidence for heterogeneity of APD prolongation has been reported in a model of pacing-induced heart failure in dogs where APD prolongation in mid-myocardial cells was substantially greater than in subepicardial cells³⁹.

Alterations in Ca²⁺ handling contribute to both hypertrophic signaling and electrical remodeling. For example, L-type Ca²⁺ current (I_Ca,L) is a major mechanism of Ca²⁺ influx in cardiac myocytes. As a general rule, I_Ca,L density correlates inversely with disease progression; in models of mild-to-moderate hypertrophy, I_Ca,L is often increased, whereas in severe hypertrophy and failure, I_Ca,L often manifests significant declines. Importantly, as membrane impedance is relatively high during phase 2 of the action potential, small changes in I_Ca,L can have significant effects on action potential morphology and duration. Further, entry of small amounts of extracellular Ca²⁺ triggers release of much larger amounts of Ca²⁺ from intracellular stores. As a consequence, modest changes in inward Ca²⁺ flux are amplified within the cell.

Electrical activity within the myocardium hinges critically on electrotonic cell-cell coupling, such that depolarization in one cell is transmitted seamlessly to neighboring cells. This coupling is mediated through gap junctions, such as connexin 43, which can become disorganized in the hypertrophied or failing heart, disrupting normal impulse conduction⁴⁰. It is worth remembering that the atria are also touched by remodeling events in cardiac hypertrophy and failure. Reduced contractility, development of fibrosis, and chamber enlargement can occur, leading to heterogeneity of conduction velocity⁴¹ and propensity to atrial fibrillation⁴². This arrhythmia, in turn, eliminates the component of ventricular filling provided by atrial contraction. Further, it promotes decreases in atrial myocyte effective refractory period and shortened APD, which together promote sustained atrial fibrillation⁴³.

Metabolic remodeling

The metabolic demands of the myocardium are exceptionally high. As a continuously working pump, the heart consumes robust quantities of adenosine triphosphate (ATP), more than any other organ in the body. And myocardial energy reserves are remarkably low, sufficient for fewer than 10 contractions. The cardiomyocyte derives ATP largely from fatty acid oxidation. That being said, the heart is a metabolic omnivore which can flexibly burn fuel derived from a wide range of sources. In the end, cardiomyocytes must generate energy continuously, consuming nutrients constantly and without interruption⁴⁴.

Among the characteristic changes occurring with cardiac hypertrophy is a shift in energy substrate utilization^45-47. This metabolic remodeling response entails up-regulation of glucose uptake and glycolysis, while β-oxidation of fatty acids is reduced. Approximate numbers are that glycolysis accounts for 10% of ATP production in the normal heart and 20% in the hypertrophied heart. Conversely, ATP production from fatty acid metabolism drops from 70% to 50%. These shifts are consistent with the over-arching process of cellular dedifferentiation in the pathologically stressed myocardium, as the shift in metabolism mimics the metabolic program in the fetal myocardium.

The impact of the partial switch to a fetal program of glucose metabolism from preferential utilization of lipid remains uncertain. Arguably the greatest impact derives from improvements in oxygen efficiency⁴⁸. However, concerns have been raised due to evidence that cardiomyocytes cultured in high glucose media are dysfunctional⁴⁹. To accomplish the synthesis of macromolecules and organelles required for cardiomyocyte hypertrophic growth, exogenous nutrients, such as glucose, cannot be metabolized exclusively for ATP production. Rather, metabolic intermediates must be channeled to support anabolic pathways. Here, interplay between the plasticity of metabolic pathways and protein quality control is critical to providing intermediate metabolites that feed into the tricarboxylic acid (TCA) cycle for ATP production, as well as serving to promote macromolecule synthesis⁵⁰. Overall, currently available information implicates metabolic reprogramming in hypertrophied heart as an ultimately maladaptive adaptation⁵¹.

Inflammation

Activation of immune mechanisms participates in ventricular remodeling, contributing to long-term cardiac injury in certain contexts. In the case of heart failure, a variety of inflammatory molecules and pathways are activated^52-55. In pre-clinical models, pressure overload triggers myocardial inflammation as demonstrated by increased expression of several proinflammatory cytokines and leukocyte infiltration within the myocardium^56-58. This and other evidence, both in animals and patients, highlights the role of inflammation as an important component of pathological hypertrophy and points to the relationship between a pro-inflammatory state and load-induced ventricular remodeling^59-62. Interestingly, myocardial cytokine levels, a surrogate marker of inflammation, are often higher in patients with elevated afterload and preserved left ventricular function compared with those patients with reduced ejection fraction⁶³. These data challenge the concept that cytokine activity and an inflammatory response are necessarily detrimental in chronic pressure overload.

A crucial element within the inflammatory response observed in models of pressure overload involves macrophages that infiltrate ventricular tissue, triggering myocardial expression of NF-кB and inflammatory cytokines⁶⁴. One of the mechanisms whereby macrophages are involved includes actions of macrophage-derived miR-155 which functions as a modulator of cardiomyocyte hypertrophy via paracrine mechanisms⁶⁵.

These observations, coupled with the effects of anti-inflammatory therapies in animals, have prompted efforts to translate these insights into patients through clinical trials^{56, 66}. However, to date, anti-inflammatory therapies for heart failure in humans have disappointed^67-69. Looking to the future, the possible role of novel anti-inflammatory therapies using selective approaches targeting specific cellular and molecular elements of pathological cardiac remodeling and heart failure could be elucidated. Recently, a novel connection between autophagy, afterload stress, and inflammation was reported, where lack of autophagy-mediated removal of mitochondrial DNA promoted an inflammatory response contributing to depressed cardiac contractile performance⁷⁰.

Vascular remodeling

The vasculature that supplies the ventricle itself remodels in both physiological and pathological hypertrophy. Hypertrophic growth of the heart involves parallel increases in myocardial mass and vascular supply, the latter occurring through proliferation of endothelium, smooth muscle cells, and blood vessels^{71, 72}. Indeed, it is possible to distinguish two interdependent phenomena involving heart vessels: 1) remodeling of pre-existing vasculature with modifications of endothelium, smooth muscle cells, and interstitial matrix, and 2) de novo myocardial angiogenesis involving an intricate network of molecules secreted from cardiomyocytes, leukocytes and fibroblasts (reviewed^{73, 74}). Paracrine signaling through mediators released by endothelial cells may modify vascular smooth muscle cell function and extracellular matrix components^{75, 76}. Also, in the setting of hypertension, vascular smooth muscle proliferates and hypertrophies, culminating in vascular wall thickening. In some settings, flow reserve is compromised.

Dysfunction of vascular endothelial cells occurring in pressure overload-induced heart failure may represent a crucial node in vascular remodeling occurring in pathological hypertrophy^{77, 78}. Indeed, one model holds that capillary growth in pathological hypertrophy does not keep up with myocyte growth, leading to inadequate oxygen diffusion capacity. Along those lines, it has also been postulated that an imbalance in the capillary-to-cardiomyocyte ratio contributes to the transition from compensatory hypertrophy to decompensated heart failure⁷⁴. Consistent with this notion, enhancing angiogenesis in a model of afterload stress can be protective⁷⁹. In fact, therapeutic strategies aimed to restore the capillary network in heart failure have been evaluated in both preclinical models and in humans^80-83. Despite some favorable findings, therapeutic myocardial angiogenesis has failed to achieve clinical significance, at least partly because these approaches typically involve delivery of genetic material or growth factors with attendant complications.

Functional role of pathological hypertrophy

Three stages of hypertrophic transformation of the heart were initially proposed by Meerson and coworkers⁸⁴. This model emphasizes that the duration of pressure overload dictates the progression of events, including hypertrophic growth and ultimately ventricular systolic function. In this model, “short-term” hypertrophy represents a beneficial event which serves to normalize wall stress, whereas prolonged hypertrophy is detrimental, provoking increased oxygen consumption and cardiomyocyte death (Figure 1).

Figure 1.

Canonical “three stages” model of hypertrophic transformation of the heart. This model posits “short-term” hypertrophy as a beneficial event and “long-term” hypertrophy as detrimental.

Considerable evidence, from both preclinical and clinical contexts, indicates that chronic, unremitting stress, as occurs in hypertension or valvular heart disease, leads inevitably to systolic dysfunction^85-89. Moreover, the Framingham Heart Study established an association between ventricular hypertrophy and increased cardiac mortality⁵. However, the inevitability of the transition from hypertrophy to failure has been questioned. Drazner and colleagues^{90, 91} have identified three challenge to this point of view: 1) in animal models of pressure overload, hypertrophy can be blocked without development of heart failure^92-94; 2) in humans, concentric hypertrophy does not uniformly progress to failure in the absence of myocardial infarction^95-97; 3) some hypertensive subjects develop dilated cardiomyopathy apparently without antecedent concentric hypertrophy and without clinical evidence of myocardial infarction⁹⁸. These important caveats are based on the notion that the time course of transition to heart failure in patients, if it occurs, is relatively uniform and can be captured within the time span of epidemiological studies.

In an effort to address these issues, magnetic resonance imaging data from the Dallas Heart Study have been used to propose a 4-tiered classification of left ventricular hypertrophy⁹⁹. This scheme seeks to overcome the limitations of the concentric versus eccentric hypertrophy model by incorporating left ventricle end-diastolic volume as a categorical variable. Future work will determine whether this 4-tiered classification conveys prognostic information regarding prognosis or therapy.

Despite these caveats, the presence of left ventricular hypertrophy per se is unequivocally associated with adverse cardiovascular outcomes independent of the underlying cause⁵. As noted earlier, whereas ventricular hypertrophy may be beneficial in the short-term to minimize wall stress, in the long term it promotes progression to heart failure and other cardiovascular disorders. Numerous epidemiological studies have clearly demonstrated that left ventricular hypertrophy is not benign but rather represents a major risk factor for cardiovascular morbidity and mortality, more robust than other conventional risk factors^{5, 100-103}. Studies from the Framingham Heart Study have demonstrated marked increases in coronary heart disease, heart failure, and sudden cardiac death associated with left ventricular hypertrophy, revealed by either electrocardiographic and echocardiographic approaches^{104,105, 106}.

Together these data clearly identify left ventricular hypertrophy as an important risk factor of cardiovascular disease. However, it is important to highlight that these observations are strictly correlative, and no mechanism(s) of hypertrophy-dependent increase in risk can be inferred. Our knowledge of potential mechanism(s) by which ventricular hypertrophy confers increased cardiovascular risk is limited by the fact that transition from the early, compensatory stage of hypertrophy to the maladaptive phase is poorly characterized.

Is load-induced hypertrophy ever truly “compensatory”?

Traditionally, ventricular hypertrophy has been viewed as an obligatory initial response to pathological stress; only with time does it transition to a disease-promoting event. Consistent with this view, the term “compensatory” connotes an adaptive role of hypertrophy as a consequence of pressure stress based on the premise that it helps normalize ventricular wall stress and myocardial oxygen demand. However, evidence from genetic and pharmacological studies in mouse models of pressure overload have demonstrated that hypertrophic growth is not universally required to preserve cardiac function^{92, 94}.

Shortly after the report¹⁰⁷ that calcineurin is capable of triggering a robust cardiac growth response, we set out to determine whether calcineurin signaling is required for afterload-induced growth of the heart. Mice were exposed to thoracic aortic constriction and injected daily with cyclosporine or vehicle, delivered in an investigator-blinded manner. After three weeks, hearts were evaluated by echocardiography and by post mortem gravimetric analyses. In the animals exposed to cyclosporine, the load-induced growth response was abolished, leading us to conclude that calcineurin activation was, in fact, required for load-induced cardiac hypertrophy in this context. Very surprising to us, however, was the fact that ventricular size and function by echo was normal in the cyclosporine-treated animals. This was true despite the imposition of 70-80mmHg of afterload stress which was persistently present until the end of the entire study; ventricular volumes and contractile parameters were normal, and the animals behaved normally. This astonishing finding was published⁹², and we estimate that this observation, which often goes unnoticed in the reporting of studies, has been replicated independently at least 100 times since then.

These findings, demonstrating that load-induced hypertrophy is not always required to maintain ventricular size and performance, raise the prospect that the hypertrophic growth response may be a relevant target for therapeutic targeting⁴. It is important, however, to recognize that these observations derive largely from genetically engineered mice and, therefore, are based on models with important limitations. Heart rate in humans is substantially slower than that in mice, and left ventricular ejection fraction is lower. In addition, surgical banding of the aorta triggers an acute increase in pressure stress on the left ventricle and therefore does not mimic conditions of chronic, progressively increasing hemodynamic load as occurs in humans. Most preclinical studies in mice were conducted over a 3-4 week period, which amounts to approximately 3 years in humans; perhaps blocking hypertrophy for a longer period of time would be harmful. Unfortunately, the critical study in large animals has not been performed to date.

How can all this be?

Increases in afterload elicit early changes in myocyte biology across a wide range of mechanisms. Frank-Starling-related adaptations ensue, triggered by stretch. In addition, contractility continues to increase for over 10-15 minutes following the initial stretch, a response initially described over 100 years ago by Gleb Von Anrep and now termed the Anrep effect¹⁰⁸. Initially attributed to increases in circulating catecholamines released by the adrenal gland, the Anrep effect was later demonstrated in isolated ventricular myofilaments and was attributed to increases in myofilament calcium sensitivity¹⁰⁹. Thus, the Anrep effect, a response which remains poorly characterized, is a rapid, pro-contractile response that occurs in the setting of abrupt increases in afterload. It is possible that this response, coupled with hypertrophic growth, represents a load adaptation intrinsic to cardiac muscle.

Initial, rapid mechanisms whereby cardiac muscle responds to increased load are, however, insufficient to maintain cardiac contractility for an extended period if the inciting stress is not eliminated. Therefore, increases in cardiac mass ensue but which ultimately lead to untoward consequences. Indeed, with persistent pressure stress, the heart undergoes apparently irreversible decompensation, resulting in chamber dilatation and reduced systolic function. This “maladaptive” hypertrophy, as emerges when the inciting stress is not abated, represents a common feature in virtually all forms of heart failure. Therefore, the concept of transition from a short-term compensatory response to maladaptation points to cardiac hypertrophy as a novel therapeutic target.

Translational studies in large animals

Our understanding of mechanisms that govern the transition between compensated ventricular hypertrophy to heart failure remains incomplete. This stems partly from lack of longitudinal analyses of ventricular structure and function in patients with increased afterload (e.g. aortic stenosis, arterial hypertension). To date, most evidence that provides insight into the effects of suppressing cardiac hypertrophy on cardiac function derives from studies in rodents¹¹⁰. However, due to intrinsic differences in contractile performance, Ca²⁺ handling, myosin isoform distributions, heart rate and lifespan between rodents and humans, these models do not recapitulate faithfully human disease. In this regard, large animal models could be more informative.

Pressure overload-induced cardiac hypertrophy has been studied in non-human primates, where left ventricular hypertrophy and myocardial fibrosis comparable to that seen in patients with aortic stenosis are seen¹¹¹. Studies to evaluate molecular mechanisms and pathophysiological adaptations in pressure overload-induced heart failure have been conducted primarily in dogs^{112, 113,114}. In view of the considerable remaining gaps in our understanding of this biology, additional exploration of the role of compensatory hypertrophy in large animal models is warranted.

How might hypertrophy be targeted?

Clinical management of pathological left ventricular hypertrophy currently focuses on the underlying growth cues (e.g. hypertension, valve disease) and typically involves a wide spectrum of pharmacologic agents that have shown safety and efficacy in reducing hypertrophy. These pharmacological agents are mostly directed against the crucial neurohormonal axes activated in response to stress. Adrenergic and renin-angiotensin-aldosterone systems (RAAS) represent two mechanisms recruited to increase contractility in the early phases of stress, becoming deleterious in the chronic context. β-adrenergic receptor blockers (β-blockers), ACE inhibitors, and angiotensin receptor blockers are effective in reducing left ventricular mass and are associated with a favorable clinical outcome in clinical settings ^{115, 116}.

Pharmacotherapy to reduce blood pressure can lead to regression of ventricular hypertrophy³⁶. Although all antihypertensive drugs promote hypertrophy regression to some extent, current evidence suggests the RAAS-targeting drugs are the most efficacious at regressing ventricular hypertrophy. Based on robust evidence that regression of hypertrophy is independently associated with improved cardiovascular outcome, numerous clinical trials have documented the favorable impact of antihypertensive drug therapy. For example, the Losartan Intervention for Endpoint Reduction in a Hypertension (LIFE) study reported greater reduction in left ventricular mass index in the losartan-treated cohort compared with an atenolol-based regimen¹¹⁷. Several other trials reached a similar conclusion, pointing to a class effect of RAAS-targeting antihypertensive agents at promoting hypertrophy regression^118-121. Together, these data highlight the role of current therapies targeting chronic neurohormonal activation in the prevention of heart failure and limiting progression of ventricular hypertrophy, lending further support to the notion that inhibiting ventricular hypertrophy is an attractive therapeutic option in patients.

All of these trials were conducted using compounds targeting cardiomyocyte cell-surface receptors. More recently, efforts have focused on anti-hypertrophic effects of molecules that act inside the cardiomyocyte, targeting crucial signaling cascades that alter gene expression and protein function. These agents include histone deacetylase (HDAC) inhibitors, a wide spectrum of pro-hypertrophic microRNAs (miRs) (reviewed¹²²), and several other small molecules (reviewed¹²³ [Table]).

Table 1.

Novel compounds and targets with confirmed or potential anti-hypertrophic activity.

	Target(s)/Mechanism(s) of action	Refs. (PMID)
HDAC inhibitors
Apicidin	Class I HDACs	18697792
Vorinostat	Class I, II HDACs	20139990-17211407
Romidepsin	Class I, II HDACs	20139990-21699444
Trichostatin A	Pan-HDACs	16380549-16735673- 12975471
Valproic Acid	Class I HDACs (weak)	16380549
Scriptaid	Pan-HDACs	16735673
SK-7041	Pan-HDACs	16380549
Pro-hypertrophic microRNAs as therapeutic targets
miR-199a	Hif1α/Sirtuin 1	20458739
miR-199b	Dyrk1a	21102440
miR-208a	THRAP1 and myostatin	19726871-21900086
miR-23a	MuRF1	19574461
miR-499	DRP1	22752967
miR-21	Sprouty2 and Spry1	17234972-19043405
miR-24	E2F2	23307820-19748357
miR-27b	PPAR-γ	21844895
miR-350	MAPK11/14 and MAPK8/9	23000971
miR-195	HMGA1	17108080-25100012
miR-221	p27	22275134
miR-212/132	Fox03	23011132
miR-25	SERCA2a	24670661
miR-155	Socs1	23956210
Other compounds
CPG-ODN c274	TLR9 agonist	23638055
Curcumin	P300 inhibitor	18292809
FTY720	Sphingolipid/NFAT inhibitor	23753531
BAY94-8862	MR antagonist	22791416
Cinaciguat	sGC stimulator	22778174
LCI699	Aldosterone synthase inhibitor	21986283
Resveratrol	Antioxidant	23784505
MTP-131	Antioxidant	21620606
TRV120027	AT1 receptor	22891045
LCZ696	Angiotensin/neprilysin inhibitor	23731190
HA-1077(Fasudil)	ROCK	14500337-15840407-15096457
Y-27632	ROCK	12623300-12176125
Sildenafil	PDE5	25139994
Ruboxistaurin	PKCb	15878171

Abbreviations: HDAC: Histone deacetylase; Hif1α: Hypoxia inducible factor1α; DyrkIA: Dual specificity tyrosine-phosphorylation-regulated 1A; THRAP1: Thyroid Hormone Receptor Associated Protein 1; MuRF1: Muscle RING-finger protein-1; THRAP1: Dynamin-related proteini; E2F2: E2F transcription factor 2; PPAR: Peroxisome proliferator-activated receptor; MAPK: Mitogen-activated protein kinase; HMGA1: High mobility group A1; Fox03: Forkhead box 03; SERCA2A: Sarcoplasmic reticulum Ca²⁺ ATPase; Socs1: Suppressor of cytokine signaling 1; TLR9: Toll-like receptor 9; NFAT: Nuclear factor of activated T-cells; MR: Mineralocorticoid receptor; AT1: Angiotensin II receptor type 1; ROCK: Rho-associated protein kinase; PDE5: Phosphodiesterase type 5; PKCβ: Protein kinase C β.

Our group has focused on reversible protein acetylation as a tractable means of governing cardiomyocyte growth^124-127, autophagy¹²⁶, and disease-responsiveness¹²⁸. Among those studies, we have noted that inhibition of HDACs is capable of blunting load-induced growth¹²⁷. Beyond that, HDAC inhibitors can promote regression of ventricular hypertrophy despite the persistent presence of afterload stress¹²⁶ (and with preservation – even improvement – in contractile performance.) HDAC inhibitors, three of which are FDA approved as third-line therapy for Sézary syndrome, may emerge as a novel means of targeting ventricular hypertrophy. Indeed, we have speculated that it might be possible to employ HDAC inhibitors to “sculpt” the hypertrophied ventricle in patients with heart failure with preserved ejection fraction (HFpEF), trimming left ventricular wall thickness progressively, to afford clinical benefit.

Comprehensive understanding of molecular events involved in maladaptive cardiac hypertrophy and the remodeling that culminates in decompensated heart failure is the first step toward developing novel treatments with clinical potential. This is especially true for patients with HFpEF, most of whom harbor ventricular hypertrophy which contributes to symptoms and clinical outcomes¹²⁹ and for which evidence-based therapy is lacking. Although these pathways manifest redundancy in their effects, hypertrophy often remains present in models in which one pathway is suppressed, suggesting that serial pharmacological brakes may be required for clinical gain.

A note of caution

Considerable evidence points to a progression of remodeling events in which stress elicits a 1) hypertrophic growth response which has beneficial features; 2) maladaptive hypertrophy, which is a clear-cut marker for untoward events; and 3) the clinical syndrome of heart failure. However, assuming this model has validity, little is known regarding specific markers of these different phases of disease pathogenesis or transition points separating them. A large number of preclinical studies have demonstrated that it is possible to blunt load-induced hypertrophy, even in the setting of persistent afterload stress, without affecting contractile function^{4, 92, 94}. These studies, then, have delineated a strategy wherein one might target maladaptive hypertrophy and thus obviate the untoward consequences of continued progression of this process. Consistently, in both preclinical studies and clinical trials, inhibition of cardiac hypertrophic growth usually results in the amelioration of left ventricular dysfunction^{5, 9, 130}.

Caution is warranted. Not all forms of pathological cardiac hypertrophy can be blocked without provoking ventricular dysfunction^131-134. Whereas the great majority of studies – again, exclusively in rodents to date – demonstrate that the load-induced growth response can be inhibited without untoward effects, there are examples where hypertrophy elicited by a specific signaling cascade appears to be required^135-137. Also, ventricular mass alone may not provide the full picture of the myriad remodeling events involved in heart growth¹³⁸. Therefore, further work is required to determine whether accompanying alterations (e.g. contractile function, coronary hemodynamics) associated with ventricular hypertrophy can be reversed by treatment and in which category of patients.

We wish to highlight that whereas cardiac hypertrophy in response to pathological stimuli manifests common characteristics irrespective of the triggering stress, it is likely that several subtypes of pathological hypertrophy exist. Thus, whereas efforts to block the inciting stimuli are warranted, we do not know whether all forms of maladaptive hypertrophy should be prevented.

Stress-induced cardiac growth: A model

How is it that hypertrophy can be both beneficial and deleterious? We propose a model in which hypertrophic transformation of the heart under conditions of disease-related stress is similar to many other processes in biology. There has been significant evolutionary pressure to promote organismal survival until procreation can occur and young can be fostered to independence. However, our species did not evolve to live 9 decades! In the settings of life-threatening stress, activation of both the β-adrenergic cascade and the RAAS axis occurs, and the end-result is enhanced cardiac performance and improved survival. These pathways allow for survival in the setting of sublethal injury, for example. However, chronic activation of these processes, life-saving in the short term, clearly conveys markedly enhanced risk in the long term. Indeed, chronic suppression of those evolutionarily ancient neurohumoral responses is fundamental to present day, evidence-based therapy for heart failure. Again, a large number of studies have demonstrated robust benefits from therapeutic strategies in which evolutionary adaptations to stress are blocked.

We suggest that ventricular hypertrophy is the same. It provides short-term benefit and long-term harm (Figure 2). If this model holds true, then suppression of ventricular hypertrophy – a therapeutic target currently not under widespread consideration – emerges as a novel and potentially important target for consideration of drug development going forward.

Figure 2.

Model of the hypertrophic response to toxic stress. We propose that hypertrophic transformation of the heart under stress is similar to other biological processes which evolved to provide short-term benefit. These include activation of the β-adrenergic and renin-angiotensin-aldosterone axes. Cornerstone therapy for heart failure involves interruption of these responses, and we suggest that judicious suppression of hypertrophy may provide similar benefit.

Summary and perspective

Heart failure is defined as a syndrome in which cardiac output is unable to meet the metabolic needs of peripheral tissues. In this setting, when cardiac output is depressed, the myocardium has a limited repertoire of responses, and, as a result, so does the treating physician. The heart is capable of responding in 4 ways: increases in heart rate, ventricular filling, contractility, and mass. Clinically, we have a full spectrum of means to regulate heart rate and optimize ventricular filling pressures. Contractility is a mechanism we target routinely in the acute setting, but no safe and effective therapies are available which promote contractility chronically. This leaves ventricular hypertrophy as the “final frontier” of heart failure therapy, a target which has never been developed and which, in fact, seems counterintuitive on initial consideration.

That said, it is critical to recognize that hypertrophic transformation of the ventricle is just that – a transformation involving cellular dedifferentiation and comprehensive reprogramming of the cardiac myocyte and other cellular elements within the ventricle. Whereas increases in myocyte size, and consequent increases in ventricular mass, are hallmark features, a wide range of additional events occurs in these stressed cells. Here, we present an overview of the large body of robust epidemiological and preclinical data pointing to the untoward consequences of hypertrophic transformation of the myocardium. These data, at the very least, raise the prospect of targeting hypertrophy therapeutically.

In recent years, significant strides have been achieved in our understanding of, and therapeutic targeting of, pathological hypertrophic remodeling. We view hypertrophic transformation as a fundamental, potentially indispensable step in the myocardial response to pressure overload which is coupled with significant detrimental consequences. This response may be beneficial in the short term, but when maintained chronically it becomes significantly detrimental. It is analogous to the responses of β-adrenergic and RAAS activation, which we now go to great lengths to block and whose suppression seemed counter-intuitive at first. We argue that hypertrophy is comparable, and patients may benefit from judicious attenuation of this chronic, long-term hypertrophic response.