Introduction

Taurine is a ubiquitous β-amino acid found in high concentrations in excitable tissues. Within the cardiomyocyte, abundant levels of taurine are found in the cytosol and the mitochondria (Jong et al. 2012). The cellular uptake of taurine by the β-amino acid transporter is required for maintenance of the high taurine content of the cardiomyocyte. Both inhibition and downregulation of the taurine transporter decreases total cellular taurine content, with the largest decreases occurring in the cytosol (Ito et al. 2008; Jong et al. 2012). Cellular taurine content also declines in the ischemia-reperfused heart; however, the reduction in taurine results from a massive release of the amino acid from the heart (Schaffer et al. 2014a). In some animal species, dietary deficiency also leads to myocardial taurine depletion (Pion et al. 1987).

Taurine exerts several actions on the heart, with its physiological functions depending upon the size and location of the taurine pool. In the cytosol, taurine functions as an osmoregulator and a modulator of protein phosphorylation (Thurston et al. 1981; Schaffer et al. 2002; Ramila et al. 2015). Two proteins regulated by cytosolic taurine are phospholamban and the sarcoplasmic reticular Ca2+ ATPase (SERCA2), both of which play central roles in excitation–contraction coupling (Ramila et al. 2015). In the mitochondria, taurine helps maintain normal respiratory chain function (Jong et al. 2012). According to Jong et al. (2012), taurine deficiency not only reduces cardiomyocyte respiration, but also promotes the generation of reactive oxygen species by the mitochondria. These effects of taurine deficiency have been attributed to taurine-mediated conjugation of the wobble position uracil residue of tRNALeu(UUR) (Suzuki et al. 2002; Schaffer et al. 2014b). According to Kirino et al. (2004), the formation of the 5-taurinomethyluridine conjugate strengthens the interaction of the AAU anticodon of tRNALeu(UUR) with the UUG codon of mitochondria encoded proteins, a property that improves UUG decoding. Because UUG-dependent proteins serve as subunits of respiratory chain complexes, taurine indirectly regulates the activity of the respiratory chain and the generation of ATP (Rosca and Hoppel 2010; Schaffer et al. 2014b).

The taurine-deficient phenotype resembles that of the mitochondrial disease, MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes) (Schaffer et al. 2014b). In MELAS, a mutation in the DNA for tRNALeu(UUR) decreases the formation of the taurine conjugate, 5-taurinomethyluridine-tRNALeu(UUR), while taurine deficiency reduces the availability of substrate for the conjugation reaction (Schaffer et al. 2014b). Because a defect in UUG decoding develops in both MELAS and the taurine-deficient phenotype, it is relevant that the pathology of both conditions is remarkably similar (Schaffer et al. 2014b). Therefore, the taurine-deficient rodent provides useful information on the pathophysiology of mitochondrial diseases. To date, there is a paucity of information on the regulation of energy metabolism by MELAS, with only a few studies published on the characteristics of cybrids and fibroblasts carrying the MELAS mutation and one study examining PET imaging of MELAS patients (James et al. 1996, 1999; Gajewski et al. 2003; Arakawa et al. 2010; Kami et al. 2012). Accordingly, the present study tests the hypothesis that taurine deficiency suppresses ATP generation by decreasing respiratory chain function.

Materials and methods

Animal models of taurine deficiency

Animal handling and experimental procedures followed the Animal Welfare Act and the Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committee of the University of South Alabama.

Homozygous taurine transporter knockout (TauTKO) mice and wild-type (WT) control mice were produced by breeding heterogenous taurine transporter knockout (TauTKO+/−) C57BL/6 mice. The TauTKO and WT mice were fed Purina rat chow and maintained on normal tap water. This study was conducted using 3- to 4-month-old mice. Mitochondrial taurine content of hearts from TauTKO mice were 60 % less than that of the WT mice (97 ± 12 vs. 38 ± 5 nmol/mg protein).

Male Wistar rats (240–260 g) were fed Purina rat chow ad libitum and maintained for 3 weeks on tap water containing 3 % β-alanine, a taurine transport inhibitor. Over a 3-week period, β-alanine feeding decreased myocardial taurine content from 98.5 to 59.3 μmol/g dry wt. Hearts, obtained from control and β-alanine-treated rats, were perfused on a standard working heart apparatus with perfusate consisting of Krebs–Henseleit buffer containing 5 mM glucose, 5 mM acetate and 2.5 U/l insulin. All hearts were paced at 300 beats/min and perfused for a 20-min stabilization period before beginning the experiment. Oxygen consumption was monitored with a Clark oxygen electrode (Schaffer et al. 1986). The formation of reducing equivalents by the citric acid cycle was calculated from the rate of acetate oxidation, glucose oxidation and endogenous palmitate oxidation. The rate of acetate oxidation was determined from the conversion of [2-14C]-acetate to 14CO2 (Kramer et al. 1983). The formation of reducing equivalents by the citric acid cycle was calculated from the rate of acetate oxidation, glucose oxidation and palmitate oxidation. The rate of glucose utilization was determined from the rate of tritium release from [3-3H]-glucose into water, lactate production and pyruvate output (Schaffer et al. 1986). Because lactate dehydrogenase is an extremely active enzyme, the lactate/pyruvate ratio is proportional to the cytosolic NADH/NAD+ ratio. Coronary effluent samples were assayed for the rates of lactate and pyruvate production (Schaffer et al. 1986). ATP and creatine phosphate content were measured from extracts of freeze-clamped hearts (Schaffer et al. 1986). Left ventricular pressure was measured with a Statham P23Gb pressure transducer by inserting a 22-gauge needle through the ventricular wall (Schaffer et al. 1985).

Isolation of mitochondrial proteins

The isolation of mitochondrial proteins was conducted according to the method described by Chen et al. (2003). Briefly, hearts were homogenized in mitochondrial buffer containing 70 mM sucrose, 190 mM mannitol, 20 mM HEPES, 1 mM EDTA, 1 mM PMSF, 1 mM NaV2O5, 1 mM NaF and 1× protease inhibitor cocktail. Homogenates were centrifuged at 600g for 10 min. The supernatant was then centrifuged for 12,000g for 30 min, producing a pellet that was defined as the mitochondrial fraction. The pellet was washed and finally re-suspended in appropriate buffers supplemented with 1 mM PMSF, 1 mM NaV2O5, 1 mM NaF and 1× protease inhibitor cocktail. Unless stated otherwise, 10–20 µg proteins were used in the experiments.

Western blotting

Total lysates were prepared by homogenizing hearts in radioimmuno-precipitation assay (RIPA) lysis buffer (50 mM Tris Base, pH 8.0, 150 mM NaCl, 0.5 % deoxycholic acid, 1 % NP-40, 0.1 % sodium dodecyl sulfate). Homogenates were centrifuged at 10,000g for 10 min and supernatants were collected as total lysates. Isolated mitochondria were prepared as described earlier. The mitochondrial pellets were suspended in RIPA lysis buffer. The protein concentration was measured by the bicinchoninic assay (BCA). Protein (20–30 µg) was mixed with an equal volume of 5× sample buffer (1.25 mM Tris HCl, pH 6.8, 1 % sodium dodecyl sulfate, 10 % glycerol, 5 % β-mercaptoethanol) and then boiled for 5 min. Proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a nitrocellulose membrane. The membranes then were blocked in blocking buffer (5 % milk in Tris-buffered saline with Tween 20) and incubated with an appropriate primary antibody overnight at 4 °C. The next day, membranes were washed before incubating with an appropriate secondary antibody. After washing, Western blots were analyzed by enhanced chemiluminescent reagents.

The antibodies used in this study were specific for the following proteins: PPARα, β-actin, carnitine palmitoyltransferase I (CPTI-M), malonyl CoA decarboxylase, phosphorylated-acetyl CoA carboxylase (ACC), pyruvate dehydrogenase kinase-4 (PDK-4) and phosphorylated-pyruvate dehydrogenase.

Statistical analyses

All results are reported as means ± SEM. Statistical significance was determined using the Student’s t test for comparison within groups or ANOVA followed by the Newman–Keuls test for comparison between groups. Values of p < 0.05 were considered statistically significant.

Results

Effect of taurine deficiency on energy metabolism

It has been shown that the failing heart is energy deficient (Ingwall and Weiss 2004; Ventura-Clapier et al. 2011). To determine whether the taurine-deficient heart is also energy starved, the rates of glycolysis, lactate and pyruvate production, glucose oxidation, acetate oxidation and oxygen consumption were determined in taurine-deficient rat hearts perfused with buffer containing 5 mM glucose, 5 mM acetate and 2.5 U/l insulin. Acetate is directly converted to acetyl CoA by the enzyme acetyl CoA synthetase (Krebs and Eggleston 1948). Acetyl CoA is also produced by β-oxidation of fatty acids and the oxidation of pyruvate via pyruvate dehydrogenase. Therefore, addition of acetate to the perfusion medium reduces both glucose oxidation and β-oxidation of fatty acids.

As seen in Fig. 1a, taurine deficiency was associated with a 17 % elevation in glucose utilization, but a 25 % decrease in glucose oxidation. The observed reduction in glucose oxidation was caused by a 2.5-fold rise in lactate output associated with a decrease in pyruvate output (Fig. 1b). Since lactate dehydrogenase is an extremely active enzyme in the heart, the cytosolic NADH/NAD+ ratio is the primary determinant of the lactate/pyruvate ratio and of lactate production. Thus, the significant increase in lactate production by the taurine-deficient heart, coupled with a decrease in pyruvate output, supports the view that an elevation in the NADH/NAD+ ratio diminished the availability of pyruvate for the generation of acetyl CoA by pyruvate dehydrogenase (Fig. 1b).

Fig. 1
figure 1

Effect of taurine deficiency on glucose metabolism. Taurine-deficient and control rat hearts were perfused for 20 min with Krebs–Henseleit buffer containing 5 mM glucose, 5 mM acetate and 2.5 U/l insulin. a Glucose utilization was determined from the generation of 3H2O from [3-3H]-glucose measured in perfusate collected at 1-min intervals. Values represent means ± SEM of 4–6 hearts. Glucose oxidation was calculated from the rate of flux through pyruvate dehydrogenase. b Lactate and pyruvate content of coronary effluent were determined spectrophotometrically. Values represent means ± SEM of 4–6 hearts. Asterisks denote a significant difference between control and taurine-deficient hearts (p < 0.05)

Full size image

Acetate oxidation is a measure of citric acid cycle flux. Because the utilization of acetate by the taurine-deficient heart was 23 % lower than that of the control heart, the generation of reducing equivalents by the citric acid cycle is significantly diminished in the taurine-deficient heart (Fig. 2).

Fig. 2
figure 2

Effect of taurine deficiency on acetate metabolism. Taurine-deficient and control rat hearts were perfused for 20 min with Krebs–Henseleit buffer containing 5 mM glucose, 5 mM acetate and 2.5 U/l insulin. Acetate utilization was determined from the generation of 14CO2 from [2-14C]-acetate measured in perfusate collected at 1-min intervals. Values represent mean ± SEM of 4 hearts. Asterisk denotes a significant difference between control and taurine-deficient hearts (p < 0.05)

Full size image

The rate of oxygen consumption by the control and taurine-deficient heart was 5137 ± 115 and 3873 ± 48 μmol O2/g dry wt/h, respectively. Using the method described in Table 1, we calculated the rates of glucose oxidation, oxidation of glycolytic NADH, acetate oxidation, β-oxidation of endogenous fatty acids and citric acid cycle flux. As seen in Table 1, the rate of β-oxidation of endogenous fatty acids is 31 % lower in the taurine-deficient heart than in that of the control heart (Table 1).

Table 1 Metabolic rates in taurine-deficient and normal rat hearts
Full size table

The total rate of ATP generation by the control and taurine-deficient hearts was 28,173 and 21,435 μmol ATP/g dry wt/h, respectively (Table 1). Despite the dramatic reduction in total ATP generation, taurine deficiency did not significantly affect the percentage of ATP generated from exogenous acetate and glucose. In perfused hearts from both taurine-deficient and control rats, ~55 and 32 % of total ATP generated by the heart came from exogenous acetate and glucose, respectively (Fig. 3). However, the percentage of total ATP generated from exogenous fatty acids (referred to as palmitate) was 11.9 % by the control heart compared to 8.1 % by the taurine-deficient heart (Fig. 3). Although taurine deficiency had only a small effect on the percentage of ATP generated by each substrate, it significantly diminished the amount of ATP generated by each substrate, suggesting that a common or related mechanism probably suppressed the metabolism of all three substrates. Figure 4 shows that the decrease in the total rate of ATP generation caused a significant decrease in the energy state of the taurine-deficient heart, as represented by the 30 % decrease in the phosphocreatine/ATP ratio (Fig. 4). Contractile function was also diminished in the taurine-deficient heart; the rate pressure product fell 14 % from 46,500 ± 1000 to 40,500 ± 850 cm H2O bpm.

Fig. 3
figure 3

Contribution of substrates toward total rate of ATP synthesis. All values represent averages from the data generated in Table 1

Full size image
Fig. 4
figure 4

Effect of taurine deficiency on creatine phosphate/ATP ratio. Hearts from taurine-deficient and control rats were perfused as described in Fig. 1. Following 20 min of perfusion, hearts were frozen with aluminum tongs placed in liquid nitrogen. After extraction, creatine phosphate and ATP content were determined enzymatically. Values shown represent means ± SEM of 5 hearts. The asterisk denotes a significant difference between control and taurine-deficient hearts (p < 0.05)

Full size image

Taurine deficiency, PPARα and energy metabolism

One of the factors that modulates the relationship between glucose and fatty acid metabolism in the failing heart is the level of the nuclear receptor transcription factor, PPARα (Barger and Kelly 2001). Like the failing heart, the taurine-deficient heart contains reduced PPARα levels (Fig. 5).

Fig. 5
figure 5

Effect of taurine deficiency on myocardial PPARα content. In the upper panel are representative Western blots of PPARα from wild-type and taurine-deficient hearts using β-actin as the loading control. Values shown in the lower panel represent means ± SEM of 6–9 hearts. The asterisk denotes a significant difference between wild-type and taurine-deficient hearts (p < 0.05)

Full size image

One of the key rate-limiting proteins of fatty acid metabolism in the heart is carnitine palmitoyltransferase I (mCPT-1), which catalyzes the formation of long chain fatty acyl carnitine from long chain fatty acyl CoA. Because long chain fatty acyl carnitine, but not long chain fatty acyl CoA, can be transported via a translocase into the mitochondria, mCPT-1 is required for oxidation of long chain fatty acids by the mitochondria. Therefore, we examined the levels of mCPT-1 in isolated mitochondria of wild-type and taurine-deficient hearts. As seen in Fig. 6a, the levels of mCPT-1 are reduced in the taurine-deficient heart, a finding consistent with the observed decrease in PPARα. However, PPARα also modulates mCPT-I activity by regulating the myocardial content of malonyl CoA, a metabolite that inhibits mCPT-1 activity. The enzyme catalyzing the biosynthesis of malonyl CoA, acetyl CoA carboxylase, is inhibited by phosphorylation while the degradation of malonyl CoA to acetyl CoA is catalyzed by the enzyme, malonyl CoA decarboxylase, whose expression is upregulated by PPARα (Barger and Kelly 2001; Campbell et al. 2002). In accordance with the effect of taurine deficiency on myocardial PPARα content, we found that the content of malonyl CoA decarboxylase is reduced in the TauTKO heart (Fig. 6b). Also acting to elevate malonyl CoA content of the taurine-deficient heart is a decrease in the phosphorylated state of acetyl CoA carboxylase (Fig. 6c). The net effect of these changes is a reduction in long chain fatty acid metabolism.

Fig. 6
figure 6

Modulation of CPT-1 by taurine deficiency. a Effect of taurine deficiency on CPT-1 expression. In the upper panel are representative Western blots of mCPT-1 from wild-type and taurine-deficient hearts using β-actin as the loading control. Values shown in the lower panel represent means ± SEM of 6–9 hearts. The asterisk denotes a significant difference between wild-type and taurine-deficient hearts (p < 0.05). b Effect of taurine deficiency on malonyl CoA decarboxylase expression. In the upper panel are representative Western blots of malonyl CoA decarboxylase from wild-type and taurine-deficient hearts using β-actin as the loading control. Values shown in the lower panel represent means ± SEM of 6–9 hearts. The asterisk denotes a significant difference between wild-type and taurine-deficient hearts (p < 0.05). c Effect of taurine deficiency on phosphorylation status of acetyl CoA carboxylase (ACC). In the upper panel are representative Western blots of phosphorylated-acetyl CoA carboxylase from wild-type and taurine-deficient hearts using β-actin as the loading control. Values shown in the lower panel represent means ± SEM of 6–9 hearts. The asterisk denotes a significant difference between wild-type and taurine-deficient hearts (p < 0.05)

Full size image

Effect of taurine deficiency on glucose oxidation

A rate-limiting event in glucose oxidation is the rate of pyruvate oxidation by pyruvate dehydrogenase (PDH). The PDH complex is highly regulated, with PPARα, substrates and products playing important roles in the modulation of PDH activity (Stanley et al. 2005). In the heart, the PDH complex consists of three enzymes, one that catalyzes the irreversible conversion of pyruvate to acetyl CoA and two enzymes that regulate the activity of PDH through phosphorylation–dephosphorylation. Pyruvate dehydrogenase kinase (PDK) catalyzes the phosphorylation and inactivation of PDH and is upregulated by PPARα (Stanley et al. 2005). PDK4 is the primary PDK isoform in the heart; therefore, we examined the effect of taurine deficiency on the myocardial levels of PDK4. In agreement with the PPARα findings, we observed that both PDK4 and phosphorylated PDH levels in the taurine-deficient heart are significantly reduced in comparison to those of the wild-type heart (Fig. 7a, b). Because the phosphorylated form of PDH is less active than the dephosphorylated form, the reduction in glucose oxidation seen in the taurine-deficient heart is not mediated by changes in the phosphorylated state of PDH.

Fig. 7
figure 7

Effect of taurine deficiency on the pyruvate dehydrogenase complex. a Effect of taurine deficiency on PDK-4 expression. In the upper panel are representative Western blots of PDK-4 from wild-type and taurine-deficient hearts using β-actin as the loading control. Values shown in the lower panel represent means ± SEM of 6–9 hearts. The asterisk denotes a significant difference between wild-type and taurine-deficient hearts (p < 0.05). b Effect of taurine deficiency on phosphorylated-pyruvate dehydrogenase (phospho-PDH) content. In the upper panel are representative Western blots of phospho-PDH from wild-type and taurine-deficient hearts using β-actin as the loading control. Values shown in the lower panel represent means ± SEM of 6–9 hearts. The asterisk denotes a significant difference between wild-type and taurine-deficient hearts (p < 0.05)

Full size image

Discussion

Effect of taurine on mitochondrial function

The present study reveals that a major function of taurine in the heart is to ensure adequate oxidative metabolism and ATP production. Thus, the taurine-deficient heart is energy starved, as the suppression of oxidative metabolism is not adequately compensated by an increase in anaerobic metabolism. Although the rate of glycolysis is elevated, glucose oxidation is depressed in the taurine-deficient heart. The present study suggests that several factors might account for this phenomenon: (a) reductions in PDH activity decrease flux through the enzyme; (b) pyruvate is diverted into lactate rather than undergoing oxidation by PDH and (c) elevations in the NADH/NAD+ ratio slow flux through the citric acid cycle.

The regulation of the pyruvate dehydrogenase complex is complicated, involving mechanisms that modulate all three enzymes of the complex: (1) PDH phosphatase, which enhances the dephosphorylation and activation of PDH, is stimulated by increases in Ca2+ and Mg2+ (Stanley et al. 2005); (2) PDH kinase (PDK) catalyzes the phosphorylation and inactivation of PDH and is inhibited by pyruvate and ADP and activated by elevations in the acetyl CoA/CoA and the NADH/NAD+ ratios (Stanley et al. 2005). Also regulating PDK is PPARα, which promotes the upregulation of PDK4 (Campbell et al. 2002; Stanley et al. 2005).

In the taurine-deficient heart, PPARα content is reduced, which increases PDH activity by promoting its dephosphorylation (Campbell et al. 2002; Stanley et al. 2005). Moreover, the decrease in ATP levels should free up Mg2+ (Suleiman 1994), which should increase PDH phosphatase activity to further stimulate the dephosphorylation and activation of PDH. Nonetheless, the dominant factor(s) influencing flux through PDH in the taurine-deficient heart are the levels of PDH’s substrates and products in the mitochondrial matrix. The rise in the mitochondrial NADH/NAD+ ratio provides an obvious link between taurine deficiency and flux through PDH. Because the taurine deficiency-mediated decrease in complex I (NADH ubiquinone oxidoreductase) activity exceeds the inhibition of oxidative metabolism (Jong et al. 2012), the mitochondrial NADH/NAD+ ratio rises, mediating a dual effect on PDH activity. First, elevations in the mitochondrial NADH/NAD+ ratio activate PDK4, which in turn reduces PDH activity (Stanley et al. 2005). Second, the increase in the mitochondrial NADH/NAD+ ratio leads to an increase in the cytosolic NADH/NAD+ ratio, which favors the generation of large amounts of lactate via lactate dehydrogenase. In support of the second scenario, we found that the rate of pyruvate production by the isolated, taurine-deficient heart was diminished 35 % while lactate production increased 2.5-fold, suggesting that NADH generated by glycolysis is preferentially used in the formation of lactate at the expense of pyruvate production and oxidation. Thus, as pointed out by Stanley et al. (2005) the myocardium becomes a net lactate producer only when there is accelerated glycolysis in the face of impaired oxidation of pyruvate, a condition that describes the taurine-deficient heart.

Taurine deficiency diminishes citric acid cycle flux and palmitate oxidation

In the present study, the major source of oxidative metabolism and ATP generation of the isolated, perfused hearts from both taurine-deficient and control animals was the acetate added to the perfusion medium. Control and taurine-deficient hearts utilized acetate at a rates of 1.55 ± 0.02 and 1.25 ± 0.01 μmol/g dry wt/h, respectively, which translates into rates of oxygen utilization of 3104 ± 175 and 2490 ± 141 μmol O2/g dry wt/h. Because the total rate of oxygen consumption of the control and taurine-deficient hearts were 5137 ± 115 and 3873 ± 48 μmol O2/g dry wt/h, respectively, the oxidation of acetate represented more than 55 % of total oxygen consumed by both groups of hearts. The rapid rate of acetate oxidation is not surprising because acetate is directly fed into the citric acid cycle (Krebs and Eggleston 1948). Thus, the observed reduction in acetate utilization by the taurine-deficient heart is attributed to the inhibition of the citric acid cycle.

Several factors contribute to diminished citric acid cycle activity in the taurine-deficient heart. First, taurine-mediated inhibition of the respiratory chain and the resulting elevation in the NADH/NAD+ ratio, inhibit several NADH-sensitive citric acid cycle enzymes (α-ketoglutarate dehydrogenase, isocitrate dehydrogenase and citrate synthase) (Weitzman and Dunmore 1969; Lawlis and Roche 1981; Gabriel and Plaut 1984). Second, taurine deficiency is associated with an increase in the generation of reactive oxygen species by the mitochondria (Jong et al. 2012). Both pyruvate dehydrogenase and α-ketoglutarate dehydrogenase depend upon lipoic acid for activity. Because lipoic acid can exist in either an oxidized or reduced state, oxidative stress influences the redox balance of lipoic acid, which alters dehydrogenase activity. Two other citric acid enzymes, whose activity is sensitive to oxidative stress, are aconitase and succinate dehydrogenase. Both enzymes contain Fe-S sites, which can be damaged by reactive oxygen species. Jong et al. (2012) has previously reported that aconitase activity is reduced by taurine deficiency. It is also relevant that oxidative stress is also involved in ventricular remodeling and the development of heart failure, a condition that further suppresses energy metabolism (Ingwall and Weiss 2004). Third, the substrate of the taurine transporter is Na+–taurine; therefore, inhibition of taurine transport reduces cardiomyocyte uptake of both taurine and Na+ (Chapman et al. 1993). Upon coupling with the Na+/Ca2+ exchanger, taurine transporter inhibition reduces both [Na+]i and [Ca2+]i (Chapman et al. 1993; Schaffer et al. 2002). The fall in intracellular Ca2+ contributes to the taurine deficiency-mediated decrease in contractile function (Ramila et al. 2015), which in turn reduces the need for energy production. Calcium plays an important role in maintaining a balance between energy supply and energy demand, with several of the citric acid cycle enzymes being regulated by Ca2+ (Wan et al. 1989). Magnesium also provides a possible link between energy supply and energy demand. According to Suleiman (1994), taurine depletion reduces [Mg2+]i, as the need for Mg2+ is diminished by the decline in ATP biosynthesis, which in turn has been implicated in impaired contractile function. Among the enzymes adversely affected by the decrease in mitochondrial Mg2+ content is PDH phosphatase, whose decrease in activity would contribute to the reduction in pyruvate dehydrogenase activity (Stanley et al. 2005).

The increase in the NADH/NAD+ ratio and citric acid cycle activity also slows both fatty acid β-oxidation and glucose oxidation (Bartlett and Eaton 2004; Stanley et al. 2005). However, the utilization of fatty acids in the taurine-deficient heart is altered by a specific reduction in PPARα content. According to Stanley et al. (2005), one of the dominant events in mild heart failure is remodeling from an adult to a fetal phenotype, a change that is accompanied by a reduction in PPARα. PPARα is capable of regulating several steps in fatty acid metabolism, although its most important action in the taurine-deficient heart likely involves diminished transport of long chain fatty acids into the mitochondria. Three proteins contribute to the transport of long chain fatty acids into the mitochondria (Stanley et al. 2005). mCPT-1 catalyzes the conversion of long chain fatty acyl CoA into long chain fatty acyl carnitine, which is transported across the mitochondrial membrane by a translocase. In the mitochondrial matrix, long chain fatty acyl carnitine is converted back to long chain fatty acyl CoA by CPT-II. The rate-limiting step in the transport process is mCPT-1, which is under regulation by PPARα and the metabolic intermediate, malonyl CoA (Stanley et al. 2005; Jaswal et al. 2011). Like the PPARα−/− heart, the taurine-deficient heart exhibits diminished expression of mCPT-1, an effect consistent with the observed reduction in PPARα content (Figs. 5, 6a). In the PPARα−/− heart, an increase in malonyl CoA content is responsible for a further reduction in mCPT-1 activity (Campbell et al. 2002; Stanley et al. 2005). Malonyl CoA, which is a potent inhibitor of mCPT-1, is synthesized by acetyl CoA carboxylase and degraded by malonyl CoA decarboxylase. In the taurine-deficient heart, the expression of malonyl CoA decarboxylase is diminished while acetyl CoA carboxylase exists largely in a dephosphorylated, active state, a condition that elevates malonyl CoA and reduces mCPT-1 activity (Stanley et al. 2005).

Suppression of ATP biosynthesis by the taurine-deficient heart

The preferred substrate in the present study is exogenous acetate, which was employed because it is converted directly to acetyl CoA for use by the citric acid cycle. In the control heart, the oxidation of acetate accounted for 60 % of total O2 consumption and 55 % of total ATP biosynthesis, values that were only slightly lower than those seen in the taurine-deficient heart. Because the percent contribution of citric acid cycle flux toward ATP biosynthesis (not shown) is also comparable between the two groups of hearts, the reduction in acetate metabolism by the taurine-deficient heart was specifically related to impaired oxidation by the citric acid cycle. On the other hand, significant differences between the two groups of hearts were noted relative to the contribution of endogenous fatty acid metabolism towards total ATP generation.

The rates of glycolysis and fatty acid oxidation are highly regulated by the heart, with the outcome of that regulation determining the preferred substrate of the heart (Stanley et al. 2005). In the control adult heart, exogenous fatty acids are the preferred substrate for ATP generation while glucose metabolism assumes a more significant role in the failing heart. Like the failing heart, the rate of glycolysis is increased in the taurine-deficient heart while the rate of endogenous fatty acid oxidation decreases. Although glycolytic flux was 17 % higher in the taurine-deficient heart than in the control heart, over 50 % of the reducing equivalents generated by glycolysis in the taurine-deficient heart are utilized in the production of lactate, leaving only 384 reducing equivalents for the generation of ATP. By comparison, despite lower rates of glycolysis, 518 reducing equivalents are available for the generation of ATP in the control heart, as <25 % of glycolytic NADH is used for the reduction of pyruvate to lactate. Thus, oxidation of pyruvate and glycolytic NADH is significantly higher in the control heart than in the taurine-deficient heart.

In order to produce ATP from glycolytic NADH, the reducing equivalents generated at the glyceraldehyde-3-phosphate dehydrogenase step must be transported into the mitochondria via the malate–aspartate shuttle. If the activity of the shuttle is inhibited, the reducing equivalents remain in the cytosol, where they contribute to the increase in the lactate/pyruvate ratio. In the present study, we have attributed the increase in the lactate/pyruvate ratio, which is proportional to the cytosolic NADH/NAD+ ratio, to inhibition of the respiratory chain. In the taurine-deficient heart, inhibition of complex I increases the mitochondrial NADH/NAD+ ratio, which in turn inhibits the malate–aspartate shuttle, causing an increase in the cytosolic NADH/NAD+ ratio. Whether taurine deficiency directly alters the function of the malate–aspartate shuttle remains to be determined. However, it is possible that the osmotic imbalance caused by taurine depletion might trigger a compensatory response in the levels of other organic osmolytes, such glutamate and aspartate. Because glutamate and aspartate are components of the malate–aspartate shuttle, the compensatory response could alter the activity of the malate–aspartate shuttle.

Glucose is a worse source of ATP in the taurine-deficient heart than in the control heart, as more than half of the NADH formed by glyceraldehyde-3-phosphate dehydrogenase of the taurine-deficient heart is employed in the reduction of pyruvate to lactate. Moreover, the rate of pyruvate oxidation (also called glucose oxidation) by pyruvate dehydrogenase and the citric acid cycle is significantly less in the taurine-deficient heart than in the control heart. Therefore, the transition from control to taurine deficiency is associated with a 25 % drop in the rate of myocardial ATP synthesis from glucose.

By comparison, the actual rate of ATP biosynthesis from endogenous fatty acids is nearly 50 % less (11.9 vs. 8.1 %) in the taurine-deficient heart than in the control heart perfused with medium containing both glucose and acetate. The major factors decreasing the generation of reducing equivalents by endogenous fatty acids are reduced carnitine palmitoyl transferase-1 activity, diminished β-oxidation and lower rates of citric acid cycle flux. Although the oxidation of both glucose and palmitate is reduced in the taurine-deficient heart, glucose oxidation assumes a more important role in the taurine-deficient heart while the role of endogenous palmitate oxidation declines.

Clinical significance of energy starvation in taurine-deficient heart

The taurine-deficient phenotype resembles the phenotype of the mitochondrial disease, MELAS. Based on PET imaging, MELAS patients exhibit reduced rates of citric acid cycle flux (Arakawa et al. 2010), while in the present study we found that taurine deficiency reduced citric acid cycle flux 24 %. Taurine deficiency also diminished both fatty acid and glucose oxidation, with initiation of a metabolic shift favoring glucose metabolism. However, the dominant defect in both MELAS and taurine deficiency is impaired respiratory chain activity, which ensures significant declines in respiration and ATP generation. We predict that similar changes in energy metabolism dominate the bioenergetics of both the taurine-deficient and MELAS hearts. Because impaired respiratory chain function in both taurine deficiency and MELAS appear to involve impaired conjugation of tRNALeu(UUR), MELAS patients may benefit from taurine therapy. In this regard, it is significant that taurine is approved for the treatment of congestive heart failure in Japan (Azuma et al. 1992).