Regulation of mammalian pyruvate dehydrogenase

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.     

Synthetic peptide substrates for mammalian pyruvate dehydrogenase kinase and pyruvate dehydrogenase phosphatase

The specificities of pyruvate dehydrogenase kinase and pyruvate dehydrogenase phosphatase were probed using synthetic peptides corresponding to the sequence around phosphorylation sites 1 and 2 on pyruvate dehydrogenase [Tyr-His-Gly-His-Ser(P1)-Met-Ser-Asp-Pro-Gly-Val-Ser(P2)-Tyr-Arg]. The dephosphotetradecapeptide containing aspartic acid at position 8 was a better substrate for the kinase than was the tetradecapeptide containing asparagine at position 8. The apparent Km and V values for the two peptides were 0.43 and 6.1 mM and 2.7 and 2.4 nmol of 32P incorporated/min/mg, respectively. Methylation of the aspartic acid residue also increased the apparent Km of the tetradecapeptide about 14-fold. These results indicate that an acidic residue on the carboxyl-terminal side of phosphorylation site 1 is an important specificity determinant for the kinase. Phosphate was incorporated only into site 1 of the synthetic peptide by the kinase. The phosphatase exhibited an apparent Km of 0.28 mM and a V of 2.3 mumol of 32P released/min/mg for the phosphorylated tetradecapeptide containing aspartic acid. Methylation of the aspartic acid residue had no significant effect on dephosphorylation. The octapeptide and phosphooctapeptide produced by cleavage of the aspartyl-prolyl bond by formic acid were poorer substrates for the kinase and phosphatase than were the tetradecapeptide and phosphotetradecapeptide, respectively. Modification of the amino terminal by acetylation or lysine addition had only a slight effect on the kinase and phosphatase activities.     

Similarity of the E1 subunits of branched-chain-oxoacid dehydrogenase from Pseudomonas putida to the corresponding subunits of mammalian branched-chain-oxoacid and pyruvate dehydrogenases

The genes encoding proteins responsible for activity of the E1 component of branched-chain-oxoacid dehydrogenase of Pseudomonas putida have been subcloned and the nucleotide sequence of this region determined. Open reading frames encoding E1 alpha (bkdA1, 1233 bp) and E1 beta (bkdA2, 1020 bp) were identified with the aid of the N-terminal sequence of the purified subunits. The Mr of E1 alpha was 45,158 and of E1 beta was 37,007, both calculated without N-terminal methionine. The deduced amino acid sequences of E1 alpha and E1 beta had no similarity to the published sequences of the E1 subunits of pyruvate and 2-oxoglutarate dehydrogenases of Escherichia coli. However, there was substantial similarity between the E1 alpha subunits of Pseudomonas and rat liver branched-chain-oxoacid dehydrogenases. In particular, the region of the E1 alpha subunit of the mammalian branched-chain-oxoacid dehydrogenase which is phosphorylated, was found to be highly conserved in the Pseudomonas E1 alpha subunit. There was also considerable similarity between the E1 beta subunits of Pseudomonas branched-chain-oxoacid dehydrogenase and human pyruvate dehydrogenase.     

Short-term regulation of the mammalian pyruvate dehydrogenase complex

In this minireview the main mechanism of control of mammalian pyruvate dehydrogenase complex (PDHC) activity by phosphorylation-dephosphorylation is presented in the first place. The information recently obtained in several laboratories includes new data about isoforms of the PDH converting enzymes (kinase and phosphatase) and their action in view of short-term regulation of PDHC. Moreover, interesting influence of exogenous thiamine diphosphate (TDP) and some divalent cations, especially Mn(2+), on the kinetic parameters of PDHC saturated with endogenous tightly bound TDP, is discussed. This influence causes a shortening of the lag-phase of the catalyzed reaction and a strong decrease of the K(m) value of PDHC mainly for pyruvate. There are weighty arguments that the effects have an allosteric nature. Thus, besides reversible phosphorylation, also direct manifold increase of mammalian PDHC affinity for the substrate by cofactors seems an important aspect of its regulation.     

A model for the spatial location of pyruvate dehydrogenase phosphatase in mammalian pyruvate dehydrogenase complex

Recent experimental findings on the structural--functional features of pyruvate dehydrogenase phosphatase (PDP) isolated from various sources are compared. Two alternative mechanisms (a and b) of dephosphorylation of the E1 component in the pyruvate dehydrogenase complex (PDC) are discussed: a) the reaction occurs as a result of stochastic collisions of PDP and PDC, and the generation of an enzyme--substrate complex (PDP--E1--PDC) and dephosphorylation of the E1 component occur independently at different PDP binding sites on the PDC core; b) the dephosphorylation is performed simultaneously by a certain number of PDP molecules symmetrically bound on the PDC core. The second mechanism is suggested by the self-assembly theory of multicomponent enzyme systems and can be proved by kinetic experiments. Based on self-assembly principles and data on feasible binding sites of peripheral components of the PDC, the stoichiometry and mutual location of PDP, pyruvate dehydrogenase kinase, and the E1 component on the core of mammalian PDC are postulated to provide optimal functioning of the PDC. Structural mechanisms of stimulation of PDP activity by Ca2+ and polyamines are also discussed.     

alpha-Ketoglutarate dehydrogenase complex of Escherichia coli. A hybrid complex containing pyruvate dehydrogenase subunits from pyruvate dehydrogenase complex

The alpha-ketoglutarate dehydrogenase complex of Escherichia coli utilizes pyruvate as a poor substrate, with an activity of 0.082 units/mg of protein compared with 22 units/mg of protein for alpha-ketoglutarate. Pyruvate fully reduces the FAD in the complex and both alpha-keto[5-14C]glutarate and [2-14C]pyruvate fully [14C] acylate the lipoyl groups with approximately 10 nmol of 14C/mg of protein, corresponding to 24 lipoyl groups. NADH-dependent succinylation by [4-14C]succinyl-CoA also labels the enzyme with approximately 10 nmol of 14C/mg of protein. Therefore, pyruvate is a true substrate. However, the pyruvate and alpha-ketoglutarate activities exhibit different thiamin pyrophosphate dependencies. Moreover, 3-fluoropyruvate inhibits the pyruvate activity of the complex without affecting the alpha-ketoglutarate activity, and 2-oxo-3-fluoroglutarate inhibits the alpha-ketoglutarate activity without affecting the pyruvate activity. 3-Fluoro[1,2-14C]pyruvate labels about 10% of the E1 components (alpha-ketoacid dehydrogenases). The dihydrolipoyl transsuccinylase-dihydrolipoyl dehydrogenase subcomplex (E2E3) is activated as a pyruvate dehydrogenase complex by addition of E. coli pyruvate dehydrogenase, the E1 component of the pyruvate dehydrogenase complex. All evidence indicates that the alpha-ketoglutarate dehydrogenase complex purified from E. coli is a hybrid complex containing pyruvate dehydrogenase (approximately 10%) and alpha-ketoglutarate dehydrogenase (approximately 90%) as its E1 components.     

R-lipoic acid inhibits mammalian pyruvate dehydrogenase kinase

The four pyruvate dehydrogenase kinase (PDK) and two pyruvate dehydrogenase phosphatase (PDP) isoenzymes that are present in mammalian tissues regulate activity of the pyruvate dehydrogenase complex (PDC) by phosphorylation/dephosphorylation of its pyruvate dehydrogenase (E1) component. The effect of lipoic acids on the activity of PDKs and PDPs was investigated in purified proteins system. R-lipoic acid, S-lipoic acid and R-dihydrolipoic acid did not significantly affect activities of PDPs and at the same time inhibited PDKs to different extents (PDK1>PDK4 approximately PDK2>PDK3 for R-LA). Since lipoic acids inhibited PDKs activity both when reconstituted in PDC and in the presence of E1 alone, dissociation of PDK from the lipoyl domains of dihydrolipoamide acetyltransferase in the presence of lipoic acids is not a likely explanation for inhibition. The activity of PDK1 towards phosphorylation sites 1, 2 and 3 of E1 was decreased to the same extent in the presence of R-lipoic acid, thus excluding protection of the E1 active site by lipoic acid from phosphorylation. R-lipoic acid inhibited autophosphorylation of PDK2 indicating that it exerted its effect on PDKs directly. Inhibition of PDK1 by R-lipoic acid was not altered by ADP but was decreased in the presence of pyruvate which itself inhibits PDKs. An inhibitory effect of lipoic acid on PDKs would result in less phosphorylation of E1 and hence increased PDC activity. This finding provides a possible mechanism for a glucose (and lactate) lowering effect of R-lipoic acid in diabetic subjects.     

The oxidation of methylglyoxal by mammalian pyruvate dehydrogenase

Mammalian pyruvate dehydrogenase actively catalyzed the oxidation of methylglyoxal to acetyl-CoA. The reaction was fully enzymatic with an estimated Km of 1.89 mM. On the other hand, methylglyoxal was a competitive inhibitor of the enzyme for pyruvate, the Ki being in the 1 mM range. The reaction was inhibited in the presence of HgCl2. The reaction products were quantitatively identified as acetyl-CoA and formic acid. A mechanism for the reaction is proposed.     

Cryoelectron microscopy of mammalian pyruvate dehydrogenase complex.

Cryoelectron microscopy has been performed on frozen-hydrated pyruvate dehydrogenase complexes from bovine heart and kidney and on various subcomplexes consisting of the dihydrolipoyl transacetylase-based (E2) core and substoichiometric levels of the other two major components, pyruvate dehydrogenase (E1) and dihydrolipoyl dehydrogenase (E3). The diameter of frozen-hydrated pyruvate dehydrogenase complex (PDC) is 50 nm, which is significantly larger than previously reported values. On the basis of micrographs of the subcomplexes, it is concluded that the E1 and E3 are attached to the E2-core complex by extended (4-6 nm maximally) flexible tethers. PDC constructed in this manner would probably collapse and appear smaller than its native size when dehydrated, as was the case in previous electron microscopy studies. The tether linking E1 to the core involves the hinge sequence located between the E1-binding and catalytic domains in the primary sequence of E2, whereas the tether linking E3 is probably derived from a similar hinge-type sequence in component X. Tilting of the E2-based cores and comparison with model structures confirmed that their overall shape is that of a pentagonal dodecahedron. The approximately 6 copies of protein X present in PDC do not appear to be clustered in one or two regions of the complex and are not likely to be symmetrically distributed.     

Function of phosphorylation sites on pyruvate dehydrogenase.

Evidence is presented that dephosphorylation of the three phosphorylation sites on bovine kidney pyruvate dehydrogenase by pyruvate dehydrogenase phosphatase is random. The relative rates of dephosphorylation were in the order site 2 > site 3 > site 1. Phosphorylation site 2, and possibly site 3, function, in addition to site 1, as inactivating sites. However, the presence of phosphoryl groups at sites 2 and 3 did not significantly affect the rate of dephosphorylation at site 1 or the rate of reactivation of the enzyme by the phosphatase. The rate-limiting step in the reactivation of phosphorylated pyruvate dehydrogenase is apparently the dephosphorylation at site 1.     

Active-site modification of mammalian pyruvate dehydrogenase by pyridoxal 5'-phosphate

The pyruvate dehydrogenase multienzyme complex from bovine kidney and heart is inactivated by treatment with pyridoxal 5'-phosphate and sodium cyanide or sodium borohydride. The site of this inhibition is the pyruvate dehydrogenase (E1) component of the complex. Inactivation of E1 by the pyridoxal phosphate-cyanide treatment was prevented by thiamin pyrophosphate. Equilibrium binding studies showed that E1 contains two thiamin pyrophosphate binding sites per molecule (alpha 2 beta 2) and that modification of E1 increased the dissociation constant (Kd) for thiamin pyrophosphate about 5-fold. Incorporation of approximately 2.4 equiv of 14CN per mole of E1 tetramer in the presence of pyridoxal phosphate resulted in about a 90% loss of E1 activity. Radioactivity was incorporated predominantly into the E1 alpha subunit. Radioactive N6-pyridoxyllysine was identified in an acid hydrolysate of the E1-pyridoxal phosphate complex that had been reduced with NaB3H4. The data are interpreted to indicate that in the presence of sodium cyanide or sodium borohydride, pyridoxal phosphate reacts with a lysine residue at or near the thiamin pyrophosphate binding site of E1. This binding site is apparently located on the alpha subunit.     

Immunocapture and microplate-based activity measurement of mammalian pyruvate dehydrogenase complex

Altered pyruvate dehydrogenase (PDH) functioning occurs in primary PDH deficiencies and in diabetes, starvation, sepsis, and possibly Alzheimer's disease. Currently, the activity of the enzyme complex is difficult to measure in a rapid high-throughput format. Here we describe the use of a monoclonal antibody raised against the E2 subunit to immunocapture the intact PDH complex still active when bound to 96-well plates. Enzyme turnover was measured by following NADH production spectrophotometrically or by a fluorescence assay on mitochondrial protein preparations in the range of 0.4 to 5.0 micro g per well. Activity is sensitive to known PDH inhibitors and remains regulated by phosphorylation and dephosphorylation after immunopurification because of the presence of bound PDH kinase(s) and phosphatase(s). It is shown that the immunocapture assay can be used to detect PDH deficiency in cell extracts of cultured fibroblasts from patients, making it useful in patient screens, as well as in the high-throughput format for discovery of new modulators of PDH functioning.     

Biosynthetic studies of several of the nuclear-encoded subunits of mammalian NADH dehydrogenase

An investigation into the biogenesis of several of the nuclear-encoded subunits of the iron-protein fragment of mitochondrial NADH dehydrogenase was undertaken utilising a bovine kidney cell line (NBL-1). Inhibition of import was achieved by treating the cells with the uncoupler carbonylcyanide p-trifluoromethoxyphenylhydrazone (FCCP) and it was demonstrated that the 75-kDa, 51-kDa and 49-kDa components of the enzyme were synthesised as larger polypeptides of 76-kDa, 52-kDa and 53-kDa, respectively. The precursors could subsequently be processed to the mature subunits by reversing the FCCP treatment and chasing for 45 min at 37 degrees C. Subcellular localisation studies using the detergent digitonin illustrated that the 76-kDa, 52-kDa and 53-kDa precursor forms were almost exclusively located in the soluble fraction of the cell, whereas the mature and pulse-chased proteins fractionated with the particulate portion of the cell. Although the mature 30-kDa and 24-kDa subunits of NADH dehydrogenase could be visualised, their precursor forms went undetected in this system.     

Reaction mechanism for mammalian pyruvate dehydrogenase using natural lipoyl domain substrates

The pyruvate dehydrogenase (E1) component of the pyruvate dehydrogenase complex (PDC) catalyzes a two-step reaction. Recombinant production of substrate amounts of the lipoyl domains of the dihydrolipoyl transacetylase (E2) component of the mammalian PDC allowed kinetic characterization of the rapid physiological reaction catalyzed by E1. Using either the N-terminal (L1) or the internal (L2) lipoyl domain of E2 as a substrate, analyses of steady state kinetic data support a ping pong mechanism. Using standard E1 preparations, Michaelis constants (Km) were 52 +/- 14 microM for L1 and 24.8 +/- 3.8 microM for pyruvate and k(cat) was 26.3 s(-1). With less common, higher activity preparations of E1, the Km values were > or =160 microM for L1 and > or =35 microM for pyruvate and k(cat) was > or =70 s(-1). Similar results were found with the L2 domain. The best synthetic lipoylated-peptide (L2 residues 163-177) was a much poorer substrate (Km > or =15 mM, k(cat) approximately equals 5 s(-1); k(cat)/Km decreased >1,500-fold) than L1 or L2, but a far better substrate in the E1 reaction than free lipoamide (k(cat)/Km increased >500-fold). Each lipoate source was an effective substrate in the dihydrolipoyl dehydrogenase (E3) reaction, but E3 had a lower Km for the L2 domain than for lipoamide or the lipoylated peptides. In contrast to measurements with slow E1 model reactions that use artificial acceptors, we confirmed that the natural E1 reaction, using lipoyl domain acceptors, was completely inhibited (>99%) by phosphorylation of E1 and the phosphorylation strongly inhibited the reverse of the second step catalyzed by E1. The mechanisms by which phosphorylation interferes with E1 activity is interpreted based on accrued results and the location of phosphorylation sites mapped onto the 3-D structure of related alpha-keto acid dehydrogenases.     

Mechanisms underlying regulation of the expression and activities of the mammalian pyruvate dehydrogenase kinases

The mechanisms that control mammalian pyruvate dehydrogenase complex (PDC) activity include its phosphorylation (inactivation) by a family of pyruvate dehydrogenase kinases (PDKs 1 - 4). Here we review new developments in the regulation of the activities and expression of the PDKs, in particular PDK2 and PDK4, in relation to glucose and lipid homeostasis. This review describes recent advances relating to the acute and long-term modes of regulation of the PDKs, with particular emphasis on the regulatory roles of nuclear receptors including peroxisome proliferator-activated receptor (PPAR) alpha and Liver X receptor (LXR), PPAR gamma coactivator alpha (PGC-1alpha) and insulin, and the impact of changes in PDK activity and expression in glucose and lipid homeostasis. Since PDK4 may assist in lipid clearance when there is an imbalance between lipid delivery and oxidation, it may represent an attractive target for interventions aimed at rectifying abnormal lipid as well as glucose homeostasis in disease states.     

Inhibition of mammalian pyruvate dehydrogenase complex by metabolites of 3-mercaptopropanoic acid

3-Mercaptopropanoyl-CoA and S-acetyl-3-mercaptopropanoyl-CoA, physiological metabolites of the known convulsant 3-mercaptopropanoic acid, were found to be inhibitors of purified pyruvate dehydrogenase complex from porcine and bovine heart. Under optimal conditions, 50% inhibition was obtained at 12.6 microM 3-mercaptopropanoyl-CoA or 5 microM S-acetyl-3-mercaptopropanoyl-CoA. The inhibition caused by S-acetyl-3-mercaptopropanoyl-CoA was irreversible. Maximal inhibition of the complex was observed when it was preincubated with the inhibitor under conditions which promote reduction of the endogenous lipoate.     

Sequence conservation in the alpha and beta subunits of pyruvate dehydrogenase and its similarity to branched-chain alpha-keto acid dehydrogenase

Amino acid sequence comparison of 8 alpha and 6 beta subunits of the alpha-keto acid dehydrogenase (E1) component of the pyruvate dehydrogenase complex and branched-chain alpha-keto acid dehydrogenase complex form multiple species was performed by computer analysis. In addition to 2 previously recognized regions of homology in the alpha subunit, a 3rd region of extensive homology was identified in E1 alpha, and may be one of the sites involved in subunit interaction. E1 beta contains 4 regions of extensive homology. Region 1 contains 10 amino acids that are homologous to a 10-amino acid stretch in Escherichia coli E1. Regions 2 and 3 have sequence homologies with other dehydrogenases suggesting that these regions may be involved in catalysis.