HuR brings the heat: linking adipose tissue to cardiac dysfunction

Obesity is a major risk factor for cardiometabolic diseases including hypertension, diabetes, coronary artery disease, and heart failure (HF). Evidence of the contributions of adipose tissue location (visceral vs. subcutaneous) and adipocyte composition (white vs. brown fat) to cardiovascular risk is increasing rapidly. Visceral fat is associated with many of the major risk factors for HF including hypertension, diabetes, inflammation, and left ventricular (LV) hypertrophy. In contrast, brown adipose tissue (BAT) has been associated with favorable cardiometabolic health including a lower risk for HF (1). Brown adipose tissue drives energy expenditure through thermogenesis and white adipose tissue (WAT) serves as a storage site for excess energy, which is stored as triglycerides. Thus, WAT, which often correlates with energy excess and obesity, is associated with inflammation and adverse cardiometabolic health. However, less is known about the flux between WAT and BAT, cross talk between adipose tissue and the heart, and how these interactions may affect cardiac remodeling or the risk for HF.

In this issue of the American Journal of Physiology-Heart and Circulatory Physiology, Guarnieri et al. (2) evaluated the role of adipose tissue-specific deletion of the RNA-binding protein human antigen R (Adipo-HuR^−/−) and effects on LV hypertrophy, fibrosis, and inflammation. They previously showed that Adipo-HuR^−/− reduced BAT-mediated thermogenesis (3). They have extended this work and found that Adipo-HuR^−/− mice developed LV hypertrophy and fibrosis, which was accompanied by proinflammatory gene expression in both cardiac and subcutaneous WAT (scWAT) as well as increases in circulating inflammatory markers (TNF-α and IL-6). They also found evidence for enrichment of vesicle-mediated transport-associated genes and genes related to apoptosis and inflammation further supporting the potential inflammatory contribution to their findings. It is important to point out that the apparent inflammation-mediated cardiac findings were driven by adipocyte-specific signaling pathways in the absence of obesity and related cardiometabolic derangements. Adipo-HuR^−/− mice are lean and as previously reported, have no differences in plasma lipids or glucose tolerance compared with control mice. They also observed that findings (i.e., increasing in uncoupling protein-1, UCP-1) in scWAT in Adipo-HuR^−/− mice were suggestive of compensatory beiging of scWAT. This functional change in WAT biology may be a compensation for BAT dysfunction, which the same laboratory had previously demonstrated (3).

Although the findings of this study suggest a plausible role for the link between reduced adipose tissue HuR expression and cardiovascular disease (CVD), there are a few important limitations to consider. Much of the emphasis on the cardiac findings in Adipo-HuR^−/− mice is focused on LV hypertrophy and a “hypercontractile state.” This conclusion is based on echocardiographic findings of reduced LV systolic dimensions and increased LV ejection fraction (hypercontractile state), and LV hypertrophy based on mildly increased LV wall thickness. Overall, LV mass was not significantly different in the Adipo-HuR^−/− mice; however, there was increased concentric remodeling. Blood pressure (cardiac afterload) is one of the main determinants of LV hypertrophy. Unfortunately, in this experiment, femoral blood pressure was measured during a terminal procedure while the mice were anesthetized and the potential chronic effects of blood pressure cannot be adequately determined. The measured values of blood pressure in Adipo-HuR^−/− mice and control mice averaged only 75.5 and 74.4 mmHg, respectively, considerably below normal blood pressure for mice and likely reflecting the effects of anesthesia and surgery. Thus, it is difficult to rule out potential indirect effects of adipocyte HuR deletion on LV wall thickness due to increased blood pressure. Excluding blood pressure effects on LV hypertrophy requires accurate 24 h/day measurements during daily activities. This is particularly relevant given HuR expression in adipocytes may influence energy expenditure and cardiac contractile function—both often sympathetically mediated processes, which along with blood pressure could be attenuated during anesthesia, although it is important to note that serum epinephrine and norepinephrine levels were lower in Adipo-HuR^−/− mice. Thus, further hemodynamic investigation, particularly focused on chronic blood pressure, is warranted in this model to more completely investigate the links to LV hypertrophy. They also assessed associations with LV weight to body weight ratios, which could be problematic given Adipo-HuR^−/− mice actually weigh less than controls.

Visceral adiposity, rather than subcutaneous adiposity, is the major factor associated with most of these cardiometabolic derangements. Local visceral fat including pericardial fat may have detrimental cardiac effects including inflammation and fibrosis, which ultimately increase the risk for coronary artery disease and HF. Recently, pericardial fat volume measured using cardiac computed tomography in 6,785 participants of the Multi-Ethnic Study of Atherosclerosis (MESA) without prevalent CVD was associated with an increased risk of incident HF (4). Elevated pericardial fat volume increased the risk for HF with preserved ejection fraction (HFpEF) rather than HF with reduced ejection fraction (HFrEF). The relationship between high pericardial fat volume and incident HF was stronger in women (HR = 2.06) compared with men (HR = 1.53). These associations remained significant after adjustment for abdominal subcutaneous and visceral fat suggesting an adverse local effect of pericardial fat. Pericardial fat may have protective cardiac effects including mechanical protection of the coronary vessels, fatty acid homeostasis and myocardial energy balance, and secretion of anti-inflammatory cytokines (5). However, excess pericardial fat deposition can lead to cardiac inflammation, fibrosis, and direct myocardial lipotoxic effects. The detrimental effects may be even more pronounced in women who are more predisposed to intramyocardial lipid accumulation and adipocyte-associated proinflammatory mediators compared with men (6). Epicardial fat was also most strongly associated with impaired myocardial perfusion at peak dose of dobutamine in women but not in men (7). These sex differences in adipose tissue biology and effects on the heart may help explain differences observed in HFpEF pathophysiology in obese women who are often affected. Interestingly, Guarnieri and colleagues did not observe cardiac hypertrophy or fibrosis in female Adipo-HuR^−/− mice. They speculate that sex differences in BAT-mediated cardiac effects play a role in cardioprotection, particularly in premenopausal women. In line with this, BAT is more prevalent in women, and BAT activity is higher in women (1).

HuR deletion in adipose cells leads to BAT dysfunction (3). BAT is thermogenic and increases energy expenditure while increasing glucose and lipid utilization. BAT has been linked to many cardiometabolic improvements including lower body mass index and lower blood glucose levels. A large retrospective analysis of ¹⁸F-fluordeoxyglucose positron-emission tomography-computed tomography scans from 52,487 patients recently evaluated BAT levels and found that individuals with BAT had a lower prevalence of cardiometabolic diseases including diabetes, hypertension, coronary artery disease, and HF (1). The beneficial effects of BAT were also more pronounced in overweight or obese individuals suggesting that BAT may mitigate the adverse effects of obesity. Beige fat, or WAT that has taken on some of the characteristics of BAT, is also thermogenic and is associated with a more favorable cardiometabolic profile than WAT. Gene expression in BAT of Adipo-HuR^−/− mice did not show any significant correlations with cardiac fibrosis (2). However, gene expression of UCP-1 and other markers of thermogenesis were increased in scWAT of Adipo-HuR^−/− mice suggesting that beiging of this fat depot may be a compensatory effect, albeit these changes in gene expression in scWAT were associated with inflammatory processes and increases in circulating TNF-α and IL-6.

Interactions between the heart, adipose tissue, and other organs such as the kidneys play important physiological roles for homeostasis; however, cross talk between these organ systems has also been implicated in the pathophysiology of HFpEF. For example, natriuretic peptides released from the heart lead to natriuresis, diuresis, and vasodilation, but they also enhance lipolysis and lipid mobilization. Natriuretic peptides have also been shown to promote beiging of WAT (8). Mouse models of HFpEF (e.g., aldosterone-induced hypertension and transverse aortic constriction) were associated with decreased WAT and increased BAT volumes. Adipocytes in WAT were smaller and had features of beiging including higher mitochondrial content and increased expression of UCP-1. Overall, these findings suggest that HFpEF activates beiging of WAT. However, these changes in WAT were not associated with changes in myocardial fibrosis that are often observed in human HFpEF. Animal experiments of HFpEF are often limited and do not completely mimic the human phenotype, which is often associated with older age as a major risk factor. Further investigation of the potential role of beiging or browning of WAT in HF and whether this is a compensatory or, less likely, a causal mechanism is warranted.

As obesity is a major risk factor for cardiometabolic diseases including HFpEF, strategies aimed at reducing visceral obesity, reducing WAT volume, and potentially increasing BAT are being evaluated for prevention and treatment of CVD (1, 9). Lifestyle modifications including diet, exercise, and reduced sedentariness remain the cornerstone of this strategy. Bariatric surgery reduces epicardial fat and may activate BAT. Novel strategies such as BAT activation or possibly even BAT transplantation may also hold potential as therapies for obesity-induced diseases; however, BAT is reported to make up <0.1% of body weight in humans. Pharmacological therapies including antidiabetic medications such as metformin, PPAR-γ agonists, and glucagon-like peptide (GLP)-1 receptor agonists also hold promise for treating obesity-induced cardiometabolic diseases (10). Recently, clinical trials of GLP-1 receptor agonists have demonstrated significant long-term weight loss with favorable cardiometabolic effects; however, there are limited clinical data investigating the potential effects of these drugs on BAT biology. Although the investigation of the potential beneficial effects of GLP-1 receptor agonists is being investigated in patients with HF, further study is needed to determine if/how these medications may influence cardiac structure and function potentially by modifying WAT or BAT biology.

Although BAT dysfunction is associated with diabetes, obesity, and CVD including HF, the loss of functional BAT has not been conclusively shown to play a direct causal role in CVD. Guarnieri et al. (2) demonstrated that adipocyte-specific deletion of HuR is associated with adverse cardiac remodeling, inflammation, and fibrosis, which are major factors involved in CVD, specifically HfpEF. This work highlights a novel pathway that requires further experimental and epidemiological investigation to unravel the complex interplay between adipose tissue and the heart (Fig. 1) that may ultimately lead to CVD.

Figure 1.

Overview of the potential role of RNA-binding protein human antigen R (HuR) on the cross talk between adipose tissue and the heart. Adipose tissue expression of HuR is associated with activation of metabolically favorable brown adipose tissue and potentially cardioprotection. Conversely, deletion of HuR in adipocytes is associated with brown adipose tissue dysfunction and white adipose tissue inflammation, which lead to cardiac inflammation and fibrosis, myocardial hypercontractility, and cardiac hypertrophy. However, sex differences may exist in this pathway as female mice with adipocyte HuR deletion did not exhibit the same adipose tissue-mediated cardiac effects (hypertrophy and fibrosis). Created with Mind the Graph and published with permission.

GRANTS

M. E. Hall is supported by National Institute of General Medical Sciences Grant 5U54GM115428.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

M.E.H. and R.K. drafted manuscript; edited and revised manuscript; and approved final version of manuscript.