Search

Competition for respiration between substrates in animal tissues has been known for at least 80 years. The most important interaction, quantitatively is between glucose and fatty acids. The starting point in 1963 for the so called Glucose Fatty Acid Cycle was the realisation that the metabolic relationship between glucose and fatty acids is reciprocal and not dependent. Glucose provision promotes glucose oxidation and glucose and lipid storage, and inhibits fatty acid oxidation. Provision of free fatty acids promotes fatty acid oxidation and storage, inhibits glucose oxidation and may promote glucose storage if glycogen reserves are incomplete. This review is concerned predominantly with evidence in man in vivo. In the authors opinion the evidence for inhibitory effects of fatty acids on whole body glucose utilization ad oxidation (predominantly muscles) is decisive and enzyme mechanisms mediating these effects are well established. There is also much evidence that fatty acid oxidation inhibits glucose oxidation and stimulates glucose formation in liver and again enzyme mechanism are known. A permissive role for fatty acids in the insulin secretory response of islet beta-cells has now been firmly established and can be visualised as a mechanism to protect continuing provision of respiratory substrate. Longer term exposure of islet beta-cells to fatty acids impairs the insulin secretory response to glucose and mechanisms are known. There is compelling evidence that fatty acid oxidation may impair glucose oxidation in uncontrolled Type 1 and Type 2 diabetes, but no convincing evidence that fatty acids have a role in diminished glucose storage (glycogen deposition) in Type 2 diabetes. The inhibition of glucose storage which may follow prolonged elevation of plasma FFA in man and experimental animals is associated with glycogen repletion whereas the inhibition of glucose storage in Type 2 diabetes is associated with glycogen depletion. The precise role of fatty acids in disturbed carbohydrate metabolism in Type 2 diabetes is an area where future progress is confidently predicted.

Fasting inhibits glucose-induced insulin secretion. We investigated the role of a glucose fatty acid cycle for such inhibition and its molecular basis in pancreatic islets from 48-h fasted rats. The fasting-impaired insulin response to 27 mM glucose was restored by 41% with a carnitine palmitoyltransferase I inhibitor, etomoxir. Etomoxir also restored (by 50%) impaired glucose oxidation in islets from fasted rats and increased the ratio of oxidation to glycolytic flux from 33 to 43%. Fasting decreased total pyruvate dehydrogenase (PDH) activity (active, unphosphorylated plus inactive, phosphorylated form) by 29%, as well as the percentage of active form (54 +/- 5 vs. 79 +/- 2% in fed rats, P < 0.001). Fasting increased islet PDH kinase activity as follows: PDH-bound activity by 36% and free (not PDH bound) PDH kinase by 70%. Fasting failed to affect PDH kinase content when assayed by an enzyme-linked immunoabsorbent assay with antibodies raised against 45 kDa PDH kinase alpha-chain. We conclude that fasting impairs B cell function to a major extent through the operation of a glucose fatty acid cycle and that decreased PDH activity resulting from increased specific activity of PDH kinase constitutes an important molecular mechanism.

The Glucose Fatty Acid Cycle as formulated 30 years ago and reviewed in the Minkowski lecture in 1966 described short term effects of fatty acids (minutes) to decrease uptake, glycolysis and oxidation of glucose in heart and skeletal muscles. Such short term effects have since been extended to include inhibition of glucose uptake and glycolysis and stimulation of gluconeogenesis in liver and these effects have also been convincingly demonstrated in man in vivo. More recently a longer term effect of fatty acid metabolism to decrease glucose oxidation (hours) has been shown in heart and skeletal muscle and liver. This effect increases the specific activity of pyruvate dehydrogenase kinase, which in turn results in enhanced phosphorylation and inactivation of the pyruvate dehydrogenase complex. Activity of the pyruvate dehydrogenase complex is the major determinant of glucose oxidation rate. It seems likely that longer term effects of fatty acids on this and other aspects of glucose metabolism could be important in the development of insulin resistance in diabetes mellitus in man.

Antibodies were raised in rabbits to free rat liver pyruvate dehydrogenase (PDH) kinase alpha-chain and shown to react with PDH kinase alpha-chain in rat heart and liver PDH complexes, in purified pig heart PDH complex and in bovine kidney dihydrolipoamide acetyltransferase-protein X-PDH kinase subcomplex. E.l.i.s.a for PDHE1 (pyruvate dehydrogenase) and PDH kinase have been developed and applied to assays of these proteins in extracts of rat liver and rat heart mitochondria; the measured immunoreactivities for PDHE1 (heart > liver) and for PDH kinase alpha-chain (liver > heart) paralleled known differences in PDH complex and PDH kinase activities respectively. The results of e.l.i.s.a of PDH kinase alpha-chain in extracts of rat liver mitochondria showed that the effects of starvation to increase PDH kinase activity in vivo, and the effects of dibutyryl cyclic AMP or palmitate to increase PDH kinase activity in hepatocytes cultured in vitro, are due largely (> 90%) to an increase in the specific activity of PDH kinase. The effect, in cultured hepatocytes, of dibutyryl cyclic AMP to increase PDH kinase activity was blocked by cycloheximide; the effect of palmitate was blocked by an inhibitor of carnitine palmitoyltransferase I (Etomoxir), but not by cycloheximide.

Glucose is essential for the energy metabolism of some cells and conservation of glucose is obligatory for survival during starvation. The principal site of this glucose conservation is the mitochondrial pyruvate dehydrogenase (PDH) complex, which is regulated by reversible phosphorylation (phosphorylation is inactivating). In cells in which glucose oxidation is switched off during starvation, fatty acids are used as fuel, and acetyl CoA and NADH formed by beta-oxidation promote phosphorylation of PDH complex by activation of PDH kinase. A longer-term mechanism further increases PDH kinase activity in response to cAMP and products of beta-oxidation of fatty acids. Coordinated inhibition of glycolytic flux mediated by effects of citrate on PFK1 and PFK2 in muscles and liver results in an associated inhibition of glucose uptake. Similar mechanisms lead to impaired glucose oxidation in diabetes.

Rat liver pyruvate dehydrogenase (PDH) kinase activator protein (KAP), a free PDH kinase readily separable from PDH complex and its intrinsic kinase, has been purified to apparent homogeneity from liver mitochondria of fed and 48-h starved rats. On SDS-PAGE an apparently single band of M(r) 45 kDa was obtained. N-Terminal amino acid sequence analyses (8-10 cycles) confirmed the presence of a single peptide in each case. The specific activity of the purified KAP from 48-h starved rats (14,413 U/mg protein) was 4.5-fold greater than that from fed rats.

Starvation increased pyruvate dehydrogenase (PDH) kinase activity in extracts of freshly excised rat soleus 2.2-fold (from 0.6 min-1 in fed rats to 1.31 min-1 in 48-h-starved rats). In fed rats, activities were unchanged following 24 h of culture in medium 199, but increased 2.1-fold on 24 h of culture with 50 microM dibutyryl cAMP plus 1 mM n-octanoate and 1.6-1.7-fold with either agent alone. Approx. 70% of the increase in PDH kinase induced by starvation was lost following 24 h of culture in medium 199; the loss was prevented by 50 microM dibutyryl cAMP plus 1 mM n-octanoate. cAMP concentrations in fresh soleus muscle were 1 nmol/g (fed rats) and 1.6 nmol/g (starved rats). After 20-60 min of culture the fed-starved difference disappeared and [cAMP] fell to 0.4 nmol/g. Calcitonin-gene-related peptide (CGRP) increased cAMP 3-fold; the increase was maintained throughout 24 h of culture, but was readily reversed at 30 min or 24 h of culture by 60-min incubation with CGRP-free medium. Starvation of the rat (48 h) had no effect on the sensitivity of soleus towards the [cAMP]-increasing effect of CGRP. It is concluded that culture may reverse effects of starvation on PDH kinase activity by lowering cAMP and by removal from the in vivo effects of circulating free fatty acids; and that starvation and CGRP had no detectable long-term effects on the cAMP system in soleus muscle.

It is shown here that rat liver pyruvate dehydrogenase (PDH) kinase activator protein (KAP) catalyses ATP-dependent inactivation and [32P]phosphorylation of pig heart PDHE1 and of yeast (Saccharomyces cerevisiae) PDH complex devoid of PDH kinase activity, that fluorosulphonylbenzoyladenosine inactivates rat liver KAP and the intrinsic PDH kinase of rat liver PDH complex, and that KAP, like PDH kinase, is inactivated by thiol-reactive reagents. It is concluded that KAP is a free PDH kinase.

Rat soleus muscle strips cultured for 24 h in medium 199 were well preserved in terms of electron microscopy; ATP and creatine phosphate concentrations; rates of glucose utilization, glycogen and protein synthesis, and effects of insulin thereon. Culture led to modest changes in fluid spaces and intracellular (K+); increased basal glucose utilization up to two-fold; had no effect on the maximum response to insulin; and had no effect on sensitivity to insulin except in the presence of adenosine deaminase. Thus in vitro neither denervation nor absence of insulin had any marked effects in 24 h to decrease responses to insulin.

The increased activity of pyruvate dehydrogenase (PDH) kinase induced in hearts of rats by starvation for 48 h was maintained following preparation of cardiac myocytes, and it was also maintained, though at a decreased level, after 25 h of culture in medium 199. This loss of PDH kinase activity was not prevented by n-octanoate, dibutyryl cyclic AMP or glucagon. The PDH kinase activity of myocytes from fed rats was increased to that of starved rats after 25 h of culture with n-octanoate, dibutyryl cyclic AMP or both agents together.

1. Pig heart pyruvate dehydrogenase phosphate complex in which all three sites of phosphorylation were completely phosphorylated was re-activated at a slower rate by phosphatase than complex predominantly phosphorylated in site 1. The ratio of initial rates of re-activation was approx. 1:5 with a comparatively crude preparation of phosphatase and with phosphatase purified by gel filtration and ion-exchange chromatography. 2. The ratio of apparent first-order rate constants during dephosphorylation of fully phosphorylated complex averaged 1/3.8/1.3 for site 1/site 2/site 3. Only site-1 dephosphorylation was linearly correlated with re-activation of the complex throughout dephosphorylation. Dephosphorylation of site 3 was linearly correlated with re-activation after an initial burst of dephosphorylation. 3. Because dephosphorylation of site 1 was always associated with dephosphorylation of site 2, it is concluded that dephosphorylation cannot be purely random. 4. The ratio of apparent first-order rate constants for dephosphorylation of site 1 (partially/fully phosphorylated complexes) averaged 1.72. This ratio is smaller than the ratio of approx. 5 for the initial rates of re-activation. Possible mechanisms involved in the diminished rate of re-activation of fully phosphorylated complex are discussed.

The [gamma-32P]ATP-back-titration method of estimating occupancy in vivo of the three phosphorylation sites in the pyruvate dehydrogenase complex was improved in precision by specific analysis with trypsin/formic acid, by more effective prevention of site-2 dephosphorylation during purification with NaF, and by other refinements. Disproportionation of phosphorylated complexes during purification was excluded. With this improved method it was shown that the relationship between occupancy of sites and the proportion of complex in the inactive form in rat heart in vivo is closely similar to that measured directly in heart mitochondria by incorporation of [32P]Pi. In the heart in vivo (as in mitochondria), occupancy of site 1 correlated linearly with the proportion of inactive complex. Occupancy of sites 2 and 3 only approached equivalence to that of site 1 when 99% of the complex was inactive (starved or diabetic rats). When 70% or less of the complex was inactive (resting or exercising fed normal rats), occupancy of sites 2 and 3 was minimal (3 less than 2) relative to site 1. The initial rate of re-activation by phosphatase of phosphorylated complex from hearts of resting or exercising fed normal rats was approximately three times that of complex from 48 h-starved rats.

Urine contains non-dialysable conjugates of glucose and galactose from which the free sugars are released by acid hydrolysis. In 14 non-diabetic subjects the 24-h outputs (mean +/- SEM) were 25 +/- 3 mumol/24 h (glucose) and 109 +/- 15 mumol/l (galactose). In collections from 23 diabetic patients output of conjugated glucose was increased to 177 +/- 96 mumol/24 h but conjugated galactose was unchanged (119 +/- 10 mumol/24 h). The concentration ratio of glucosyl/galactosyl allows relative 24-h outputs of conjugated glucose to be estimated on random samples. The ratio (mean +/- SEM) was 0.24 +/- 0.01 in 27 normal men and 0.37 +/- 0.04 in 30 normal women. The mean ratio was increased to 0.72 +/- 0.20 in 56 male and to 0.51 +/- 0.07 in 17 female diabetic patients. In individual diabetic patients, the ratio was increased significantly in 28 out of 56 men and in 5 out of 30 women. In male diabetic patients the ratio was increased in 17 out of 23 Type 1 (insulin-dependent) but in only 3 out of 15 Type 2 (non-insulin-dependent) diabetic patients. There was no correlation between the glucosyl/galactosyl ratio and patient age, known duration of diabetes, or urinary excretion of free glucose or protein. Urine samples showing increased glucosyl/galactosyl ratios did not yield bacteria on culture and were negative for Candida albicans cell-wall mannan antigen. It is concluded that 80% of male Type 1 diabetic patients show increased urinary excretion of non-dialysable conjugated glucose. In women, there is much greater variation in non-diabetic subjects which may obscure an increased excretion in diabetic patients.

Activity of the pyruvate dehydrogenase complex determines the rate of glucose oxidation in animals including man. The complex is regulated by reversible phosphorylation, phosphorylation resulting in inactivation. Activity is therefore dependent upon the activities of pyruvate dehydrogenase kinase and phosphatase. Activity of the complex is reduced in diabetes and starvation as a result of insulin deficiency. The mechanism involves activation of pyruvate dehydrogenase kinase by short-term effects of products of fatty acid oxidation and by longer term effects involving specific protein synthesis; in hepatocytes the signals may include lipid fuels and glucagon. Activity of the branched chain ketoacid dehydrogenase complex determines the rate of degradation of branched chain aminoacids which is adjusted according to dietary supply. The complex is regulated by reversible phosphorylation, phosphorylation being inactivating. In liver and kidney, but not in muscles a protein activator (free E1 component) may reactivate phosphorylated complex without dephosphorylation and facilitate hepatic oxidation of branched chain ketoacids. Metabolic adjustments induced by diet and diabetes include loss of activator protein, loss of total complex activity in liver but not muscles, and enhanced inactivation by phosphorylation in liver.

The activity of branched-chain 2-oxo acid dehydrogenase kinase was 3-fold greater in extracts of heart mitochondria than in extracts of liver mitochondria from rats fed on normal diet. Feeding rats on a 0%-casein diet for 10 days increased the activity of branched-chain kinase 4-fold in extracts of liver mitochondria and in branched-chain dehydrogenase complex purified from such extracts; starvation (48 h) was without effect. In extract of heart mitochondria, kinase activity was increased 2-fold by feeding on 0%-casein diet and 1.5-fold by 48 h of starvation.

1. Evidence is given for three sites of phosphorylation in the alpha-chains of the decarboxylase component of purified rat heart pyruvate dehydrogenase complex, analogous to those established for procine and bovine complexes. Inactivation of rat heart complex was correlated with phosphorylation of site 1. Relative initial rates of phosphorylation were site 1 greater than site 2 greater than site 3. 2. Methods are described for measurement of incorporation of 32Pi into the complex in rat heart mitochondria oxidizing 2-oxoglutarate + L-malate (total, sites 1, 2 and 3). Inactivation of the complex was related linearly to phosphorylation of site 1 in mitochondria of normal or diabetic rats. The relative initial rates of phosphorylation were site 1 greater than site 2 greater than site 3. Rates of site-2 and site-3 phosphorylation may have been closer to that of site 1 in mitochondria of diabetic rats than in mitochondria of normal rats. 3. The concentration of inactive (phosphorylated) complex was varied in mitochondria from normal rats by inhibiting the kinase reaction with pyruvate at concentrations ranging from 0.15 to 0.4 mM. The results showed that the concentration of inactive complex is related linearly to incorporation of 32Pi into site 1. Inhibition of 32Pi incorporations with pyruvate at all concentrations over this range was site 3 greater than site 2 greater than site 1. 4. With mitochondria from diabetic rats, pyruvate (0.15-0.4 mM) inhibited incorporation of 32Pi into site 3, but it had no effect on the concentration of inactive complex or on incorporations of 32Pi into site 1 or site 2. It is concluded that site-3 phosphorylation is not required for inactivation of the complex in rat heart mitochondria. 5. Evidence is given that phosphorylation of sites 2 and 3 may inhibit reactivation of the complex by dephosphorylation in rat heart mitochondria.

Intravenous administration of the fatty acid oxidation inhibitor 2-tetradecylglycidic acid had no effect on the proportion of pyruvate dehydrogenase complex in the active form in heart, diaphragm or gastrocnemius muscles or in liver, kidney or adipose tissue of fed normal rats. The compound reversed the effect of 48h starvation (which decreased the proportion of active complex) in heart muscle, partially reversed the effect of starvation in kidney, but had no effect in the other tissues listed. The compound failed to reverse the effect of alloxan-diabetes (which decreased the proportion of active complex) in any of these tissues. In perfused hearts of fed normal rats, 2-tetradecylglycidate reversed effects of palmitate (which decreased the proportion of active complex), but it had no effect in the absence of palmitate. In perfused hearts of 48h-starved rats the compound increased the proportion of active complex to that found in fed normal rats in the presence or absence of insulin. In perfused hearts of diabetic rats the compound normalized the proportion of active complex in the presence of insulin, but not in its absence. Palmitate reversed the effects of 2-tetradecylglycidate in perfused hearts of starved or diabetic rats. Evidence is given that 2-tetradecylglycidate only reverses effects of starvation and alloxan-diabetes on the proportion of active complex in heart muscle under conditions in which it inhibits fatty acid oxidation. It is concluded that effects of starvation and alloxan-diabetes on the proportion of active complex in heart muscle are dependent on fatty acid oxidation. Insulin had no effect on the proportion of active complex in hearts or diaphragms of fed or starved rats in vitro. In perfused hearts of alloxan-diabetic rats, insulin induced a modest increase in the proportion of active complex in the presence of albumin, but not in its absence.

The effect of ischaemia on the concentration of active pyruvate dehydrogenase complex has been investigated in glucose perfused hearts of normal rats fed a normal diet or a high fat diet or starved for 48 h; and in hearts from alloxan-diabetic rats. Global ischaemia induced by low flow (approx. 1 ml/min) lowered the concentration of active complex under most of the experimental conditions employed. Parallel studies showed that anoxia and K+ arrest of the heart had effects similar to that of ischaemia and suggested that hypoxia and decreased mechanical activity of the heart may be responsible for effects of low flow ischaemia. Evidence is reviewed that the effects of low flow ischaemia, K+ arrest and anoxia may be mediated through activation of pyruvate dehydrogenase kinase by increased reduction of mitochondrial NAD+. In hearts of normal rats on a normal diet, global ischaemia induced by zero flow and regional ischaemia induced by coronary artery ligation increased the concentration of active complex. Evidence is given that this may result from a combination of anoxia and acidosis. In aerobic perfusions at 60 mmHg, concentrations of active complex were ranked in the order: normal diet greater than high fat diet greater than 48 h starved greater than alloxan diabetic. This order was maintained when the concentration of active complex was increased by perfusion at 120 mmHg or lowered by global ischaemia induced by zero flow.

1. Sodium dichloroacetate (1mM) inhibited glucose production from L-lactate in kidney-cortex slices from fed, starved or alloxan-diabetic rates. In general gluconeogenesis from other substrates was no inhibited. 2. Sodium dichloracetate inhibited glucose production from L-lactate but no from pyruvate in perfused isolated kidneys from normal or alloxan-diabetic rats. 3. Sodium dichloroacetate is an inhibitor of the pyruvate dehydrogenase kinase reaction and it effected conversion of pyruvate dehydrogenase into its its active (dephosphorylated) form in kidney in vivo. In general, pyruvate dehydrogenase was mainly in the active form in kidneys perfused or incubated with L-lactate and the inhibitory effect of dichloroacetate on glucose production was not dependent on activation of pyruvate dehydrogenase. 4. Balance data from kidney slices showed that dichloroacetate inhibits lactate uptake, glucose and pyruvate production from lactate, but no oxidation of lactate. 5. The mechanism of this effect of dichloroactetate on glucose production from lactate has not been fully defined, but evidence suggests that it may involve a fall in tissue pyruvate concentration and inhibition of pyruvate carboxylation.

Thiophosphorylation of the pig heart pyruvate dehydrogenase complex with ATP [35S] (adenosine 5'[gamma-thio]-triphosphate) was analogous to phosphorylation with [gamma-32P]ATP except that thiophosphorylation of sites 2 and 3 was more rapid. With pyruvate dehydrogenase phosphatase the rate of dethiophosphorylation was 0.5% of that of dephosphorylation. Thiophosphorylation of sites 2 and 3 in complex phosphorylated in site 1 reduced the rate of dephosphorylation of site 1.

In heart muscle regulation of pyruvate dehydrogenase (PDH) complex activity by reversible phosphorylation is the major determinant of glucose oxidation under physiological conditions and in diabetes. Altered mitochondrial concentrations of effectors of PDH kinase and phosphatase (metabolites, Ca2+, H+) appear to explain effects of oxidation of lipid fuels, myocardial contraction and ischaemia on PDH complex activity. The effects of diabetes and starvation are mediated in addition by protein(s) which increase the activity of PDH kinase. End product inhibition by NADH may be important in ischaemia.

In mammalian tissues, two types of regulation of the pyruvate dehydrogenase complex have been described: end product inhibition by acetyl CoA and NADH: and the interconversion of an inactive phosphorylated form and an active nonphosphorylated form by an ATP requiring kinase and a specific phosphatase. This article is largely concerned with the latter type of regulation of the complex in adipose tissue by insulin (and other hormones) and in heart muscle by lipid fuels. Effectors of the two interconverting enzymes include pyruvate and ADP which inhibit the kinase, acetoin which activates the kinase and Ca2+ and Mg2+ which both activate the phosphatase and inhibit the kinase. Evidence is presented that all components of the pyruvate dehydrogenase complex including the phosphatase and kinase are located within the inner mitochondrial membrane. Direct measurements of the matrix concentration of substrates and effectors is not possible by techniques presently available. This is the key problem in the identification of the mechansims involved in the alterations in pyruvate dehydrogenase activity observed in adipose tissue and muscle. A number of indirect approaches have been used and these are reviewed. Most hopeful is the recent finding in this laboratory that in both adipose tissue and heart muscle, differences in activity of pyruvate dehydrogenase in the intact tissue persist during preparation and subsequent incubation of mitochondria.

The effect of ischemia on the concentration of active pyruvate dehydrogenase (PDH) complex has been investigated in glucose-perfused hearts of normal rats fed a normal diet or a high-fat diet or starved for 48 hr and in hearts from alloxan-diabetic rats. Global ischemia induced by low flow (approximately equal to 1 ml/min) lowered the concentration of active complex under most conditions employed. Parallel studies of the effect of anoxia and of potassium arrest of the heart indicated that the effect of low-flow ischemia may result from decreased mechanical activity of the heart as a consequence of tissue hypoxia; the enzymatic mechanism may be activation of PDH kinase by increased reduction of mitochondrial NAD. In hearts of normal rats fed a normal diet, global ischemia induced by zero flow increased the concentration of active complex. Evidence is given that this may result from a combination of anoxia and acidosis. In aerobic perfusions, concentrations of active complex were ranked in the order: normal diet greater than high-fat diet greater than 48-hr starved greater than alloxan-diabetic. This order was maintained when the concentration of active complex was lowered by global ischemia induced by zero flow.

Extracts of heart mitochondria from fed and from 48 h starved rats subjected to gel filtration on Sephacryl S-300 gave 4 major protein peaks. Pyruvate dehydrogenase complex eluted in the void volume and was assayed for intrinsic pyruvate dehydrogenase kinase activity which was increased approximately 3-fold by 48 h starvation of the rat. A second fraction, containing peaks 2 and 3 which overlapped, enhanced the activity of the intrinsic kinase and corresponds to kinase/activator protein described previously. Its activity was increased 1.5-fold by starvation.

The branched-chain 2 oxoacid dehydrogenase complex has been purified from well-washed ox-kidney mitochondria together with branched-chain dehydrogenase kinase. The complex was inactivated and phosphorylated by ATP in about 5 min at 30 degrees C. After hydrolysis of ATP by a contaminating ATPase (5-10 min) the complex was dephosphorylated and reactivated. Dephosphorylation and reactivation were linearly correlated. Reactivation was dependent upon Mg2+ (K0.5 greater than 1 mM) and inhibited completely by 50 mM fluoride. Reactivation and dephosphorylation are attributed to a mitochondrial branched-chain dehydrogenase phosphatase.

Starvation of rats for 48 h increased the activity of PDH (pyruvate dehydrogenase) kinase 2.2-fold in extracts of liver mitochondria, 2.9-fold in PDH complex partially purified therefrom by fractional precipitation, and 5-fold in PDH complex partially purified by gel filtration on Sephacryl S-300. A protein fraction was separated from PDH complex in extracts of rat liver mitochondria by gel filtration or fractional precipitation, which increased the activity of PDH kinase in rat liver and pig heart PDH complexes. The activity of this protein fraction was increased approx. 2.5-fold by 48 h starvation of rats. With highly purified pig heart PDH complex it was shown that the protein fraction increased the Vmax. of the PDH kinase reaction 35-fold (fraction from fed rats) or 82-fold (fraction from starved rats); starvation had no effect on the concentration of protein fraction required to give 0.5 Vmax. Evidence is given that the increase in PDH kinase activity effected in extracts of liver mitochondria by starvation is due to increased activity of kinase activator protein, which is tightly bound by rat liver PDH complex and not removed by a single gel filtration. With pig heart PDH complex, increased PDH kinase activity was retained after gel filtration of an admixture with kinase activator protein from starved rats, but was restored to the control value by a second gel filtration; the alterations in PDH kinase activity were associated with obvious changes in protein bands in SDS gels.

The pyruvate dehydrogenase (E1) and acetyltransferase (E2) components of pig heart and ox kidney pyruvate dehydrogenase (PDH) complex were separated and purified. The E1 component was phosphorylated (alpha-chain) and inactivated by MgATP. Phosphorylation was mainly confined to site 1. Addition of E2 accelerated phosphorylation of all three sites in E1 alpha and inactivation of E1. On the basis of histone H1 phosphorylation, E2 is presumed to contain PDH kinase, which was removed (greater than 98%) by treatment with p-hydroxymercuriphenylsulphonate. Stimulation of ATP-dependent inactivation of E1 by E2 was independent of histone H1 kinase activity of E2. The effect of E2 is attributed to conformational change(s) induced in E1 and/or E1-associated PDH kinase. PDH kinase activity associated with E1 could not be separated from it be gel filtration or DEAE-cellulose chromatography. Subunits of PDH kinase were not detected on sodium dodecyl sulphate/polyacrylamide gels of E1 or E2, presumably because of low concentration. The activity of pig heart PDH complex was increased by E2, but not by E1, indicating that E2 is rate-limiting in the holocomplex reaction. ATP-dependent inactivation of PDH complex was accelerated by E1 or by phosphorylated E1 plus associated PDH kinase, but not by E2 plus presumed PDH kinase. It is suggested that a substantial proportion of PDH kinase may accompany E1 when PDH complex is dissociated into its component enzymes. The possibility that E1 may possess intrinsic PDH kinase activity is considered unlikely, but may not have been fully excluded.