Cardiac Atrophy and Remodeling

Overview of Atrophic Cardiac Remodeling

Cardiac remodeling encompasses the many biochemical and molecular adaptations that the heart initiates in response to altered demand ( Figure 3.1 ). Atrophic remodeling occurs in response to ventricular or hemodynamic unloading caused by such stimuli as placement of an LVAD, surgical correction of cardiac valvular disease, bariatric surgery or pharmacological intervention in patients with hypertension. Upon unloading, the myocardium decreases in size but contractile function is generally maintained. Reduced cardiomyocyte size is responsible for the decrease in myocardial mass; cell death does not significantly contribute to the phenotype. Because of the interest in the molecular and functional pathology involved in reversing cardiomyocyte size, the mechanisms responsible for atrophic remodeling have been studied in both patients and in experimental animal models of ventricular unloading. In particular, the hearts of patients receiving LVADs and experimental animals receiving a heterotopic heart transplant have been examined.

Surprisingly, many of the mechanisms and expression of molecular hallmarks that result from these stimuli are shared between hypertrophic and atrophic cardiac remodeling despite the opposing trophic adaptation. Cellular processes regulating cardiac gene expression are altered including increased signaling through pathways that promote proteolysis as well as either increased or decreased activity of pro-growth pathways. Proteolytic pathways are generally more highly activated and therefore, degradative processes are overall favored within atrophic cardiomyocytes. The mechanisms mediating cardiac atrophy are highly stimulus-specific. For example, the ubiquitin-proteasome system (UPS), which is principally responsible for protein degradation in the heart, is up-regulated while mTOR-mediated protein translation is simultaneously activated in heterotopic heart transplantation in rats. However, in rabbits that exhibit cardiac atrophic remodeling due to fasting, global protein degradation is increased while synthesis of myofibrillar proteins is inhibited. Finally, fetal gene expression is up-regulated, metabolic genes mediating glucose transport are down-regulated, and expression of sarcomeric proteins is dysregulated similarly in atrophy and hypertrophy.

We will present an overview of general mechanisms mediating cardiac atrophy followed by several examples of atrophic remodeling with different stimuli. We will focus first on mechanisms mediating the ability of the heart to detect alterations in hemodynamic load and the morphological and histological changes observed in atrophic remodeling. We will then discuss the biochemical and molecular pathways responsible for protein degradation and synthesis and for metabolism in the atrophic myocardium. Finally, we will present clinical and molecular data specific to LVAD- and cancer-induced atrophic remodeling.

Models of Atrophic Remodeling

Although initially developed to examine the process of immune rejection of transplanted organs, heterotopic heart transplantation has become the main model of ventricular unloading in experimental animals. In this model, a healthy heart is transplanted into the abdominal cavity through anastomosis of the ascending aorta and pulmonary artery associated with the donor heart to the recipient abdominal aorta and inferior vena cava, respectively (Figure 3 . 2 A ). The effect of the transplantation is similar to aorto-caval fistula in that blood is diverted into the donor heart, thus reducing pressure entering the heart of the recipient. This technique is commonly used to examine the temporal, molecular, and biochemical responses to ventricular unloading that parallels the cardiac unloading seen in patients with LVADs that mechanically unloads the heart.

Permanent implantation of portable LVADs began in the 1980s as a bridge to heart transplantation for patients diagnosed with refractory heart failure. Recently, LVADs have been transplanted for longer periods with the aim of promoting reversal of cardiac pathology and removing the LVAD as an alternative to heart transplantation (see ‘Atrophic remodeling as a potential therapeutic,’ below for additional details). LVADs are battery-operated pumps that are implanted into the patient’s abdominal wall or peritoneal cavity; battery packs are worn externally (Figure 3 . 2 B). Two conduits are implanted: one into the left ventricular apex and one into the ascending aorta to unload the left ventricle by moving blood out and bypassing the systemic circulation. Generally, LVADs can respond to cardiac demand and are tolerated well; complications are mostly limited to infection, bleeding, and stroke. Much of what is known about atrophic cardiac remodeling in patients was learned from studying patients receiving LVADs.

Cardiac Workload Determines Cardiac Size

Although the heart has a large capacity to increase in mass, there is an upper limit mainly due to the lack of significant proliferative capacity of terminally differentiated adult cardiomyocytes. Contractility may be changed, muscle mass may be altered, or actin–myosin crossbridge formation may be adjusted through differential expression of sarcomeric proteins. Each of these mechanisms is in place to offset abnormal transmural pressure that causes or is a result of inefficient cardiomyocyte contractility. In atrophic remodeling, cardiomyocytes respond to reduced cardiac load by utilizing these mechanisms.

Individual myocardial cells must adapt to a new pattern of force and distortion during ventricular unloading. This response is mediated by complex interactions between the extracellular matrix (ECM) and cells in the myocardium. Components of the complex ECM that surround both myofibrils and cardiomyocytes are physically connected to the interior of the cardiomyocyte to transduce force, which also allows the translation of force into biochemical signals through the association of adaptor and signaling proteins. The signal is therefore translated through altered protein–protein interactions and is conveyed to the nucleus where gene expression is modified and functional alterations are initiated.

Numerous proteins participate in mechanosensation and mediate initiation of signaling pathways that alter cardiomyocyte form and function. At the costamere, melusin and integrin-linked kinase (ILK) interact with the cytoplasmic tail of β1-integrin, a predominant component of the cardiac ECM. ILK recruits several adaptor proteins to the costamere and activates downstream signaling proteins to alter gene expression, thus contributing to regulation of cardiomyocyte size. Integrin activation also initiates focal adhesion kinase (FAK) signaling to induce transcriptional activity through mitogen-activated protein kinase (MAPK) and c-Jun N-terminal kinase (JNK) as well as Rho, which promotes actin reorganization. Activation of these signaling pathways induces alterations in gene expression that allow efficient use of metabolic substrates and maintain cardiac contractility (discussed below). As such, mechanical stress largely determines the morphological and functional plasticity of the cardiomyocyte.

Although the ability of clinicians to directly measure regional transmural pressure would assist in predicting cardiac remodeling and potentially avert functional decompensation and heart failure, accurate measurements of transmural pressure in vivo are limited by the mechanical complexity of the heart. However, mathematical modeling to quantify morphological alterations dictating intraventricular pressure and wall stress is possible using the law of Laplace. This principle demonstrates that wall thickness is indirectly proportional to ventricular wall stress and therefore makes accurate predictions about cardiomyocyte size in response to pressure changes; as wall stress is reduced through a stimulus like unloading, cardiomyocyte size is reduced to normalize (increase) transmural pressure.

The law of Laplace has been exploited in the development of medical devices that are currently used to mimic increased hemodynamic load by physically changing the shape of the heart. For example, artificial alteration of transmural pressure using an Acorn support device that constricts the heart with a mesh cylinder, or the Myocor Myosplint that pulls the sides of the ventricle toward one another, augments signal transduction consistent with increased mechanical stress in cardiomyocytes. These studies reported promising results suggesting that physically manipulating heart size and, by extension, altering transmural pressure can limit morphological and molecular dysfunction caused by atrophic remodeling.

Morphological Features of the Atrophic Heart

Interestingly, in contrast to the mechanisms mediating ventricular wall thinning and chamber dilation such as cell death and cardiomyocyte elongation, atrophic remodeling is a reversible process that maintains cardiac function by reducing cardiomyocyte mass. Cardiac atrophy can be identified in vivo using echocardiography to visualize loss of myocardial mass or histologically by measuring myocyte cross-sectional area. With mechanical unloading, reduced myocardial size is attributable to a reduction in cardiomyocyte mass rather than to cell death. Reduced cardiomyocyte volume has been attributed to degradation of sarcomeric proteins in parallel, rather than preferential loss of a subset of proteins. Accompanying this reduction is an alteration in the innervation of the left ventricle such that fewer than half the number of ramifying axons are observed compared to a healthy heart. Myofibrillar disarray caused by the altered relationship between cardiomyocytes and the ECM is also present along with interstitial fibrosis, both considered to be deleterious to cardiac function. The net effect is reduced myocardial mass and increased wall stiffness in the atrophic heart.

Extracellular Matrix Remodeling with Cardiac Atrophy

The ECM of cardiac tissue is comprised of type I collagen, elastin, proteoglycans, and glycoproteins and is integral to the maintenance of tensile strength and myofibrillar organization of cardiac muscle. However, this collagenous matrix is not merely a passive structural component of the myocardium; it participates in mechanosensation by physically connecting the ECM to intracellular cytoskeletal elements in cardiomyocytes, converting mechanical information to biochemical signals that promote functional alterations in cardiac output, and organizing cardiomyocytes into functional units responsible for efficient contraction. Because intimate connections between the cardiomyocyte and ECM are required to coordinate contraction of the myocardium, cardiac structural remodeling cannot be limited to the cardiomyocytes; decreased cardiomyocyte size alters the relationship between the cell and the ECM. For this type of architectural change to take place, degradation of existing matrix proteins followed by expression of new matrix proteins must occur. Expression of matrix metalloproteinases (MMPs) that degrade ECM proteins is regulated transcriptionally in response to alterations in myocardial stretch but do not act unopposed. Tissue inhibitors of metalloproteinases (TIMPs) form complexes with MMPs and inactivate the catalytic domain of MMPs. Therefore, TIMP expression in the myocardium plays an important role in modulating the activity of MMPs. Interestingly, although MMPs are up-regulated in heart failure, expression of MMPs is decreased and TIMPs are increased in rats following heterotopic heart transplant.

Several soluble regulatory factors are released in the atrophic myocardium by cardiomyocytes, fibroblasts, and infiltrating immune cells, and induce increased production of the cardiac matrix with cardiac remodeling. Among these factors, the cytokine osteopontin is required for fibroblast differentiation into α-smooth muscle actin-expressing myofibroblasts that produce ECM components, which contribute to cardiac fibrosis. Expression of osteopontin is increased in the mechanically unloaded human heart. Expression of the fetal genes atrial natriuretic factor (ANF) or B-type natriuretic peptide (BNP, discussed below), which are also up-regulated in atrophic remodeling, has been shown to oppose myofibroblast activity, supporting the conclusion that initial responses to hemodynamic unloading may be cardioprotective in nature.

Activated myofibroblasts enhance the synthesis and secretion of type I fibrillar collagen. Matrix production is required to preserve alignment of cardiomyocytes and maintain the structural integrity of the myocardium. However, increased expression of matrix proteins in atrophic remodeling causes myofibrillar disarray ( Figure 3.3 ), and excessive ECM negatively affects cardiac function through its contribution to tissue stiffness. The transition from atrophic remodeling to heart failure is likely attributable to the development of fibrotic foci in the myocardium. In fact, fibrotic tissue emanating from scarred myocardium entangles cardiomyocytes and can exacerbate atrophy and actively promote progression to heart failure.

Excessive myocardial matrix production also negatively impacts cardiac function and propensity for arrhythmias due to conduction slowing or block (electrical remodeling). The stiffness of cross-linked type I collagen may also restrict arterioles responsible for delivering sufficient amounts of oxygen to the highly metabolic cardiac tissue, leading to a metabolic crisis secondary to mild hypoxia. Because myofibroblasts are physically connected to cardiomyocytes via gap junctions, abnormal electrical coupling between myofibroblasts and cardiomyocytes can cause depolarizing currents in neighboring cardiomyocytes. Although it may seem logical to disrupt the development of fibrosis as a therapeutic intervention to prevent increased stiffness, myofibrillar disarray, and contractile dysfunction, it is important to emphasize that the development of fibrosis is thought to be important in healing; attempts thus far to inhibit fibrosis have led to increased lethality in experimental animals.