Number 25, 2004
Heart failure in diabetes

Peroxisome proliferator-activated receptors and the cardiovascular system

Back to the Summary

Lazaros A. Nikolaidis, T. Barry Levine
Division of Cardiology, Department of Medicine, Drexel University College of Medicine, Allegheny General Hospital, Pittsburgh, USA
Correspondence: Dr Lazaros A. Nikolaidis, Division of Cardiology, Department of Medicine, Drexel University College of Medicine, Allegheny General Hospital, 320 East North Avenue, Pittsburgh, PA 15212, USA.
Tel: +14123598701, fax: +14123598964, e-mail: lazaros@pol.net

Introduction
Peroxisome proliferator-activated receptors (PPARs) are nuclear transcription factors of the hormone receptor family, predominantly regulating the expression of metabolic enzymes [1,2]. Three known isoforms (PPARs α, γ, and β/δ), which have discrete tissue distribution and metabolic properties, become activated after binding to either natural (physiologic) or synthetic (pharmacologic) ligands (Table I).

Table I. Tissue distribution and cardinal action of the three known peroxisome proliferator-activated receptor (PPAR) isoforms.

Peroxisome proliferator-activated receptor-α
PPARs-α regulate lipid metabolism and are expressed in tissues with active lipid turnover, where they promote mitochondrial transport and β-oxidation of free fatty acids (FFAs) and decrease FFA esterification into triglycerides. PPARs-α also upregulate apolipoproteins A-I and A-II genes and increase high-density lipoprotein concentrations. The antilipemic effect of fibrates involves activation of PPAR-α [3].

Peroxisome proliferator-activated receptor-γ

Metabolic effects
PPARs-γ are predominantly expressed in adipocytes, but also in skeletal muscle, liver, macrophages, T cells, myocardium, and vascular endothelium. When activated by ligands, PPARs-γ modulate lipid storage and redistribution away from visceral organs and into adipose tissue [4] by promoting catabolic over anabolic utilization of FFA in the liver and skeletal muscle, and modulating adipokines (adiponectin upregulation, leptin downregulation). PPARs-γ also enhance insulin signaling [5] by upregulating proteins necessary for insulin action (insulin receptor substrate-1, the regulatory kinase Akt, and glucose transporter 4), accounting for the antidiabetic effects of PPAR-γ agonists (thiazolidinediones [TZDs]).

Cardiovascular effects
Beyond affecting lipid and carbohydrate metabolism, PPARs-γ inhibit nuclear transcription factor kappa B (NFκB), which is implicated in atherogenesis, endothelial dysfunction, vascular growth and proliferation, expression of adhesion molecules (vascular cell adhesion molecule, intercellular adhesion molecule-1, E-selectin), and oxidation of low-density lipoprotein in atherosclerotic plaques [6]. Activation of NFκB by endothelin, catecholamines, and angiotensin II is involved in hypertrophic, proinflammatory, and cytotoxic pathways [7], promoting myocardial remodeling, cardiac hypertrophy, and CHF. Expression of PPAR-γ by T cells reduces the proinflammatory cytokines tumor necrosis factor (TNF)-α and interleukins-1, -2, -6, and -8 [8].

Thiazolidinediones

Cardiovascular effects of thiazolidinediones
As activators of PPAR-γ, TZDs exert insulinotropic and insulin-sensitizing cellular effects and improve the lipid profile. In addition, they exhibit pleiotropic cardiovascular effects independent of metabolism (Table II), via inhibition of NFκB at tissues expressing PPAR-γ [9]. TZDs promote regression of left ventricular hypertrophy [10,11] and improve systemic hemodynamics [12], left ventricular systolic and diastolic function, and experimental mitral regurgitation, inhibit myocardial collagen synthesis in experimental models [13], and exert antioxidant and anti-inflammatory effects via downregulation of TNF-α, transforming growth factor-β, adhesion molecules, and proinflammatory interleukins [8]. They have cardioprotective effects in ischemia–reperfusion in diabetic [14] and nondiabetic [15] animals, decrease infarct size, and attenuate postinfarct ventricular remodeling [16]. Vascular effects include systemic and coronary vasodilatation, improvement of endothelial function [17], prevention of atherosclerosis [18], attenuation of vascular remodeling and re-stenosis after angioplasty [19], and mitigation of posttransplant arteriosclerosis [20].

Table II. Cardiovascular effects of thiazolidinedione (TZD) treatment.

Congestive heart failure as insulin resistant state
Diabetes or the metabolic syndrome frequently accompanies ischemic cardiomyopathy. Diabetic cardiomyopathy develops in the absence of epicardial coronary stenosis, as a result of impaired coronary microcirculation and flow reserve or myocardial autonomic dysfunction [21]. The potential applicability of TZDs in CHF is intriguing, even in the absence of diabetes, in the light of emerging evidence of progressive insulin resistance developing as a result of cardiomyopathy. Glucose intolerance develops in nondiabetic patients with CHF, irrespective of the etiology, and signifies a poor prognosis [22,23]. Clinical studies have demonstrated impaired systemic and myocardial glucose uptake in CHF [24], in contrast to intact myocardial glucose uptake in patients with diabetes who have coronary artery disease but a preserved left ventricular ejection fraction [25].
CHF is characterized by myocardial energetic dysequilibrium with high myocardial oxygen demands, benefiting from utilization of glucose rather than FFA as preferred metabolic substrate. Although the expression of metabolic enzymes shifts to an “embryonic pattern” favoring glucose oxidation in chronic CHF [26], this “new equilibrium” can be jeopardized by excess catecholamine- and cytokine-mediated overproduction of FFA (“lipotoxicity”) and cellular events impeding glucose uptake and oxidation in endstage, decompensated CHF. Such mechanisms extend beyond competitive substrate inhibition and involve the deleterious effects of metabolic intermediaries [27] such as diacyl glycerol, ceramides, inactivation of insulin receptors by angiotensin II or endothelin, and distal cellular deficits in glucose transporter 4 transporters or key regulatory proteins of insulin signaling, such as impaired phosphorylation of Akt [28], a survival kinase also implicated in apoptosis.

Limitations of the clinical use of thiazolidinediones in congestive heart failure
In spite of potential benefits and salutary experimental studies, utilization of TZDs is restricted even in patients with diabetes who have CHF [29,30]. The first-generation TZD, troglitazone, has been withdrawn because of hepatotoxicity. Currently, two second-generation, nonhepatotoxic TZDs (rosiglitazone, pioglitazone) are used clinically. The main limitations restricting the use of TZDs in patients with diabetes who have CHF relate primarily to an increased incidence of peripheral edema, and secondarily to a less well defined risk of exacerbation of CHF, attributable to volume expansion [30]. These side effects are more frequent (Table III) when TZDs are combined with insulin [31]. Although peripheral edema is a frequent adverse event, it is unlikely to be mediated by detrimental effects of TZDs on central hemodynamics [32,33]; indeed, the findings of a retrospective study [33] suggested that TZDs improved central hemodynamics in patients with CHF who were diabetic. Peripheral edema is usually reversible, dose-dependent and responsive to diuretics or angiotensin-converting enzyme inhibitors [30]. Suggested mechanisms (Table IV) include calcium channel blockade [34], effects on renal microcirculation and permeability, and increased renal reabsorption of sodium [35], attributed to compensatory activation of the renin–angiotensin–aldosterone system in response to a vasodilatory effect. In contrast, TZDs may improve pulmonary capillary function [36]. True exacerbation of CHF is rare (33 incidents reported up to July 2004), and is not causally or exclusively attributable to TZDs per se, with all fatal cases involving plausible comorbid etiologies [37]. Because of these concerns, American College of Cardiology/American Heart Association/American Diabetes Association guidelines advocate against treatment with TZDs in patients with diabetes who have New York Heart Association III–IV CHF, and against the use of TZD–insulin combinations in patients with diabetes who have known left ventricular dysfunction or risk factors for CHF [30].

Table III. Incidence of edema associated with thiazolidinedione (TZD) monotherapy and combination therapy with other antidiabetic modalities in patients with diabetes. (From Nesto et al [30], with permission.)

Table IV. Plausible mechanisms associated with thiazolidinedione (TZD)-mediated peripheral edema.

Future directions
The safety and efficacy of TZDs in patients with diabetes at different stages of left ventricular dysfunction require further investigation to define the risk–benefit ratio for every subgroup of patient. Evidence-based therapeutic algorithms may allow patients with CHF to derive benefit from treatment with TZDs. However, these drugs should not be initiated or administered during acutely decompensated, fluid-overloaded states (akin to the β-blocker paradigm). Precise delineation of mechanisms of TZD-mediated edema will have a positive influence on the design of safer alternatives for this population of patients. Such drugs may exploit combined PPAR-γ/PPAR-α strategies, yet be devoid of adverse effects. Whether the benefits of TZDs represent class effects, are more pronounced in ischemic than in nonischemic cardiomyopathy, or persist in combination with oral hypoglycemic agents, in spite of greater rates of edema, remain to be investigated. Most intriguing is the concept of potential salutary effects of TZDs in patients with CHF who do not have diabetes but who are at risk of developing insulin resistance as a result of cardiomyopathy. ▪

Back to the Summary

REFERENCES

Back to the Summary