Phosphotransfer and Nucleotidyltransfer

Phosphotransferases, phosphatases, and nucleotidyltransferases catalyze nucleophilic substitution at phosphorus. They constitute a dominant class of enzymes in intermediary metabolism, energy transduction, nucleic acid biosynthesis and processing, and regulation of many cellular processes, including replication, cellular development, and apoptosis. The mechanisms of the action of these enzymes have been studied intensively at several levels, ranging from the biosynthesis of metabolites and nucleic acids to unmasking signaling networks to elucidating the molecular mechanisms of catalysis. We focus on the chemical mechanisms of the reactions of biological phosphates. More than 40 years of research on this chemistry reveals that the mechanisms can be grouped into two classes: the phosphoryl group (PO3−) transfer mechanisms and the nucleotidyl or alkylphosphoryl group (ROPO2−) transfer mechanisms. Because the fundamental chemical mechanisms of these reactions are not treated in textbooks, we begin by considering this chemistry and then move on to the enzymatic reaction mechanisms. Phosphomonoesters, phosphoanhydrides, and phosphoramidates undergo substitution at phosphorus by transfer of the phosphoryl (PO3–) group, that is, by P—O and P—N cleavage. The current description of a typical phosphoryl group transfer mechanism is one in which the phosphoryl donor and acceptor interact weakly with the phosphoryl group in flight in a transition state in which the total bonding to phosphorus is decreased relative to the ground state. The bonding is weak between phosphorus and the leaving group R–X and between phosphorus and the accepting group Y in the transition state of. Because of decreased bonding to phosphorus, this is a loose transition state that has been described as dissociative. The latter should not be confused with the dissociative mechanism, which is considered later. To avoid confusion, we use the term loose transition state. Detailed studies indicate that the bonding denoted by the dashed lines in represents partial covalency on the order of 10% to 20% of the strength of a full covalent bond, or a bond order of 0.1 to 0.2.

Alkyltransferases

A number of enzymes catalyze alkylation reactions, most of which are reactions of S-adenosyl-L-methionine (SAM) as a methylating agent in the biosynthesis of hormones, modification of DNA, and methyl esterification of proteins involved in signal transduction. Other examples of enzymatic alkylation include prenyl transfer reactions, adenosyltransfer from ATP to methionine in the biosynthesis of SAM, and adenosyltransfer from ATP to cob(I)alamin in the biosynthesis of adenosylcobalamin. Methyl group transfer is also the essential step in the reaction of methionine synthase, which uses 5-methyltetrahydrofolate as an alkylating agent. In an analogous reaction, an analog of 5-methyltetrahydrofolate is the methyl group donor in the methylation of coenzyme M to form methyl coenzyme M, the proximate precursor of methane in methanogenesis (see chap. 4). Glysosyl transfer is an alkylation reaction catalyzed by a large class of enzymes, the glycosyltransferases and glycosidases. The special nature of the glycosyl compounds and their potential for undergoing glycosyltransfer places them in their own class in biochemistry (see chap. 12). The reactivity of glycosyl compounds can be attributed to the contribution of the oxygen atom directly bonded to the glycosyl carbon, the locus of alkylation. In this chapter, we consider other enzymatic alkylations. Alkylation consists of the transfer of a carbon from a leaving group to a nucleophilic acceptor, as in eq.15-1, where R is H or an organic group. The rate is controlled by the reactivity of the nucleophile X:, the stability of the leaving group Y:, and the electrophilic reactivity of the central carbon atom. Alkylation may be regarded as one of the simplest organic chemical reactions because there are few complications in the mechanism. It is the reaction of a nucleophilic molecule with an electrophilic molecule to displace a leaving group. Enzymatic alkylations proceed by polar and not radical mechanisms. In organic chemistry, polar alkylation can occur either by an associative or one-step mechanism, as in fig. 15-1A, or by a dissociative or two-step mechanism through a carbocationic intermediate, as in fig. 15-1B. The chemical nature of the alkylating agent, the propensity of the leaving group to leave, and the polarity of the solvent determine the mechanism.

Nitrogen and Sulfur Transferases

Unlike other group transfer reactions in biochemistry, the actions of nitrogen transferring enzymes do not follow a single unifying chemical principle. Nitrogen-transferring enzymes catalyze aminotransfer, amidotransfer, and amidinotransfer. An aminotransferase catalyzes the transfer of the NH2 group from a primary amine to a ketone or aldehyde. An amidotransferase catalyzes the transfer of the anide-NH2 group from glutamine to another group. These reactions proceed by polar reaction mechanisms. Aminomutases catalyze 1,2-intramolecular aminotransfer, in which an amino group is inserted into an adjacent C—H bond. The action of lysine 2,3-aminomutase, described in chapter 7, is an example of an aminomutase that functions by a radical reaction mechanism. Tyrosine 2,3-aminomutase also catalyzes the 2,3-amino migration, but it does so by a polar reaction mechanism. In this chapter, we consider NH2-transferring enzymes that function by polar reaction mechanisms. Transaminases or aminotransferases are the most extensively studied pyridoxal-5'-phosphate (PLP)–dependent enzymes, and many aminotransferases catalyze essential steps in catabolic and anabolic metabolism. In the classic transaminase reaction, aspartate aminotransferase (AAT) catalyzes the fully reversible reaction of L-aspartate with α-ketoglutarate according to fig. 13-1 to form oxaloacetate and L-glutamate. Like all aminotransferases, AAT is PLP dependent, and PLP functions in its classic role of providing a reactive carbonyl group to function in facilitating the cleavage of the α-H of aspartate and the departure of the α-amino group of aspartate for transfer to α-ketoglutarate (Snell, 1962). PLP in the holoenzyme functions in essence to stabilize the α-carbanions of L-aspartate or L-glutamate, the major biological role of PLP discussed in chapter 3. The functional groups of the enzyme catalyze steps in the mechanism, such as the 1,3-prototropic shift of the α-proton to C4' of pyridoxamine 5'-phosphate (PMP). The steady-state kinetics corresponds to the ping pong bi bi mechanism shown at the bottom of fig. 13-1. This mechanism allows L-aspartate to react with the internal aldimine, E=PLP in fig. 13-1, to produce an equivalent of oxaloacetate, with conversion of PLP to PMP at the active site (E.PMP), the free, covalently modified enzyme in the ping pong mechanism.

Addition and Elimination

Most elimination and addition reactions in biochemistry proceed by α,β-elimination/addition mechanisms. In the case of elimination, the leaving group is β to an activating functional group in the substrate. The activating group may be the carbonyl group of a ketone or aldehyde, the iminium group derived from an aldehyde or ketone, or the acyl-carbonyl of a carboxylic acid or ester, and the proton is α to the activating group. Addition reactions in this class are the same reactions in reverse, and they follow the course of the Michael addition in organic chemistry. The generic process is illustrated in scheme 9-1. Substituents among the activating and leaving groups are diverse and are presumed to account for the significant variations among enzymes in the class. A few enzymes in this class catalyze elimination/addition without the assistance of a coenzyme or cofactor. They presumably incorporate sufficiently acidic (A—H) or basic (:B) amino acid side chains to catalyze the proton transfer processes, or they may stabilize carbanionic intermediates by low-barrier hydrogen bonding. Others employ divalent metal ions, pyridoxal-5'-phosphate (PLP), [4Fe–4S] centers, or NAD+ to facilitate the reactions. Cofactors and coenzymes increase the acidity of Cα—H or improve the propensity of the leaving group Y to depart. In most cases, the major barrier consists of increasing the acidity of the Cα—H group, which decreases the pKa. In a few cases, as when the leaving group is a carboxylic acid or a phosphate, no catalysis is required for it to depart. Limited space prevents discussion of the many enzymes that catalyze cofactor-independent α, β-eliminations. We address the actions of fumarase and crotonase because of the historic emphasis on the biochemical significance of these enzymes. Many other dehydratases and ammonia lyases also belong in this group. In the tricarboxylic acid cycle, fumarate arises from the action of succinate dehydrogenase, and fumarase (EC 4.2.1.2) catalyzes the addition of water to form S-malate. The reaction can be monitored in either direction, and in various studies, the kinetic parameters may be quoted as such (e.g., fumarate formation, or malate formation). The body of knowledge about the action of fumarase is surprisingly incomplete, given the importance of the enzyme in metabolism.

Decarboxylation and Carboxylation

Decarboxylation is an essential process in catabolic metabolism of essentially all nutrients that serve as sources of energy in biological cells and organisms. The most widely known biological process leading to decarboxylation is the metabolism of glucose, in which all of the carbon in the molecule is oxidized to carbon dioxide by way of the glycolytic pathway, the pyruvate dehydrogenase complex, and the tricarboxylic acid cycle. The decarboxylation steps take place in thiamine pyrophosphate (TPP)–dependent α-ketoacid dehydrogenase complexes and isocitrate dehydrogenase. The latter enzyme does not require a coenzyme, other than the cosubstrate NAD+. Many other decarboxylations require coenzymes such as pyridoxal-5'-phosphate (PLP) or a pyruvoyl moiety in the peptide chain. Biological carboxylation is the essential process in the fixation of carbon dioxide by plants and of bicarbonate by animals, plants, and bacteria. Carboxylation by enzymes requires the action of biotin or a divalent metal cofactor, and it requires ATP when the carboxylating agent is the bicarbonate ion. The most prevalent enzymatic carboxylation is that of ribulose bisphosphate carboxylase (rubisco), which is responsible for carbon dioxide fixation in plants. The basic chemistry of decarboxylation is illustrated by mechanisms A to D in fig. 8-1. The mechanisms all require some means of accommodation for the electrons from the cleavage of the bond linking the carboxylate group to the α-carbon. In mechanism A, an electron sink at the β-carbon provides a haven for two electrons. Acetoacetate decarboxylase functions by this mechanism (see chap. 1), as well as PLP- and TPP-dependent decarboxylases (see chap. 3). In mechanism B, a leaving group at the β-carbon departs with two electrons. Mevalonate-5-diphosphate decarboxylate functions by mechanism B and is discussed in a later section. In mechanism C, a leaving group replaces the α-carbon and departs with a pair of electrons. A biological example is formate dehydrogenase, in which the leaving group is a hydride that is transferred to NAD+. In mechanism D, a free radical center is created adjacent to the α-carbon and potentiates the homolytic scission of the bond to the carboxylate group. Mechanism D requires secondary electron transfer processes to create the radical center and quench the formyl radical.

Coenzymes II: Metallic Coenzymes

The original coenzymes were small organic molecules that activated enzymes and participated directly in catalyzing enzymatic reactions. Most of them were derived from vitamins and were known as biologically “activated” forms of vitamins such as niacin, riboflavin, thiamine, and pyridoxal. Heme was in a separate category, perhaps because of its widespread biological role as an oxygen carrier, and because it was not a vitamin, it was not widely regarded as a coenzyme. However, heme was clearly an enzymatic prosthetic group in enzymes such as peroxidases and catalase, and it was known to participate in catalysis. Today, heme takes its place among the coenzymes. Other, more recently discovered metallic cofactors round out this chapter on metallocoenzymes. Most of the detailed mechanisms of metallocoenzyme-dependent reactions are not known. Hypothetical mechanisms can often be written, and some of them are supported by a few experiments. Emerging principles are emphasized here for several of the more extensively studied metallocoenzymes. In other cases, the detailed mechanisms that we include in figures and schemes must be regarded as conjectural. We do not regard them as fanciful, but they have not been proved and are referred to as “a mechanism for” in recognition that other possible mechanisms have not been excluded. Space does not permit all conceivable mechanisms to be aired, and we hope that those shown here will stimulate discussion and experimentation. Vitamin B12 coenzymes may be regarded as transitional from traditional coenzymes, in that the parent cyanocobalamin is a true vitamin, and its biologically activated forms adenosylcobalamin and methylcobalamin, with their covalent cobalt-carbon bonds, are organometallic compounds. For these reasons, we begin by discussing the vitamin B12 coenzymes. The structure in fig. 4-1 is that of adenosylcobalamin, the first B12 coenzyme to be discovered. The molecule consists of the tetradentate corrin ring, cobalt in its 3+ oxidation state held within the corrin ring, the lower axial dimethylbenzimidazole α-ribotide ligand linked by a phosphodiester group to the corrin, and the 5'-deoxyadenosyl moiety covalently bonded to cobalt. The corrin ring is structurally and biosynthetically related to heme, but it differs in a number of respects, including that it is more highly reduced and incorporates extensive stereochemistry.

Oxidases and Oxygenases

An oxidase catalyzes the oxidation of a substrate by O2 without incorporating an oxygen atom into the product. A monooxygenase catalyzes oxidation by O2 with incorporation of one oxygen atom into the product, and oxidation by a dioxygenase proceeds with incorporation of both atoms of O2 into the product. These reactions generally require an organic or metallic coenzyme, with few exceptions, notably urate oxidase. Mechanisms of action of phenylalanine hydroxylase, galactose oxidase, and ascorbate oxidase are provided in chapter 4 in connection with the introduction of metallic coenzymes. In this chapter, we present cases of well-studied coenzyme and metal-dependent oxidases and oxygenases, and we consider one example of an oxidase that does not require a cofactor. Biochemical diversity may be a characteristic of oxidases, which include flavoproteins, heme proteins, copper proteins, and quinoproteins. The actions of copper and topaquinone-dependent amine oxidases are presented in chapter 3, and in chapter 4, two copper-dependent oxidases are discussed. In this chapter, we discuss flavin-dependent oxidases, a mononuclear iron oxidase, and a cofactor-independent oxidase. Flavin-dependent oxidases catalyze the reaction of O2 with an alcohol or amine to produce the corresponding carbonyl compound and H2O2. Examples include glucose oxidase, which produces gluconolactone and H2O2 from glucose and O2 according to. A D-Amino acid oxidase (EC 1.4.3.3) catalyzes a formally similar reaction to produce an α-keto acid from the corresponding α-D-amino acid. The oxidation of an amino acid by an oxidase produces ammonium ion in addition to hydrogen peroxide and the ketoacid, and so it is formally more complex. It proceeds in the three phases described in, the reduction of FAD to FADH2 by the amino acid, hydrolysis of the resultant α-iminoacid to the corresponding α-ketoacid and NH4, and oxidation of FADH2 by O2 to form H2O2. D-Amino acid oxidase is a thoroughly studied example of a flavoprotein oxidase. The enzyme is a 84-kDa homodimer containing one molecule of FAD per subunit. The mechanisms of the hydrolysis of imines and of the oxidation of dihydroflavins are discussed in chapters 1 and 3.

Oxidoreductases

Oxidoreductases constitute a very large class of enzymes. They are dehydrogenases and reductases that catalyze the removal or addition of the elements of molecular hydrogen to or from substrates. Enzymatic dehydrogenation is sometimes linked to auxiliary functions such as decarboxylation, deamination, or dehydration of the substrate, as in the actions of isocitrate dehydrogenase (decarboxylation), glutamate dehydrogenase (deamination), and ribonucleotide reductase (deoxygenation). The best known oxidoreductases are the NAD-dependent dehydrogenases, and a thorough discussion of the actions of these enzymes could easily fill a volume the size of this book. For this reason, this discussion must focus on the salient aspects of reaction mechanisms that represent the major classes of oxidoreductases. Authoritative reviews on the kinetics and structures of the main dehydrogenases are available (Banaszak et al., 1975; Brändén et al., 1975; Dalziel, 1975; Harris and Waters, 1976; Holbrook et al., 1975; Rossman et al., 1975; Smith et al., 1975; Williams, 1976). In this chapter, we emphasize the diverse oxidoreduction mechanisms and place less emphasis on auxiliary functions such as decarboxylation, the mechanisms of which are similar to the actions of enzymes discussed in earlier chapters of this book. Discussions of several dehydrogenases not included in this chapter can be found in other chapters. These include methanol, glucose, and methylamine dehydrogenases in chapter 3, dimethylsulfoxide reductase in chapter 4, and dihydrofolate reductase and β-hydroxymethylglutaryl CoA reductase in chapter 5. Pyruvate and α-ketoglutarate dehydrogenases are discussed in chapter 18. Enzymatic addition or removal of the elements of hydrogen to or from an organic molecule generally requires the action of a coenzyme. In principle, the process may proceed by any of several mechanisms, including the formal transfer of a hydride and a proton; or the transfer of two electrons and two protons; or the transfer of a hydrogen atom, an electron, and a proton; or any of several other sequences. Proteins alone do not efficiently catalyze these processes; coenzymes and cofactors generally provide the essential chemistry for catalysis by oxidoreductases. Many enzymes catalyze the dehydrogenation of an alcoholic group to a ketone or aldehyde coupled with the reduction of NAD+ to NADH.

Carbon-Carbon Condensation and Cleavage

In chemistry, many methods are available to synthesize carbon-carbon bonds, and the reactions proceed by both polar and radical mechanisms. However, enzymatic ligation of two molecules through carbon-carbon bond formation invariably proceeds by a polar mechanism. Often, the reaction involves a carbanionic intermediate or a carbanion-equivalent species such as an enamine, but carbenium ion intermediates also participate in terpene biosynthesis. The only well-known enzymatic processes leading to carbon-carbon bonding by radical mechanisms are the adenosylcobalamin-dependent isomerization reactions discussed in chapter 7. The basic mechanisms illustrated in fig. 14-1 lead to the ligation of molecules through the synthesis of carbon-carbon bonds. Fig. 14-1A depicts the addition of a stabilized carbanion to an aldehyde or ketone to form an adduct. The carbanion can itself be derived from an aldehyde or ketone, as it is in the reactions of aldolases and transketolase. In chapter 1, we discuss the mechanisms of aldolase reactions in connection with the catalytic power of metal ions and of iminium ions formed between substrate carbonyl groups and the lysyl-ε-amino groups of enzymes. In the actions of class I aldolases, X=C in fig. 14-1A is an iminium group formed between a lysyl residue of the enzyme and an aldehyde or ketone group of a substrate. In this case, the carbanion is more accurately described as an enamine, a resonance form in which the charges are not separated. In the actions of class II aldolases, X=C in fig. 14-1A is a carbonyl group (i.e., C=O) coordinated to a divalent metal ion, usually Zn2+, which facilitates carbanion formation through enolization. In this case, the carbanion may be more accurately described as an enolate ion with the charge localized on metal-coordinated oxygen. Iminium ion formation and divalent metal ion ligation both lower the pKa value of the α-C(H) by 7-10 units, thereby facilitating enolization and carbanion formation (see chap. 1). An enolate carbanion may also be derived from a CoA-thioester such as acetyl CoA in the reaction of citrate synthase. Once the carbanion or carbanion-equivalent is formed in an active site, its addition to an aldehyde or ketone group in an adjacent cosubstrate proceeds rapidly.

Glycosyl Group Transferases

Glycosyl group transfer underlies the biosynthesis and breakdown of all nucleotides, polysaccharides, glycoproteins, glycolipids, and glycosylated nucleic acids, as well as certain DNA repair processes. Glycosyl transfer consists of the transfer of the anomeric carbon of a sugar derivative from one acceptor to another, as in, which describes the transfer of a generic pyranosyl ring between nucleophilic atoms :X and :Y of acceptor molecules. The stereochemistry at the anomeric carbon is not specified in eq. 12-1, but the leaving group occupies the axial position in an α-anomer or the equatorial position in a β-anomer. The overall transfer can proceed with either retention or inversion of configuration. In biochemistry, the acceptor atoms can be oxygen, nitrogen, sulfur, or in the biosynthesis of C-nucleosides even carbon. The great majority of biological glycosyl transfer reactions involve transfer between oxygen atoms of different acceptor molecules. Enzymes catalyzing glycosyl transfer are broadly grouped according to whether the acceptor :Y–R2 in is water or another molecule. In the actions of glycosidases, the acceptor is water, and glycosyl transfer results in hydrolysis of a glycoside, a practically irreversible process in dilute aqueous solutions. In the action of glycosyltransferases, the acceptors are molecules with hydroxyl, amide, amine, sulfhydryl, or phosphate groups. The simplest nonenzymatic glycosyl transfer reaction is the hydrolysis of a glycoside, and early studies revealed the fundamental fact that glycosides are much less reactive toward hydrolysis in basic solutions than in acidic solutions. This fact underlies much that is known about the mechanism of glycosyl transfer; that is, the anomeric carbon of a glycoside is remarkably unreactive toward direct nucleophilic attack, but it becomes reactive when one of the oxygens is protonated by an acid, as illustrated in fig. 12-1 for the acid-catalyzed hydrolysis of a generic glycoside. The reaction by both mechanisms in fig. 12-1 proceeds by pre-equilibrium protonation of the glycoside to form oxonium ion intermediates, which are subject to hydrolysis by water. The two mechanisms in fig. 12-1 are of interest. The mechanism proceeding through exocyclic cleavage of the glycoside has historically been regarded as the more likely, and for this reason, the route through endocyclic cleavage has received little consideration.