Lydia M. Le Page,1 Oliver J. Rider,2 Andrew J. Lewis,2 Vicky Ball,1 Kieran Clarke,1

Edvin Johansson,3 Carolyn A. Carr,1 Lisa C. Heather,1 and Damian J. Tyler1

Increasing Pyruvate Dehydrogenase Fluxas a Treatment for Diabetic Cardiomyopathy:A Combined 13C Hyperpolarized MagneticResonance and Echocardiography StudyDiabetes 2015;64:2735–2743 | DOI: 10.2337/db14-1560

Although diabetic cardiomyopathy is widely recognized,there are no specific treatments available. Altered myocar-dial substrate selection has emerged as a candidatemechanism behind the development of cardiac dysfunctionin diabetes. As pyruvate dehydrogenase (PDH) activityappears central to the balance of substrate use, we aimedto investigate the relationship between PDH flux andmyocardial function in a rodent model of type 2 diabetesand to explore whether or not increasing PDH flux, withdichloroacetate, would restore the balance of substrate useand improve cardiac function. All animals underwent in vivohyperpolarized [1-13C]pyruvate magnetic resonance spec-troscopy and echocardiography to assess cardiac PDH fluxand function, respectively. Diabetic animals showed signif-icantly higher blood glucose levels (10.8 6 0.7 vs. 8.4 6 0.5mmol/L), lower PDH flux (0.0056 0.001 vs. 0.0176 0.002 s-1), and significantly impaired diastolic function (transmitralearly diastolic peak velocity/early diastolic myocardial ve-locity ratio [E/E9] 12.26 0.8 vs. 206 2), which are in keepingwith early diabetic cardiomyopathy. Twenty-eight days oftreatment with dichloroacetate restored PDH flux to normallevels (0.018 6 0.002 s-1), reversed diastolic dysfunction(E/E9 14 6 1), and normalized blood glucose levels (7.5 6

0.7 mmol/L). The treatment of diabetes with dichloroa-cetate therefore restored the balance of myocardialsubstrate selection, reversed diastolic dysfunction, andnormalized blood glucose levels. This suggests thatPDH modulation could be a novel therapy for the treat-ment and/or prevention of diabetic cardiomyopathy.

It is now firmly established that type 2 diabetes contrib-utes to an increased risk for the development of heart

failure (1). Although some of this risk can be attributed toincreased coronary artery disease and hypertension, it isbecoming clear that patients with type 2 diabetes are alsoat risk for the development of “diabetic cardiomyopathy”(2–5), which manifests across a spectrum from subclinicalleft ventricular (LV) diastolic dysfunction (i.e., transmitralearly diastolic peak velocity/early diastolic myocardialvelocity ratio [E/E9]) to overt systolic failure (6). As theincidence of type 2 diabetes is rapidly increasing, under-standing the pathophysiology behind diabetic cardiomy-opathy and developing new treatment strategies is ofincreasing clinical importance.

Cardiac metabolism and altered substrate use are nowemerging as candidate mechanisms underpinning diabeticcardiomyopathy and, as such, are targets for noveltreatments (7,8). The cardiac metabolic changes in type2 diabetes are linked to an increase in circulating fattyacid levels that results from insulin insensitivity and a fail-ure to suppress adipose tissue hormone-sensitive lipase(9). This increase in fatty acid availability, and conse-quently increased cardiac usage, is thought to result ina loss of efficiency between substrate use and ATP pro-duction in the diabetic heart (10). Changes in cardiac sub-strate use precede functional changes in the diabetic heart(11). As a result, metabolic therapies aimed at restoringthe balance of substrate use are an attractive target toimprove, or even prevent, diabetic cardiomyopathy (12).Specifically, diastolic dysfunction could be a good initialindicator of the effect of metabolic therapy, given thatdysfunction is seen in up to 60% of diabetic patients(13), precedes systolic dysfunction (14), and is more

1Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford,U.K.2Division of Cardiovascular Medicine, University of Oxford, Oxford, U.K.3AstraZeneca, Mölndal, Sweden

Corresponding author: Damian J. Tyler, damian.tyler@dpag.ox.ac.uk.

Received 20 October 2014 and accepted 16 March 2015.

© 2015 by the American Diabetes Association. Readers may use this article aslong as the work is properly cited, the use is educational and not for profit, andthe work is not altered.

Diabetes Volume 64, August 2015 2735

METABOLISM

sensitive to changes in myocardial energetics than systolicfunction (15).

One of the potential targets for metabolic therapy ispyruvate dehydrogenase (PDH), a key enzyme thatregulates the balance between carbohydrate and fatmetabolism in the heart. In diabetes, PDH activity isdecreased and pyruvate oxidation is impaired (16,17),resulting in a lack of metabolic flexibility in substrateselection, contributing to the overuse of fatty acids.Therefore, it follows that the restoration of PDH activ-ity may re-establish a normal fuel balance, thereby re-storing cardiac function.

PDH kinase (PDK) is responsible for the phosphoryla-tion, and consequent deactivation, of PDH. Dichloroace-tate (DCA) is a pyruvate mimetic that inhibits all isoformsof PDK, leading to an increase in PDH activity and glucoseoxidation (18). The mechanism of action and isoform-specific inhibition kinetics of DCA have been exten-sively studied and reported (19,20). Previous studies(21,22) have shown that increasing glucose use viaDCA, or via carnitine palmitoyltransferase-1 inhibition,improved LV function in the isolated perfused diabeticrat heart. As yet, there have not been any in vivo stud-ies investigating whether DCA improves cardiac metab-olism or function as a potential therapy for diabeticcardiomyopathy.

Until recently, we have been limited in our ability tomeasure metabolism in vivo. However, advances in hyper-polarized 13C magnetic resonance (MR) spectroscopy (23)have been made that now enable us to obtain a real-time,in vivo assessment of carbohydrate metabolism (24). Incombination with an echocardiographic assessment of car-diac function, hyperpolarized 13C MR spectroscopy pro-vides an ideal tool to investigate the effects of systemictreatments on diabetic myocardial metabolism and func-tion in vivo.

In this study, we aimed 1) to use hyperpolarized[1-13C]pyruvate MR spectroscopy to assess cardiac PDHflux, and echocardiography to assess cardiac function, ina rodent model of type 2 diabetes (25); and 2) to inves-tigate the effects of increasing PDH flux using DCA oncardiac substrate selection and function in the diabeticrodent myocardium.

RESEARCH DESIGN AND METHODS

Twenty-four male Wistar rats (initial body weight ;200 g;Harlan U.K.) were housed on a 12:12-h light/dark cycle inanimal facilities at the University of Oxford. All metabolicstudies were performed between 7:00 A.M. and 1:00 P.M.

with animals in the fed state. All procedures conformedto the Home Office Guidance on the Operation of the Ani-mals (Scientific Procedures) Act of 1986 and to University ofOxford institutional guidelines. To generate a model of type2 diabetes, two groups of eight rats were fed a high-fat dietfor 3 weeks (60% fat, 35% protein, 5% carbohydrate; SpecialDiet Services, Essex, U.K.). After 12 days, these high-fatdiet–fed rats were fasted overnight before receiving an

intraperitoneal injection of 25 mg/kg streptozotocin (STZ;freshly prepared in citrate buffer, 0.655 g citric acid, 0.552 gsodium citrate in 100 mL double-distilled H2O, pH 4) (25).These animals then continued on the high-fat diet for theremainder of the study. This model aimed to generatea moderate type 2 diabetic phenotype by using thehigh-fat feeding to induce peripheral insulin resistance,followed by a low dose of the pancreatic b-cell toxin STZ.STZ is traditionally used at high doses to induce type 1diabetes, as it results in impaired insulin secretion fromthe b-cell. Reed et al. (26) proposed that if a low dose ofSTZ was used after high-fat feeding, the function of theb-cell mass would be modestly impaired without com-pletely compromising insulin secretion, resulting ina moderate impairment in glucose tolerance (i.e., hyper-glycemia in the absence of hypoinsulinemia). The finalgroup of animals (n = 8) were fed a standard chow dietand received no injection of STZ so that they could act ascontrols.

After the initial 3-week period to establish the diabeticmodel, echocardiography was carried out in both thecontrol and diabetic groups. Following this, one group ofdiabetic animals began treatment with DCA for a period of28 days. Dichloroacetic acid (Sigma-Aldrich, Dorset, U.K.)was dissolved in the animals’ drinking water (to a finalconcentration of 1 mmol/L) and neutralized to pH 7.2with NaOH. At the end of the 28-day treatment period,all animals underwent metabolic and functional analysisbefore being sacrificed, and plasma and tissue sampleswere collected. For all in vivo studies, animals were anes-thetized using 2% isoflurane with 2 L/min oxygen.

EchocardiographyEchocardiographic indices were obtained according to therecommendations of the British Society of Echocardiogra-phy. Transthoracic echocardiography was performed incontrol and diabetic animals, by blinded observers, with theuse of a commercially available Vivid I echocardiographysystem (GE Healthcare) using an 11.5 MHz phased array10S-RS pediatric echo probe. Wall thickness and LVdimensions were obtained from a short-axis view at thelevel of the papillary muscles. LV fractional shortening wascalculated as (LVd 2 LVs)/LVd 3 100, where LVs is LVend-systolic diameter and LVd is the LV end-diastolic di-ameter. Two-dimensional guided pulsed-wave Dopplerrecordings of LV inflow were obtained from the apicalfour-chamber view to measure maximal E and late peakvelocity (A). The E parameter provides a measure of thepeak early diastolic mitral inflow velocity, and is affected byleft atrial pressure, LV relaxation, and LV systolic pressure.This results in it being preload sensitive. Tissue Dopplerimaging was therefore also recorded from the medial mitralvalve annulus to record E9 and active LV diastolic myocar-dial velocity (A9). E9 provides a measure of the velocity ofthe mitral medial annulus and is considered a noninvasivesurrogate for LV relaxation. It is significantly less preloaddependent, and the E/E9 ratio is therefore assumed to

2736 PDH Activation in the Diabetic Rat Heart Diabetes Volume 64, August 2015

overcome the influence of ventricular relaxation on Eand provide a preload independent reflection of LV fill-ing pressure.

MR SpectroscopyRats were positioned in a 7T horizontal bore MR scannerinterfaced to a Direct Drive console (Varian MedicalSystems, U.K.), and a home-built 1H/13C butterfly coil (loopdiameter 2 cm) was placed over the chest, localizing thesignal from the heart (27). Correct positioning was con-firmed by the acquisition of an axial proton FLASH image(echo time 1.17 ms; repetition time 2.33 ms; matrix size 643 64; field of view 60 3 60 mm; slice thickness 2.5 mm;excitation flip angle 15°). An electrocardiogram-gated shimwas used to reduce the proton linewidth to ;120 Hz.Hyperpolarized [1-13C]pyruvate (Sigma-Aldrich) was pre-pared by 40 min of hyperpolarization at;1 K, as describedby Ardenkjaer-Larsen et al. (23), before being rapidly dis-solved in a pressurized and heated alkaline solution. Thisproduced a solution of 80 mmol/L hyperpolarized sodium[1-13C]pyruvate at physiological temperature and pH, witha polarization of ;30%. One milliliter of this solution wasinjected over 10 s via a tail vein cannula (dose ;0.32mmol/kg). Sixty individual electrocardiogram-gated 13CMR slice-selective, pulse-acquired cardiac spectra were ac-quired over 60 s after injection (repetition time 1 s; exci-tation flip angle 5°; slice thickness 10 mm; sweep width13,593 Hz; acquired points 2,048; frequency centered onthe C1 pyruvate resonance) (28). The combination of sur-face coil localization with slice selection ensured that thesignals measured were accurately localized to the cardiaclumen (pyruvate) and the front wall of the myocardium(bicarbonate, lactate, alanine). Minimal contribution wasprovided from the skeletal muscle because of the largedifferential in metabolic rate between the contractingmyocardium and the resting skeletal muscle.

Tissue CollectionAnimals were sacrificed in the fasted state with anoverdose of isoflurane (5% isoflurane with 2 L/minoxygen). The heart was removed, washed briefly in PBS,and frozen in liquid nitrogen. Epididymal fat pads wereremoved and weighed, with weights reported relative tobody weight. Blood samples were taken and placed inheparinized tubes before centrifugation at 13,000 rpm for10 min at 4°C. The plasma was then frozen in liquidnitrogen for later biochemical analyses.

Biochemical Analyses

Western BlottingFrozen tissue was crushed and lysis buffer was addedbefore tissue was homogenized; a protein assay estab-lished the protein concentration of each lysate. The sameconcentration of protein from each sample was loadedonto 12.5% SDS-PAGE gels and separated by electropho-resis (29). Primary antibodies for PDK1 and PDK2 werepurchased from New England Biolabs and Abgent, respec-tively; an antibody for PDK4 was donated by Professor

Mary Sugden (Queen Mary University of London, Lon-don, U.K.); primary antibodies for uncoupling protein 3(UCP3) and medium chain acyl-CoA dehydrogenase(MCAD) were purchased from Abcam. Even protein load-ing and transfer were confirmed by Ponceau staining(0.1% weight for volume in 5% volume for volume aceticacid; Sigma-Aldrich), and internal standards were used toensure homogeneity between samples and gels. Bandswere quantified using UN-SCAN-IT gel software (Silk Sci-entific), and all samples were run in duplicate on separategels to confirm results.

Blood AnalysesFasting blood glucose levels were assessed using anOptium Blood Glucose Monitor. Fasting insulin levelswere assessed on postmortem plasma using an insulinELISA kit (Mercodia, Uppsala, Sweden). Plasma metabo-lites (nonesterified fatty acids, LDLs, HDLs, triglycerides,and cholesterol) were assessed using an ABX Pentra 400(Horiba ABX Diagnostics).

Spectroscopic Data AnalysisAll cardiac 13C spectra were analyzed using the AMARESalgorithm in the jMRUI software package (30). Spectrawere direct current offset corrected based on the lasthalf of acquired points. The peak areas of [1-13C]pyruvate,[1-13C]lactate, [1-13C]alanine, and [13C]bicarbonate ateach time point were quantified and used as input datafor a kinetic model based on that developed by Zierhutet al. (31) and Atherton et al. (24). PDH flux was quanti-fied as the rate of 13C label transfer from pyruvate tobicarbonate (32). The rate of 13C label transfer from py-ruvate to lactate and alanine was used as a marker oflactate dehydrogenase activity and alanine aminotransfer-ase activity, respectively.

Statistical AnalysesValues are reported as the mean 6 SEM. Differencesamong the three groups were assessed using ANOVAwith a Holm-Šídák post hoc correction. A Pearson corre-lation was used to assess the correlation between cardiacPDH flux and diastolic dysfunction (E/E9). Statistical sig-nificance was considered at P # 0.05.

RESULTS

Blood Metabolites and Protein ExpressionAt the end of the 7-week study period, the diabetic grouphad significantly elevated fasting blood glucose levelscompared with control animals (Fig. 1A). Fasting plasmainsulin levels, epididymal fat pad weight, and bodyweights were not significantly different between the con-trol and diabetic groups (Fig. 1B and C). Diabetic animalstreated with DCA for 28 days demonstrated fastingplasma glucose concentrations that were significantlylower than those of untreated diabetic animals (Fig. 1A).Fasting plasma insulin levels were not different in treateddiabetic animals compared with untreated (P = 0.30) andcontrol (P = 0.13) animals (Fig. 1B). Epididymal fat pads

diabetes.diabetesjournals.org Le Page and Associates 2737

were similarly unaffected in comparison with both un-treated diabetic (P = 0.39) and control (P = 0.20) animals(Fig. 1C).

The plasma metabolites (Table 1) showed a significantreduction in the plasma triglyceride levels in the untreateddiabetic and DCA-treated diabetic groups relative to con-trols. This finding potentially supports a greater rate of fattyacid oxidation in the diabetic animals that was unaffected byDCA treatment. The only other difference in the plasmalipid profile was a small decrease in the LDL levels measuredin the DCA-treated diabetic animals relative to those ofcontrols. Taken together, these results provide little evidencefor a significant effect of the DCA treatment on fatty acidmetabolism in the diabetic animals.

Cardiac PDK4 expression was elevated (Fig. 2A) in di-abetic compared with control animals, but PDK1 and PDK2expression was unaffected by diabetes (Fig. 2B and C).Cardiac expression of PDK1, PDK2, and PDK4 was signif-icantly decreased in DCA-treated diabetic animals com-pared with untreated diabetic animals (P = 0.023, 0.009,and 0.002, respectively). PDK4 returned to a level not sig-nificantly different from that of controls (P = 0.2) (Fig. 2A).The protein expressions of the peroxisome proliferator–activated receptor-a target genes UCP3 and MCAD were

elevated (Fig. 2D and E) in the untreated and DCA-treateddiabetic animals relative to controls, indicating an in-creased rate of fatty acid oxidation in the diabetic animalsthat was unaffected by the DCA treatment.

In Vivo Cardiac Carbohydrate MetabolismThe rate of 13C label transfer from pyruvate to bicarbon-ate in vivo was used as a real-time assessment of PDH flux(Fig. 3). Control animals showed a mean PDH flux of0.017 6 0.002 s-1, compared with a significantly de-creased cardiac PDH flux in diabetic animals of 0.005 60.001 s-1 (P = 0.0002; Fig. 4A). 13C label transfer to lactatein diabetic animals was not significantly different fromthat in controls; however, 13C label transfer to alaninewas significantly increased (0.0080 6 0.0003 s-1) in sub-jects with diabetes compared with control subjects at0.0049 6 0.0005 s-1 (P = 0.0001; Fig. 4B and C). CardiacPDH flux was significantly increased by DCA treatment(0.018 6 0.002 s-1, P , 0.0001; Fig. 4A), compared withuntreated diabetic animals, to the extent that the rate offlux was not different from that in control animals (P =0.6). The rate of 13C label transfer to lactate was un-changed, but 13C label transfer to alanine was decreasedin treated diabetic animal hearts compared with untreatedsubjects with diabetes (0.0038 6 0.0004 vs. 0.0080 60.0003 s-1, P , 0.0001), to a level not significantly dif-ferent from that in controls (P = 0.06; Fig. 4B and C).

Cardiac FunctionCardiac function, both systolic and diastolic, was assessedby echocardiography to investigate the differences be-tween control and diabetic animals at both 3 and 7 weeks(3-week data not shown; see 7-week data in Fig. 5). Atboth time points, there was no difference in ejection frac-tion between groups or in E/A ventricular filling rates. Afurther measure of diastolic function, E/E9, was shown tobe significantly increased in the diabetic animals at both 3and 7 weeks, indicating diastolic dysfunction (Fig. 5C). At7 weeks, the ejection fraction and E/A ventricular fillingrates were not different in DCA-treated diabetic animalscompared with either control or untreated diabetic ani-mals (Fig. 5A and B). Treatment with DCA did, however,result in a decreased E/E9 ratio compared with diabeticanimals (14 6 1 vs. 20 6 2, respectively; P = 0.019; Fig.5C), restoring diastolic function back to that found incontrol rats (P = 0.4). Thus, cardiac PDH flux and diastolic

Figure 1—Biochemical characterization of the type 2 diabeticmodel: fasting blood glucose concentration (in mmol/L) (A), fastingplasma insulin concentration (in pmol) (B), and epididymal fat padweight normalized to body weight (C). *P # 0.05.

Table 1—Plasma lipid metabolites

Plasma metabolites(mmol/L) Control Diabetic Diabetic DCA

NEFA 0.33 6 0.01 0.32 6 0.01 0.30 6 0.03

LDL 0.44 6 0.04 0.38 6 0.05 0.29 6 0.02*

HDL 0.65 6 0.04 0.62 6 0.04 0.61 6 0.05

Triglycerides 0.98 6 0.10 0.51 6 0.06** 0.53 6 0.09**

Cholesterol 1.9 6 0.2 1.6 6 0.2 1.5 6 0.2

NEFA, nonesterified fatty acid. *P , 0.05 vs. control; **P , 0.01vs. control.

2738 PDH Activation in the Diabetic Rat Heart Diabetes Volume 64, August 2015

dysfunction (as assessed by E/E9 ratio) were significantlynegatively correlated (P = 0.02).

SummaryOverall, this suggests that, similar to the findings inhumans, this diabetic model is characterized by hypergly-cemia, reduced PDH flux, elevated PDK4 expression, anddiastolic, but not systolic, dysfunction. Treatment withDCA reversed hyperglycemia, decreased PDK4 expression,improved PDH flux, and reversed diastolic dysfunction.

This suggests that modulating substrate selection may bea potential therapeutic target for the treatment and/orprevention of diabetic cardiomyopathy.

DISCUSSION

Type 2 diabetes is a growing global health concern. Itis now well established that type 2 diabetes markedlyelevates the risk of the development of heart failure (1).Given the fact that the incidence of type 2 diabetes israpidly increasing, it is likely that the rates of diabeticcardiomyopathy and heart failure will also continue torise. As a result, further understanding of the etiologyof diabetic cardiomyopathy and the development of noveltherapeutic strategies aimed at preventing or treating di-abetic cardiomyopathy is of great clinical importance.

Using the combination of hyperpolarized [1-13C]pyruvatespectroscopy and echocardiography, we have shown thefollowing for the first time: 1) that the diastolic dys-function associated with diabetes is linked with a reduc-tion in PDH flux; and 2) that short-term treatment withthe PDK inhibitor DCA can restore PDH flux, reverse di-astolic dysfunction, and improve whole-body glycemiccontrol in a rodent model of type 2 diabetes.

Linking PDH Activity and Diastolic Function in DiabetesSubstrate selection is a fundamental step in myocardialmetabolism. In the normal resting heart, most (60–90%)of the acetyl CoA that enters the Krebs cycle comes fromthe b-oxidation of free fatty acids with 10–40% of acetyl

Figure 2—Protein expression of the three cardiac isoforms of PDK, along with UCP3 and MCAD: cardiac PDK4 expression (A), cardiacPDK1 expression (B), cardiac PDK2 expression (C), cardiac UCP3 expression (D), and cardiac MCAD expression (E). *P # 0.05. a.u.,arbitrary units.

Figure 3—An example of in vivo spectral time course taken froma control Wistar rat heart showing the injection and subsequentdecay of the hyperpolarized [1-13C]pyruvate due to recovery backto thermal equilibrium, and exchange with [1-13C]lactate, [1-13C]alanine, and [13C]bicarbonate.

diabetes.diabetesjournals.org Le Page and Associates 2739

CoA coming from the oxidation of pyruvate, which itselfis derived from either glycolysis or lactate oxidation(8,33). However, the heart displays great flexibility inthis choice of substrate, depending on the prevailingmetabolic conditions (34). In the diabetic myocardium,because of the combined effects of insulin resistance,high circulating free fatty acid levels, and PDH inhibi-tion, fatty acids are used almost exclusively to supportATP synthesis (22). This shift toward fatty acid metab-olism and reduced carbohydrate oxidation appears to bedue to a combination of several factors, including re-duced insulin-induced GLUT4-mediated glucose uptake,suppressed glycolysis, and reduced mitochondrial PDHactivity involving PDK4 (35,36). Further, the inhibitionof glucose oxidation by fatty acids at the level of thePDH complex is universally reported (37–39).

The crucial importance of this increase in fatty acidmetabolism lies in the fact that the mitochondrial redoxstate and, as a result, also the free energy of hydrolysis ofATP are negatively affected by a change in the balanceof substrate use (40). In agreement with this, a de-creased efficiency of substrate use to create ATP has been

demonstrated in diabetes, and reduced myocardial ener-getics and diastolic dysfunction have been shown in mul-tiple studies (41,42). This is in line with the concept thatan impairment in high-energy phosphate metabolism ini-tially affects the ability of the sarcoplasmic reticular Ca2+

ATPase, the most energetically demanding of all enzymesinvolved in contractile function (43), to lower cytosolicCa2+, impairing diastolic function.

This is the first study to show a relationship betweencardiac substrate selection and diastolic function in vivo.We have shown that, in the heart, the presence ofdiabetes is associated with reduced PDH flux (as a resultof elevated PDK4 levels) and, importantly, that thisreduction in PDH flux is related to impaired diastolicfunction. This establishes PDH as a potential therapeutictarget for improving diastolic function in diabetes.

PDH as a Potential Therapeutic TargetDCA inhibits PDK, which results in an increase in theproportion of active PDH (18). Although the PDK isoformsPDK1, PDK2, and PDK4 are present in the heart (19), ithas been widely shown in diabetes that the inhibition ofthe PDH complex and an inability to metabolize glucose are

Figure 4—Assessment of in vivo cardiac carbohydrate metabolismusing hyperpolarized [1-13C]pyruvate MR spectroscopy: cardiacPDH flux (A), cardiac 13C label transfer to lactate (B), and cardiac13C label transfer to alanine (C ). *P # 0.05.

Figure 5—Assessment of cardiac systolic and diastolic functionusing echocardiography: LV ejection fraction (A), E/A ratio (B), andpreload-independent E/E9 ratio (C). *P # 0.05.

2740 PDH Activation in the Diabetic Rat Heart Diabetes Volume 64, August 2015

mediated by the upregulation of PDK4 (44). In agreementwith this, we have shown that diabetes was associated withincreased PDK4 expression and with reduced PDH flux(Figs. 2A and 4A).

In addition, we have shown that treatment with DCAresulted in a restoration of PDH flux to a level seen innormal, nondiabetic animals, along with the downregula-tion of all isoforms of PDK in the heart (Figs. 2A and 4A).It is likely that the restoration of PDH flux was a result ofboth the downregulation of PDK expression and the di-rect effect of DCA on PDK activity, although PDK isoformactivity was not directly assessed in this work (19). Fur-thermore, in association with this increase in PDH flux wehave shown not only a reversal of diastolic dysfunctionbut also, crucially, that PDH flux remains related to di-astolic function after DCA treatment. This suggests thatrestoration of the balance of cardiac substrate selection indiabetes by increasing flux through PDH, and increasingglucose oxidation, may be a central mechanism behind theobserved restoration of diastolic function. This providesa therapeutic target in the form of cardiac substrate se-lection and, more specifically, the PDK/PDH interactionfor the treatment of, or potentially the prevention of,diabetic cardiomyopathy.

Several other investigators have explored the physio-logical relevance and therapeutic potential of the PDK/PDH interaction using a PDK4-deficient mouse model inaddition to pharmacological modulation with DCA. Jeounget al. (45) showed that starvation lowered blood glucoselevels more in mice lacking PDK4 than in wild-type mice.They further showed that the activity state of PDH wasgreater in the kidney, gastrocnemius muscle, diaphragm,and heart, but not in the liver of starved PDK42/2 mice,indicating that the upregulation of PDK4 in tissues otherthan the liver was clearly important during starvation forthe regulation of glucose homoeostasis. Jeoung and Harris(46) went on to show that fasting blood glucose levels werelower, glucose tolerance was slightly improved, and insulinsensitivity was slightly greater in PDK42/2 mice comparedwith wild-type mice subjected to diet-induced obesity.Work by Ussher et al. (47) demonstrated that direct stim-ulation of PDH in mice with DCA significantly decreasedinfarct size after temporary ligation of the left anteriordescending coronary artery. These results were then re-capitulated in PDK4-deficient mice, which had enhancedmyocardial PDH activity. Finally, Mori et al. (48) showedthat the deletion of PDK4 prevented diastolic dysfunctionand normalized blood glucose levels in a rodent model ofcardiac hypertrophy.

While fatty acid oxidation rates were not specificallymeasured in this study, the protein expression levels ofthe peroxisome proliferator–activated receptor-a targetgenes UCP3 and MCAD were shown to be elevated inthe diabetic animals irrespective of treatment with DCA.This would suggest an increase in fatty acid oxidationrates in diabetes that was unaffected by treatment withDCA. This finding is supported by the work of Ussher

et al. (47), who have shown that palmitate oxidation ratesare unaffected by DCA treatment in isolated perfusedmouse hearts during ischemia/reperfusion experiments.

Beneficial Effects Outside the HeartIn addition to the cardiac glucose oxidation increase seenin this study as a result of DCA treatment, we observeda more systemic effect, in the reduction in circulatingglucose levels. This is likely to be related to the directeffects of DCA on the liver (49). We have also showna significant increase in cardiac 13C label transfer to ala-nine in diabetic animals (Fig. 4C), which is reduced tonormal levels after DCA treatment, potentially demon-strating a reduction in the supply of this gluconeogenicprecursor to the liver, facilitated by the glucose-alaninecycle. It is likely that this is at least partially involved inthe restoration of normal blood glucose levels that is seenwith DCA treatment. This also suggests that targeting thePDK family in diabetes is likely to result in positive effectsin blood glucose management.

While our data support very strongly a direct effect ofDCA on the heart (i.e., metabolic changes that correlatewith the change in function), we cannot say that thewhole-body effects of DCA are not having a contributoryeffect on the changes in cardiac function that we haveobserved. In support of the hypothesis that the directcardiac effects of DCA are causing the changes in cardiacfunction is the study by Nicholl et al. (21), in which DCAwas provided to an isolated working heart preparationfrom an STZ-induced diabetic rat. In this study, they dem-onstrated that isolated diabetic hearts provided with DCArevealed a marked improvement in function (both interms of heart rate and rate pressure product). Thiswork provides evidence of a direct effect of enhancedglucose oxidation on cardiac function but cannot excludethe potential effect of changes in whole-body metabolismfrom having made a contribution to the results observedin our study.

Clinical RelevanceThe incidence of diabetes is continually increasing, with.600 million people worldwide expected to have type 2diabetes by 2030 (50). Further, subclinical diastolic dys-function has been linked to an increased risk of heartfailure (51). Therefore, a therapeutic target that bothimproves generalized glycemic control and reverses cardiacdysfunction is likely to be of great clinical importance. DCAhas previously been the subject of clinical trials as a diabetictreatment (52). However, despite the benefits in glycemiccontrol, it has not been widely used as a diabetic treatmentbecause of side effects, which include peripheral neuropa-thy (53). Although other agents, such as phenylbutyrate(54), are now available to increase PDH activity, their usehas, in general, been limited to the treatment of rare con-ditions of inborn errors in metabolism, including congen-ital PDH deficiency (55). The findings in our study supportfurther work into the development of PDK and PDH mod-ulators as targets for the treatment of type 2 diabetes.

diabetes.diabetesjournals.org Le Page and Associates 2741

Finally, with clinical hyperpolarization studies imminent,studies that can identify reversible changes in cardiac sub-strate selection, which, if targeted, could improve cardiacdysfunction in diabetes, are likely to become a reality in thenear future (56).

Study LimitationsAs with any study, the limitations of the techniques usedneed to be taken into consideration when interpreting theresults. One limitation of the current study was that theuse of hyperpolarized pyruvate only probes one side ofthe metabolic process, namely, carbohydrate metabolism, andis unable to report directly on fatty acid oxidation because ofthe low solubility and short hyperpolarized lifetimes of thephysiologically relevant long-chain fatty acids.

Another limitation was that because of the echocardiog-raphy equipment used, we were able to assess only thediastolic parameters E/A and E/E9. The use of a higher-frequency echo probe operating at .12 MHz would haveallowed a more in-depth study of the diastolic dysfunctionobserved in our diabetic animals (e.g., isovolumic relaxationtime, myocardial performance index, propagation of mitralinflow, and flow propagation velocity). However, we feel thatthe use of the load-independent parameter E/E9 offereda sensitive and valuable assessment of diastolic dysfunctionin our model and its normalization under DCA treatment.

No direct assessment of either glucose tolerance orcardiac insulin resistance was undertaken in this work, andso we cannot make any comment about the contributionthat improved cardiac insulin sensitivity may have playedin the restoration of PDH flux and the improvement indiastolic function that we observed. However, previouswork by Lloyd et al. (57) has demonstrated distinct effectsof insulin and DCA in the isolated perfused rat heart, withDCA specifically enhancing pyruvate oxidation, while insu-lin was shown to increase glucose uptake and glycolyticflux. It would, therefore, seem likely that the results ob-served in our study occur via the direct effect of DCA onPDH flux and pyruvate oxidation rather than via a second-ary effect on cardiac insulin resistance.

ConclusionDiabetes is a significant global health burden and isstrongly linked to the development of widespread cardio-vascular problems, including heart failure. We have shownhere that by specifically targeting the PDK/PDH control ofsubstrate selection in diabetes, benefits in both diastolicfunction and general glycemic control can be achieved.

Acknowledgments. The authors thank Lucia Giles, Lucy Ambrose, LattMansor, and Jack Miller (all from the University of Oxford) for their technicalassistance and Professor Mary Sugden (Queen Mary University of London) for thedonation of a primary antibody for pyruvate dehydrogenase kinase 4 (PDK4).Funding. This study was funded by the British Heart Foundation (grants FS/10/002/28078 and FS/14/17/30634) and Diabetes UK (grant 11/0004175), andequipment support was provided by GE Healthcare.Duality of Interest. L.M.L.P. was supported in the form of a partialcontribution to her DPhil studies by AstraZeneca PLC, London, U.K. D.J.T. has

previously received grant support from GE Healthcare. E.J. is an employee ofAstraZeneca PLC, London, U.K. No other potential conflicts of interest relevant tothis article were reported.Author Contributions. L.M.L.P. undertook the experiments, analyzed thedata, and wrote the manuscript. O.J.R. undertook the experiments, analyzed thedata, and reviewed and edited the manuscript. A.J.L., V.B., C.A.C., and L.C.H.undertook the experiments and reviewed and edited the manuscript. K.C. and E.J.reviewed and edited the manuscript. D.J.T. conceived the study, analyzed the data,wrote the manuscript, and reviewed and edited the manuscript. D.J.T. is theguarantor of this work and, as such, had full access to all the data in the study andtakes responsibility for the integrity of the data and the accuracy of the data analysis.

References1. Morrish NJ, Wang SL, Stevens LK, Fuller JH, Keen H. Mortality and causesof death in the WHO Multinational Study of Vascular Disease in Diabetes. Dia-betologia 2001;44(Suppl. 2):S14–S212. Aneja A, Tang WH, Bansilal S, Garcia MJ, Farkouh ME. Diabetic cardio-myopathy: insights into pathogenesis, diagnostic challenges, and therapeuticoptions. Am J Med 2008;121:748–7573. Fein FS. Diabetic cardiomyopathy. Diabetes Care 1990;13:1169–11794. Regan TJ, Lyons MM, Ahmed SS, et al. Evidence for cardiomyopathy infamilial diabetes mellitus. J Clin Invest 1977;60:884–8995. Cosyns B, Droogmans S, Weytjens C, et al. Effect of streptozotocin-induceddiabetes on left ventricular function in adult rats: an in vivo Pinhole Gated SPECTstudy. Cardiovasc Diabetol 2007;6:306. Rajan SK, Gokhale SM. Cardiovascular function in patients with insulin-dependent diabetes mellitus: a study using noninvasive methods. Ann N Y AcadSci 2002;958:425–4307. Isfort M, Stevens SC, Schaffer S, Jong CJ, Wold LE. Metabolic dysfunction indiabetic cardiomyopathy. Heart Fail Rev 2014;19:35–488. Stanley WC, Lopaschuk GD, McCormack JG. Regulation of energy substratemetabolism in the diabetic heart. Cardiovasc Res 1997;34:25–339. Reaven GM, Hollenbeck C, Jeng CY, Wu MS, Chen YD. Measurement ofplasma glucose, free fatty acid, lactate, and insulin for 24 h in patients withNIDDM. Diabetes 1988;37:1020–102410. Mazumder PK, O’Neill BT, Roberts MW, et al. Impaired cardiac efficiencyand increased fatty acid oxidation in insulin-resistant ob/ob mouse hearts. Di-abetes 2004;53:2366–237411. Chatham JC, Forder JR. Relationship between cardiac function and sub-strate oxidation in hearts of diabetic rats. Am J Physiol 1997;273:H52–H5812. Heather LC, Clarke K. Metabolism, hypoxia and the diabetic heart. J Mol CellCardiol 2011;50:598–60513. Bell DS. Diabetic cardiomyopathy. Diabetes Care 2003;26:2949–295114. Schilling JD, Mann DL. Diabetic cardiomyopathy: bench to bedside. HeartFail Clin 2012;8:619–63115. Neubauer S. The failing heart—an engine out of fuel. N Engl J Med 2007;356:1140–115116. Hall JL, Stanley WC, Lopaschuk GD, et al. Impaired pyruvate oxidation butnormal glucose uptake in diabetic pig heart during dobutamine-induced work.Am J Physiol 1996;271:H2320–H232917. Seymour AM, Chatham JC. The effects of hypertrophy and diabetes oncardiac vs. activity. J Mol Cell Cardiol 1997;29:2771–277818. Stacpoole PW, Greene YJ. Dichloroacetate. Diabetes Care 1992;15:785–79119. Bowker-Kinley MM, Davis WI, Wu P, Harris RA, Popov KM. Evidence forexistence of tissue-specific regulation of the mammalian pyruvate de-hydrogenase complex. Biochem J 1998;329:191–19620. Kato M, Li J, Chuang JL, Chuang DT. Distinct structural mechanisms forinhibition of pyruvate dehydrogenase kinase isoforms by AZD7545, dichloroacetate,and radicicol. Structure 2007;15:992–100421. Nicholl TA, Lopaschuk GD, McNeill JH. Effects of free fatty acids and di-chloroacetate on isolated working diabetic rat heart. Am J Physiol 1991;261:H1053–H1059

2742 PDH Activation in the Diabetic Rat Heart Diabetes Volume 64, August 2015

22. Wall SR, Lopaschuk GD. Glucose oxidation rates in fatty acid-perfused isolatedworking hearts from diabetic rats. Biochim Biophys Acta 1989;1006:97–10323. Ardenkjaer-Larsen JH, Fridlund B, Gram A, et al. Increase in signal-to-noiseratio of . 10,000 times in liquid-state NMR. Proc Natl Acad Sci U S A 2003;100:10158–1016324. Atherton HJ, Schroeder MA, Dodd MS, et al. Validation of the in vivo as-sessment of pyruvate dehydrogenase activity using hyperpolarised 13C MRS.NMR Biomed 2011;24:201–20825. Mansor LS, Gonzalez ER, Cole MA, et al. Cardiac metabolism in a new ratmodel of type 2 diabetes using high-fat diet with low dose streptozotocin. Car-diovasc Diabetol 2013;12:13626. Reed MJ, Meszaros K, Entes LJ, et al. A new rat model of type 2 diabetes:the fat-fed, streptozotocin-treated rat. Metabolism 2000;49:1390–139427. Tyler DJ, Schroeder MA, Cochlin LE, Clarke K, Radda GK. Application ofhyperpolarized magnetic resonance in the study of cardiac metabolism. ApplMagn Reson 2008;34:523–53128. Schroeder MA, Cochlin LE, Heather LC, Clarke K, Radda GK, Tyler DJ. In vivoassessment of pyruvate dehydrogenase flux in the heart using hyperpolarizedcarbon-13 magnetic resonance. Proc Natl Acad Sci U S A 2008;105:12051–1205629. Boehm EA, Jones BE, Radda GK, Veech RL, Clarke K. Increased uncouplingproteins and decreased efficiency in palmitate-perfused hyperthyroid rat heart.Am J Physiol Heart Circ Physiol 2001;280:H977–H98330. Naressi A, Couturier C, Devos JM, et al. Java-based graphical user interfacefor the MRUI quantitation package. MAGMA 2001;12:141–15231. Zierhut ML, Yen YF, Chen AP, et al. Kinetic modeling of hyperpolarized 13C1-pyruvate metabolism in normal rats and TRAMP mice. J Magn Reson 2010;202:85–9232. Atherton HJ, Dodd MS, Heather LC, et al. Role of pyruvate dehydrogenaseinhibition in the development of hypertrophy in the hyperthyroid rat heart:a combined magnetic resonance imaging and hyperpolarized magnetic reso-nance spectroscopy study. Circulation 2011;123:2552–256133. Stanley WC, Recchia FA, Lopaschuk GD. Myocardial substrate metabolismin the normal and failing heart. Physiol Rev 2005;85:1093–112934. Taegtmeyer H, Golfman L, Sharma S, Razeghi P, van Arsdall M. Linkinggene expression to function: metabolic flexibility in the normal and diseasedheart. Ann N Y Acad Sci 2004;1015:202–21335. Kolter T, Uphues I, Eckel J. Molecular analysis of insulin resistance in isolatedventricular cardiomyocytes of obese Zucker rats. Am J Physiol 1997;273:E59–E6736. Randle PJ, Kerbey AL, Espinal J. Mechanisms decreasing glucose oxidationin diabetes and starvation: role of lipid fuels and hormones. Diabetes Metab Rev1988;4:623–63837. Bryson JM, Cooney GJ, Wensley VR, Phuyal JL and Caterson ID. The effects ofthe inhibition of fatty acid oxidation on pyruvate dehydrogenase complex activity intissues of lean and obese mice. Int J Obes Relat Metab Disord 1996;20:738–74438. Priestman DA, Orfali KA, Sugden MC. Pyruvate inhibition of pyruvate de-hydrogenase kinase. Effects of progressive starvation and hyperthyroidism invivo, and of dibutyryl cyclic AMP and fatty acids in cultured cardiac myocytes.FEBS Lett 1996;393:174–17839. Tsutsumi E, Takenaka F. Inhibition of pyruvate kinase by free fatty acids inrat heart muscle. Biochim Biophys Acta 1969;171:355–357

40. Rider OJ, Cox P, Tyler D, Clarke K, Neubauer S. Myocardial substratemetabolism in obesity. Int J Obes (Lond) 2013;37:972–97941. Diamant M, Lamb HJ, Groeneveld Y, et al. Diastolic dysfunction is associatedwith altered myocardial metabolism in asymptomatic normotensive patients withwell-controlled type 2 diabetes mellitus. J Am Coll Cardiol 2003;42:328–33542. Lamb HJ, Beyerbacht HP, van der Laarse A, et al. Diastolic dysfunction inhypertensive heart disease is associated with altered myocardial metabolism.Circulation 1999;99:2261–226743. Smith IC, Bombardier E, Vigna C, Tupling AR. ATP consumption by sarco-plasmic reticulum Ca²⁺ pumps accounts for 40-50% of resting metabolic rate inmouse fast and slow twitch skeletal muscle. PLoS One 2013;8:e6892444. Wu P, Sato J, Zhao Y, Jaskiewicz J, Popov KM, Harris RA. Starvation anddiabetes increase the amount of pyruvate dehydrogenase kinase isoenzyme 4 inrat heart. Biochem J 1998;329:197–20145. Jeoung NH, Wu P, Joshi MA, et al. Role of pyruvate dehydrogenase kinaseisoenzyme 4 (PDHK4) in glucose homoeostasis during starvation. Biochem J2006;397:417–42546. Jeoung NH, Harris RA. Pyruvate dehydrogenase kinase-4 deficiency lowersblood glucose and improves glucose tolerance in diet-induced obese mice. Am JPhysiol Endocrinol Metab 2008;295:E46–E5447. Ussher JR, Wang W, Gandhi M, et al. Stimulation of glucose oxidationprotects against acute myocardial infarction and reperfusion injury. CardiovascRes 2012;94:359–36948. Mori J, Alrob OA, Wagg CS, Harris RA, Lopaschuk GD, Oudit GY. ANG IIcauses insulin resistance and induces cardiac metabolic switch and inefficiency:a critical role of PDK4. Am J Physiol Heart Circ Physiol 2013;304:H1103–H111349. Blackshear PJ, Holloway PA, Alberti KG. The metabolic effects of sodiumdichloroacetate in the starved rat. Biochem J 1974;142:279–28650. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes:estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:1047–105351. Zile MR, Brutsaert DL. New concepts in diastolic dysfunction and diastolicheart failure: part I: diagnosis, prognosis, and measurements of diastolic func-tion. Circulation 2002;105:1387–139352. Stacpoole PW, Moore GW, Kornhauser DM. Metabolic effects of di-chloroacetate in patients with diabetes mellitus and hyperlipoproteinemia. N EnglJ Med 1978;298:526–53053. Calcutt NA, Lopez VL, Bautista AD, et al. Peripheral neuropathy in ratsexposed to dichloroacetate. J Neuropathol Exp Neurol 2009;68:985–99354. Ferriero R, Manco G, Lamantea E, et al. Phenylbutyrate therapy for pyruvatedehydrogenase complex deficiency and lactic acidosis. Sci Transl Med 2013;5:175ra3155. Leonard JV, Morris AA. Diagnosis and early management of inborn errors ofmetabolism presenting around the time of birth. Acta Paediatr 2006;95:6–1456. Rider OJ, Tyler DJ. Clinical implications of cardiac hyperpolarized magneticresonance imaging. J Cardiovasc Magn Reson 2013;15:9357. Lloyd S, Brocks C, Chatham JC. Differential modulation of glucose, lactate,and pyruvate oxidation by insulin and dichloroacetate in the rat heart. Am JPhysiol Heart Circ Physiol 2003;285:H163–H172

diabetes.diabetesjournals.org Le Page and Associates 2743