Tissue-specific pyruvate dehydrogenase complex deficiency causes cardiac hypertrophy and sudden death of weaned male mice

Abstract

Pyruvate dehydrogenase complex (PDC) plays an important role in energy homeostasis in the heart by catalyzing the oxidative decarboxylation of pyruvate derived primarily from glucose and lactate. Because various pathophysiological states can markedly alter cardiac glucose metabolism and PDC has been shown to be altered in response to chronic ischemia, cardiac physiology of a mouse model with knockout of the α-subunit of the pyruvate dehydrogenase component of PDC in heart/skeletal muscle (H/SM-PDCKO) was investigated. H/SM-PDCKO mice did not show embryonic lethality and grew normally during the preweaning period. Heart and skeletal muscle of homozygous male mice had very low PDC activity (∼5% of wild-type), and PDC activity in these tissues from heterozygous females was ∼50%. Male mice did not survive for >7 days after weaning on a rodent chow diet. However, they survived on a high-fat diet and developed left ventricular hypertrophy and reduced left ventricular systolic function compared with wild-type male mice. The changes in the heterozygote female mice were of lesser severity. The deficiency of PDC in H/SM-PDCKO male mice greatly compromises the ability of the heart to oxidize glucose for the generation of energy (and hence cardiac function) and results in cardiac pathological changes. This mouse model demonstrates the importance of glucose oxidation in cardiac energetics and function under basal conditions.

the healthy adult mammalian heart derives 60–90% of ATP from fatty acid oxidation, 10–40% from glucose and lactate oxidation in the tricarboxylic acid (TCA) cycle [through the pyruvate dehydrogenase (PDH) complex (PDC)], and <2% from glycolysis (9, 21, 22). The pathways of uptake and oxidation of fatty acids and glucose are tightly regulated because the heart has limited storage capacity for fatty acids and glucose, and it needs to respond to the changes in fuel availability and energy demands. The switch in the fuel selection provides for constant ATP production despite different developmental, dietary, and pathophysiological conditions. For example, the fetal heart relies more on glucose metabolism, whereas fatty acid oxidation is the primary energy source for the adult heart (15). Different pathological conditions such as cardiac hypertrophy, hypoxia, and ischemia change cardiac metabolism toward glucose utilization, whereas diabetes shifts metabolism toward fatty acid oxidation. Exercise and fasting conditions also lead to increases in fatty acid oxidation and decreases in glucose oxidation (9, 21, 22).

PDC plays a key role in glucose metabolism by linking glycolysis and the TCA cycle. PDC is a highly organized multienzyme complex composed of three catalytic components (PDH, dihydrolipoamide acetyltransferase, and dihydrolipoamide dehydrogenase), one binding protein (dihydrolipoamide dehydrogenase-binding protein), and two regulatory enzymes (a family of PDH kinases and a family of PDH phosphatases) (18). Activity of PDC is regulated through phosphorylation/dephosphorylation of the α-subunit of the PDH component of the complex catalyzed by PDH kinases and PDH phosphatases (10, 23). The PDH component is a heterotetramer composed of two α- and two β-subunits. The human PDHα subunit is encoded by two genes as follows: PDHA1 located on chromosome X (Pdha1 on chromosome X for mouse) is expressed in somatic cells, and PDHA 2 located on chromosome 4 (Pdha2 on chromosome 19 for mouse) is expressed in testis only (6, 16, 25). Deletion of Pdha1 (as was performed in this study) and the consequent absence of PDHα results in the elimination of the PDH reaction, hence causing deficiency of PDC. Systemic PDC deficiency is a metabolic disorder resulting in severe damage of the central nervous system and lactic acidosis (14). About 80% of the reported cases of PDC deficiency involve PDH deficiency (14).

Although glucose and lactate oxidation in the heart provide only a small fraction of its total energy requirements under basal conditions (9, 21, 22), the importance of this minor component in cardiac energy homeostasis is not fully understood. Furthermore, increased glucose oxidation is a common metabolic manifestation of cardiac pathology (7, 22), and recent studies have shown that PDC is regionally downregulated in viable, chronically dysfunctional myocardium (hibernating myocardium) (17). Recently, Zhao et al. (27) observed a marked reduction in the active form of PDC activity in hearts of transgenic mice overexpressing PDH kinase 4 in heart only. Although there was a marked decrease in glucose oxidation, no overt cardiomyopathy was observed in these transgenic mice (27). To investigate the role of PDC in cardiac energy homeostasis, a heart/skeletal muscle-specific PDC knockout (H/SM-PDCKO) mouse model with complete absence (null male mice) and 50% reduction (heterozygous female mice) of PDC activity was developed. Our results show that PDC null mutation in male mice resulted in death in ∼7 days after weaning on a rodent laboratory chow diet. Although weaning of these male mice on a high-fat diet prevented death, myocyte hypertrophy and left ventricular dysfunction still developed. Heterozygous female mice survived normally on a rodent laboratory chow diet but displayed an intermediate level of myocyte hypertrophy and left ventricular dysfunction. Thus this study shows the importance of glucose oxidation via PDC in cardiac energetics and function under basal conditions.

MATERIALS AND METHODS

Generation of heart/skeletal muscle-specific PDC-deficient mice.

All experiments were conducted in accordance with the Guide for the Use and Care of Laboratory Animals and approved by the Institutional Animal Care and Use Committee of the University at Buffalo. To generate H/SM-specific deletion of the Pdha1 gene, females [genotype Pdha1^flox8/Pdha1^flox8 with Pdha1^flox8 alleles having two loxP sites in the intronic sequences flanking exon 8 as described (11)] were bred with transgenic male mice that carried an autosomally integrated Cre gene driven by the heart/skeletal muscle-specific muscle creatine kinase (Mck) promoter (provided by Dr. C. Ronald Kahn, Joslin Diabetes Center) (5). Previous studies have documented that expression of Mck is highly restricted to skeletal muscle and heart (5, 26), and a 6.5-kb fragment containing the Mck promoter and enhancers 1 and 2 can confer the regional specificity to a downstream coding sequence (such as Cre) under transgenic conditions (5, 26). During development, expression of the Mck gene is initiated at embryonic day 17 in the mouse, is ∼40% of maximum at birth, has maximal levels at day 10, and thereafter maintains a consistently high level of expression through life (5).

The progeny were tested for the presence or absence of the Cre transgene by PCR analysis using tail DNA samples around postnatal day 15. The progeny had genotype Pdha1^Δex8/Y, Cre^heart+ (H/SM-PDCKO, homozygous males) and Pdha1^wt/Pdha1^Δex8, Cre^heart+ (heterozygous females). Wild-type control animals were raised by breeding wild-type males and Pdha1^flox8/Pdha1^flox8 females. The progeny had the following genotype: Pdha1^flox8/Y, Cre⁻ (wild-type males) and Pdha1^wt/Pdha1^flox8, Cre⁻ (wild-type females). Pups were nursed by their natural dams and were weaned on a high-fat diet (%calorie distribution: 18 carbohydrate, 67 fat, and 15 protein) ad libitum. All mice were fed a rodent laboratory chow diet (%calorie distribution: 70 carbohydrate, 10.9 fat, and 19.1 protein) and water ad libitum. When indicated, mice were switched to a standard rodent laboratory chow ad libitum. In the postweaning period, the mice were killed, and tissue-specific genotype was determined by PCR analysis using tissue DNA samples. All studies were performed with 2- to 3-mo-old mice.

The genotyping of mice was performed in the following way: genomic DNA was isolated with the Omni Prep kit (Bio-WORLD, Dublin, OH) either from tail clips or from heart, skeletal muscle, and liver of killed animals. The presence of Pdha1 alleles (Pdha1^wt, Pdha1^flox8, and Pdha1^Δex8) and Cre^heart transgene was determined by PCR analysis of the genomic DNA and the specific sets of primers as described previously (19). Tail blood glucose was measured with a glucometer (Ascensia Elite; Bayer, Mishawaka, IN). Plasma insulin concentrations were measured by RIA using a commercially available kit as per the manufacturer's instructions (Linco Research, St. Louis, MO).

Determination of PDC activity.

Total PDC activity was determined in tissue homogenates as described previously (12). Pieces of tissues were homogenized in 50 mM 3-(N-morpholino)propanesulfonic acid buffer, pH 7.4, 80 mM KCl, 2 mM MgCl₂, 0.5 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride, and 0.1 μg/ml leupeptin; frozen-thawn in liquid nitrogen three times; and centrifuged at 600 g for 10 min. The supernatant was used to measure protein by Bradford assay (4) and PDC activity. To measure total PDC activity, PDC was dephosphorylated (to completely activate PDC) with purified recombinant rat PDH phosphatase 1. For this purpose, 50–250 μg of protein (50 μg for heart, 150 μg for skeletal muscle, and 250 μg for liver) were incubated with 10 mM MgCl₂, 1 mM CaCl₂, 4.4 mM dichloroacetate (inhibitor of PDH kinase), and 5 μg of PDH phosphatase 1 for 30 min at 30°C. Total PDC activity was determined by production of ¹⁴CO₂ from [1-¹⁴C]pyruvic acid (Amersham), as described (12).

Echocardiogram analysis.

Transthoracic echocardiograms were performed on conscious mice by a single operator with a 10-MHz phased array transducer (System 5; GE Medical Systems). Studies were performed on wild-type mice (n = 16), heterozygous females (n = 21), and homozygous males (n = 12) ∼8 wk after birth while maintained on a high-fat diet. Studies were repeated 7 ± 1 days after the diet was changed to a rodent laboratory chow. The anterior chest of the mouse was shaved, and parasternal and short-axis two-dimensional images of the left ventricle were obtained to determine correct M-mode cursor positioning. The maximum left ventricular chamber dimension was defined as end-diastole (LVIDd), and end-systole (LVIDs) was defined as minimum chamber dimension, as recommended by the American Society of Echocardiography (20). Left ventricular function was quantified with fractional shortening, defined as [100 × (LVIDd − LVIDs)/LVIDd], calculated from three consecutive beats. Before the diet switch, left ventricular mass was estimated as previously described (24).

Pathologic and morphometric analyses.

Euthanasia was performed after the second echocardiogram. The heart was excised, and left ventricular mass was quantified and expressed relative to body weight (mg/g). Tissues from wild-type mice (n = 4), heterozygous animals (n = 4), and homozygous animals (n = 5) were fixed in formalin and sectioned for morphometric analysis at ×600 magnification with light microscopy, as previously described (3, 13). Nuclear density [N(n)_A] was determined in 69 ± 7 transversely oriented sections per animal with a calibrated ocular reticle (13). Myocyte diameter, d, and nuclear length, D(n), were determined in 80–100 longitudinally oriented myocytes/animal. Myocyte nuclei per unit volume of myocardium, N(n)_v, was obtained using the equation: N(n)_v = N(n)_A/D(n) (3). The total number of nuclei in the ventricle [N(n)_T] was calculated from N(n)_v and volume of the ventricle (V): N(n)_T = N(n)_v × V (3). Myocyte apoptosis was evaluated using terminal uridine nick-end labeling (TUNEL; Chemicon) and epifluorescence with a fluorescein isothiocyanate filter, as previously described (8, 13).

Statistical analysis.

Statistical comparisons between groups were performed with a two-way ANOVA to account for both sex and PDC status. When significant differences were present, post hoc Holm-Sidak testing was performed for PDC-deficient mice vs. control animals (SigmaStat 3.0; SPSS). Paired t-tests were used to compare parameters between the echocardiographic studies. Statistical significance was defined as P < 0.05. With the exception of body weight and left ventricular diastolic dimension, there were no significant differences between male and female control animals so the groups have been combined in Figs. 1- to improve clarity.

Fig. 1.

PCR analysis of the genomic DNA isolated from hearts (H), skeletal muscle (M), and liver (L) of heart/skeletal muscle pyruvate dehydrogenase complex knockout (H/SM-PDCKO) and wild-type (control) mice. The presence of Pdha1^wt allele is indicated by 700-bp DNA fragment, the presence of Pdha1^flox8 allele by 800-bp DNA fragment, and the presence of deletion of exon 8 Pdha1^Δex8 by 400-bp fragment. The Cre^heart transgene was detected by the presence of 240-bp fragment. +, Lane for Pdha1^flox8 positive control (top) and Cre positive control (bottom); −, wild-type (control) mouse DNA sample (presence of Pdha1^wt 700-bp product and absence of Cre product).

RESULTS

Generation of H/SM-PDCKO mice.

There was no noticeable embryonic lethality (as evident from normal litter size). Newborn H/SM-PDCKO mice were indistinguishable from the wild-type animals (generated separately as described in the materials and methods section with the absence of Cre) and apparently grew normally during the preweaning period. Figure 1, top, shows the presence of Pdha1^flox8 (800-bp fragment) for all tissues of H/SM-PDCKO males and Pdha1^Δex8 (400-bp fragment) for hearts and skeletal muscle, as expected, but not for livers from males. The H/SM-PDCKO female mice showed the presence of two alleles (Pdha1^flox8 and Pdha1^wt) in all tissues and had deletion (Pdha1^Δex8) in hearts and skeletal muscle only (Fig. 1, top). The presence of the Cre transgene was indicated by a 240-bp amplification product in all three tissues (Fig. 1, bottom).

When H/SM-PDCKO male mice were weaned on a rodent laboratory chow on the postnatal day 21, they survived for ∼7 days and died suddenly (observed for 10 male mice from 3 mothers). Heterozygous female H/SM-PDCKO mice when weaned on a rodent laboratory chow did well without any mortality for up to 6 mo (until experiment terminated). To improve survival, all mice (both males and females, wild-type and experimental) were initially weaned on a high-fat diet. A switch from a high-fat diet to a rodent laboratory chow was performed later and resulted in death of H/SM PDCKO male mice (survived for only 12 days after diet switch). At age 9 wk both H/SM-PDCKO males and females were significantly heavier than their age- and sex-matched wild-type mice (Fig. 2A). The heart weights of both the H/SM-PDCKO males and females were significantly heavier than their corresponding wild-type mice (Fig. 2B). However, the heart weight-to-body weight ratios were significantly higher for the H/SM-PDCKO male mice only (Fig. 2C). Random blood glucose levels in both heterozygous female and homozygous male H/SM-PDCKO mice fed a laboratory chow diet were not significantly different from that of wild-type mice (females: wild type 171 ± 8 mg/dl and PDCKO 177 ± 14; males: wild type 207 ± 10 and PDCKO 204 ± 10). Random plasma insulin levels in both heterozygous female and homozygous male H/SM-PDCKO mice fed a laboratory chow diet were significantly increased compared with sex-matched wild-type mice (females: wild type 0.656 ± 0.046 ng insulin/ml and PDCKO 1.090 ± 0.160, P < 0.05, and males: wild type 0.810 ± 0.043 and PDCKO 1.330 ± 0.238, P < 0.05).

Fig. 2.

Body weights (A), heart weights (B), and heart weight-to-body weight ratios (C) of 9 wk-old wild-type (control) (open bars) and H/SM-PDCKO (filled bars) mice. Values are given as means ± SE (n = 5–17 mice). **P < 0.01 and ***P < 0.005.

PDC activity in H/SM-PDCKO mice.

Total PDC activity was measured in hearts, skeletal muscle, and livers from wild-type and experimental males and females. Compared with wild-type male mice, PDC activity in hearts from H/SM-PDCKO male mice was at an undetectable level (Fig. 3), and PDC activity was reduced to <5% in skeletal muscle from H/SM-PDCKO male mice (possibly because of the presence of other cell types in skeletal muscle) (Fig. 3). These results confirm the knockout of PDH (and hence PDC) in hearts and skeletal muscle of experimental male animals. In H/SM- PDCKO female mice, PDC activity was reduced by ∼50% in both the heart and skeletal muscle compared with the corresponding tissues from the wild-type mice. This level of reduction in PDC activity is consistent with random inactivation of one of the two X chromosomes in females. As expected, PDC activity in livers from H/SM-PDCKO mice was not different from that of livers from wild-type mice.

Fig. 3.

Total pyruvate dehydrogenase complex (PDC) activity in tissues of H/SM-PDCKO and wild-type mice. Wild-type males and females had similar total PDC activity in different tissues (and hence the results were combined). Open bars, wild type (males + females); hatched bars, females; filled bar, males. UD, undetectable activity. The wild-type PDC activities (mU/mg protein) were as follows: 79, heart; 14, muscle; and 8.2, liver. Values are expressed as %wild-type PDC activity and given as means ± SE of 6–8 determinations (**P < 0.005 and ***P < 0.001).

Cardiac physiology in PDC-deficient mice.

Echocardiographic parameters for all groups of animals are shown in Table 1. Initial studies on a high fat-diet revealed a larger left ventricular diastolic dimension in male mice (both wild type and H/SM-PDCKO). As illustrated in Fig. 4A, left ventricular systolic function as assessed by fractional shortening was reduced in H/SM-PDCKO homozygous male mice, whereas the heterozygous female mice had normal fractional shortening compared with wild-type mice. The systolic dysfunction was the result of an increase in the left ventricular systolic dimension (Table 1).

Table 1.

Echocardiographic parameters

Animals	n	Left Ventricle
		High-Fat Diet			Lab Chow Diet
		Diastole, mm	Systole, mm	FS, %	Diastole, mm	Systole, mm	FS, %
Wild-type female	6	2.3±0.1	0.9±0.1	61±3	2.4±0.1‡	0.9±0.1	64±5
Heterozygotes	21	2.3±0.1	0.9±0.1	62±2	2.5±0.1‡	1.1±0.1‡	57±1‡
Wild-type male	10	2.7±0.1§	1.2±0.1	58±2	2.5±0.1	1.0±0.1	62±2
Homozygotes	12	2.8±0.2§	1.7±0.2*†	41±5*†	3.3±0.2*†‡	2.4±0.3*†‡	31±5*†‡

Values are given as means ± SE; n, no. of animals. FS, fractional shortening.

^*P < 0.05 vs. wild type,

^†vs. heterozygotes,

^‡vs. high-fat diet,

^§and vs. females.

Fig. 4.

Left ventricular fractional shortening in mice maintained on a high-fat diet and ∼7 days after the switch to a rodent chow diet. Measurements were performed as described in materials and methods. Open bar, high-fat diet; shaded bar, rodent chow diet. Values are given as means ± SE; n = 16 wild type (both female and male mice), 21 hetrozygotes (females), and 12 homozygotes (males). *P < 0.05 vs. high-fat diet.

Following the transition to a rodent laboratory chow diet, there was evidence for ventricular compromise among both groups of PDC-deficient mice. Fractional shortening declined in both the heterozygous and homozygous mice compared with their function on a high-fat diet (Fig. 4B), with function in the homozygous male mice significantly worse than in heterozygous female mice. In all but the homozygous PDC-deficient mice there was an increase in left ventricular diastolic dimension, consistent with overall growth. Furthermore, in both groups of PDC-deficient mice, the diet change was associated with an increase in the left ventricular systolic dimension, consistent with the reduction in systolic function (Table 1).

Cardiac pathology in PDC-deficient mice.

As shown in Fig. 5, the alterations in left ventricular dimensions and function in homozygous H/SM-PDCKO male mice were associated with marked myocyte and left ventricular hypertrophy (normalized to body weight). However, left ventricular hypertrophy was not evident when left ventricular mass was estimated by echocardiography while mice were being fed a high-fat diet (male control 83 ± 3 vs. homozygous 74 ± 8 mg, P = 0.36; female control 50 ± 5 vs. heterozygous 54 ± 5 mg, P = 0.69). Morphometric analysis (Table 2) revealed significant increases in myocyte diameter in both PDC-deficient groups, and the myocyte diameter was even greater in the homozygous male than heterozygous female mice (Fig. 5). Cellular hypertrophy in these two groups was confirmed by commensurate reductions in myocyte nuclear density (Fig. 5B). Estimates of total myocyte nuclei in the left ventricle were similar in all groups of animals (Table 2), arguing against a role of myocyte loss with compensatory hypertrophy. Furthermore, TUNEL staining did not identify apoptosis in any sample. As illustrated in the representative images in Fig. 5A, there was no difference in connective tissue staining among the groups of mice.

Fig. 5.

Myocardial micrographs (A) and left ventricular (LV) mass (left), myocyte diameter (middle), and nuclear density (right) (B) in wild-type and PDC-deficient heterozygous female and homozygous male mice. A: representative periodic acid-Schiff stained slides from the three groups of animals (n = 4–5 animals/group). Original magnification ×600. B: left ventricular mass was corrected for body weight. Open bar, wild-type mice; shaded bar, heterozygous females; filled bar, homozygous males. Values are given as means ± SE (n = 4–5/group). P < 0.05 vs. wild type (*) and vs. heterozygous females (†).

Table 2.

Myocyte morphometry

Cellular Parameters	PDC Deficient		Wild Type
	Homozygotes	Heterozygotes
Nuclear density, nuclei/mm2	576±12*	721±79*	1,000±45
Nuclear length, μm	11.6±0.5	11.0±0.4	11.3±0.4
Cell diameter, μm	19.0±0.3*†	17.2±0.2*	15.3±0.2
Cell volume, μm3/nucleus	20,263±1,152*	15,688±1,479	11,382±866
Cell length, μm/nucleus	72±4	67±5	62±4
Total nuclei, ×106	6.9±0.4	5.9±0.5	7.0±0.6

Values are given as means ± SE; n = 4–5 animals. PDC, pyruvate dehydrogenase complex.

^*P < 0.05 vs. wild type

^†and vs. heterozygotes.

DISCUSSION

In this study, we used Cre-loxP gene targeting to inactivate the Pdha1 gene selectively in the heart and skeletal muscle to investigate its effects on heart development and function. This murine model has allowed us for the first time to investigate the possible obligatory requirement of glucose and lactate oxidation in the developing heart under normal dietary and physiological conditions. Interestingly, there was no embryonic or immediate postnatal lethality of H/SM-PDCKO male mice. This is most likely due to 1) the late initiation of Cre expression around fetal day 17 and 2) increased availability of fatty acids derived from milk fats during the suckling period. However, weaning of male H/SM-PDCKO mice on a rodent laboratory chow on postnatal day 21 resulted in mortality of all affected male mice in ∼7 days. In contrast, when male H/SM-PDCKO mice were weaned on a high-fat diet, they survived for several months (at which time experiments were terminated). Heterozygous PDC-deficient female mice experienced no mortality upon weaning on a rodent chow diet. The major findings of this study are 1) impairment of left ventricular systolic function of adult H/SM-PDCKO male mice irrespective of their dietary regimen during the postweaning period, 2) the development of myocyte and left ventricular hypertrophy when H/SM-PDCKO male mice are switched to a rodent laboratory chow diet, and 3) the development of an intermediate degree of cardiac hypertrophy without any mortality in heterozygous H/SM-PDCKO female mice.

Myocardial metabolism is dependent on the arterial concentrations of substrates and hormones, and its work load is supported by coordinated carbon flux through the major metabolic pathways. The mitochondrial fuel selection is reciprocally modulated by 1) malonyl-CoA generated from acetyl-CoA derived from glucose carbons and its effect on fatty acyl-CoA transport via the action of carnitine-palmitoyl-CoA transferase I and 2) by β-oxidation of fatty acyl-CoAs to acetyl-CoA and NADH and their inhibitory action exerted on the PDC by activating PDH kinases (10, 23). We suggest that metabolic flexibility in substrate switching under normal physiological conditions is lost in hearts of H/SM-PDCKO male mice. Glucose and lactate oxidation via PDC was nearly absent, and any increase in the rate of anaerobic glycolysis appeared to be insufficient for ATP production in affected male mice consuming a rodent chow diet (high carbohydrate/low fat), resulting in cardiac failure within a few days. Interestingly, although active PDC activity was markedly reduced (to a one-tenth level of wild-type controls) in hearts of the transgenic mice overexpressing PDH kinase 4 (27), these mice showed no sign of overt cardiomyopathy, suggesting that the residual active PDC activity was sufficient to afford protection against the development of cardiomyopathy. It should be emphasized that our PDCKO male mice had no detectable PDC activity in the heart. This observation clearly demonstrates the obligatory requirement for carbon flux through the PDC reaction in hearts from mice under normal dietary and physiological conditions. Furthermore, the findings suggest that levels of plasma free fatty acids in affected male mice consuming a rodent chow diet were not sufficient to generate energy for cardiac function. This notion is supported by the observation that H/SM-PDCKO male mice when weaned on a high-fat diet survived for months.

Because expression of muscle creatine kinase begins at embryonic day 17 in rodents (5), the developing fetal hearts in H/SM-PDCKO male mice are deprived gradually of ATP production from glucose oxidation from embryonic day 17 until birth. This would affect the developmental processes in the heart during the late fetal period. Even during the postnatal suckling period, although mouse milk is rich in fat-derived calorie, the contribution from glucose oxidation to the total ATP production by the heart would still be restricted, leading to further impairment in the early postnatal development of the heart. This scenario would account for an early death of H/SM-PDCKO male mice within 7 days after weaning them on rodent laboratory chow.

Compared with normal conditions, a high rate of cardiac work load results in the increase in the flux through PDC by activating this enzyme complex and causes expansion of the TCA cycle pool size at the level of 4-carbon intermediates that is largely independent of cardiac fatty acid oxidation (21). In isolated working hypertrophied hearts perfused with mixed substrates (0.4 mM palmitate, 0.5 mM lactate, and 11 mM glucose), the contribution of glycolysis to ATP production was increased significantly (19%) in hypertrophied hearts at low workloads (aortic afterload of 60 mmHg) than in control hearts, whereas the oxidation of glucose and lactate did not differ between the two groups (1). Although palmitate oxidation remained the primary energy fuel in both groups, the rate of palmitate oxidation was reduced by 47% in hypertrophied hearts than in control hearts (1). At higher workloads (120 mmHg), the extra ATP production was produced primarily from an increase in the oxidation of glucose and lactate in both groups, and the contribution of palmitate oxidation to overall ATP production was reduced. In that study, a reduced contribution of fatty acid oxidation to ATP production in hypertrophied hearts was accompanied by a compensatory increase in glycolysis at a low working load. Because H/SM-PDCKO male mice were unable to generate sufficient energy for cardiac function under normal dietary and resting conditions, they would not be able to perform even for a short duration under a high rate of cardiac work load. In another study using cardiomyocyte-selective insulin receptor knockout mice, there was increased expression of glucose transporter 4 and increased glucose uptake and increased glycolysis in hearts with diminished fatty acid oxidation and impaired cardiac performance (2). These findings indicated an important developmental role of insulin signaling in gene expression and the switching of substrate utilization in myocytes (2). Possible adaptations in gene expression patterns of enzymes in the glycolytic pathway in PDCKO male mice remain to be investigated.

We recognize that our H/SM-PDCKO male mice have PDC deficiency in both the heart and skeletal muscle, resulting in impairment in glucose (and lactate) oxidation in both tissues. However, impairment in glucose oxidation in skeletal muscle could not be the primary reason for early death observed in all H/SM-PDCKO male mice weaned on a rodent laboratory chow diet. Rather, we contend that mortality on a rodent laboratory chow diet was due to impairment in cardiac metabolism in these mice. However, any possible contribution from impaired glucose metabolism in skeletal muscle to the development of heart failure and premature death cannot be ruled out at the present time. The altered metabolism resulted in myocyte and ventricular hypertrophy, which eventually led to reduced systolic function and heart failure. This is consistent with our recent observation of regionally reduced protein levels for multiple components of PDC and a ∼50% reduction in total PDC activity in hibernating myocardium (17). This state is also associated with regional myocyte cellular hypertrophy from apoptosis-induced myocyte loss arising from chronic repetitive ischemia (13). While speculative, the present results raise the possibility that the metabolic adaptations of hibernating myocardium may arise as a common response to cellular myocyte hypertrophy that is accompanied by impaired contractile function. The mouse model presented here demonstrates the importance of glucose oxidation in cardiac energetics and function under basal dietary conditions.

GRANTS

This study was supported in part by National Institutes of Health Grants DK-20478 (M. S. Patel), HL-81722 (J. A. Fallavollita), and HL-55324 and HL-61610 (J. M. Canty, Jr.). G. Suzuki was the recipient of a Buswell Fellowship from the University at Buffalo.