Keywords

1 Introduction

The heart has a very high energy demand because of the continuous work performed by the cardiac muscle. Under physiological conditions, the main substrates for energy production are fatty acids (60–90 % of ATP production), glucose, and lactate. Under nonischemic conditions, more than 95 % of the energy requirement is supplied by the oxidation of fatty acids and carbohydrates, and myocardial energy expenditure can be estimated from oxygen consumption. The contribution of these substrates to overall energy production is a dynamic process, and physiological adaptations such as fetal-to-newborn transition [1–3], and alterations related to disease states have been well established [4–8]. The heart exerts a metabolic flexibility, and myocardial substrate utilization depends on substrate availability, nutritional status, and exercise level. With glucose as the more energetically efficient substrate, the healthy heart is able to switch to glucose under stress conditions, such as ischemia, pressure overload, or in heart failure. Interestingly, interventions such as increasing fatty acid uptake [9, 10], or fatty acid oxidation [11, 12], result in alterations that resemble diabetic cardiomyopathy, and normalization of substrate metabolism in diabetic models reverses these alterations [13]. These studies indicate an important contributing role for substrate metabolism in the development of diabetic cardiomyopathy.

Although excessively available during diabetes, utilization of carbohydrates is compromised. Thus, the myocardium is forced to use alternative sources for ATP production such as fatty acids, which are high in diabetes as well [14–18]. The consequences of this pathological switch occur at multiple levels, as is discussed here.

From the mechanical efficiency point of view, an increase in fatty acid usage at the expense of carbohydrates results in ATP hydrolysis for noncontractile purposes and is attributed to a lower phosphate/oxygen (P/O) ratio for fatty acid metabolism, increased uncoupling of mitochondrial oxidative phosphorylation, and futile cycling (reviewed in [4]).

High levels of fatty acids further lower glucose usage as first defined by Randle et al. [19]. We now know that intracellular changes occur during diabetes and contribute to the aforementioned switch as well as development of diabetic cardiomyopathy.

2 Alterations in Cardiac Carbohydrate Utilization

Two glucose transporters, GLUT1 and GLUT4, are involved in basal and insulin-mediated glucose uptake. GLUT1 shows sarcolemmal localization and represents basal cardiac uptake. GLUT4, on the other hand, is located in the intracellular pool, and insulin facilitates the localization of this transporter to the sarcolemma [20]. More recently, AMP-dependent protein kinase (AMPK)-mediated and insulin-independent uptake of glucose by this transporter has been documented [21]. A decrease in the number and translocation of GLUT4 to the sarcolemma has been suggested to play a major role in the reduction of glucose metabolism in diabetes. A reduction in both glycolysis and glucose oxidation was reported in db/db mice in which cardiac dysfunction has been already defined [13]. Because both metabolic parameters and cardiac function were normalized in transgenic mice that overexpressed GLUT4, it has been concluded that there was a causative relationship between impaired substrate metabolism and diabetic cardiomyopathy [13]. A key enzyme in the regulation of glycolysis is phosphofructokinase (PFK)-I, the enzyme that catalyzes the phosphorylation of fructose-6-phosphate to produce fructose-1,6-bisphosphate. PFK-I activity is inhibited by citrate and acetyl CoA and activated by a low ATP/ADP ratio [22]. An increase in citrate levels as a result of increased fatty acid oxidation in the diabetic heart probably contributes to the inhibition of PFK-I and therefore glycolysis. At the transcriptional level of glucose uptake and metabolism, Finck et al. reported that both GLUT4 and PFK expression were lower and that PDK4 expression was higher in transgenic mice that overexpress peroxisome proliferator-activated receptor (PPAR)-α [12]. The inhibition of GLUT4 and PFK was probably not a direct result of PPAR-α overexpression, but associated with alterations in substrate metabolism that were mediated by PPAR-α. The increase in PDK4, on the other hand, was likely linked to PPAR-α overexpression because PPAR-α ligands were previously shown to activate this enzyme [23]. Another member of the PPAR family of transcription factors is PPAR-δ. PPAR-δ is the predominant form in the heart and regulates cardiac substrate metabolism [24]. A decrease in PPAR-δ expression was reported in diabetic heart [25]. However, another similar study reported that mice overexpressing PPAR-β/δ did not accumulate myocardial lipid and had normal cardiac function [26]. Conversely, cardiac glucose transport and glycolytic enzymes were activated in PPAR-β/δ transgenics.

Another limiting step in myocardial glucose metabolism is pyruvate dehydrogenase complex (PDH), which catalyzes the irreversible conversion of pyruvate to acetyl CoA. The amount of active dephosphorylated PDH is reduced when phosphorylated by PDH kinase (PDK) and induced also by PDH phosphatase. The rate of pyruvate oxidation depends not only on the phosphorylation state but also on the concentrations of its substrates (pyruvate, NAD+, and CoA) and products (NADH and acetyl CoA). Thus, an increase in mitochondrial acetyl CoA, through an increase in fatty acid oxidation, for example, inhibits pyruvate oxidation. Indeed, the active dephosphorylated form of the PDH is reduced in diabetes models [7]. Moreover, PDK-4 is one of the targets of PPAR-α, and upregulation of PDK-4 in mice overexpressing PPARα is associated with reduced glucose oxidation [27].

Inhibition of pyruvate conversion into acetyl CoA results in accumulation and diversion of glycolytic intermediates into diacylglycerol biosynthesis, which contributes to the activation of diacylglycerol-sensitive protein kinase C (PKC) isoforms. Recently, inhibition of one of the PKC isoforms, PKC-beta, was shown to preserve cardiac function in a transgenic mice model of diabetic diastolic failure [28].

The reports on carbohydrate utilization in the human diabetic heart are controversial. Studies in type 1 diabetic patients reported lower [29, 30] or unchanged [31] uptake of carbohydrates by the myocardium. In type 2 diabetes, GLUT4 protein levels in diabetics were about 30 % lower compared to control patients [32]. However, other studies reported that cardiac glucose uptake was not compromised in type 2 diabetes [33–35] and reduced only in type 2 diabetics with hypertriglyceridemia [36] and increased plasma fatty acids. Because glucose can still enter the cell through mass action, as evidenced by a high glucose pool in the type 1 diabetic heart [22], glucose metabolism is unlikely to be regulated at the level of uptake in diabetes despite a deficiency in or resistance to insulin action.

Lactate is another potential substrate for myocardial ATP production in vivo [37], but the information on diabetes-related alterations in lactate oxidation is relatively scant. A greater decrease in lactate oxidation relative to glucose oxidation in hearts from diabetic rats was observed when lactate and glucose were the only substrates in the perfusate for ATP production. Under these conditions, a specific inhibition of lactate oxidation independent of pyruvate dehydrogenase was suggested [38]. Hearts from ZDF rats also showed lower lactate oxidation [39, 40]. The contribution of lactate in diabetic cardiomyopathy clearly needs further study.

3 Alterations in Cardiac Fatty Acid Utilization

An increase of fatty acids that are supplied as free acids bound to albumin and as esters in chylomicrons and very low density lipoproteins have been reported in diabetes [7]. The effects of elevated levels of lipoproteins on myocardial fatty acid metabolism are not clear, and neither is the relative contribution of cardiac lipoprotein lipase (LPL) activity to the delivery of free fatty acids to the diabetic heart. Unchanged, increased, and decreased levels of LPL protein and activity were reported in the diabetic heart, and this discrepancy was suggested to be related to the diversity in rat strains, the dosage of diabetogenic agent, and the duration of diabetes [41].

Free fatty acids enter the cardiac myocyte either by passive diffusion or via a protein carrier-mediated pathway. These protein carriers include fatty acid translocase (FAT)/CD36, the plasma membrane isoform of fatty acid-binding protein (FABPpm), and fatty acid transport protein (FATP) 1/6. FAT/CD36 plays a major role in the translocation of fatty acid across the sarcolemmal membrane of cardiac myocytes as this protein was shown to mediate 50–60 % of fatty acid and transport of the heart. Also, FAT/CD36 is able to translocate between intracellular endosomes and the sarcolemmal membrane and thereby regulate fatty acid uptake [4].

Fatty acid uptake is increased in diabetes and leads to increased fatty acid oxidation and triacylglycerol (TAG) storage. In the streptozotocin (STZ)-induced model of type 1 diabetes, this increase is facilitated by fatty acid translocase (FAT/CD36) [42]. In type 2 diabetic models, an increase in FAT/CD36 and fatty acid-binding protein (FATP1) [43] and a permanent relocation of FAT/CD36 to the cardiomyocyte membrane was shown to increase fatty acid uptake [44, 45]. Interestingly, insulin was suggested to upregulate FAT/CD36 and translocate it to the sarcolemma [46].

The majority (70–90 %) of the fatty acids that enter cardiomyocyte is oxidized for energy production; the rest is converted to TAGs [47]. Excessive accumulation of lipids, or lipotoxicity, within nonadipose tissue provides substrates for nonoxidative processes such as ceramide and diacylglycerol synthesis, which can lead to apoptosis [48, 49]. Accumulation of TAG within the myocardium of insulin-resistant rats was associated with contractile dysfunction [50]. We also have shown that insulin-resistant rats have increased TAG accumulation, which reduced insulin-stimulated glucose metabolism [51]. Although the exact mechanism of lipotoxicity-induced cardiac dysfunction is not known, it seems to be related to a combination of apoptotic cell death and impaired substrate metabolism.

The most important step in the regulation of fatty acid oxidation is the transport of fatty acids into the mitochondria for further metabolism. The activation of short- and medium-chain fatty acids occurs in the matrix and does not require carnitine. However, long-chain fatty acids are shuttled into the mitochondria by three carnitine-dependant enzymes. Carnitine palmitoyltransferase (CPT)-I catalyzes the conversion of long-chain acyl CoA to long-chain acylcarnitine. Carnitine:acylcarnitine translocase (CAT) transports long-chain acylcarnitine across the inner mitochondrial membrane, and CPT-II regenerates long-chain acyl-CoA in the mitochondrial matrix [52]. Of these, CPT-I is the master regulator of mitochondrial uptake of fatty acids and is allosterically inhibited by malonyl CoA [53]. The turnover of malonyl CoA in the heart is fast. Therefore, myocardial malonyl CoA concentrations are dependent on the balance between its synthesis from acetyl CoA via acetyl CoA carboxylase (ACC) and its degradation via malonyl CoA decarboxylase (MCD) [4]. A good correlation has been established between levels of malonyl CoA and rates of fatty acid oxidation, and a reduction in malonyl CoA levels is almost consistent in situations in which fatty acid oxidation is increased [18]. The decrease in malonyl CoA levels seems to be related to increased degradation of malonyl CoA by MCD [18]. MCD is transcriptionally regulated by PPAR-α [54, 55]. In addition to diabetes, activity and expression of cardiac MCD was increased in fasting, high-fat-fed, and newborn hearts [56–59]. Moreover, PPAR-α null mice had increased rates of glucose oxidation and decreased expression and activity of MCD [56].

Increase in circulating fatty acids directly modifies the enzymes in substrate metabolism because fatty acids and their derivatives are ligands for the PPAR family of nuclear receptors, of which PPAR-α and its coactivator peroxisome proliferator-activated receptor-gamma coactivator (PGC)-1 are particularly important in the heart [60–62]. PPAR-α signaling was increased in 15-week-old ob/ob and db/db mice [60]. Other studies reported an increase in the expressions of PPAR-α, PGC-1, and their targets in models of insulin resistance and type 2 diabetes [12, 63–65].

Once in the mitochondrial matrix, long-chain fatty acyl CoAs pass through the β-oxidation enzyme system to produce one acetyl CoA at each cycle, one NADH, and one FADH. The key enzyme in the β-oxidation pathway is β-hydroxyacyl-CoA dehydrogenase. The activity of this enzyme was shown to be normal [66] or high [67] in diabetic rat mitochondria. Higher expression of another enzyme, 3-ketoacyl-CoA thiolase, has also been shown in streptozotocin-diabetic rat hearts [8]. Thus, a combination of high circulating levels of fatty acids, decreased inhibition of fatty acid uptake by the mitochondria, and a normal or accelerated β-oxidation pathway results in a large proportion of acetyl CoA for the tricarboxylic acid (TCA) cycle being supplied from fatty acid oxidation.

4 Alterations in Diabetic Mitochondria

Acetyl CoA derived from both β-oxidation of fatty acids and PDH enters the TCA cycle. The myocardial TCA cycle does not appear to be altered in diabetes because measurements of TCA cycle enzyme activities are similar in diabetic and control hearts [7].

Mitochondrial proteins are produced in the mitochondria and are regulated under the mitochondrial transcriptional factor A (TFAM) by mitochondrial DNA. A recent study reported that overexpression of TFAM improved ATP content and SERCA2a content, both of which were deteriorated by exposing neonatal rat myocytes to hyperglycemic conditions [68].

5 Ketone Body Metabolism

Plasma ketone bodies (β-hydroxybutyrate and acetoacetate) are formed from fatty acids in the liver, and their plasma concentration is normally very low. Thus, ketone bodies are normally a minor substrate for the myocardium. When plasma levels increase during extreme conditions, such as poorly controlled diabetes and starvation, the heart extracts and oxidizes ketone bodies [29]. Ketone bodies inhibit the uptake and metabolism of other energetic substrates [69–74]. The biochemical mechanisms responsible for the inhibition of glucose, lactate, or fatty acid oxidation is not clearly understood.

6 Potential Drugs with Metabolic Effects in the Treatment of Diabetes

Our knowledge of the adipose tissue-originated hormones—adipokines—and their biological effects has increased tremendously since the discovery of the OB gene product, leptin, in 1994 [75]. Two of these, leptin and adiponectin, have direct effects on substrate metabolism as well as insulin secretion. Despite the presence of some conflicting reports, leptin seems to regulate pancreatic cell function and thus insulin secretion [76–79]. Acute exposure of myocytes [80] and isolated working hearts [51] to leptin stimulates fatty acid oxidation. In contrast, type 2 diabetic models, ob/ob and db/db mice, also display increased myocardial fatty acid β-oxidation [60, 81, 82], which is probably related to other alterations in these genetic models. Moreover, leptin administration to subjects with lipodystrophy, abnormal distribution of adipose tissue in which leptin levels are usually decreased [83, 84], has been shown to improve metabolic abnormalities such as hypertriglyceridemia and impaired glucose control, which often are resistant to maximum doses of insulin sensitizers or very high doses of insulin [85–87]. Taken together, these studies suggest that leptin has the potential to become an effective antidiabetic agent with effects on substrate metabolism.

Described by Scherer et al. in 1995 [88], adiponectin is another antidiabetic adipokine. Adiponectin possibly affects β-cell function as adiponectin receptors are expressed in β cells [89–91]. More recently, adiponectin was shown to protect against caspase-8-mediated apoptosis in β cells [92]. As for the effects on substrate metabolism, globular head domain of adiponectin (gAd) stimulates fatty acid oxidation in isolated working 1-day-old rabbit heart [93]. Similar effects on fatty acid oxidation with both the high molecular weight hexameric form of adiponectin and gAd have been reported for cardiac myocytes [94].

Finally, it is worth mentioning other drugs that have beneficial effects on substrate metabolism. Beta-adrenergic receptor antagonists (β-blockers) are used in the treatment of many disease states such as hypertension, ischemic heart disease, arrhythmia, heart failure, glaucoma, and migraine, as well as to reduce the symptoms related to anxiety and hyperthyroidism. Depending on their receptor subtype specificity, β-blockers have different effects on substrate usage. As clearly shown by clinical trials, long-term use of nonselective β-blockers such as proporanolol increases the incidence of diabetes [95, 96], and cardioselective β-blockers such as atenolol and metoprolol reduce insulin sensitivity [97, 98], whereas carvedilol improves serum lipid profile, reduces insulin resistance, and decreases the risk for diabetes [99, 100]. Based on these reports, nonselective and cardioselective β-blockers are accepted as “pro-insulin resistant” whereas carvedilol, the drug that also blocks α1-adrenergic receptors, is known to have opposite effects (reviewed in [101, 102]). The beneficial properties of carvedilol seem to be related to its direct effects on substrate metabolism because carvedilol was shown to reduce the myocardial uptake of long-chain fatty acids and improved left ventricular ejection fraction in heart failure patients [103]. More recently, carvedilol treatment reduced fatty acid oxidation and increased glycolysis in C2C12 cells [104]. The possible effects of carvedilol on diabetic cardiomyopathy need future investigations. However, the fact that carvedilol has beneficial effects on insulin resistance warrants an improvement in substrate utilization and hence diabetic cardiomyopathy.

7 Conclusion

Diabetes is a complex disorder. Its contributors exist at multiple levels. Hormones predominantly determine the levels of energetic substrates, and their major impact on energetic substrate utilization has been well documented. However, modulation of subcellular steps that are involved in ATP-producing processes seems to be just as important in the regulation of substrate metabolism and is expected to have beneficial effects on the mechanical function of the heart. Future studies will reveal such interventions that improve the hormonal milieu or deteriorated substrate metabolism, and leptin and adiponectin agonists are good candidates.