Molecular Mechanisms Underlying the Long-term Impact of Dietary Fat to Increase Cardiac Pyruvate Dehydrogenase Kinase: Regulation by Insulin, Cyclic AMP and Pyruvate

J Mol Cell Cardiol 29, 1867–1875 (1997)

Molecular Mechanisms Underlying the Long-term Impact of Dietary Fat to Increase Cardiac Pyruvate Dehydrogenase Kinase: Regulation by Insulin, Cyclic AMP and Pyruvate Mary C. Sugden, Karen A. Orfali, Lee G. D. Fryer, Mark J. Holness and David A. Priestman Department of Biochemistry, Basic Medical Sciences, St Bartholomew’s and the Royal London School of Medicine and Dentistry, Queen Mary & Westfield College (University of London), London E1 4NS, UK (Received 28 October 1996, accepted in revised form 6 March 1997) M. C. S, K. A. O, L. G. D. F, M. J. H  D. A. P. Molecular Mechanisms Underlying the Long-term Impact of Dietary Fat to Increase Cardiac Pyruvate Dehydrogenase Kinase: Regulation by Insulin, Cyclic AMP and Pyruvate. Journal of Molecular and Cellular Cardiology (1997) 29, 1867–1875. Previous studies have demonstrated that pyruvate dehydrogenase kinase (PDHK) activity in extracts of rat cardiac mitochondria is increased ≈two-fold by providing a high-fat diet for 28 days. The present study sought to establish the factor(s) that might underlie the response of cardiac PDHK to the provision of a high-fat diet. ELISA assays of PDHKII, conducted over a range of PDHK activities, demonstrated that the increase in cardiac PDHK activity was not due to an increase in mitochondrial immunoreactive PDHKII concentration. The pyruvate concentration giving 50% active PDHC (PDHa) in mitochondria incubated with respiratory substrates was unaffected by high-fat feeding, demonstrating a dissociation between increased PDHK activity and altered sensitivity of PDHK to suppression by pyruvate. In cardiac myocytes cultured (25 h) with n-octanoate (1 m) plus dibutyryl cAMP (50 l), insulin at 12.5 lU/ml, 25 lU/ml and 75 lU/ml, suppressed PDHK activities in cells prepared from control rats, but insulin at concentrations <100 lU/ml failed to suppress PDHK activities in cardiac myocytes prepared from high-fat-fed rats. In vivo, cardiac insulin sensitivity (assessed by euglycaemic hyperinsulinaemic clamp in combination with 2-[3H] deoxyglucose administration) was suppressed after high-fat feeding. A sustained (24 h) two- to four-fold elevation in plasma insulin concentration (achieved by insulin infusion via osmotic pumps) did not affect PDHK activity in hearts of control rats. In contrast, PDHK activity in hearts of high-fat-fed rats was suppressed to values not significantly different from (insulin-infused) control rats. Basal and agonist-stimulated cAMP concentrations were unaffected by high-fat-feeding or insulin. Furthermore, rates of palmitate oxidation (to CO2) in cardiac myocytes (in the absence or presence of insulin or adrenergic agonists) were not statistically significantly affected by high-fat-feeding. The results indicate that an impaired action of insulin to suppress PDHK participates in the mechanism by which increased PDHK activity is achieved in response to high-fat feeding, but insulin does not act through decreasing cAMP concentrations or suppressing fatty acid oxidation.  1997 Academic Press Limited

K W: Pyruvate dehydrogenase kinase; Pyruvate inhibition; Cardiac myocytes; Fatty acids; Cyclic AMP; Rat heart.

Please address all correspondence to: M. C. Sugden, Department of Biochemistry, Basic Medical Sciences, Queen Mary and Westfield College, Mile End Road, London E1 4NS, UK.

0022–2828/97/071867+09 $25.00/0

mc970425

 1997 Academic Press Limited

1868

M. C. Sugden et al.

Introduction In starvation, glucose conservation is achieved by suppression of pyruvate dehydrogenase complex (PDHC) activity (reviewed by Sugden and Holness, 1994). Phosphorylation (inactivation) of PDHC is catalysed by pyruvate dehydrogenase kinase (PDHK). Increasing mitochondrial (acetyl-CoA)/ (CoA) and (NADH)/(NAD+) ratios activate PDHK, whereas increasing mitochondrial pyruvate concentrations suppress PDHK (Kerbey et al., 1976). Prolonged starvation leads to a long-term stable increase in cardiac PDHK specific activity (reviewed by Randle et al., (1994) in association with decreased sensitivity of cardiac PDHK to inhibition by pyruvate (Hutson and Randle, 1978; Priestman et al., 1996). Culture of cardiac myocytes prepared from fed rats with a fatty acid (n-octanoate) and dibutyryl cyclic AMP (Bt2cAMP) for 24 h mimics the long-term effect of starvation to increase PDHK activity (Marchington et al., 1990; Orfali et al., 1993) and to desensitise cardiac PDHK to inhibition by pyruvate (Priestman et al., 1996). The effect of addition of a fatty acid in combination with Bt2cAMP to increase cardiac PDHK in culture is opposed by the further addition of insulin at a high physiological concentration (100 lU/ml) (Orfali et al., 1995). The administration of a high-fat diet leads to a stable increase in cardiac PDHK activity (Orfali et al., 1993; Sugden et al., 1995). The characteristics of PDHK in cardiac myocytes from high-fat-fed rats in culture resemble those found when the cells are prepared from starved rats, with attenuated increases in PDHK activity in response to the addition of either fatty acid or Bt2cAMP (Orfali et al., 1993). However, the response of cardiac PDHK to high-fat feeding is much more sluggish than that to starvation (28 days compared with 24–48 h) (Orfali et al., 1993). To date, the mechanisms underlying the response of cardiac PDHK to high-fat feeding have not been thoroughly investigated and remain incompletely understood. The present study examined whether the increase in PDHK activity observed in response to high-fat feeding is, like that of starvation, due to an increase in PDHK specific activity in association with a reduced sensitivity of PDHK activity to acute suppression by pyruvate. In addition, in view of the proposed role of increased fatty acid oxidation in mediating the long-term increase in PDHK activity evoked by starvation, experiments were conducted to investigate whether high-fat feeding influences the capacity or regulation of cardiac fatty acid oxidation. Finally, the study investigated whether the long-term en-

hancement of PDHK activity evoked by high-fat feeding is associated with cardiac insulin resistance, altered cardiac cAMP concentrations and/or altered responses to agonists which increase cAMP, and whether a sustained elevation in plasma insulin concentration could overcome the effects of highfat feeding. The overall aim of these studies was to define potential mechanisms by which cardiac PDHK activity is increased by high-fat feeding.

Materials and methods Materials Kits for determination of cyclic AMP concentrations were from Amersham International, Amersham Bucks, UK; kits for determination of plasma insulin concentrations were from Phadeseph Pharmacia, Uppsala, Sweden; medium 199 was from Gibco and collagenase from Lorne Laboratories; female Wistar rats and Alzet mini pumps were purchased from Charles River Ltd, Margate, Kent, UK; other biochemicals and chemicals were from Boehringer Corp. or from Sigma Chemical Corp., Poole, Dorset, UK.

Rats and diets Female albino Wistar rats were maintained on a 12 h light/12 h dark cycle (light from 10.00 h). Rats were sampled in the absorptive state at the end of the dark phase. Rats were permitted free access to either standard rodent diet (8% fat, 72% carbohydrate and 20% protein, by calories) or a high-fat/low-carbohydrate experimental diet (47% fat, 33% carbohydrate and 20% protein, by calories) containing lard as the major source of energy (43% of total calories) (Orfali et al., 1993; Fryer et al., 1995). Corn oil (4% of total calories) was included in the high-fat diet to prevent essential fatty acid deficiency. The daily energy intake was not significantly affected by the type of diet consumed (results not shown).

Enzyme assays Three isoenzymic forms of PDHK (PDHKI, II and III) have been identified in the heart (Popov et al., 1993; Ramavedi et al., 1995), although it remains to be established whether PDHK III exists in rodents. In tissues tested thus far, the expression of PDHKII

The Impact of Dietary Fat on the Regulation of Cardiac PDHK

mRNA is higher than that of either PDHKI (rat, human) or PDHKIII (human) (Popov et al., 1993; Ramavedi et al., 1995). Antibodies were raised in New Zealand White rabbits to purified recombinant PDHKII (provided by Zeneca Pharmaceuticals). The priming dose was 10 lg of protein in Freund’s complete adjuvant given subcutaneously at four sites on the back of the rabbit. After 6 weeks, boosting injections (10 lg) in Freund’s incomplete adjuvant were given at 4-weekly intervals. Blood was removed from an ear vein at 10 days after boosts. The serum obtained was screened for antibodies by Western blotting from SDS-PAGE using pre-immune serum as control. The antiserum was specific to PDHKII with negligible cross-reaction to PDHKI as assessed by Western blots (results not shown). ELISA assays were performed using clarified extracts of rat heart mitochondria as described in Priestman et al. (1994). Active PDHC (PDHa) was assayed in isolated mitochondria as described by Priestman et al. (1994; 1996). Total PDHC was assayed as active complex after incubation of mitochondria for 10 min in the absence of respiratory substrate. PDHK activities were assayed at pH 7.0 in extracts of heart mitochondria by the rate of ATP-dependent inactivation of PDHa and computed as apparent first-order rate constants for ATP-dependent PDHa inactivation (Kerbey and Randle, 1982). To test the effects of pyruvate, heart mitochondria were incubated in KCl media in the presence of 5 m 2-oxoglutarate/0.5 m -malate, together with the concentrations of pyruvate indicated. Incubations were terminated by centrifugation after 5 min, and assayed for PDHa (Priestman et al., 1996).

Studies with cardiac myocytes Calcium-tolerant ventricular cardiac myocytes were isolated by collagenase digestion (see Orfali et al., 1993). Cardiac myocyte preparations were 85–95% viable as assessed by rod-shaped morphology and Trypan Blue exclusion. Protein content was determined by the Biuret method. Myocyte culture and assay of PDHK in myocyte extracts were undertaken as described by Orfali et al. (1993). Rates of palmitate oxidation by freshly-prepared myocytes were estimated over a 1 h period, as described in Awan and Saggerson (1993). The incubation medium contained 0.5 m albumin-bound [1-14C] palmitate (1 lCi/lmol) and the additions specified. To measure cAMP accumulation, freshly-prepared cardiac myocytes were incubated with the agonists specified at 37°C for 2 min. The reaction was then

1869

stopped by the addition of ice-cold ethanol to a final concentration of 65% (v/v). cAMP concentrations were estimated in cell extracts using a RIA kit.

Insulin action in vivo Acute insulin action in vivo was assessed using the euglycaemic-hyperinsulinaemic clamp technique in awake and freely-moving rats, as described previously (James et al., 1985; Sugden and Holness, 1995). Rats were studied in the post-absorptive state. Insulin was given at a fixed rate of 4.2 mU/ kg/min for 2.5 h. Blood glucose concentrations in the basal state and during insulin infusion were monitored using a glucose analyser (YSI, Yellow Springs, OH, USA). Cardiac glucose utilisation (transport/phosphorylation) indices were measured in the basal state and during the last hour of insulin infusion using 2-[1-3H]deoxyglucose (Sugden and Holness, 1995, see also James et al., 1985). In studies to assess the effects of longer-term exposure to insulin, insulin was administered via a subcutaneously implanted Alzet osmotic pump at a rate of 2 U per day. Rats were allowed free access to the relevant diet throughout, sampled in the fed state at 24 h after the start of insulin infusion.

Presentation of results Results are means±standard error (..) for the numbers of observations indicated. Statistical significance of differences between groups was assessed by Student’s unpaired t-test. Curve-fitting was carried out using Fig P software.

Results Cardiac PDHKII specific activity after high-fat feeding PDHK activity measured in extracts of cardiac mitochondria was increased from 0.507± 0.062 min−1 (n=6) to 1.009±0.119 min−1 (n=6) after provision of the high-fat diet for 28 days. We examined whether this activity increase was due to an increase in the concentration of PDHKII, the isoform of PDHK most highly expressed in heart (Popov et

M. C. Sugden et al.

0.9

100

A405

0.6

C/E

1.5 0.3

1.1 0.7 0 1 2 3 4 PDH complex (m-units/well)

0 0

1 2 3 PDH complex (m-units/well)

4

Figure 1 ELISA. of PDHKII in extracts of rat heart mitochondria. Heart mitochondria from control rats (open symbols) and high-fat-fed rats (closed symbols) were extracted by freezing and thawing (three times) in mitochondrial extraction buffer. The extracts were clarified by centrifugation and ELISAs performed as described by Priestman et al. (1994). Results are means±.. for three ELISAs (each four wells per assay). Inset shows ratios of fed control/high-fat-fed (C/E; means ±...) for individual points in the ELISA

al., 1993). Figure 1 shows the results for ELISA assays of PDHKII in clarified extracts of heart mitochondria from control rats (provided with standard high-carbohydrate/low-fat diet ad libitum) and highfat-fed rats. The assay was conducted over a range of PDHC activities (from 0.03–4 m-units/well). Total PDHC activity was unaffected by high-fat feeding for 28 days (see Orfali et al., 1993). The amount of immunoreactive PDHKII in the extracts from control fed rats was ≈10% greater than for the high-fat-fed rats, but there was no significant difference between the individual values (see the inset to Fig. 1). In both sets of extracts, a plateau was reached at ≈1 m-unit PDHC/well. It is concluded that the concentrations of PDHKII in extracts of heart mitochondria from control and high-fat-fed rats are similar.

Characteristics of inhibition of PDHK by pyruvate after high-fat feeding We examined whether the increase in PDHK activity evoked by high-fat feeding was accompanied by impaired suppression of PDHK activity by pyruvate. PDHa activities were measured in heart mitochondria incubated with respiratory substrate in the presence or absence of pyruvate. The effects of increasing concentrations of sodium pyruvate

Active PDH complex (% of total complex)

1870

75

50

25

0 0.01

0.1 1 Pyruvate concentration (mM)

10

Figure 2 Effect of pyruvate concentrations on steadystate PDH-complex activity in cardiac mitochondria. Mitochondria (0.5–1 mg of protein) prepared from control rats (open circles) or high-fat-fed rats (closed circles) were incubated for 5 min at 30°C in 0.5 ml of KCl medium containing 5 m 2-oxoglutarate/0.5 m -malate with the concentrations of pyruvate shown. Each point is the mean of eight observations for four mitochondrial preparations.

(20 l–10 m) on PDHa activities are shown in Figure 2. As observed previously (Cooper et al., 1974, 1975; Priestman et al., 1996) sodium pyruvate inhibited PDHK in rat heart mitochondria with activation of PDHC. The sodium pyruvate concentration giving 50% active complex (EC50) in mitochondria from fed control rats was ≈0.3 m. The EC50 for PDHC activation was unaffected by high-fat feeding. The maximum activation attainable by inclusion of pyruvate was slightly increased (by ≈8%) by high-fat feeding in vivo.

The response of cardiac PDHK to insulin in cultured cardiac myocytes Earlier studies have demonstrated that insulin at a high concentration (100 lU/ml) can suppress the effects of addition of n-octanoate (1 m) in combination with Bt2cAMP (50 l) to increase PDHK in cardiac myocytes from control and high-fat-fed rats (Fryer et al., 1995; Orfali et al., 1995). In the present study, plasma insulin concentrations in control and high-fat-fed rats in the fed state were 32±7 lU/ml and 35±3 lU/ml, respectively. We therefore aimed to determine whether suppression of PDHK activity by insulin could be observed at

The Impact of Dietary Fat on the Regulation of Cardiac PDHK 140

PDH kinase activity (% of maximum)

120

100

80

60 * * *

40

*

20

0

20 40 60 80 Insulin concentration ( U/ml)

100

Figure 3 Effect of insulin in culture on PDHK activities in cardiac myocytes from control and high-fat-fed rats. Cardiac myocytes prepared from control rats (open symbols) and high-fat fed rats (closed symbols) were cultured (24 h) with n-octanoate (1 m)+BT2cAMP (50 l) in the absence or presence of insulin at the concentrations indicated. Full details of procedures for measurement of PDHK activities in extracts of cardiac myocytes are given in Orfali et al. (1993). Results are means±.. for four myocyte preparations, with cultures performed in duplicate. Results are expressed as percentages of PDHK activities observed after culture with n-octanoate (1 m)+BT2cAMP (50 l) in the absence of insulin, namely 0.21±0.03 min−1 for myocytes prepared from control rats (closed symbols) and 0.18±0.02 min−1 for myocytes prepared from high-fat-fed rats (open symbols).

the physiological insulin concentrations observed in vivo. Cardiac myocytes prepared from control and high-fat-fed rats were cultured for 25 h with n-octanoate (1 m) plus B†2cAMP (50 l) in the presence of insulin at 12.5 lU/ml, 25 lU/ml, and 75 lU/ml (Fig. 3). In cardiac myocytes prepared from control rats, insulin at 12.5 lU/ml, 25 lU/ ml and 75 lU/ml significantly suppressed PDHK activities. In contrast, insulin at these concentrations failed to suppress PDHK activities significantly in cultured cardiac myocytes prepared from high-fat-fed rats.

Cardiac insulin sensitivity in vivo Two series of experiments were undertaken to test whether cardiac insulin resistance was evident in

1871

the intact high-fat-fed rat and might participate in the mechanism by which high-fat feeding enhances cardiac PDHK activity. First, we assessed the response of cardiac glucose utilisation (transport and phosphorylation) to insulin at a concentration in the high physiological range in vivo using the euglycaemic hyperinsulinaemic clamp technique in combination with the administration of 2-[13 H]deoxyglucose. Secondly, in view of the impaired response of PDHK to insulin in cultured cardiac myocytes prepared from high-fat-fed rats at insulin concentrations (<75 lU/ml) in the range found in the fed state in vivo, we examined whether a sustained elevation of insulin in vivo could reverse the effect of high-fat feeding to increase cardiac PDHK activity. Cardiac glucose utilisation indices (which estimate rates of glucose transport plus phosphorylation in vivo) were significantly lower in the high-fat-fed group in the basal (post-absorptive) state (≈25% of control) (Fig. 4). In control postabsorptive rats, short-term (2.5 h) insulin infusion at a constant rate of 4.2 mU/min/kg body weight elicited a marked (5.9-fold) increase in plasma insulin concentrations to values in the high physiological range (c. 80 lU/ml). In high-fat-fed rats, short-term insulin infusion also elicited a marked (5.5-fold increase) in plasma insulin concentrations, such that circulating insulin concentrations in control and high-fat-fed rats after short-term insulin infusion were not statistically different. In both groups of rats, steady-state glucose concentrations during short-term insulin infusion were maintained at ≈4 m by glucose infusion (see legend to Fig. 4). In the control rats, the acute elevation of circulating insulin concentrations resulted in a significant (56%; P <0.005) increase in cardiac glucose utilisation indices. Cardiac glucose utilisation indices were also significantly increased by short-term insulin infusion in high-fat-fed rats (4.6-fold; P <0.01), but cardiac glucose utilisation indices during insulin infusion were significantly lower in the high-fat-fed rats than in the controls (by 37%; P<0.01) (Fig. 4). We have shown previously (Fryer et al., 1995) that switching from high-fat diet to standard diet for 24 h reverses the effect of 28 days of high-fat feeding to increase cardiac PDHK activity. Further experiments therefore examined the effects of a sustained (24 h) elevation in plasma insulin on cardiac PDHK activities in vivo. Rats previously maintained on high-fat diet for 28 days were infused with insulin for 24 h via a surgically-implanted osmotic pump. Control rats (maintained on standard diet) were also infused with insulin for 24 h.

1872

M. C. Sugden et al.

Glucose transport/phosphorylation index (ng/min/mg)

120

Table 1 Effects of adrenergic agonists and insulin on cAMP accumulation by cardiac myocytes from control or high-fat-fed rats

90

Addition

cAMP accumulation (% of control) Control High-fat

* 60

30 †

0

Control

High-fat

Figure 4 Cardiac glucose transport/phosphorylation indices for control and high-fat-fed rats in the post-absorptive state and during euglycaemic-hyperinsulinaemic clamps. Cardiac glucose utilisation (transport/phosphorylation) indices were measured in awake, unstressed, unrestrained rats as described by Sugden and Holness (1995). Full details are given in the Materials and Methods section. Results are means±.. for six observations. Values obtained in the basal (post-absorptive) state are shown as filled bars and values obtained during hyperinsulinaemia are shown as open bars. Blood glucose concentrations in the basal state were 4.2±0.1 m (n= 6) and 3.9±0.2 m (n=6), respectively, for control and high-fat-fed rats. Plasma insulin concentrations in the basal (post-absorptive) state were 14±2 lU/ml (n=6) and 15±2 lU/ml (n=6), respectively, for control and high-fat-fed rats. Steady-state blood glucose concentrations during hyperinsulinaemia were 4.0±0.3 m (n=6) and 4.2±0.2 m (n=6), respectively, for control and high-fat-fed rats. Steady-state plasma insulin concentrations during hyperinsulinaemia were 82±5 lU/ ml (n=6) and 83±4 lU/ml (n=6), respectively, for control and high-fat-fed rats. Statistically significant effects of high-fat feeding are indicated by: ∗P<0.05; †P<0.01.

Although insulin infusion rates were identical in the two groups, plasma insulin concentrations at the end of the long-term insulin infusions were significantly higher in the high-fat-fed rats (concentrations of 62±8 lU/ml (n=7) and 121±4 lU/ ml (n=4) in control and high-fat-fed rats respectively, P<0.001), suggesting that insulin clearance may be impaired by high-fat feeding. Insulin infusion via osmotic pump did not influence cardiac PDHK activities in control rats [PDHK activities of (n=6) and 0.510± 0.507±0.062 min−1 0.174 min−1 (n=6), respectively, in control rats and insulin-infused control rats]. In contrast, PDHK activities measured in extracts of cardiac mitochondria from insulin-infused high-fat-fed rats [0.589±0.140 min−1 (n=6)] were significantly

None 100±10 (14) Noradrenaline 277±26 (13)† (5 l) Adrenaline (5 l) 269±43 (6)∗ Insulin (1 mU/ml) 103±6 (6) Noradrenaline+ 270±57 (4)∗ insulin

100±14 (10) 286±13 (10)† 313±41 (7)† 98±9 (4) 294±28 (4)†

Values are mean±.. for the number of separate preparations indicated in parentheses. cAMP concentrations were determined in extracts of myocytes prepared from control or high-fat-fed rats. Full details of the procedures used are given in the Materials and Methods section. Statistically significant effects of agonists are indicated: ∗P<0.01; †P<0.001. There were no statistically significant effects of high-fat feeding.

lower than those found in mitochondria from highfat-fed rats not infused with insulin [1.009±0.119 min−1 (n=6)] and only 15% (..) higher than those of control rats infused with insulin.

Effects of high-fat feeding on cardiac cAMP concentrations An attenuated response of PDHK to Bt2cAMP in cultured cardiac myocytes from high-fat-fed rats (Orfali et al., 1993) raised the possibility that an increase in the intracellular cAMP concentration (unopposed by insulin) might be involved in the stable increase in cardiac PDHK activity evoked by high-fat feeding. We therefore examined whether the effect of high-fat feeding to increase cardiac PDHK in vivo might reflect an increase in steadystate cAMP concentrations or an enhanced response of the heart to agents known to increase cardiac cAMP concentrations. cAMP concentrations measured in freshly-prepared cardiac myocytes from control fed rats and high-fat-fed rats were not significantly different [2.83±0.20 (n= 35) pmol/mg protein and 2.55±0.22 (n=30) pmol/mg protein, respectively]. Furthermore, there was no effect of high-fat feeding on the response of cardiac cAMP concentrations to noradrenaline and adrenaline (Table 1). Pre-incubation of cardiac myocytes for 5 min with insulin (100 lU/ml) together with inclusion of insulin during the 2 min incubation with noradrenaline was without significant effect on either basal or agonist-stimulated cAMP accumulation (Table 1).

The Impact of Dietary Fat on the Regulation of Cardiac PDHK Table 2 Effects of glucose and insulin on the oxidation of (1–14C)palmitate by cardiac myocytes from control and high-fat-fed rats Addition

Glucose (5 m) Insulin (1 mU/ml) Noradrenaline (5 l) Glucose+insulin Glucose+noradrenaline Glucose+insulin+ noradrenaline

1873

insulin or noradrenaline in the absence or presence of glucose did not influence rates of palmitate oxidation by cardiac myocytes from high-fat-fed rats.

Palmitate oxidation (% of no additions) Control

High-fat

62±4 (7) 100±6 (6) 105±13 (4) 58±2 (7) 62±6 (4) 58±3 (4)

70± 4 (5) 99±3 (5) 112±8 (4) 60±5 (5) 59±6 (4) 59+3 (4)

Myocytes were incubated as described in the Materials and Methods section for 1 h with 0.5 m (1−14C)palmitate and fatty-acid poor BSA (40 mg/ml). Values are means±.. for the number of separate myocyte preparations indicated in parenthesis. Rates of (1-14C)palmitate oxidation to 14CO2 were 0.94±0.11 (n=7) lmol 14CO2/h per mg protein for myocytes from control rats and 0.79±0.17 (n=5) lmol 14CO2/h per mg protein for myocytes from high-fat-fed rats (P>0.05). Rates of 14 CO2/h production are expressed relative to those in the absence of additions.

Effects of insulin on palmitate oxidation by freshlyprepared cardiac myocytes Studies with cultured cells have demonstrated that fatty acids in vitro, like starvation in vivo, increase the specific activity but not the concentration of PDHK (Priestman et al., 1994; Randle et al., 1994). The effect of fatty acids is dependent on their mitochondrial b-oxidation (Priestman et al., 1994). The effects of glucose, insulin and noradrenaline on palmitate oxidation to CO2 by freshly-prepared cardiac myocytes from control and high-fat-fed rats are shown in Table 2. In agreement with previous results (Awan and Saggerson, 1993), glucose (5 m) significantly decreased oxidation of palmitate to CO2 (by approx. 38%) in cells prepared from control fed rats. The effect of glucose was not significantly influenced by the further addition of either insulin or noradrenaline, and no significant effect of these hormones was seen in the absence of glucose. Since 28 days of high-fat feeding suppresses the response of PDHK to fatty acids in cultured cardiac myocytes (Orfali et al., 1993), we examined whether prolonged high-fat feeding altered the characteristics of palmitate oxidation by isolated cells (Table 2). Rates of palmitate oxidation to CO2 were not significantly influenced by high-fat feeding (see legend to Table 2). Glucose (5 m) addition to cells prepared from high-fat-fed rats suppressed palmitate oxidation by approximately 30%. The inclusion of

Discussion The current study has revealed several important new findings regarding the mechanisms underlying the long-term regulation of PDHK in the heart. First, the long-term administration of a high-fat diet increases PDHK activity without any increase in the concentration of PDHKII, the isoform of PDHK whose expression is highest in heart (Popov et al., 1993; Ramavedi et al., 1995). Secondly, in the absence of any change in cardiac cAMP concentrations or any major enhancement in the response of the heart to physiological agonists (noradrenaline and adrenaline) that might be expected to participate in cAMP-mediated events in cardiac metabolism, the study indicates that the increase in PDHK activity evoked by prolonged high-fat feeding (unlike that of starvation) occurs through a cAMP-independent mechanism. Our data also demonstrate that the increase in PDHK activity elicited by high-fat feeding is not linked with decreased sensitivity of PDHK to inhibition by pyruvate. The lack of any major effect of highfat feeding of the sensitivity of cardiac PDHK to inhibition by pyruvate contrasts with the effect of prolonged (48 h) starvation to decrease the EC50 for sodium pyruvate and to decrease the maximum activation attainable by inclusion of pyruvate at high concentrations (Hutson and Randle, 1978; Priestman et al., 1996), and is consistent with with the concept that different signals or mechanisms may be involved in the cardiac PDHK activity increases evoked by starvation and high-fat feeding. Finally, a range of approaches has provided good evidence that the increase in cardiac PDHK activity evoked by high-fat feeding is likely to be a consequence of cardiac insulin resistance. Thus, the effect of high-fat feeding to increase cardiac PDHK is associated both with decreased suppression of PDHK activity by insulin at submaximal concentrations (<75 lU/ml) in cultured cardiac myocytes, and with decreased stimulation of cardiac glucose uptake and phosphorylation by insulin in vivo, whereas a sustained elevation in the plasma insulin concentration (through insulin infusion) counteracts the effect of high-fat feeding to increase cardiac PDHK in vivo. The study also eliminates potential mechanisms by which insulin may exert its effects. Insulin failed to influence cAMP accumulation after challenge

1874

M. C. Sugden et al.

of cardiac myocytes with b-adrenergic agonists. Similarly, insulin (either in the absence or the presence of an elevated cardiac cAMP concentration) was without significant influence on rates of palmitate oxidation to CO2. These results strongly suggest that the effect of insulin to suppress the increase in cardiac PDHK activity elicited by fat feeding in vivo is not due to a decline in intracellular cAMP concentrations or to any major alteration in fatty acid disposal at the level of acute modulation of fatty acid partitioning between oxidation and esterification. This conclusion is supported by the observation that insulin can suppress the effects of the addition of n-octanoate+Bt2cAMP to increase PDHK in cultured myocytes. Awan and Saggerson (1993) observed in earlier studies that insulin had a variable and inconsistent effect to suppress palmitate oxidation. Their observation, taken in conjunction with the present studies, suggest that acute regulation of long-chain fatty acid oxidation by insulin is not a major feature of cardiac metabolism. However, in confirmation of the results obtained by Awan and Saggerson (1993), the presence of glucose (5 m) suppressed fatty acid oxidation in freshly-prepared cardiac myocytes. This effect of glucose has been attributed to an increase in cardiac malonyl-CoA concentrations and inhibition of carnitine palmitoyltransferase I (Anwan and Saggerson, 1993). Previous studies from our group have shown that in our hands glucose uptake by cardiac myocytes from rats maintained on standard diet is increased approximately two-fold by insulin (Orfali et al., 1995). The failure of insulin to augment the effect of glucose to suppress palmitate oxidation in control rats in the present experiments thus suggests that malonyl-CoA concentrations are already sufficiently high to suppress fatty acid oxidation completely. Contrary to the findings of Awan and Saggerson (1993) we did not observe any effects of noradrenaline to increase palmitate oxidation to CO2, nor did noradrenaline oppose the effect of glucose to suppress palmitate oxidation. Nevertheless, we observed clear increases in intracellular cAMP concentrations in response to noradrenaline. The basis for the difference between the responses of palmitate oxidation to noradrenaline observed in the current and the previous study may reflect differences in the sexes and/or strains of the rats used. However, our results suggest that it is unlikely that cAMP-dependent phosphorylation has a consistent role in the acute regulation intracellular fatty acid partitioning in the adult female rat heart. Within the specific context of the long-term regulation of cardiac PDHK activity, our results clearly indicate that the effects

of cAMP to increase PDHK in cultured cardiac myocytes are not achieved through changes in rates of fatty acid oxidation. In summary, high-fat feeding increases cardiac PDHK activity without any change in PDHKII protein concentration, sensitivity of PDHK to pyruvate inhibition or cardiac cyclic AMP concentration. The mechanism by which the effect of high-fat feeding is achieved may be related to, or co-ordinate with, impaired sensitivity of cardiac glucose utilisation to stimulation by insulin at the insulin concentrations pertaining in vivo. The insulin resistance at the levels of cardiac glucose transport/phosphorylation and PDHK activity that characterises the response to high-fat feeding may be part of a strategy to promote glucose conservation under conditions where the supply of dietary carbohydrate is reduced but dietary lipid is provided. This insulin resistance may not necessarily impact adversely on heart function under conditions of increased workload, which stimulates cardiac glucose uptake independently of insulin. Since the enhanced PDHK activity observed after high-fat feeding in vivo is not associated with diminished sensitivity of the PDHK reaction to pyruvate inhibition, conditions leading to increased glucose uptake and pyruvate generation, such as increased workload, may be able to counteract the longer-term increase in PDHK activity evoked by high-fat feeding. Thus, increased pyruvate production secondary to increased heart work would be predicted to be accompanied by PDHC activation, increased pyruvate oxidation and increased ATP production.

Acknowledgements These studies were supported by project grants from the British Heart Foundation, for which we are grateful.

References A MM, S ED, 1993. Malonyl-CoA metabolism in cardiac myocytes and its relevance to the control of fatty acid oxidation. Biochem J 295: 61–66. C RH, R PJ, D RM, 1974. Regulation of heart muscle pyruvate dehydrogenase kinase. Biochem J 143: 625–641. C RH, R PJ, D RM, 1975. Stimulation of phosphorylation and inactivation of pyruvate dehydrogenase by physiological inhibitors of the pyruvate dehydrogenase reaction. Nature 257: 808–809. F LGD, O KA, H MJ, S ED, S MC, 1995. The long-term regulation of skeletal

The Impact of Dietary Fat on the Regulation of Cardiac PDHK muscle pyruvate dehydrogenase kinase by dietary lipid is dependent on fatty acid composition. Eur J Biochem 229: 741–748. H NJ, R PJ, 1978. Enhanced activity of pyruvate dehydrogenase kinase in rat heart mitochondria in alloxan-diabetes or starvation. FEBS Lett 92: 73–76. J DE, J AB, K EW, 1985. Heterogeneity of insulin action in individual muscle in vivo: euglycemic studies in rats. Am J Physiol 248: E567–E574. K AL, R PJ, C RH, W S, P HT, D RM, 1976. Regulation of pyruvate dehydrogenase in rat heart. Mechanism of regulation of proportions of dephosphorylated and phosphorylated enzyme by oxidation of fatty acids and ketone bodies and of effects of diabetes: role of coenzyme A, acetylcoenzyme A and reduced and oxidized nicotinamideadenine dinucleotide. Biochem J 154: 327–348. K AL, R PJ, 1982. Pyruvate dehydrogenase kinase/activator in rat heart mitochondria. Assay, effect of starvation, and effect of protein-synthesis inhibitors in starvation. Biochem J 206: 103–111. M DR, K AL, R PJ, 1990. Longerterm regulation of pyruvate dehydrogenase kinase in cultured rat cardiac myocytes. Biochem J 267: 245– 247. O KA, F LG, H MJ, S MC, 1993. Long-term regulation of pyruvate dehydrogenase kinase by high-fat feeding. Experiments in vivo and in cultured cardiomyocytes. FEBS Lett 336: 501–505. O KA, F LGD, H MJ, S MC, 1995. Interactive effects of insulin and triiodothyronine on

1875

pyruvate dehydrogenase kinase activity in cardiac myocytes. J Mol Cell Cardiol 27: 901–908. P KM, K NY, Z Y, S Y, C DW, H RA, 1993. Primary structure of pyruvate dehydrogenase kinase establishes a new family of eukaryotic protein kinases. J Biol Chem 268: 26602– 26606. P DA, M SC, H A, R PJ, 1994. Role of protein synthesis and of fatty acid metabolism in the longer-term regulation of pyruvate dehydrogenase kinase. Biochem J 300: 659–664. P DA, O KA, S MC, 1996. Effects of progressive starvation and hyperthyroidism in vivo, and of dibutyryl cyclic AMP and fatty acids in cultured cardiac myocytes. FEBS Lett 393: 174–178. R G, B-K MM, K NY, Z Y, P KM, 1995. Diversity of the pyruvate dehydrogenase kinase gene family in humans. J Biol Chem 270: 28989–28994. R PJ, P DA, M S, H A, 1994. Mechanisms modifying glucose oxidation in diabetes mellitus. Diabetologia 37: S155–S161. S MC, O KA, H MJ, 1995. The pyruvate dehydrogenase complex: nutrient control and the pathogenesis of insulin resistance. J Nutr 125: 1746S– 1752S. S MC, H MJ, 1994. Interactive regulation of the pyruvate dehydrogenase complex and the carnitine palmitoyltransferase system. FASEB J 8: 54–61. S MC, H MJ, 1995. Modulation of in vivo insulin action by dietary protein during pregnancy. Am J Physiol 268: E722–E729.