Functions of autophagy in pathological cardiac hypertrophy.

Pathological cardiac hypertrophy is the response of heart to various biomechanical and physiopathological stimuli, such as aging, myocardial ischemia and hypertension. However, a long-term exposure to the stress makes heart progress to heart failure. Autophagy is a dynamic self-degradative process necessary for the maintenance of cellular homeostasis. Accumulating evidence has revealed a tight link between cardiomyocyte autophagy and cardiac hypertrophy. Sophisticatedly regulated autophagy protects heart from various physiological and pathological stimuli by degradating and recycling of protein aggregates, lipid drops, or organelles. Here we review the recent progresses concerning the functions of autophagy in cardiac hypertrophy induced by various hypertrophic stimuli. Moreover, the therapeutic strategies targeting autophagy for cardiac hypertrophy will also be discussed.

Introduction

Pathological cardiac hypertrophy refers to the enlargement of heart in a variety of diseases e.g. aging, myocardial ischemia and hypertension 1. At cellular level, pathological cardiac hypertrophy is characterized by an increase in cardiomyocyte size, disarrangement of sarcomeric structure, enhanced protein synthesis and re-expression of fetal genes 2. Although pathological cardiac hypertrophy initially compensates for the biomechanical and physiopathological stimuli, it progresses generally to detrimental cardiac remodeling and heart failure, which is a clinical risk factor for death 3, 4. The mechanisms of pathological cardiac hypertrophy have been studied at multiple levels, such as genetic mutations in structural proteins, alterations of molecular signaling pathways, detrimental changes in subcellular organelles and communications among various cell types in hearts 5-8. Further understanding the mechanisms underlying pathological cardiac hypertrophy would provide novel therapeutic strategies for the treatment of heart failure.

Autophagy is a conserved catabolic process by which cellular components are transported to and degraded in the lysosomes 9. Autophagy occurs in response to various stimuli, such as nutrient limitation and cellular stress 10, 11. Not only does autophagy play essential roles in tissue homeostasis by degradating long-lived proteins and unwanted organelles 12, it also acts as a recycling system to produce new energy and materials for the renovation of cells 13. Constitutive autophagy is required for cell survival, while excessive autophagy triggers autophagic cell death, indicating dual roles of autophagy in pro- or anti-survival potential 14. Particularly, studies by using autophagy-defective animal models have elucidated the significant physiopathological functions of autophagy at whole-organism or tissue-specific levels. Dysregulation of autophagy is associated with many clinical disorders, such as cancer, neurodegeneration, infection and metabolic syndromes 12, 15.

Recently, a number of studies have begun to show that autophagy plays important roles in cardiac hypertrophy under various pathological stimuli, thus indicating a possible link between autophagy dysregulation and cardiac hypertrophy. In this review, we focus on how autophagy is regulated and what are the effects of altered autophagy on cardiac hypertrophy. In addition, the possibility of autophagy as a therapeutic target for cardiac hypertrophy will also be highlighted.

Molecular Basis of Autophagy Process

Autophagy could be either a non-specific degradation of cytoplasmic cargo (general autophagy) or a specific removal of unwanted organelles (selective autophagy), such as the clearance of mitochondria (mitophagy), peroxisomes (pexophagy), and lipid droplets (lipophagy) 12. In both processes, intracellular components are first surrounded by double membrane-bound autophagic vesicles which then fuse with lysosomes to form autolysosomes for the degradation 16. Autophagy process is governed by distinct functional complexes of autophagy-related gene (ATG) products. The induction of autophagy is relied on the unc-51 like kinase (ULK) complex stimulated by upstream nutrient-sensing kinases. Starvation or growth factor deprivation leads to the activation of ULK1/2 kinase, thus promoting autophagy induction 17-19. After induction, the class III PI3K-Beclin1-VPS34 complex is activated by ULK complex 20 and recruited to the mitochondria, endoplasmic reticulum (ER), ER-mitochondria contact sites, Golgi apparatus, recycling endosome or plasma membrane to form an isolated membrane, which is called phagophore nucleation 21-25. Apparently, the phagophore originates differently in different cell types and different physiologic or stress-related conditions. The phagophore elongation and formation of complete autophagosomes require two ubiquitin-like protein conjugation systems: the one is the ATG12-ATG5 conjugate which functions as a dimeric complex together with ATG16L1; the other is phosphatidylethanolamine (PE)-conjugated LC3, which is cleaved by ATG4 and subsquencely activated by ATG7 and E2-like enzyme ATG3. The mature autophagosome then fuse with lysosome, which is facilitated by the lysosomal membrane protein LAMP2 and the small GTPase RAB7. After fusion, lysosomal acid hydrolases, including Cathepsin-B, Cathepsin-D and Cathepsin-L, participate in the degradation of its contents 14, 26, 27.

Autophagy usually occurs at low basal levels to maintain the cellular homeostasis, and could be stimulated in response to various cellular stresses, such as nutrient limitation, oxidative stress and ER stress. Nutrient limitation exists in many pathological conditions and serves as one of the most typical triggers of autophagy. In mammalian cells, there exist two kinds of energy sensors, the short-term kinases and long-term transcriptional factors, to make accurate autophagic responses according to the status of energy metabolism. The AMP-activated protein kinase (AMPK) and mammalian target of rapamycin (mTOR) are two major short-term energy sensors for the rapid regulation of autophagy. AMPK is activated in response to low ATP/AMP ratio and promotes the induction of autophagy 28, 29. mTORC1 is another important energy sensor to both amino acids and glucose, and acts as a central mediator of multiple autophagy-regulating signals. Nutrient inadequacy promotes autophagy by the inhibition of mTORC1 29, 30. In addition, autophagy is also regulated by long-term transcriptional factors. The nuclear-sensing receptors peroxisome proliferator-activated receptor-α (PPARα), cAMP response element-binding protein (CREB) and farnesoid X receptor (FXR), share common binding sites in many autophagic gene promoters and collaborate to regulate autophagy as key physiological switches 31, 32.

Oxidative stress, caused by overproduction of reactive oxygen species (ROS) excess to cellular antioxidant ability, plays vital roles in activating autophagy 33. On one hand, ROS inhibits ATG4 activity via oxidation of an essential cysteine residue, which blocks the cleavage of PE from PE-conjugated LC3 and promotes autophagy 34. On the other hand, the increased ROS levels result in the loss of membrane potential and the subsequent activation of Parkin-dependent mitophagy 35, 36.

ER stress is another trigger of autophagy in mammalian cells. ER stress happens upon a wide variety of physiopathological stimuli which disrupt protein folding capacity of ER and cause subsequent accumulation of unfolded or misfolded proteins 37. ER stress promotes autophagy by two means in accordance with its dual roles in regulating unfolded protein response (UPR) and Ca2+ handling. UPR activates autophagy via regulation of autophagy- related kinases or by ATF4-dependent transcriptional regulation 37, 38. Ca2+ release from ER promotes autophagy via activation of several Ca2+-dependent signaling pathways 39.

Roles of Autophagy in Cardiac Hypertrophy

Autophagy plays complicated but indispensable roles in the maintenance of cardiac homeostasis 40. On one hand, autophagy can act as a pro-survival factor. Firstly, autophagy promotes the degradation and recycling of long-lived proteins, lipid drops, or even damaged organelles, demonstrating a survival response to energy deprivation during cardiac hypertrophy. Secondly, the autophagy-lysosome system collaborates with ubiquitin-proteasome system in the protein quality control, which is essential for preventing cells from cytotoxicity caused by misfolded protein aggregates. Thirdly, autophagy participates in the clearance of damaged mitochondria accumulated in hypertrophic hearts, thus eliminating the ill effects of mitochondria-derived ROS on cardiomyocytes. On the other hand, autophagy can trigger type-II programmed cell death, which makes it a pro-death factor (Fig.1).

Figure 1

Dual roles of autophagy in cardiac hypertrophy. Baseline autophagy maintains cardiac structure and function by energy supply, ROS clearance and protein quality control, while excess autophagy is detrimental to the heart by enhancing cell death.

Pathological cardiac hypertrophy is accomplished by dysfunction of numerous hypertrophic related signaling pathways, many of which are also broadly involved in the regulation of cardiomyocyte autophagy 41-46. Ca2+/Calcineurin signaling pathway, which promotes cardiac hypertrophy by synergistically activating hypertrophic gene transcription, has been shown to inhibit cardiomyocyte autophagy in an AMPK-depedent manner 41, 42. PI3K/AKT pathway are involved in cardiac hypertrophy by two well-established downstream targets, glycogen synthase kinase-3 (GSK-3) and mTOR, both of which are main regulators of cardiomyocyte autophagy 43-45. Mitogen-activated protein kinases (MAPKs) singaling, another significant hypertrophic signaling pathway, is also implicated in cardiomyocyte autophagy and mitophagy 46. These studies suggest the link between cardiomyocyte autophagy and cardiac hypertrophy.

Furthermore, the experiments of disruption of basal autophagy in mouse heart establish the direct link between autophagy and cardiac hypertrophy 47-49. Deletion of Atg5 specifically in adult hearts causes severe cardiac hypertrophy and dysfunction 47. Similar results are obtained by ablation of Vps34 in the hearts, which show increased left ventricular wall thickness and mass, and decreased cardiac contractility with lower ejection fraction and fractional shortening 48. In addition, loss of Mcl1, an anti-apoptotic BCL2 protein, inhibits autophagy and leads to rapid contractile dysfunction, cardiac hypertrophy and early mortality accompanied by cardiac fibrosis and inflammation 49.

Autophagy in aging-related cardiac hypertrophy

Autophagy acts as a defense mechanism to protect cardiomyocytes from aging, and thus plays pivotal roles in the amelioration of aging-related cardiac hypertrophy (Fig.2). The heart develops cardiac hypertrophy and diastolic dysfunction with aging when the autophagic activity is inhibited 50, 51. Growing evidence suggests that the mechanisms involve signaling controlling cardiac aging, including NAD+-dependent class III histone deacetylases Sirtuins (SIRT), and endothelin-1 (ET-1) 52, 53. The impaired SIRT1 activity might be an important cause for the inhibition of autophagy in aging hearts, as SIRT1-regulated AKT signaling and ATG protein activity are all necessary for the normal autophagy maintenace 52, 54-56. In addition, the increased plasma level of endothelin-1 (ET-1) with aging also triggers a decrease in autophagy 53.

Figure 2

The effects of autophagic alteration on hearts under different hypertrophic stimuli. The autophagic activity is decreased in aging hearts but is increased in ischemic hearts. Inhibition of autophagy in mice is detrimental to, while induction of autophagy is beneficial for hearts suffering aging or ischemia. In the mouse model of hypertensive cardiac hypertrophy, autophagy is inhibited in the early stage but is induced in the late stage. Induction of autophagy is demonstrated to be detrimental to hypertensive cardiac hypertrophy, while contradict results concerning the effects of inhibited autophagy on hypertensive cardiac hypertrophy are reported. (↑: induction; ↓: inhibition; KO: knockout mice; TG: transgenic mice; KD: gene knockdown; DN: dominant negative transgenic mice)

Autophagy inhibition can induce aging-related cardiac hypertrophy 57-60. Cardiac-specific Atg5-deficient mice show inhibited autophagy, shortened lifespan and aging-related cardiomyopathy, characterized by increased left ventricular dimension and decreased fractional shortening 57. Gsk3α knockout mice demonstrate suppressed autophagy in an mTOR-dependent manner and develop cardiac hypertrophy and contractile dysfunction with age 58. Deficiency of UVRGA, a protein which promotes autophagosome formation by binding directly to Beclin1-VPS34 complex and enhancing VPS34 activity, impairs autophagic flux and leads to age-related cardiomyopathy accompanied by enhanced inflammatory response 59. In addition, loss of Atrogin-1 in mice fails to degrade CHMP2B, resulting in autophagy impairment and cardiomyopathy characterized by myocardial remodeling with reduced diastolic function and arrhythmias 60.

Induction of autophagy can reverse aging-associated cardiac hypertrophy and contractile dysfunction 53, 61, 62. It has been shown that p53 can impair autophagy by disrupting Parkin translocation to damaged mitochondria and subsequent mitophagy. Parkin-overexpression or p53-deficiency ameliorates the functional decline in aging hearts accompanied by decreased senescence-associated β-galactosidase activity 61, 62. Cardiomyocyte endothelin-1 receptor ETA deletion induces autophagy and alleviates aging-associated cardiac hypertrophy 53.

All these studies demonstrate the essential roles of autophagy in the maintenance of cardiac homeostasis during aging, suggesting the therapeutic potential of activation of autophagy for cardiac aging.

Autophagy in myocardial ischemia-induced cardiac hypertrophy

Autophagy contributes to cardiomyocyte survival and protects hearts from myocardial ischemia-induced cardiac pathological remodeling and dysfunction (Fig.2). Myocardial ischemia impairs cardiac function, which is due to inadequate nutrient and oxygen supply caused by the blockage of the coronary arteries. The ischemic hearts are exposed to starvation and hypoxia, which stimulate autophagy via AMPK activation and Bnip3 upregulation respectively 63-66. Furthermore, the oxidative stress and ER stress, two significant triggers of autophagy mentioned aboved, are also induced by ischemia 67, 68.

Inhibition of autophagy could promote injury in myocardial ischemia 69, 70. Knockdown of Atg5 or Lamp2 in vitro reduces ATP content and diminishes survival of cardiomyocytes subjected to anoxia 69. Administration of autophagy inhibitor bafilomycin A1 significantly increases infarct size in a mouse model of acute myocardial infarction 69. In addition, Parkin deficient mice are more sensitive to myocardial infarction and present decreased mitophagy, enlarged left ventricular and larger infarcts after the infarction 70.

Stimulation of autophagy plays protective roles in myocardial ischemia. STAT1 deficient hearts show increased levels of autophagy and significantly decreased infarct size following myocardial infarction, which is reversed by pre-treatment with the autophagy inhibitor 71. Mst1 suppresses autophagy via phosphorylating Beclin1 and enhancing interaction between Beclin1 and BCL-2/BCL-XL. Activation of autophagy is found in Mst1 knockout and dominant negative transgenic mice, which reduces MI scar contraction, attenuates left ventricular enlargement and improves LV function 72, 73.

Autophagy in hypertension-induced cardiac hypertrophy

There exist controversies concerning the function of autophagy in hypertension-induced cardiac hypertrophy (Fig.2). In clinic, hypertension is one of the most common causes of cardiac hypertrophy and heart failure. Hypertensive cardiac hypertrophy is due to not only the mechanical stress from elevated blood pressure but also the alternations of neurohormones, growth factors and cytokines. Autophagic level is changed during different periods of transverse aortic constriction (TAC)-induced cardiac hypertrophy 47. Autophagy is suppressed in hypertrophied hearts after TAC for 1 week, and is upregulated in failing hearts after TAC for 4 weeks, revealed by alteration of LC3-II expression levels 47. Renin-angiotensin-aldosterone system (RAAS) is a hormone system that regulates blood pressure and shows dual effects on autophagy via its two receptors, angiotensin II receptor, type 1 (AGTR1) and AGTR2 74. Moreover, complicated regulatory effects of numerous hypertrophic signalings might be responsible for the phase-dependent change of autophagy 41-46. Disruption of autophagy-regulating genes is also associated with TAC-induced cardiac hypertrophy. Cardiac-specific deficiency of Atg5 early in cardiogenesis shows no pathological cardiac phenotypes at baseline, but displays severe cardiac dysfunction and left ventricular dilatation 1 week after TAC operation 47. Cathepsin-L deficiency impairs autophagosomal content degradation and worsens ventricular remodeling and heart failure in response to TAC 75. These results suggest the protective role of autophagy in pressure overload-induced cardiac hypertrophy. However, another study on Beclin1 draws an opposite conclusion. They find that TAC induces autophagic activity from 1 day till 3 weeks. Mice haploinsufficient for Beclin1 are resistant to TAC-induced pathological remodeling of the left ventricle. While cardiomyocyte-specific overexpression of Beclin1 promotes autophagy and pathological ventricular remodeling caused by TAC 76. The study suggests that TAC-induced cardiac autophagy is a maladaptive response.

Autophagy as a Therapeutic Target for Cardiac Hypertrophy

Autophagy is a must for the maintenance of normal cardiac performance, and either inadequate or excessive autophagy in different cardiac pathological conditions is a maladaptive response. Efforts have been made to elucidate the possibility of tuning autophagy as a therapeutic target of cardiac hypertrophy.

Administration of regulators of autophagy related pathways has been shown to have the therapeutic potential for treating cardiac hypertrophy. A number of studies indicate that induction of autophagic activity is benefit for heart diseases. As the inhibitor of mTOR, rapamycin can prevent cardiac hypertrophy induced by TAC or thyroid hormone treatment, and reverse cardiac hypertrophy and dysfunction in aging mice 77-80. Administration of AICAR or metformin, activators of AMPK, in a TAC mice model induces autophagy, inhibits cardiac hypertrophy and improves heart function 81. Overexpression of Atg7 rescues the CryABR120G autophagic deficiency, and attenuates the accumulation of misfolded proteins and aggregates in cardiomyocytes 82. Furthermore, Sustained Atg7-induced autophagy in the CryABR120G hearts decreases cardiac hypertrophy and interstitial fibrosis, ameliorates ventricular dysfunction 83. Some studies show that inhibition of autophagy can also be benefit for heart diseases. Histone deacetylases inhibitor trichostatin A abolishes TAC-induced autophagic induction and blunts pressure overload-induced cardiac hypertrophy 84.

Recent studies have demonstrated that miRNAs play important roles in regulating cardiac autophagy 85-88. Targeting autophagy-regulating miRNAs by chemically engineered oligonucleotides mimics/antagomirs has been used to treat cardiac hypertrophy. Intravenously injection of specific antagomirs against miR-132, an autophagy-inhibiting miRNA, rescues pressure overload-induced cardiac hypertrophy and heart failure in mice 87. Intra-myocardial injection with miR-99a-overexpressing lentiviral vectors increases autophagy via inhibiting mTOR signaling, and improves cardiac function and survival after myocardial infarction 88. These data suggest that autophagy-regulating miRNAs might be novel therapeutic targets for stress-triggered cardiac hypertrophy.

Conclusions and Perspectives

Autophagy is a catabolic process essential for maintaining cardiac homeostasis in response to various stresses. Dysregulation of autophagy under normal or pathological conditions leads to severe cardiac hypertrophy and heart failure. More importantly, the potential of autophagy as a therapeutic target for cardiac hypertrophy has been demonstrated. Nevertheless, further study is needed to decipher the regulatory networks by which cardiac autophagy is sophisticatedly regulated. New regulators of cardiac autophagy should be explored. Long non-coding RNAs have been demonstrated to regulate heart development 89, 90 and cardiac hypertrophy 91, but their function in the regulation of cardiac autophagy needs to be further investigated. Hopefully, a better understanding of the molecular mechanisms controlling cardiac autophagy will eventually provide new promising therapeutic strategies for patients with cardiac hypertrophy.