Adiposity of the Heart*, Revisited

The unrelenting obesity epidemic is one likely explanation for the recent adverse secular trends in cardiovascular morbidity and mortality rates in the United States (1, 2). Hospitalizations for congestive heart failure have increased, and the steady decline in coronary heart diseaserelated deaths since the 1950s has leveled off (3). The recent obesity epidemic poses a major threat to human health in the United States because these persons will be predisposed to a burden of major chronic disease (1, 2). Obesity has both metabolic and cardiovascular health consequences; in particular, obese individuals are at much greater risk for type 2 diabetes and cardiovascular disease (3, 4). Obesity is traditionally considered to be an indirect cause of heart disease. Obese persons typically present with several Framingham risk factors, including hypertension, dyslipidemia, and diabetes mellitus. These risk factors predispose the patient to myocardial infarction that, in severe cases, results in ischemic cardiomyopathy (4). In addition to an elevated Framingham risk score, the hemodynamic hallmarks of obesity are increased heart rate and stroke volume (5). This hyperdynamic circulation is thought to be a compensatory adaptation to increased adipose tissue mass at the expense of eccentric left ventricular remodeling. In extreme obesity, this condition can progress to nonischemic dilated cardiomyopathy (2, 6). In contrast to these 2 rather traditional concepts, an emerging body of basic research is revisiting a previous hypothesis (7, 8): that fat is a direct cardiotoxin (9, 10). In 1933, the original autopsy studies of Smith and Willius (8) suggested that fatty degeneration of the heart is a common consequence of obesity and a possible cause of dilated cardiomyopathy in humans. After Alexander and colleagues (11) called this theory into question in the 1960s, the issue lay dormant for the next several decades (9). Now a growing body of evidence is revisiting the hypothesis that excessive deposits of lipids within myocardial tissue (that is, cardiac lipotoxicity) is an important but forgotten cause of nonischemic dilated cardiomyopathy in humans (12, 13). Under healthy conditions, most triglyceride is stored in adipocytes; the amount of triglyceride stored in nonadipocyte tissues (such as the pancreas, liver, and myocardium) is minimal and very tightly regulated. Various genetic rodent models of obesity have shown that cytosolic triglyceride accumulates excessively in these organs (termed steatosis) when this regulation is disrupted. This accumulation has been implicated in activating adverse signaling cascades that culminate in irreversible cell death (termed lipotoxicity) and lead to several well-recognized clinical syndromes (13). These include nonalcoholic hepatic steatosis; pancreatic -cell failure in type 2 diabetes; and most recently, dilated cardiomyopathy (Figure 1). Figure 1. Concept of lipotoxicity. bottom The purposes of this article are to review recent basic animal research that demonstrates direct toxic effects of lipid accumulation on the myocardium and to highlight emerging efforts to translate this work into the clinical setting by using novel cardiac magnetic resonance imaging and spectroscopy technology. The results of this research could provide insight into the pathogenesis of heart disease in obese humans and guide the development of a novel biomarker and drug target for the prevention of heart failure in these persons. Steatosis in Rodents The seminal research that showed a role for steatosis in obesity-related organ dysfunction was performed with the Zucker diabetic fatty rat, which is a genetic model of progressive type 2 diabetes (14-16). In this obese rodent, type 2 diabetes developed secondary to a loss-of-function mutation in tissue receptors for leptin, the adipocyte-derived hormone that regulates appetite and body weight (17). This model of genetic obesity is more extreme than the milder leptin resistance that commonly accompanies dietary obesity in humans (17). Initial studies demonstrated that pancreatic steatosis directly caused islet cell failure and the subsequent hyperglycemia that characterized this model (14-16). Although leptin was generally thought to act centrally to regulate caloric intake and energy expenditure (17), a series of studies provided experimental evidence that leptin also acts directly on the pancreatic islet cells to stimulate fatty acid oxidation, thereby limiting cellular triglyceride accumulation (18). These findings suggested that leptin signaling is also essential in regulating peripheral lipid stores. Furthermore, the investigators described a pathway whereby failure of the leptin receptor led to excessive cytosolic accumulation of triglyceride and its by-product, ceramide, within islet cells. This accumulation activated the inducible form of nitric oxide synthase, which accelerated cell death (apoptosis) and failure of the -cell (14, 15). Interventions that stimulated free-fatty acid oxidation, like restoration of leptin signaling or thiazolidinedione therapy, effectively attenuated triglyceride accumulation in islet cells and prevented the onset of type 2 diabetes (19). These findings provided evidence that steatosis is an integral determinant of -cell failure in the pathogenesis of obesity-associated type 2 diabetes. In addition to pancreatic -cell failure, the Zucker diabetic fatty rat experienced age-related cardiac dysfunction that was characterized by eccentric left ventricular remodeling, increased left ventricular pressure, and decreased systolic performance (9, 20). The abnormalities in cardiac structure and function are accompanied by a 2-fold increase in myocardial triglyceride content and ceramide that is similar to the accumulation seen in islet cells. Myocardial DNA laddering, which is a marker of apoptosis, is also increased (9). Of note, early administration of thiazolidinedione therapy is effective in attenuating myocardial triglyceride accumulation and normalizing left ventricular contractile performance (9, 20), as shown in Figure 2. Because reduced myocardial lipid content and improved cardiac structure and function were observed independent of changes in body weight, they strongly suggest a role for myocardial steatosis in obesity-related cardiomyopathy. Figure 2. Myocardial lipotoxicity in the Zucker diabetic fatty rat. Top panel. white bars light gray bars dark gray bars Bottom panel. The extreme obesity in the Zucker rat model makes it difficult to determine whether the cardiac maladaptations are related to excessive myocardial lipid accumulation or to increased expression of conventional risk factors for cardiovascular disease. To address this limitation, various lean genetic mouse models of cardiac-restricted steatosis have recently been developed (10, 21-29). These animals display diffuse myocardial lipid content in the absence of obesity or any other traditional cardiovascular risk factors, thereby allowing researchers to study the acute effects of myocardial steatosis on left ventricular structure and function. Overexpression of long-chain acyl-CoA synthetase, a key enzyme involved in triglyceride synthesis, produces an example of cardiac-restricted steatosis. Increased protein expression of acyl-CoA synthetase in the myocardium disrupts the balance between lipid import and export in the myocardium (Figure 3), which results in diffuse lipid accumulation and a greater than 2-fold increase in heart mass (10). The severe myocardial steatosis that is observed in this animal is associated with substantial left ventricular hypertrophy by 4 weeks of age that coincides with left ventricular dilatation and eventually progresses to heart failure. Of importance, the changes in cardiac lipid content, structure, and function develop without any change in lipid profile or body weight of the animal. This pattern of steatosis-induced heart failure has been reproduced by targeted overexpression of genes that are involved in lipid delivery (24, 26) and synthesis (10, 25) and by targeted deletion of genes that are involved in lipoprotein secretion (21) from the myocardium. Taken together, these data demonstrate that cardiac-specific steatosis, independent of systemic obesity, is a direct cause of dilated cardiomyopathy. Figure 3. Myocardial-specific lipotoxicity. Top panel. Middle panel. gray bars white bars Bottom panel. The development of cardiac-restricted transgenic murine models have also shown the therapeutic potential of several countermeasures, including adenoviral administration of leptin (Figure 3) and apolipoprotein B (26, 28), dietary replacement of long-chain triglycerides with medium-chain triglycerides (22), and blockade of production of reactive oxygen species (29). Each of these interventions has effectively ameliorated the myocardial steatosis in these mouse models and has rescued the myocardium from progression to dilated cardiomyopathy. These data reinforce the observations in the Zucker diabetic fatty rat that lipid accumulation is toxic in the myocardium. It is important to note that current thinking suggests that the cardiomyopathy is not a direct consequence of triglyceride accumulation alone, but that cardiomyopathy develops secondary to an accumulation of by-products of lipid metabolism, such as ceramide or other fatty acid derivatives that are known to interfere with intracellular signaling pathways (9, 30). This research provides convincing evidence for an acute role of steatosis in the development of left ventricular hypertrophy and dysfunction in animal models of obesity; until recently, however, few data from human research were available to support this theory. Quantification of Lipids in Human Tissues To study the role of steatosis in the clinical setting, we and others have developed a magnetic resonance imaging and spectroscopy technique that permits the precise and reproducible quantification of intracellular trig