Coenzyme A–mediated degradation of pyruvate dehydrogenase kinase 4 promotes cardiac metabolic flexibility after high-fat feeding in mice

Abstract

Cardiac energy is produced primarily by oxidation of fatty acids and glucose, with the relative contributions of each nutrient being sensitive to changes in substrate availability and energetic demand. A major contributor to cardiac metabolic flexibility is pyruvate dehydrogenase (PDH), which converts glucose-derived pyruvate to acetyl-CoA within the mitochondria. PDH is inhibited by phosphorylation dependent on the competing activities of pyruvate dehydrogenase kinases (PDK1–4) and phosphatases (PDP1–2). A single high-fat meal increases cardiac PDK4 content and subsequently inhibits PDH activity, reducing pyruvate utilization when abundant fatty acids are available. In this study, we demonstrate that diet-induced increases in PDK4 are reversible and characterize a novel pathway that regulates PDK4 degradation in response to the cardiac metabolic environment. We found that PDK4 degradation is promoted by CoA (CoASH), the levels of which declined in mice fed a high-fat diet and normalized following transition to a control diet. We conclude that CoASH functions as a metabolic sensor linking the rate of PDK4 degradation to fatty acid availability in the heart. However, prolonged high-fat feeding followed by return to a low-fat diet resulted in persistent in vitro sensitivity of PDH to fatty acid–induced inhibition despite reductions in PDK4 content. Moreover, increases in the levels of proteins responsible for β-oxidation and rates of palmitate oxidation by isolated cardiac mitochondria following long-term consumption of high dietary fat persisted after transition to the control diet. We propose that these changes prime PDH for inhibition upon reintroduction of fatty acids.

Introduction

Metabolic flexibility, the capacity for adaptive fuel source utilization, is a hallmark of healthy cardiac tissue (1–3). By adjusting the rates of glucose and fatty acid oxidation, the heart maintains consistent energy production despite fluctuations in dietary intake and energy demand. Impairment of the pathways that regulate substrate preference leads to metabolic inflexibility, evident in obese and diabetic populations (1, 2, 4, 5). Cardiac tissue from obese individuals has a higher than normal preference for fatty acid oxidation that comes at the expense of inducible glucose utilization (2, 4, 6, 7). This disconnect initiates long-term metabolic complications such as impaired insulin signaling and sensitivity (2, 8, 9). Understanding the regulatory networks that coordinate inducible substrate utilization is therefore necessary for the design of treatments targeting pathological conditions associated with metabolic inflexibility.

The rate of cardiac glucose oxidation is dictated, in large part, by the multisubunit enzyme pyruvate dehydrogenase (PDH)² (10–14). The PDH complex decarboxylates pyruvate, the product of glycolysis, forming acetyl-CoA and NADH upon reduction of NAD⁺. Acetyl-CoA can then be further oxidized by the Krebs cycle, ultimately driving ATP production through oxidative phosphorylation. PDH modulates glucose metabolism because of its sensitivity to multiple metabolic inputs. PDH activity is regulated by reversible phosphorylation of its E1 subunits resulting from the competing activities of pyruvate dehydrogenase kinases (PDK1–4) and phosphatases (PDP1–2) with which PDH stably associates. PDPs and PDKs are sensitive to the presence of numerous metabolites, enabling the integration of complex metabolic changes into a PDH phosphorylation/activation status, which reflects the nutritional and energetic needs of the cell (10–14). In general, compounds that reflect high energetic and reductive potential (i.e. ATP, acetyl-CoA, and NADH) promote PDH phosphorylation and inactivation. In contrast, compounds and ions indicative of low energetic and reductive potential or increased demand (i.e. ADP, CoASH, NAD⁺, and Ca²⁺) induce dephosphorylation and activation of PDH.

Rapid regulation of the expression of PDK4, one of the three cardiac-specific PDK isoforms, confers critical sensitivity of PDH to nutritional status (8, 15). Following a single high-fat meal, PDK4 mRNA and protein content increase, resulting in PDH inactivation and diminished mitochondrial pyruvate utilization (8). Similarly, an overnight fast leads to increased expression of PDK4 in cardiac and skeletal muscle, preserving limited glucose supplies for glycolytic tissues such as the brain (8, 16, 17). However, persistent PDK4 up-regulation results in metabolic inflexibility and loss of insulin signaling (8, 18, 19). PDK4 degradation would therefore be essential to reduce PDK4 content upon return to a control diet. Evidence for such a pathway is provided by our previous findings that PDK4 has a short half-life relative to other PDK isoforms and is specifically degraded by the mitochondrial protease Lon (20). Moreover, we demonstrated that the rate of PDK4 turnover is reduced in the presence of fatty acids, indicating that protein degradation represents a previously underappreciated facet of PDK4 regulation and, in effect, cardiac PDH activity and metabolic flexibility (20).

Dissociation of PDK4 from the PDH complex governs the susceptibility of PDK4 to degradation by the Lon protease (20). Metabolites promoting PDK4/PDH dissociation may therefore act as in vivo metabolic sensors linking the rate of PDK4 degradation to the nutritional needs of the cell. We therefore sought to identify metabolite(s) that regulate PDK4 dissociation and degradation to determine their role in reestablishing PDH activity and metabolic flexibility following changes in dietary fat content.

To distinguish between the effects of short- and long-term high dietary fat consumption on the plasticity of metabolic regulation, mice were fed a diet high in fat content (60% kilocalories from fat) for 3 or 28 days, followed by transition to a low-fat diet (10% kilocalories from fat). We provide evidence that high dietary fat–induced increases in cardiac PDK4 content are reversed upon transition to a low-fat diet irrespective of the duration of consumption of high fat. We also demonstrate that cardiac CoASH levels are significantly reduced during consumption of a high-fat diet and restored to basal levels upon return to a low-fat diet. In vitro assays demonstrate that CoASH selectively promotes PDK4/PDH dissociation and PDK4 degradation. Therefore, diet-dependent fluctuations in cardiac CoASH levels are a likely mechanism by which PDK4 degradation is linked to cellular fatty acid availability. However, long-term (28 days) relative to short-term (3 days) consumption of high dietary fat results in persistent in vitro sensitivity of PDH to fatty acid–induced inhibition upon transition to a low-fat diet. Offering one potential explanation, we provide evidence that long-term high-fat feeding results in prolonged changes in the expression of specific proteins associated with β-oxidation. These changes likely promote β-oxidation and PDH inhibition when fatty acids are reintroduced, despite reductions in PDK4 content.

Results

Transient changes in cardiac PDK4 mRNA and protein content following short- and long-term high-fat feeding

To investigate the reversibility of diet-induced increases in PDK4 content, mice were fed a high-fat diet for 3 or 28 days. A subset of these mice was then transitioned to a low-fat diet to identify transient versus persistent changes to PDK4 expression (Fig. 1A). High fat consumption for 28 days resulted in a significant increase in mouse body weight (∼30%), which remained elevated upon transition to a low-fat diet for 10 days (Fig. 1B). Short-term high-fat feeding had no significant effect on body weight. In general agreement with previous findings (8), high dietary fat resulted in an ∼4-fold increase in PDK4 mRNA (Fig. 1C) and an ∼2-fold increase in PDK4 protein (Fig. 1D) irrespective of diet duration (3 or 28 days). Following 3 days of high-fat feeding, PDK4 mRNA and protein content returned to basal levels by 3 days on the low-fat diet (Fig. 1, C and D). In contrast, 28 days of high dietary fat followed by return to a control diet for 3 days resulted in substantial variability in PDK4 mRNA expression and protein content (data not shown). Further investigation revealed that these differences were likely due to mice initially exhibiting highly variable food intake upon return to a low-fat diet. It is well known that fasting results in an increase in PDK4 mRNA and protein expression (16, 17, 21–24). However, by 10 days of low dietary fat consumption, after normalization of dietary intake, basal PDK4 mRNA and protein content were restored to the levels observed in mice fed only a low-fat diet (Fig. 1, C and D). Therefore, the heart maintains the ability to re-establish basal PDK4 content following short- or long-term exposure to high dietary fat.

Figure 1.

Increases in PDK4 mRNA and protein in response to high dietary fat are reversible upon transition to a low-fat diet. A, schematic depicting the durations of high-fat diet (HFD) and low-fat diet (LFD) feeding used in each dietary regime. B, body weights from mice fed the indicated diets. C and D, relative PDK4 mRNA (C) and protein (D) content were quantified (n = 5 and n = 4 biological replicates for mRNA and protein, respectively) by quantitative real-time PCR and Western blotting using cardiac tissue collected from mice fed the indicated diets. Scatterplot data are represented as the mean, and error bars represent ± S.D. *, p < 0.05.

CoASH-dependent PDK4/PDH dissociation promotes PDK4 degradation

Reductions in PDK4 protein content following transition to a low-fat diet indicate a role for PDK4 degradation. In a previous study, we demonstrated that degradation of PDK4 by the Lon protease requires dissociation of PDK4 from the PDH complex (20). We sought to determine metabolites that govern the interaction of PDK4 with PDH and, ultimately, degradation of PDK4. The relative levels of PDK4 associated and dissociated from the PDH complex were investigated following precipitation of the PDH complex with low concentrations of PEG as described previously (20). PDH was precipitated after preincubation of solubilized cardiac mitochondria with a series of metabolites that reflect the energetic needs of the cell (NAD⁺, NADH, CoASH, acetyl-CoA, and ADP). Of the metabolites tested, CoASH alone promoted substantial dissociation of PDK4 from the PDH complex (Fig. 2A). CoASH induced a concentration-dependent dissociation of PDK4 from PDH, with 60% dissociation at 250 μm CoASH (Fig. 2B). Interestingly, CoASH-induced PDK4 dissociation was not affected by a 10-fold excess of acetyl-CoA, suggesting no competition for binding. CoASH-induced dissociation was specific to PDK4, as the other PDK isoforms present in the heart (PDK1 and 2) remained associated with PDH at concentrations of CoASH that induced significant dissociation of PDK4 (Fig. 2B). This provides an explanation for why PDK4 was observed previously to be specifically targeted for rapid degradation (20).

Figure 2.

Cardiac CoASH regulates PDK4 degradation by promoting dissociation from the PDH complex. A, mouse heart mitochondria were isolated, lysed, and preincubated (0.3 mg/ml) for 20 min at 4 °C in the presence of the indicated metabolites. Large-molecular-weight complexes were then precipitated with PEG (3%, 1.5 h at 4°C). The presence of PDK4 and PDH (anti-lipoic acid) in the pellet (P) and soluble (S) fractions was determined by Western blotting. B, the PDH complex was precipitated by PEG as in A with the indicated concentrations of CoASH and acetyl-CoA (Ac-CoA). Quantification of Western blotting band intensities (n = 4–5 biological replicates for each condition) was used to determine the percentage of PDK1, PDK2, and PDK4 bound to the PDH complex under each condition. C, Lon protease activity was stimulated in lysed mitochondria (7.5 mg/ml) preincubated for 15 min at 4 °C in the presence or absence of CoASH (125 μm). Following reaction quenching, PDK4 expression levels were determined by Western blot analysis. D, quantification of replicates performed as in C (n = 4–5 biological replicates for each time point). All data are represented as the mean, and error bars represent ± S.D. *, p < 0.05; **, p < 0.01; ***, p < 0.001; n.s., not significant.

To test whether CoASH could directly stimulate PDK4 degradation, Lon-dependent disappearance of PDK4 was evaluated in lysed mitochondria using an assay developed previously (20). Although little to no PDK4 degradation was observed in the absence of any added CoASH, the presence of 125 μm CoASH stimulated PDK4 degradation, resulting in an ∼50% depletion of PDK4 in 120 min (Fig. 2, C and D). CoASH was therefore further investigated in vitro and in vivo as a candidate metabolite that regulates PDK4 degradation.

Fatty acid metabolism leads to diminished cardiac [CoASH]

CoASH is converted to acetyl-CoA and acyl-CoA upon oxidation of pyruvate and fatty acids. Isolated cardiac mitochondria were incubated with pyruvate or the fatty acid palmitoyl-carnitine to determine the effects of each metabolite on CoASH concentration. Fatty acid utilization resulted in a substantial reduction in CoASH content relative to that observed with pyruvate as a respiratory substrate (Fig. 3A). To determine whether in vitro effects are reflected in vivo, heart tissue was prepared from short- and long-term high-fat diet–fed mice, and mice transitioned from a high-fat to a low-fat diet. Consumption of a high-fat diet for 3 or 28 days resulted in a significant decrease in cardiac CoASH content relative to that observed in mice fed a control diet (Fig. 3B). Moreover, following short- or long-term consumption of a high-fat diet, decreases in CoASH content were reversed upon return to a low-fat diet (Fig. 3B). Therefore, cardiac CoASH levels are responsive to dietary fat content and inversely correlate with the steady-state expression of PDK4.

Figure 3.

Mitochondrial and cardiac CoASH content declines in response to fatty acid utilization and high dietary fat. A, mouse heart mitochondria were isolated and incubated (2 mg/ml) for 2 min at room temperature with pyruvate (100 μm) and malate (1.0 mm) or palmitoyl-carnitine (25 μm) and malate (1.0 mm) as respiratory substrates. CoASH was then extracted in 5% metaphosphoric acid and resolved and quantified using HPLC and electrochemical detection (n = 4 biological replicates). B, pulverized cardiac tissue collected from mice fed the indicated diets for the indicated durations were used for CoASH extraction and quantification (n = 5–6 biological replicates). HF, high-fat; LF, low-fat. Scatterplot data are represented as the mean, and error bars represent ± S.D. **, p < 0.01; ***, p < 0.001.

Long-term high-fat feeding sensitizes PDH to persistent inhibition in the presence of fatty acids

We sought to determine whether reestablishment of basal PDK4 mRNA and protein content upon return of high fat fed mice to control diet results in a reduction in the in vitro sensitivity of PDH to fatty acid-induced inhibition. To define changes in the regulation of PDH, purified mitochondria were incubated with palmitoyl-carnitine and/or pyruvate as respiratory substrates. PDH activity was assayed following induction of ADP-dependent state 3 respiration. Mitochondria from mice fed a high-fat diet for 3 days relative to a control diet exhibited an ∼50% reduction in PDH activity when respiring on palmitoyl-carnitine and pyruvate (Fig. 4A). In contrast, in the absence of palmitoyl-carnitine, an ∼20% reduction in PDH activity was evident (Fig. 4B). Thus, in line with findings reported previously, addition of the fatty acid substrate palmitoyl-carnitine exacerbated diet-induced PDH inhibition (Fig. 4, A and B) (8). As with the increase in PDK4 content, extending the duration of high-fat feeding from 3 to 28 days had no further effect on the extent of fatty acid–induced PDH inhibition. However, 3-day and 28-day high-fat feeding exerted marked differences in the ability of cardiac PDH to recover basal activity following transition to a low-fat diet. In mice fed a high-fat diet for 3 days, control levels of PDH activity were reached upon return to a low-fat diet (10 days) in mitochondria respiring with palmitoyl-carnitine and/or pyruvate (Fig. 4, A and B). In contrast, fatty acid–induced inhibition of PDH was a more persistent phenomenon in mice fed high dietary fat for 28 days. When assayed in the presence of pyruvate and palmitoyl-carnitine, inhibition of PDH persisted up to 20 days after transition to a low-fat diet (Fig. 4A). As reflected in the attainable PDH activity in mitochondria respiring on pyruvate alone (Fig. 4B), differences between dietary durations were not due to a change in the maximum activity of PDH or, as reported previously, the protein content of PDH subunits (8).

Figure 4.

The sensitivity of PDH to fatty acid–induced inhibition persists following prolonged high-fat feeding. A and B, cardiac mitochondria were isolated from mice fed each of the indicated dietary regimes. PDH activity was assayed following (2 min) induction of state 3 respiration upon addition of 0.25 mm ADP to mitochondria with the respiratory substrates pyruvate (100 μm) and malate (1 mm) in the presence (A) or absence (B) of palmitoyl-carnitine (25 μm). Data are presented as the average of two technical duplicates, each for five biological replicates. d, day; HF, high-fat; LF, low-fat. C, mitochondria respired as in A and were then transferred to Laemmli buffer for Western blot analysis using antibodies specific for the indicated PDH phosphorylation sites. HFD, high-fat diet; LFD, low-fat diet. D–F, quantification of five biological replicates using samples prepared as in C. Scatterplot data are represented as the mean, and error bars represent ± S.D. *, p < 0.05; **, p < 0.01; ***, p < 0.001; with significance indicated in A and B relative to the low-fat diet control.

Potential mechanisms responsible for persistent inhibition of PDH following long-term consumption of high dietary fat

PDH is regulated by both allosteric interactions and phosphorylation of three sites on the E1 subunit: Ser²³², Ser²⁹³, and Ser³⁰⁰ (10, 12, 13). To determine whether phosphorylation at one or more of these sites is responsible for the observed inhibition, Western blot analysis was performed on cardiac mitochondria following state 3 respiration with pyruvate and palmitoyl-carnitine. We observed no significant changes for phosphorylation of Ser²⁹³ or Ser³⁰⁰ under the conditions tested (Fig. 4, C, E, and F). However, Ser²³² exhibited a significant increase (∼3-fold) in phosphorylation following 28 days of high-fat feeding (Fig. 4D). Phosphorylation decreased but remained elevated (∼1.5-fold) relative to samples from control mice 10 days after return to a low-fat diet (Fig. 4, C and F). Previously, this site was found to be hyperphosphorylated by up-regulation of PDK4 (8), although this site is also known to be modified by other PDK isoforms (25, 26). Although significant, the elevation in phosphorylation following return to a control diet was modest (Fig. 4, C and D), raising the possibility of PDH inhibition through other means. The exacerbation of PDH inhibition by inclusion of a fatty acid metabolic fuel source (Fig. 4B) suggested a link between fatty acid metabolism and the observed inhibition of PDH. Furthermore, PDH is known to be directly inhibited by products of fatty acid metabolism (10–14). We therefore sought to determine whether there were any observable changes to fatty acid metabolism in mice fed long-term high dietary fat.

Long-term high-fat feeding leads to persistent changes in the expression of proteins involved in β-oxidation

To determine whether persistent alterations to cardiac fatty acid metabolism result from long-term high-fat feeding, selected reaction monitoring MS was used to quantify the expression of proteins involved in β-oxidation. We found that 28 days of high dietary fat resulted in significant up-regulation of a number of proteins that play a role in the oxidation of fatty acids (Fig. 5, A and B). Up-regulated proteins included those associated with fatty acid transport (Slc25a20, carnitine translocase) in addition to multiple proteins directly involved in fatty acid oxidation (Acadvl, acyl-CoA dehydrogenase; Ech1, enol-CoA hydratase; Hadha, acyl-CoA acetyltransferase). Interestingly, the only protein found to be significantly down-regulated by 28 days of high dietary fat was Acot13 (fatty acid-CoA thioesterase). The level of this enzyme returned to the control value upon transition to a low-fat diet (data not shown). Fatty acid-CoASH thioesterase regenerates CoASH from fatty acyl-CoA (27). As such, changes in the expression of this enzyme may contribute, in part, to the observed diet-induced changes in cardiac CoASH concentrations. In contrast, only one protein, acyl-CoA dehydrogenase family member 11 (Acad11), was up-regulated by 3 days of high fat consumption, returning to basal expression following transition to a low-fat diet (data not shown). Importantly, some of the observed changes following 28 days of high-fat feeding persisted following transition to a low-fat diet (Fig. 6, A–C). Ten days after transition to a low-fat diet, we observed significant up-regulation of Acadvl, Hadha, and carnitine palmitoyltransferase 2 (Cpt2), suggesting an elevated propensity for fatty acid metabolism. These data highlight a transition in cardiac metabolism that occurs following long-term but not short-term high fat consumption that would likely promote fatty acid metabolism.

Figure 5.

Prolonged high-fat feeding results in increased expression of proteins involved in β-oxidation. A, cardiac mitochondria were isolated from mice fed a high-fat diet (HFD) for 3 or 28 days (d). The levels of the indicated mitochondrial proteins involved in β-oxidation were then quantified using selected reaction monitoring MS. Values are reported relative to those obtained for mice maintained on a low-fat diet (LFD), as indicated by the dashed line (n = 5 biological replicates per group). B, schematic of pathways involved in β-oxidation of fatty acids. Numbers indicate the position of enzymes quantified in A. All data are represented as the mean, and error bars represent ± S.D. *, p < 0.05; **, p < 0.01.

Figure 6.

Up-regulation of specific proteins involved in β-oxidation persists upon return to a control diet following prolonged high-fat feeding. A–C, selected reaction monitoring MS was used to quantify the expression of the β-oxidation proteins acyl-CoA dehydrogenase (A, Acadavl), carnitine palmitoyltransferase 2 (B, Cpt2), and acyl-CoA acetyltransferase (C, Hadha) in cardiac tissue from mice fed the indicated dietary regimes. Values are reported relative to expression in cardiac tissue of control mice fed a low-fat diet (LFD, n = 5 biological replicates per group). Scatterplot data are represented as the mean, and error bars represent ± S.D. **, p < 0.05, with significance indicated relative to the low-fat diet control. d, day; HFD, high-fat diet.

Cardiac mitochondria isolated from mice following long-term high-fat feeding exhibit persistent increases in palmitate oxidation

We next sought to determine whether persistent increases in the expression of proteins involved in β-oxidation after 28 days of high dietary fat were associated with elevated rates of β-oxidation. [1-¹⁴C]palmitate oxidation was measured in isolated cardiac mitochondria. As judged by production of CO₂ and acid-soluble metabolites (Fig. 7, A and B), the rate of palmitate oxidation was elevated after 28 days of high dietary fat, an increase that endured even after transition to a low-fat diet for 10 days. These data highlight a persistent change in cardiac mitochondrial metabolism that occurs upon long-term high-fat consumption that promotes fatty acid utilization and PDH inhibition.

Figure 7.

Increased rates of cardiac mitochondrial palmitate oxidation in response to prolonged high-fat feeding persist upon return to a control diet. A and B, cardiac mitochondria were isolated from mice fed the indicated dietary regimes. [1-¹⁴C]palmitate oxidation was assayed as described under “Experimental procedures” and is presented as the etomoxir (CPT-1 inhibitor)-sensitive rate of CO₂ (A) and acid-soluble metabolite (B) production. Data are presented as the average of two technical duplicates, each for seven biological replicates. Scatterplot data are represented as the mean, and error bars represent ± S.D. *, p < 0.05; **, p < 0.005; ***, p < 0.001, with significance indicated relative to the low-fat diet (LFD) control. d, day; HFD, high-fat diet.

Discussion

The consistent production of energy for cardiac function relies on regulatory networks that link the activities of metabolic enzymes to the nutritional and energetic state of the cell, organ, and organism (1–3). A rapid increase in PDK4 expression in response to high dietary fat and subsequent inhibition of PDH is one such process limiting cardiac glucose consumption in the presence of high fatty acid availability (8, 28, 29). However, persistent up-regulation of PDK4 results in metabolic inflexibility and loss of insulin signaling (8, 18, 19), indicating the need for a means to reduce PDK4 content. Previous studies have predominantly focused on the transcriptional regulation of PDK4. However, in a recent study, we discovered an additional mode of regulation: PDK4 can be rapidly degraded by the Lon protease, a process modulated by fatty acid availability (20). In this study, we provide evidence for the in vivo reversibility of diet-induced PDK4 up-regulation and define a novel pathway in which the degradation of cardiac PDK4 is linked to dietary fat content through variations in cardiac CoASH concentrations.

In our previous study, we found that dichloroacetate, an inhibitor of PDK4, promoted degradation of PDK4 by the Lon protease by inducing dissociation of PDK4 from the PDH complex both in vitro and in vivo (20). We reasoned that association of PDK4 with PDH might also be sensitive to the concentrations of endogenous metabolites as a means by which PDK4 degradation could be tied to the nutritional state of an organism. Allosteric regulation of PDK enzymatic activities by metabolites has been defined previously (30). However, PDK regulation by small molecules in the context of PDK turnover has not yet been explored. Of the panel of metabolites tested, CoASH induced dissociation of PDK4 from the PDH complex, resulting in elevated rates of PDK4 degradation. It is interesting to note that dissociation was not observed in the presence of NAD⁺ or ADP, which, like CoASH, are known PDK4 inhibitors that signal for increased energy production (31). Therefore, CoASH exerts effects on PDK4 activity not simply through allosteric inhibition but also by promoting dissociation from the PDH complex and subsequent degradation. Intriguingly, CoASH-dependent PDH dissociation was not observed for the other cardiac PDK isoforms, PDK1 and 2. In fact, both enzymes showed higher association with PDH in the absence of CoASH, in agreement with previous reports of lower K_m values for PDK1 and 2 relative to PDK4 (32). This is consistent with our observation that PDK1 and PDK2 have substantially longer half-lives in vivo than PDK4 (20). PDK4 alone undergoes an increase in expression in the presence of high dietary fat; therefore, diet-dependent protein degradation is specific to this PDK isoform (8).

The role of CoASH as an indicator of fatty acid availability and utilization is supported by the significant decrease in cardiac CoASH levels when mice were fed a high-fat diet and restoration of CoASH to control levels when mice were returned to a low-fat diet. In vitro analyses of mitochondria respiring with pyruvate versus palmitoyl-carnitine indicate that fatty acid metabolism, relative to pyruvate, results in lower levels of CoASH, reflecting the observed in vivo data. A likely explanation is that β-oxidation results in greater incorporation of CoASH into fatty acyl-CoA derivatives, effectively decreasing the cellular pool of CoASH. Importantly, PDK4 levels exhibited an inverse relationship to CoASH content, increasing with a high-fat diet and returning to control levels upon reversion to a low-fat diet, irrespective of the duration of high fat feeding. As such, our observations support a novel function for CoASH as a metabolic signal regulating the susceptibility of PDK4 to degradation.

Ultimately, a decrease in PDK4 content would be predicted to diminish the sensitivity of PDH to fatty acid-induced inhibition. We therefore assayed PDH activity in purified cardiac mitochondria consuming the respiratory substrates palmitoyl-carnitine and/or pyruvate. High-fat feeding resulted in increased expression of PDK4 and enhanced sensitivity of PDH to palmitoyl-carnitine–dependent inhibition, the respective magnitudes of which were independent of dietary duration. However, 3 and 28 days of high dietary fat had different effects on the plasticity of PDH inhibition upon return to a low-fat diet, despite reductions in PDK4 protein levels to control values. Following 3 days of high-fat feeding, the sensitivity of PDH to inhibition by fatty acids was reversible upon transition to a low-fat diet for 10 days. In contrast, PDH remained sensitive to fatty acid-induced inhibition upon return to a low-fat diet for up to 20 days following 28 days of high dietary fat. Changes in PDH content and maximum activity were not responsible. Although persistent hyperphosphorylation of the E1 Ser²³² residue may play a role, the degree of phosphorylation after a transition to a low-fat diet was 1.5-fold higher than in mice maintained on a control diet. This value was significantly lower than the increase in phosphorylation (∼3-fold) observed at 28 days on a high-fat diet. Alternatively, PDH is known to be directly inhibited by products of fatty acid metabolism. Persistent in vitro fatty acid-induced PDH inhibition following 28 days of high dietary fat raises the possibility of a substantial reprogramming of cardiac metabolism, promoting mitochondrial fatty acid oxidation, despite the change in dietary fat intake (4). This is in agreement with a study demonstrating that overexpression of PDK4 results in reprogramming of metabolism, facilitating fatty acid utilization (19).

We observed that a number of proteins involved in β-oxidation exhibited significant increases in content by 28 days, but not 3 days, of high-fat feeding. Furthermore, the proteins Acadvl, Cpt2, and Hadha and mitochondrial palmitate oxidation remained significantly elevated 10 days after transition to a low-fat diet, supporting a reprogramming of cardiac metabolism that is not readily reversed by lowering dietary fat consumption. Nevertheless, cardiac CoASH and PDK4 returned to control levels when mice fed a high-fat diet for 28 days were transitioned to a low-fat diet. These results suggest that metabolic flexibility is reestablished in vivo. However, the persistent in vitro sensitivity of PDH to fatty acid-induced inhibition suggests that increased expression of enzymes involved in β-oxidation and associated rates of mitochondrial fatty acid oxidation promote PDH inhibition. These metabolic alterations may prime cardiomyocytes for future changes in nutrient availability.

Experimental procedures

Mice and diets

Male C57BL/6N mice (Charles River Laboratories) 8–12 weeks of age were used in this study. Mice were fed a low-fat diet (10% fat, 70% carbohydrate, and 20% protein, by kilocalories) or high-fat diet (60% fat, 20% carbohydrate, and 20% protein, by kilocalories) (Research Diets Inc.) ad libitum. Mice were euthanized by cervical dislocation, and hearts were excised. All procedures were approved by the Oklahoma Medical Research Foundation Animal Care and Use Committee.

Quantitative RT-PCR

RNA was extracted with Tripure reagent (Roche) using pulverized cardiac tissue (∼10 mg). Complementary DNA synthesis was performed using the Quantitect reverse transcription kit (Qiagen), following the manufacturer's suggested protocol. quantitative real-time PCR was performed as described previously (8). Target gene expression was normalized to three reference genes (Gapdh, Sdha, and Hspcb), which were unchanged between conditions. The primer pair used to quantify PDK4 mRNA was as follows: forward, 5′-AGGGAGGTCGAGCTGTT-CTC-3′; reverse, 5′-GGAGTGTTCACT-AAGCGGTCA-3′.

Western blot analysis

Western blot analysis was performed as described previously (8). Anti-phospho-PDH E1 (Ser²⁹³), anti-phospho-PDH E1 (Ser³⁰⁰), and anti-phospho-PDH E1 (Ser²³²) were purchased from EMD Millipore. Anti-PDK1 and anti-PDK2 were purchased from Santa Cruz Biotechnology. Rabbit polyclonal antibodies to lipoic acid (for detection of the E2 subunit of PDH) (33) and PDK4 (20) were produced by Biosynthesis, Inc. Antiserum to PDK4 was generated to a mixture of the following peptide sequences linked to keyhole limpet hemocyanin (KLH): CIPSREPKNLAKEKLA, DLVEFHEKSPEDQKALSE, and EFVDTLVKVRNRHHNVVPT. Anti-PDK4 was validated by Western blot detection of PDK4 at the appropriate molecular weight and by confirmation of immunochemically detected changes in PDK4 content with selected reaction monitoring mass spectrometric quantification of PDK4 (20). Horseradish peroxidase–conjugated secondary antibodies (Pierce) and Super Signal West Pico chemiluminescent substrate (Thermo Scientific) were used to visualize primary antibody binding. Western blotting band densitometry was quantified using ImageJ software in the linear range of band intensity. Assignment of molecular weights, as depicted in the figures, was based on utilization of the MagicMark^TM XP Western Protein Standard (Life Technologies).

Isolation of cardiac mitochondria

Mitochondria were purified as described previously. Briefly, hearts were excised from euthanized mice and perfused with ice-cold isolation buffer (10 mm MOPS-NaOH (pH 7.4), 210 mm mannitol, 70 mm sucrose, 1 mm 2K⁺–EDTA). Homogenized cardiac tissue was centrifuged at 550 × g for 5 min at 4 °C, and the supernatant was filtered through cheesecloth. Mitochondria were obtained from the supernatant by centrifugation at 10,000 × g for 10 min at 4 °C. Mitochondria were then resuspended to a concentration of 20–30 mg/ml in isolation buffer. Protein concentrations were determined using the bicinchoninic acid method (Pierce).

Evaluation of PDH activity

Purified mitochondria were resuspended in respiration buffer (10 mm MOPS-NaOH (pH 7.4), 210 mm mannitol, 70 mm sucrose, and 5 mm K₂HPO₄) to a final concentration of 0.25–0.5 mg/ml. State 2 respiration was initiated by addition of 1 mm malate along with a metabolic substrate (100 μm pyruvate and/or 25 μm palmitoyl-carnitine). Following a 2-min incubation at 22 °C, state 3 respiration was initiated by addition of ADP to a final concentration of 0.25 mm. Mitochondria were allowed to respire for 2 min, and then a 170-μl aliquot was removed and transferred to 0.8 ml of 25 mm MOPS (pH 7.4) and 0.05% Triton X-100. PDH activity was measured spectrophotometrically (Agilent, 8452A) as the rate of NADH (340 nm, e = 6,200 M⁻¹ cm⁻¹) production from NAD⁺ following addition of 2.5 mm pyruvate, 0.1 mm CoASH, 0.2 mm thiamine pyrophosphate, 1.0 mm NAD⁺, and 5.0 mm MgCl₂.

Analysis of mitochondrial palmitate oxidation

Oxidation of [1-¹⁴C]palmitic acid by isolated cardiac mitochondria was determined as described previously (34). Briefly, mitochondria were preincubated with or without 100 μm etomoxir (a CPT-1 inhibitor) for 5 min on ice. Mitochondria were then transferred to reaction buffer (100 mm sucrose, 10 mm Tris-HCl, 5 mm KH₂PO₄, 0.2 mm EDTA, 80 mm KCl, 1 mm MgCl₂, 2 mm l-carnitine, 0.1 mm malate, 0.05 mm CoA, 2 mm ATP, 1 mm DTT, and 0.7% BSA/0.1 mm palmitate/0.4 μCi [¹⁴C]palmitate). At 1 h (within the linear range for palmitate oxidation), the mitochondrial reaction mixture was transferred to tubes containing 1 m perchloric acid (1 h with gentle agitation). CO₂ was captured on Whatman filter paper treated with 1 m NaOH placed in the lids of the tubes. The filter papers were then transferred to scintillation vials for counting. Following centrifugation at 14,000 × g, the supernatant containing the acid-soluble metabolites was transferred to a scintillation vial for counting and analysis.

PEG precipitation

Mitochondria were lysed by dilution to 0.3 mg/ml in 25 mm MOPS (pH 7.4) and 0.5% IGEPAL®, followed by centrifugation at 10,000 × g for 10 min at 4 °C. The supernatant was transferred to a fresh tube, and CoASH or acetyl-CoA was added up to 250 μm. The tubes were then incubated on ice for 20 min. The pH was adjusted to 6.4–6.5 using 10% acetic acid, and then PEG was added to a final concentration of 3% (w/v, optimized for the precipitation of PDH), followed by rotation at 4 °C for 1.5 h. PDH-bound proteins were pelleted by centrifugation at 16,000 × g for 20 min at 4 °C. Unbound proteins in the supernatant were then precipitated by adding PEG to 15% (w/v), followed by 1.5-h incubation at 4 °C. The unbound fraction was then pelleted by centrifugation at 16,000 × g for 20 min at 4 °C. PDH-bound and unbound pellets were resuspended in Laemmli buffer and stored at −80 °C until use.

PDK4 degradation assays

Mitochondria were diluted to 7.5 mg/ml in respiration buffer on ice. Mitochondria were lysed by addition of 0.5% IGEPAL, followed by a 15,000 × g spin for 10 min at 4 °C. The supernatant was then brought to 125 μm CoA, followed by a 15-min incubation on ice. Degradation was initiated by addition of a 2× volume of 50 mm Tris-HCl (pH 8.0), 17 mm MgCl₂, 2 mm DTT, and 17 mm ATP. Samples were incubated at 37 °C, with time points taken at t = 0, 15, 30, 60, and 120 min. Reactions were quenched by the addition of 2× Laemmli buffer with protease inhibitor mixture (Roche). Samples were snap-frozen in liquid N₂ and stored at −80 °C until use.

Selected reaction monitoring mass spectrometry analysis

As described previously (35), quantitative proteomics was used to determine the levels of specific proteins. Mitochondrial protein was run 1.5 cm into a 12.5% SDS-PAGE gel (Criterion, Bio-Rad). The gel was then fixed and stained with GelCode Blue (Pierce). The entire lane was cut into ∼1-mm³ pieces. The samples were washed, reduced with DTT, alkylated with iodoacetamide, and digested with trypsin. The peptides were extracted with 70% methanol/5% acetic acid in water. The extract was dried and reconstituted in 1% acetic acid. The samples were analyzed using selected reaction monitoring with a triple quadrupole mass spectrometer (Thermo Scientific, TSQ Vantage) configured with a splitless capillary column HPLC system (Eksigent). The data were processed using the program Skyline (Michael MacCoss), which aligned the various collision-induced dissociation reactions monitored for each peptide and determined the chromatographic peak areas. The response for each protein was taken as the geomean of all peptides monitored. Changes in the relative abundance of the proteins were determined by normalization to the BSA internal standard, with confirmation by normalization to the housekeeping proteins.

Quantification of [CoASH]

The levels of CoASH in mitochondria and cardiac tissue were quantified using reverse-phase HPLC and electrochemical detection. CoASH was extracted from mitochondria or heart homogenate by treatment with 5% metaphosphoric acid. Protein was pelleted by centrifugation (10 min at 16,000 × g). The supernatant was filtered (0.45-μm syringe filters) prior to analysis of CoASH by HPLC and electrochemical detection (Shimadzu HPLC system, ESA Coularray electrochemical detector 5600A set at 750 mV). CoASH was eluted through a C18 column (Phenomenex Luna C18(2), 100 Å, 3 μm, 150 × 4.6 mm) at 0.5 ml/min using an isocratic mobile phase consisting of 25 mm NaH₂PO₄, and 0.5 mm 1-octane sulfonic acid, 1.5% acetonitrile (pH 2.7). CoASH concentrations were calculated employing CoASH standard curves constructed from peak areas.

Statistics

GraphPad Prism was used for statistical calculations. Data are presented as mean ± S.E. Statistical significance was determined by using the two-tailed Student's t test with p values denoted as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Author contributions

C. S., Z. T. Y., and L. I. S. conceptualization; C. S., Z. T. Y., C. A. M., A. E., C. S. K., M. K., and L. I. S. data curation; C. S., Z. T. Y., C. A. M., A. E., C. K., M. K., and L. I. S. formal analysis; C. S., Z. T. Y., C. A. M., C. K., M. K., and L. I. S. validation; C. S., Z. T. Y., C. A. M., A. E., C. K., M. K., and L. I. S. investigation; C. S., Z. T. Y., C. A. M., A. E., C. K., M. K., and L. I. S. methodology; C. S. and Z. T. Y. writing-original draft; C. S., Z. T. Y., C. A. M., A. E., C. K., M. K., and L. I. S. writing-review and editing; C. A. M. and L. I. S. resources; M. K. and L. I. S. supervision; M. K. and L. I. S. funding acquisition; L. I. S. project administration.