The-Keto Acid

The a-Keto Acid

Dehydrogenase Complexes

Prepared by

Franklin R. Leach

Department of Biochemistry and Molecular Biology

Oklahoma State University

Stillwater, OK

9/15/02

The a-Keto Acid Dehydrogenase Complexes

A Brief Review 2002

Contents

Introduction

Background

Complex Structure

Regulation

Whole Genomes

Components Parts

E1

E2

X

E3

Relation of Lipoic Acid to Oxidative Damage

Relation to Medicine

Therapeutic potential

Ischemic heart disease

Lactic acidosis and related diseases

Maple syrup urine disease and branched chain-related diseases

Leigh's necrotizing encephalomyelopathy

Alzheimer's disease

Primary biliary cirrhosis

Systemic sclerosis

Lipoic Acid Activation and the Lipoamidase Reaction

Recent Results on the Regulatory Enzymes

Web Connections

Literature Cited

Appendix

Symposium Honoring Lester Reed - A Tribute

Recollection - From lipoic acid to multienzyme complexes

Introduction

Remarkable progress in understanding the function and mechanism of action of the a-keto acid dehydrogenases has been made in the last 50 years. These complexes are the classical example of a multienzyme complexes. A conference on "a-Keto Acid Dehydrogenase Complexes: Organization, Regulation, and Biomedical Aspects" (1), held November 16-18, 1988 in Austin, Texas to honor Professor Lester J. Reed, summarized much of the early progress. This conference celebrated Dr. Reed's 65th birthday. My interest in this topic is because I did my dissertation research with Lester Reed (1953-57) and worked on the lipoic acid activating enzyme system. A Tribute to Lester Reed written by Tom Roche, Head of the Department of Biochemistry at Kansas State University and a former Reed postdoc, is in the appendix. It shows the extent to which Reed's laboratory has contributed to our understanding of the a-keto acid dehydrogenases. The complexes have been isolated, their composition and organization determined, their base sequences are being elucidated, and their amino acid sequences and crystallographic patterns are being deduced. The mechanisms of regulation of the activities of these complexes have been established. A FASEB symposium reviews the topic (2). Several other reviews have appeared (3-6). At the 1994 ASBMB meeting in Washington, DC, Reed was given the ASBMB-Merck Award and presented a review of "A Trail of Research: From Lipoic Acid to Multienzyme Complexes". An American Institute of Nutrition Symposium "a-Keto Acid Dehydrogenase Complexes: Nutrient Control, Gene Regulation, and Genetic Defects" was also held in 1994. A review paper from that symposium has appeared (7). Reed has recalled "From lipoic acid to multi-enzyme complexes" for Protein Science (7a). (See appendix). For the Journal of Biological Chemistry Centennial collection Reed (7b) traced "a trail of research from lipoic acid to a-keto acid dehydrogenase complexes".

The molecular understanding of the a-keto acid dehydrogenases began in the 1950s with the isolation and determination of the structure of lipoic acid (8-10). The next key finding was the enzymatic mechanism by which lipoic acid was converted to the enzyme-bound functional form (11, 12). From that point until the current application of molecular biology techniques the emphasis of study has been on the isolation, characterization, determination of structure/function relationships, and regulation of the a-keto acid dehydrogenase complexes (13-15). This has matured into recent and current determinations of the amino acid and base sequences for many components of these complexes (see refs. 1-7).

The reaction (sum) catalyzed by the a-keto acid dehydrogenases is:

TPP, Lipoic acid, FAD

RCOCO2H + NAD+ +CoASH &emdash;&emdash;&emdash;&emdash;&emdash;&emdash;&emdash;> RCO-S-CoA + CO2 + NADH +H+

The component steps in this overall reaction are:

CH3COCO2- + E1TPP + H+ <==> CO2 + E1 CH3C(OH)=TPP (1)

E1 CH3C(OH)=TPP + E2_LipS2 <==> E1TPP + E2-Lip(SH)-S-COCH3 (2)

E2-Lip(SH)-S-COCH3 + CoASH <===> E2_Lip(SH2) + CH3COSCoA (3)

E2-Lip(SH2) + E3FAD <===> E2-LipS2 + Dihydro-E3FAD (4)

Dihydro-E3FAD + NAD+ <===> E3FAD + NADH + H+ (5)

Lewisite (CHCl=CHAsCl2) is a poison gas that was synthesized too late to be used in World War I. Rudolph Peters headed the Oxford University laboratory that searched for antidotes to chemical warfare agents. They developed BAL (British anilewisite, 2,3 dimercaptopropanol, and for security reasons, OX 217) in 1940. The target of the arsenical was lipoic acid. See (16).

Lipoic acid, 1,2-dithiolane-3-pentanoic acid (6,8-dimercapto-octanic acid), functions in transacylation, redox, and transport reactions. It plays a central role in oxidative metabolism: the oxidative decarboxylation of pyruvate, branched chain amino acid metabolism, glycine decarboxylation, and in the citric acid cycle. Lipoic acid is formed from octanoic acid via an enzymatic S-insertion. Additional details have been learned about lipoic acid synthesis. The question of why mitochondria synthesize fatty acids has been answered: the synthesis of lipoic acid. Wada, Shintani, and Ohlrogge (17) established that pea mitochondria can acyl carrier protein and the enzymes to synthesize fatty acids. Radioactivity from labeled malonic acid was found in the H protein, a lipoyl-containing enzyme involved in glycine metabolism. In Saccharomyces cerevisiae Brody, Oh, Hoja, and Schweizer (18) found that absence of the yeast gene ACP1, resulted in a decreased lipoic acid content. They conclude that the mitochondrial ACP is invovled in the synthesis of octanoate which is a lipoic acid precursor. Jordan and Cronan (19) found that the acyl carrier protein of lipid synthesis could donate lipoic acids to the pyruvate dehydrogenase complex in both E. coli and mitochondria.

Self, Tsai, and Stadtman (20) have prepared the selenotrisulfide derivatives of lipoic acid and lipoamide. The selenotrisulfide derivative of lipoic acid was an effective substrate for thioredoxin reductase. The lipoamide derivative was reduced by dihydrolipoamide dehydrogenase. The selenium analogs of lipoic acid had been used earlier by Reed, Morris, and Cronan (21) to isolate E. coli mutants. Replacement of either the C-6 or C-8 sulfur atom with Se gave lipoic acid derivatives with unaltered biological properties. The replacement of both S's with Se producing selenolipoic acid that was a growth inhibitor of E. coli. When radioactive 75Se was used, the selenolipoic acid was found incorporated in the a-ketoacid dehydrogenases. Resistant mutants were isolated. These mutations were traced to lipoate-protein ligase and to an unknown function in the synthesis of lipoic acid.

The E. coli LipA is a lipoyl synthase that forms lipoyl groups from octanoyl-ACP (22). This enzyme as well as biotin synthase contains (2Fe-2S) centers that can combine to form a (4Fe-4S) center. The iron-sulfur center is involved in the formation of a C-S bond (23). The enzymology of sulfur activation during thiamine and biotin biosynthesis has been discussed by Begley, Xi, Kinsland, Taylor, and McLafferty (24).

The reactions for lipoic acid activation and its covalent attachment to the pyruvate dehydrogenase complex are:

E1 + ATP + Lipoic Acid &emdash;&emdash;&emdash;&emdash;&emdash;> E1-lipoyl-AMP + PP

E1-lipoyl-AMP + E2 &emdash;&emdash;&emdash;&emdash;> Lipoyl-E2 + AMP + E1

Lipoyl-E2 + apo-PDHC &emdash;&emdash;&emdash;&emdash;&emdash;&emdash;> Lipoyl -PDHC + E2

Where E1 and E2 are the two enzymes of the Streptococcus faecalis lipoic acid-activating system. The lipoic acid is covalently bound to the e-amino group of lysines of E2 of the pyruvate dehydrogenase complex (PDHC). PDHC consists of three distinct enzymes designated as E1, E2, and E3 - note that the distinction between the lipoic acid-activating enzymes and the a-keto acid dehydrogenases made by a subscript number for the former and an on the line number for the latter.

The reaction involved in lipoic acid removal (lipoamidase reaction) is:

Lipoyl-PDHC + H2O &emdash;&emdash;&emdash;&emdash;&emdash;&emdash;&emdash;> apo-PDHC + Lipoic acid

Background

The a-keto acid dehydrogenases are large enzyme complexes that serve essential roles in metabolism (25). The pyruvate dehydrogenase (PDHC) provides the link between glycolysis and the citric acid cycle and produces acetyl-CoA for the citric acid cycle and acetyl groups for acetylcholine synthesis; in omnivores 50-80% of metabolism goes through the PDHC (26). The a-ketoglutarate dehydrogenase functions in the citric acid cycle. The branched chain a-keto acid dehydrogenase is important in regulation of nitrogen metabolism (26). These enzyme complexes involve five cofactors: thiamine pyrophosphate (TPP), lipoic acid (LA) in the form of enzyme-bound lipoamide (the amide between lipoic acid and the e-amino group of lysine), nicotinamide dinucleotide (NAD+), coenzyme A (CoA), and flavin adenine dinucleotide (FAD) shown in Fig. 1 where E1 is the carboxylase (in the case of pyruvate, pyruvate dehydrogenase, EC # 1.2.4.1), E2 is the transacylase-reductase (in the case of pyruvate, dihydrolipoamide acetyltransferase, EC # 2.3.1.61), and E3 is dihydrolipoamide dehydrogenase (EC # 1.8.1.4). The activity of the mammalian forms of these enzymes is regulated by inhibition by products and by a phosphorylation-dephosphorylation cycle (involving insulin among other factors) (14).

There are a series of 5 reactions indicated by the Arabiac numbers that constitute the complete reaction sequence. Reaction 1 is the decarboxylation of pyruvate with the production of CO2 and hydroxyethythiamine pyrophosphate. The hydroxyethylthiamine pyrophosphate is then oxidzed to acetylthiamine pyrophosphate with the reduction of lipoic acid. The acetyl group is then transferred to lipoic acid yielding the 8-S-acyl compound in reaction 2. The third reaction is the transfer of the acyl group to CoA yielding the acyl CoA derivatives. Lipoic acid is now in the dihydro form and must be reoxidized. This occurs in reaction 4. The reduced dihydrolipoyl dehydrogeanse is then oxidzed in reaction 5 producing NADH.

TPP / FAD
Lipoic Acid / NAD+
CoA

The following scheme shows the reaction mechanism in detail.

Complex Structure

There are two polyhedral forms of E2: cubic and dodecahedral (8). The components of the mammalian complexes and E. coli are summarized in Table 1.

Enzyme / Abbreviation / Mr x 10-6 / Subunits / # /Complex
# / # / Mr x 10-3
Part A. Mammalian
Bovine heart
Native complex / PDHC / 8.5
Pyruvate dehydrogenase / E1 / 0.154 / 60
E1a / 2 / 41
E1b / 2 / 36
E2 / 3.1 / 60 / 60 / 1
X / 6 / 50
Dihydrolipoyl dehydrogenase / E3 / 0.11 / 2 / 55 / 12
Kinase / PDHk / 0.1
PDHka / 1 / 48
PDHkb / 1 / 48
Phosphatase / PDHp / 0.15
PDHpa / 1 / 97
PDHpb / 1 / 50
Part B. Bacteria
E. coli
Native complex / PDHC / 4.6
Pyruvate dehydrogenase / E1 / 0.19 / 2 / 99 / 24
Dihydrolipoyl transacetylase / E2 / 1.7 / 24 / 66 / 24
Dihydrolipoyl dehydrogeanse / E3 / 0.112 / 2 / 51 / 24

where k is for kinase and p is for phosphatase.

The amino acid sequence of protein X differs from that of E2, but both contain acetylable lipoamide. Protein X may contribute to assembly of the complex (27). Protein X is also called E3BP now that its function has been established

There is an unique structural organization of the Saccharomyces cerevisiae pyruvate dehydrogenase complex (28). The Reed group used truncated E2, BP and various physical techniques to determine the arrangement. The showed that there were 12 large openings in the E2 core multimer that permitted entrance of BP into the central cavity. Various model structures are depicted.

Regulation

The a-keto acid dehydrogenase complexes are regulated by end-product inhibition by NADH and the appropriate acyl CoA. In addition there is regulation by phosphorylation-dephosphorylation (14,15). This covalent modification cycle is in turn regulated by many components, as is shown in Fig. 2; this occurs for the pyruvate and branched chain ketoacid complexes. The pyruvate dehydrogeanse complex of yeast is regulation by phosphorylation (29). Olson and his group at UT San Antonio have reviewed the regulation of pyruvate dehydrogenase multienzyme complex in the Annual Review of Nutrition (30).

Figure 2. Regulation of the mammalian and yeast pyruvate dehydrogenase complexes by phosphorylation/dephosphorylation.

Whole Genomes

A cluster of genes that encode the branched-chain a-keto acid dehydrogenase from Streptomyces avermitilis has been cloned and sequenced (31). ORF1 has E1a with 1,146 nucleotides encoding a 381 amino acid protein of MW 40,969 Da. ORF2 (E1b) 1,005 nucleotides would code for a 334 amino acid protein of MW 35,577. The inner genic distance is 73 nucleotides. The ATG start codon of ORF3 overlaps the stop codon of ORF2; ORF3 has part of an E2-like sequence. The sequence and organization of the genes encoding enzymes involved in pyruvate metabolism in Mycoplasma capricolum has been analyzed (32). Three operons were found: 1) naox encoding a NADH-oxidase and lplA coding for lipoyl protein ligase, 2) odpA for E1a and odpB for E1b, and 3) odp2 encodes E2 with a single lipoyl domain and dldH a modified E3 that contains a lipoyl domain.

The cloning, structure, chromosomal localization, and promoter of human 2-oxoglutarate dehydrogenase gene has been reviewed by Koike (33). The cDNA contains a 3006-bp open reading frame encoding a 40-amino acid leader peptide and a 962-amino acid mature protein with Mr of 108,878. There are 22 exons spanning 85 kb. The gene is located on chromosome 7 at p13-p14. There are two 10-bp cis-acting elements and two trans-acting elements with a nuclear factor binding to region -63 to -24 that includes the two cis-acting elements involved in the control of synthesis.

The gene and subunit unit organization of the bacterial pyruvate dehydrogenase complexes has been reviewed by Neveling, Bringer-Meyer, and Sahm (34).

Componet Parts

E1. The E1 component contains TPP and catalyzes the decarboxylation of the a-keto acid with the generation of reduced and acylated E2. A tightly bound enzyme intermediate in the process is 2-(1-hydroxyethylidene)-thiamine pyrophosphate when the substrate is pyruvate (35). The nucleotide sequence for the ace E gene of Escherichia coli has been determined (36). The ace E structural gene contains 2,655 base pairs coding for 885 amino acids excluding the initiator. The relative molecular mass of 99,474, the amino-terminal residue, and carboxyl-terminal sequence predicted from the nucleotide sequence are in excellent agreement with published information on E1. The upstream gene A produces protein A of Mr 27,049 which has the helix-turn-helix structure characteristic of a positive regulator (37).

There is little similarity between the sequences of the E1 enzymes of E. coli and bovine heart. The amino acid sequences of the E1ab subunit [the a subunit of E1 for the branch chain complex] of rat liver (30), the E1ab subunit of human liver (39), the E1ap subunit of human liver (40-42), the E1ab subunit of the bovine liver (43), the E1ab and E1bb of Pseudomonas putida (44), and the E1bp subunit of human liver (42) have all been determined from cDNAs. All three reported human E1a cDNA sequences have significant differences still to be be resolved (4). The ace E gene encoding the E1 for pyruvate (36) and the suc A gene (45) encoding the 2-oxoglutarate dehydrogenase have been sequenced for E. coli. The yeast E1bp has 333 amino acids and 36,486 kD (46). A cDNA has been cloned and its amino acid sequence deduced for the E1ap from Arabidopsis thaliana (47). It has about 50 % sequence identity and the phosphorylation site and active site cysteines are conserved.