Generation of protein-derived redox cofactors by posttranslational modification

Abstract

Redox enzymes which catalyze the oxidation and reduction of substrates are ubiquitous in nature. These enzymes typically possess exogenous cofactors to allow them to perform catalytic functions which cannot be accomplished using only amino acid residues. It is now evident that nature also employs an alternative strategy of generating catalytic and redox-active sites in proteins by posttranslational modification of amino acid residues. This review describes the structures and functions of several of these protein-derived cofactors and the diverse mechanisms of posttranslational modification through which they are generated.

---

Victor L. Davidson

Victor Davidson received his BS in Biochemistry from the University of Illinois in 1973 and PhD in Chemistry from Texas Tech University in 1987. After postdoctoral training at Purdue University from 1982–1984 and a research position at the University of California at San Francisco from 1984–1988 he joined the faculty at the University of Mississippi Medical Center where he is currently Professor of Biochemistry.

---

I. Introduction

Redox enzymes catalyze the oxidation and reduction of substrates. Given the limited chemical functionality of amino acid side-chains, these enzymes typically employ redox-active transition metals or exogenous organic molecules as cofactors at their active sites to assist with catalysis. In some cases posttranslational modification is required to covalently attach organic cofactors, such as flavin and heme, to amino acid residues. There are also examples of significant structural modification during posttranslational assembly of some metal-binding sites, as with iron–sulfur clusters. This review describes redox enzymes which possess catalytic and redox-active centers that are formed by posttranslational modification of one or more amino acid residues on the polypeptide chain. In some cases these posttranslational modifications are autocatalytic self-processing events, while in some cases they are catalyzed by accessory enzymes. Understanding the mechanisms of biosynthesis of these protein-derived cofactors^1,2 expands our views of the diversity of types of posttranslational modifications that occur in nature, and the range of chemical and redox reactions that may be catalyzed by enzymes. The best characterized protein-derived redox cofactors are those that are derived from posttranslational modification of tyrosine³ and tryptophan residues.⁴ Many of these posttranslational modifications require the presence of a metal, usually copper or iron. However, despite this common feature, the mechanisms by which these posttranslational modifications are catalyzed are quite diverse. Since the residues which are modified are within the protein structure rather than exposed on the surface, this poses a particularly challenging task for enzymes that catalyze these posttranslational modifications.

II. Tyrosine-derived redox cofactors

Three redox cofactors have been described which are derived from posttranslational modification of tyrosine residues. Each functions in a different class of oxidase enzymes (Fig. 1). Topaquinone (2,4,5-trihydroxyphenylalanine quinone or TPQ)⁵ is present in amine oxidases which also contain copper. These enzymes are ubiquitous in nature and participate in a wide range of biological activities from metabolism in bacteria to the inflammatory response in humans. Lysine tyrosylquinone (LTQ)⁶ is present in lysyl oxidase, an enzyme required for maturation of collagen and elastin in mammals.⁷ A tyrosine–cysteine cross-linked cofactor⁸ has been reported in oxidative enzymes in yeast, fungi and mammals. For each of these tyrosine-derived cofactors the posttranslational modifications by which they are generated are autocatalytic, self-processing events that do not require accessory enzymes. In each case, copper is required for in the posttranslational modification, and then remains in the active site of the enzyme after cofactor biosynthesis where it plays a role in the catalytic mechanism of the mature enzyme.

Fig. 1 Cofactors derived from posttranslational modification of tyrosine residues. (top panel) Structures are shown of topaquinone (TPQ), lysine tyrosylquinone (LTQ) and the tyrosine–cysteine cofactor with the posttranslational modifications colored red. (bottom panel) The structures of the TPQ cofactor in A. globiformis amine oxidase (PDB entry 1ivx) and the cysteine–tyrosine cofactor in galactose oxidase from Dactylium dendroides (PDB entry 1gog) are displayed as sticks colored gray for carbon, red for oxygen, blue for nitrogen and yellow for sulfur. The copper which interacts with each cofactor is displayed as a dark blue sphere.

A Topaquinone (TPQ)

1 Occurrence and function. TPQ-bearing enzymes are ubiquitous in nature and participate in a wide range of biological activities from metabolism in bacteria to the inflammatory response in humans.⁹ These enzymes have been given various names including benzylamine oxidase, diamine oxidase, histamine oxidase, plasma amine oxidase, retinal amine oxidase, and semicarbazide-sensitive amine oxidase. These names are typically based on the tissue from which the enzyme was isolated or one of the substrates for the enzyme, but do not necessarily describe a distinct subclass of the enzyme family. In microorganisms TPQ-dependent copper amine oxidases play a nutritional role, allowing primary amines to be utilized as the sole source of carbon and nitrogen for growth. In plants they play roles in the biosynthesis of hormones, cell walls, and alkaloids.¹⁰ In animals the physiological roles of these amine oxidases are quite varied and include regulation of glucose homeostasis, lymphocyte adhesion, and adipocyte maturation.^11–13 In fact it has been shown that human endothelial vascular adhesion protein 1, whose cell surface expression is induced under inflammatory conditions, is a member of this class of TPQ-containing amine oxidases.¹¹ The overall reaction catalyzed by the TPQ-dependent copper amine oxidases is the conversion of a primary amine substrate to its corresponding aldehyde plus ammonia with molecular oxygen and water as co-substrates and H₂O₂ as an additional product.¹⁴ The catalytic mechanism consists of two half reactions. In the reductive half-reaction (eqn (1)) the amine substrate forms a covalent adduct with TPQ which is reduced by two electrons and then releases the aldehyde product yielding an aminoquinol intermediate. In the oxidative half-reaction (eqn (2)) the aminoquinol is re-oxidized to TPQ by O₂, releasing NH₃ and H₂O₂.

E=O_(quinone) + R–CH₂NH₂ → E–NH_{2(aminoquinol)} + R–CHO (1)

E–NH_{2(aminoquinol)} + O₂ + H₂O → E=O_(quinone) + NH₃ + H₂O₂ (2)

2 Cofactor biogenesis. The posttranslational modifications to create TPQ from tyrosine are the insertion of oxygens at positions 2 and 5 on the aromatic ring. The tyrosine which is modified is present in a conserved sequence, Thr-X-X-Asn-Tyr-Asp/Gln.¹⁵ The modifications are self-processing autocatalytic events that require molecular oxygen as well as copper, but no other enzymes.^16,17

Several mechanisms have been proposed for TPQ biosynthesis^18–21 and a consensus mechanism is shown in Fig. 2. While the exact chemical mechanism is not fully understood there is general agreement on the nature of the intermediate species which are indicated. It requires the participation of nearby amino acid residues, as well as a bound copper which is coordinated by three strictly conserved histidine residues.²² The tyrosine which is modified is deprotonated by a nearby basic amino acid residue and then reacts with O₂ at the C5 position to yield a dopaquinone intermediate.²³ This reaction is believed to proceed via a resonance stabilized intermediate pair; a charge transfer complex between the tyrosinate and Cu²⁺, and a Cu⁺–tyrosyl radical species. The intermediate reacts with O₂ to yield superoxide, which then reacts with the tyrosine. Then fragmentation of the oxygen–oxygen peroxide bond yields dopaquinone. A highly unusual aspect of the biogenesis of TPQ is that at this point in the biosynthetic process the side-chain of the dopaquinone intermediate rotates 180° around its C_β–C_γ bond. Evidence for the rotation of the modified tyrosine during cofactor biosynthesis is supported by crystal structures of the unmodified apoprotein, and a zinc-substituted protein in which the posttranslational modification has not occurred.^20,24 After this conformational change, hydroxide from water adds at the C2 position to yield 2,4,5-trihydroxyphenylalanine, which then is oxidized by a second molecule of O₂ to form the mature TPQ plus H₂O₂. The copper remains bound at this site and then plays a role in catalysis. In addition to copper, other residues assist in catalysis of TPQ. Site-directed mutagenesis of Tyr305 of the amine oxidase from Hansenula polymorpha revealed that this residue confers critical control of oxidation chemistry involved and provided direct evidence of peroxy intermediates in TPQ.²⁵

Fig. 2 Proposed mechanism for biosynthesis of TPQ in copper containing amine oxidases.

B Lysine tyrosylquinone (LTQ)

1 Occurrence and function. LTQ is present in mammalian lysyl oxidase which is a critical enzyme in the formation of connective tissue. It catalyzes the posttranslational modification of elastin and collagen by oxidizing selected peptidyl lysine residues to peptidyl α-aminoadipic δ-semialdehyde residues.²⁶ This initiates formation of the covalent crosslinks that render these extracellular proteins insoluble. Lysyl oxidase expression is also associated with both cell proliferation and tumor suppression.²⁷ Interestingly, tumor suppressor activity that has been shown to depend on the propeptide region derived from the initially synthesized 50 kDa lysyl oxidase precursor protein, and not on the activity of the mature 30 kDa lysyl oxidase enzyme.²⁸

2 Cofactor biogenesis. The posttranslational modifications to create LTQ from tyrosine are the insertion of oxygen at position 5 on the aromatic ring and formation of a covalent crosslink between the carbon at position 2 and the side-chain nitrogen of a lysine residue (Fig. 1).⁶ Mechanistic studies of LTQ biosynthesis have been hampered by difficulties in expressing the recombinant mammalian lysyl oxidase preprotein. The mechanism of biosynthesis is not known with certainty but it is a self-processing event.²⁹ The covalent bond with the lysine is formed at the same position on the modified tyrosine at which the second oxygen atom is inserted into the mature TPQ cofactor in the amine oxidases, which was discussed above. Given the presence of copper at the active site of both TPQ and LTQ enzymes, it has been assumed that the initial steps in the mechanism of biosynthesis of LTQ are similar to those described earlier for TPQ, but with the lysine nitrogen adding to the dopaquinone intermediate (see Fig. 2).²³ By analogy with the proposed mechanism for TPQ biosynthesis (see Fig. 2), following nucleophilic attack at the C2 position by the amino nitrogen, aromatization yields a reduced lysylquinol intermediate which is then oxidized by a second molecule of O₂ to form LTQ plus H₂O₂.

C Tyrosine–cysteine cofactor

1 Occurrence and function. A tyrosine–cysteine crosslinked cofactor (Fig. 1) is found in copper-containing galactose and glyoxal oxidases from fungi.^8,30 It was first described in the crystal structure of galactose oxidase which revealed the formation of a covalent thioether bond between the carbon adjacent to the phenyl oxygen and a sulfur from a nearby cysteine residue.⁸ Glyoxal oxidase is a secreted fungal enzyme that functions as an extracellular producer of hydrogen peroxide, fueling peroxidases that are responsible for microbial lignin degradation.³¹ The function of galactose oxidase is unclear, but it may be involved in breaking down the plant cell wall prior to invasion by the fungal host. The overall reaction catalyzed by these enzymes is the oxidation of a primary alcohol to the corresponding aldehyde, coupled to the reduction of dioxygen to hydrogen peroxide (eqn 3).

RCH₂OH + O₂ → RCHO + H₂O₂ (3)

After the completion of the posttranslational modification, the modified tyrosineprovides a ligand for the metal which is present in the active site and required for activity. While the requirement for the metal in both cofactor biosynthesis and subsequent enzymatic activity is similar to what is observed with TPQ enzymes, the active sites of enzymes with the tyrosine–cysteine bear no similarity to those harboring TPQ. A similar tyrosine–cysteine cross-linked species has been identified in the active site of the iron-dependent mammalian enzyme cysteine dioxygenase,³² however there is uncertainty as to whether it is essential for catalytic activity.³³

2 Cofactor biogenesis. The mechanism of biosynthesis of the tyrosine–cysteine cofactor is best described for the fungal galactose oxidase.³⁴ It is initially synthesized as a pro-enzyme with a pre-sequence which is cleaved during the maturation of the tyrosine–cysteine cofactor. While it was initially believed that the pre-sequence played a role in the posttranslational modification of the tyrosine,^35,36 this now seems not to be the case. A recombinant galactose oxidase precursor protein was isolated with a processed N-terminus (i.e., no pre-sequence) and no tyrosine–cysteine crosslink. The posttranslational modifications could be achieved from this precursor by addition of O₂ and Cu¹⁺.³⁴ Cu²⁺ could be used instead of Cu¹⁺, but then the maturation process required hours to complete rather than seconds. This led to the proposal of the following mechanism by Whittaker (Fig. 3).³⁴ Cu¹⁺ initially binds to the active site and becomes coordinated with the tyrosine. On addition of O₂, electron transfer from Cu¹⁺ to O₂ forms a Cu²⁺–superoxide coordinated intermediate which abstracts a hydrogen atom from a nearby cysteine to generate a thiyl radical. Alternatively, it was proposed that the process may proceed via a tyrosyl phenoxyl free radical intermediate. The thiyl radical then adds to the tyrosine aromatic ring. Loss of a proton and reduction of the metal ion with release of H₂O₂ restores the aromatic conjugation of the ring. This species corresponds to the fully reduced mature cofactor formed from substrate oxidation during catalytic turnover. Subsequent reaction with another O₂ yields an oxidized copper–tyrosyl radical complex, which is the species that is present in the resting state of the enzyme.³⁷

Fig. 3 Proposed mechanism for biosynthesis of the tyrosine–cysteine cofactor in galactose oxidase.

III. Tryptophan-derived cofactors

Tryptophan-derived quinone cofactors⁴ (Fig. 4) are present in two independently evolved bacterial dehydrogenases possessing either tryptophan tryptophylquinone (2′,4-bitryptophan-6,7-dione or TTQ)³⁸ or cysteine tryptophylquinone (CTQ).^39,40

Fig. 4 Cofactors derived from posttranslational modification of tryptophan residues. (top panel) Structures are shown of tryptophan tryptophylquinone (TTQ) and cysteine tryptophylquinone (CTQ) with the posttranslational modifications colored red. (bottom panel) The structure of the TTQ cofactor in P. denitrificans methylamine dehydrogenase (PDB entry 2bbk) is displayed as sticks colored gray for carbon, red for oxygen and blue for nitrogen.

A Tryptophan tryptophyquinone (TTQ)

1 Occurrence and function. TTQ has been identified in methylamine dehydrogenase (MADH) and aromatic amine dehydrogenase from a variety of bacterial sources.^4,41 These enzymes are inducible and allow the host bacterium to grow on the substrate amine as a sole source of carbon, nitrogen and energy. In contrast to the tyrosine-derived cofactors described above, the enzymes that possess these tryptophan-derived cofactors are dehydrogenases rather than oxidases. The overall reaction catalyzed by these dehydrogenases is the conversion of a primary amine substrate to its corresponding aldehyde plus ammonia with the two substrate-derived electrons transferred to an electron acceptor protein. The catalytic mechanism consists of reductive and oxidative reactions. In the reductive reaction the amine substrate forms a covalent adduct with the C6 of TTQ which is reduced by two electrons and then releases the aldehyde product yielding an aminoquinol intermediate (eqn (4)).⁴² The oxidative reactions involve interprotein electron transfer usually to a type 1 copper protein, amicyanin for MADH⁴³ and azurin for AADH.⁴⁴ Since these copper proteins are one-electron carriers the oxidation of the aminoquinol to TTQ requires two one-electron transfers. The product of the first electron-transfer is an aminosemiquinone (eqn (5)).⁴⁵ The product of the second electron-transfer is an aminoquinone which is hydrolyzed to release the ammonia product and regenerate the oxidized TTQ (eqn (6)).⁴⁶

E=O_(quinone) + CH₃NH₂ → E–NH_{2(aminoquinol)} + HCHO (4)

E–NH₂ + Amicyanin(Cu²⁺) → E˙–NH_{2(aminosemiquinone)} + Amicyanin(Cu¹⁺) (5)

E˙–NH₂ + Amicyanin(Cu²⁺) + H₂O → E=O_(quinone) + Amicyanin(Cu¹⁺) + NH₄⁺ + H⁺ (6)

2 Cofactor biogenesis. In contrast to the tyrosine derived cofactors described above, the posttranslational modifications which form TTQ are not self-processing events but require the action of accessory enzymes. The posttranslational modifications to create TTQ are the insertion of oxygen at positions 6 and 7 on the indole ring of a specific tryptophan and formation of a covalent crosslink to the indole ring of another tryptophan residue. The mechanism of biosynthesis of TTQ has been studied most extensively in methylamine dehydrogenase (MADH) from Paracoccus denitrificans.⁴⁷ The mechanism by which TTQ biosynthesis occurs has not been completely elucidated, but it is known to require the action of other enzymes that are subject to the same genetic regulation as the structural genes for the enzyme.^48,49 The genes that encode the two MADH subunits, mauA and mauB, are clustered in the methylamine utilization (mau) locus.⁴⁹ A role specifically in TTQ biosynthesis has been demonstrated for one of these genes, mauG. Its gene product, MauG, is a diheme enzyme.⁵⁰ Inactivation of mauG allowed isolation of a biosynthetic precursor of MADH (preMADH) that contains an incompletely synthesized TTQ with a single hydroxyl group on βTrp57 and no covalent cross-link to βTrp108.⁵¹ Incubation of this intermediate in vitro with purified MauG and oxidation equivalents provided by a reductant plus O₂, or by H₂O₂, resulted in completion of TTQ biosynthesis and formation of active MADH.^51–53 Thus, while the mechanism by which the first oxygen is inserted into βTrp57 is unknown, MauG is able to complete the biosynthetic process.

Completion of TTQ biosynthesis is a six-electron oxidation process to achieve oxygen insertion, crosslinking of the βTrp57 and βTrp108 side-chains, and oxidation to the quinone (Fig. 5A). These biosynthetic oxidation reactions proceed via a relatively stable, high valent bis-Fe(IV) MauG intermediate, with one heme as Fe(IV)=O and the other as Fe(IV) with two axial ligands from the protein.⁵⁴ The two hemes of MauG have been shown to exhibit cooperative redox behavior and behave as a single diheme unit.^54,55 The crystal structure of MauG⁵⁶ reveals that the sole axial ligand for the ferryl heme is histidine, and the axial ligands for the other heme are histidine and tyrosine. This is the first example of a c-type heme with tyrosine providing an axial ligand. Replacement of this tyrosine with histidine demonstrated that the tyrosine ligation was critical for stabilization of the Fe(IV) state on the six-coordinate heme which was critical for TTQ biosynthesis activity.⁵⁷

Fig. 5 (A) The six-electron oxidation reaction that is catalyzed by MauG TTQ biosynthesis from the preMADH precursor of methylamine dehydrogenase. [O] represents oxidizing equivalents that may be provided by H₂O₂ or a reductant plus O₂. (B) The structure of the preMADH–MauG complex. MauGs are colored pink, the preMADH α subunits are colored blue and the β subunits are colored green. The hemes and residues β57 and β108, which are converted to TTQ, are colored black and indicated by arrows. The structure is PDB entry 3L4M.

Despite the cooperative redox behavior, the heme irons are separated by 21 Å, although an intervening tryptophan residue is within 4 Å of each heme. Further surprising insight into the catalytic mechanism was obtained from the crystal structure of the MauG–preMADH protein complex.⁵⁶ This structure revealed that βTrp57 and βTrp108 of preTTQ do not make direct contact with either heme of MauG but are actually separated by very long distance requiring long range ET for catalysis (Fig. 5B). The oxygen-binding heme iron is more than 40 Å from the site of oxygen insertion into residue βTrp57. The relevance of this structure was confirmed by the demonstration that TTQ biosynthesis could be achieved in crystallo by addition of H₂O₂ to the crystal.⁵⁶ Thus, it appears that the role of MauG is to provide an extremely high potential oxidant to extract electrons from the preMADH substrate, generating reactive radical intermediates that then insert oxygen atom from solvent and form the covalent crosslink between the two side-chains.

The kinetic mechanism of initial two-electron oxidation of preMADH by bis-Fe(IV) MauG has been characterized⁵⁸ and it was demonstrated that the product of the reaction is a radical species which was formed on preMADH,⁵⁴ and has yet to be fully characterized. In contrast to other heme-dependent oxygenases such as cytochrome P450s which require oxygen binding to prime the enzyme for substrate binding,^59,60 the kinetic mechanism of MauG was shown to be random as the order of binding of the oxygen or H₂O₂ did not matter.⁵⁸ However, cycling between the bis-Fe(IV) and diferric state in the absence of the preMADH substrate did cause inactivation of MauG.⁶¹ The order of oxygen insertion and crosslink formation during TTQ biosynthesis has not yet been determined. However, it is believed that the final reaction is the two-electron oxidation of quinol MADH to quinone (TTQ) MADH, and this step has been enzymatically characterized.⁶² Thus, while the overall process of MauG-dependent TTQ biosynthesis has not yet been fully elucidated, it is clear that these unusual posttranslational modifications comprise a novel mechanism for cofactor biogenesis which requires a diheme enzyme with unprecedented physical and catalytic properties.

B Cysteine tryptophylquinone (CTQ)

1 Occurrence and function. CTQ (Fig. 4) has been identified as the protein-derived cofactor of bacterial quinohemoprotein amine dehydrogenase (QHNDH).^39,40 QHNDH contains two covalently bound c-type hemes in addition to CTQ. The overall reaction catalyzed by QHNDH is similar to that catalyzed by the TTQ-dependent dehydrogenases. The primary amine substrate is converted to its corresponding aldehyde plus ammonia. However, in QHNDH the immediate electron acceptor for the two substrate-derived electrons are the two hemes which are present in the α subunit. These electrons are then donated to either azurin or a cytochrome c, depending upon the host bacterium, to reoxidize QHNDH.^63,64

2 Cofactor biogenesis. CTQ is formed by posttranslational modifications during which two atoms of oxygen are incorporated into the indole ring of a tryptophan residue and a covalent bond is formed between the modified indole ring and the sulfur of a cysteine residue. While CTQ- and TTQ-dependent enzymes possess the same tryptophylquinone moiety at their active sites, the overall structures of the CTQ- and TTQ-dependent enzymes are quite different. The TTQ-bearing amine dehydrogenases are α₂β₂ heterodimeric proteins.⁴¹ The CTQ-bearing quinohemoprotein amine dehydrogenase is an αβγ heterotrimeric protein.^39,40 Another highly unusual structural feature of QHNDH is that in addition to CTQ, the γ subunit contains three novel thioether crosslinks. Each is formed between the sulfur of a cysteine residue and either the β-methylene carbon of an aspartic acid residue or the γ-methylene carbon of a glutamic acid residue. These crosslinked amino acid residues appear not to participate in catalysis. Instead they play a structural role in determining the tertiary structure of the γ subunit. It is interesting to note that the TTQ-dependent methylamine dehydrogenase does not have these thioether crosslinked residues, but does have six intrasubunit disulfide bonds between cysteine residues which similarly play a role in determining the tertiary structure of the TTQ-bearing β subunit of that enzyme.⁶⁵

The mechanism of biogenesis of CTQ is much less understood than that of TTQ. The gene cluster which encodes QHNDH³⁹ is completely different from that which encodes MADH. The QHNDH gene cluster does not contain a mauG equivalent. It does contains an open reading frame (ORF2) which encodes a putative radical S-adenosylmethione protein,^66,67 which is not present in the MADH gene cluster. The ORF2 protein is believed to be required for the posttranslational formation of the intra-peptidyl thioether crosslinks.⁶⁸ While an accessory gene like mauG is not present, the α subunit of QHNDH contains two c-type hemes. This may be coincidental but it raises the possibility that this diheme protein subunit may perform a role in CTQ biosynthesis similar to that of MauG during TTQ biosynthesis; but at present this is purely speculation.

IV. Other posttranslationally modified redox sites

In addition to the posttranslationally modified enzymes discussed thus far, there are also enzymes and proteins which use either copper or heme iron that possess posttranslationally modified amino acid residues in the active sites which the metals inhabit. The extent to which this phenomenon occurs is still an open question, and the mechanisms of posttranslational modifications have not yet been established for most of the modified proteins thus far described. Described below are of few of these proteins with posttranslationally modified redox sites which have been structurally characterized.

A Posttranslationally modified residues at heme sites

Posttranslationally modified amino acid residues have been identified in the proximity of the catalytic heme of certain redox enzymes (Fig. 6). Catalase-peroxidases (KatGs) are bifunctional hemoproteins that exhibit both catalatic and peroxidatic activities. Of particular interest is the KatG of Mycobacterium tuberculosis which contributes to pathogenesis through the ability to catabolize the peroxides generated by the phagocyte NADPH oxidase.⁶⁹ Crystal structures of KatG from three bacterial sources revealed that it possesses a crosslinked triad of amino acid residues at the heme active site.^70–72 In this case, there are two covalent crosslinks to a tyrosine residue, both to the aromatic ring. One is with a tryptophan residue and one is with a methionine residue. It was possible to gain insight into the mechanism of formation of these crosslinks from studies on the expression of the recombinant enzyme from M. tuberculosis. A protein lacking the crosslinks was prepared from cells deficient in heme by growth in iron-depleted media. After reconstitution with heme in vitro, formation of the crosslinks was observed after addition of peroxyacetic acid.⁷³ On the basis of these results, it was proposed that the formation of these crosslinks was catalyzed by the endogenous heme, which initially functions as a peroxidase using these amino acid residues as co-substrates in the reaction. Formation of amino acid-based radicals followed by radical coupling generates each covalent crosslink. Once the formation of the mature active site with the methionine–tyrosine–tryptophan crosslinked species is complete in the mature enzyme, the heme then reacts with its natural substrates. Crosslinked amino acid residue pairs have been observed in the heme sites of two other catalases. A crosslink between a histidine and a tyrosine residue has been identified in catalase HPII,⁷⁴ and a crosslink between a cysteine and a tyrosine residue has been identified in catalase-1.⁷⁵ In each case, the crosslink is formed at the β-carbon of the tyrosine residue rather than to the aromatic ring. The crosslinked residues are not directly coordinated to the heme iron but in close proximity. While mechanisms by which these posttranslational modifications occur are not known, a process similar to that which was elucidated for KatG is a distinct possibility.

Fig. 6 Posttranslationally modified amino acid residues at heme sites of enzymes. (top panel) Structures are shown of three sets of amino acid residues at the heme sites of their respective enzymes which possess posttranslational modifications (colored in red) that are required for the enzymatic activity. (bottom panel) The structure of the heme site of KatG from M. tuberculosis (PDB entry 1sj2) with residues Met255, Tyr229 and Trp107 and the nearby heme displayed as sticks colored gray for carbon, red for oxygen, blue for nitrogen, yellow for sulfur and dark red for iron.

B Posttranslationally modified copper ligands

Posttranslationally modified amino acid residues have been identified which provide ligands for copper at the catalytic site of certain redox enzymes (Fig. 7). A tyrosine–histidine crosslink is present in cytochrome c oxidase, which is a component of the membrane bound respiratory electron transport chains of mammals, plants and microorganisms. The crystal structures of bacterial and mammalian cytochrome c oxidase^76,77 revealed that a histidine residue which provides one of the copper ligands of the binuclear Cu_B component of the Cu_B–heme_a3 center is covalently crosslinked to a tyrosine residue. Analysis by mass spectrometry revealed that this same crosslinked amino acid species is present and a common feature of each of the three distinct families of heme–copper oxygen reductases in nature.⁷⁸ A cysteine–histidine crosslink is present in catechol oxidase and hemocyanin, two enzymes that possess type 3 dinuclear copper centers. Catechol oxidase catalyzes the oxidation of phenols by molecular oxidation. The product of the oxidation of catechol is benzoquinone, which is responsible for the browning of fruit. The crystal structure of a plant catechol oxidase revealed that the Cu_A of the dinuclear Cu_A–Cu_B metal site is coordinated by three histidine ligands, one of which is covalently cross-linked to a cysteine residue by a thioether bond.⁷⁹ Hemocyanins carry oxygen in the blood of some mollusks and crustaceans. Their biological function is similar to that of hemoglobin in animals but rather than using heme iron to bind O₂, hemocyanins possess two copper atoms that reversibly bind O₂. A cysteine–histidine bond to one of the copper ligands in the dinuclear metal site of hemocyanin has been identified in its crystal structure.⁸⁰

Fig. 7 Posttranslationally modified amino acid residues that provide ligands for copper. (top panel) Structures are shown of two sets of amino acid residues at the copper sites of their respective enzymes which possess posttranslational modifications (colored in red) that are required for the enzymatic activity. (bottom panel) The structure of modified site of catechol oxidase from sweet potato (PDB entry 1bt3) is shown with the cross-linked amino acid residues Cys92 and His109 displayed as sticks colored gray for carbon, red for oxygen, blue for nitrogen and yellow for sulfur. The coppers of the dinuclear copper center are displayed as dark blue spheres.

V. Concluding remarks

A distinctive aspect of the posttranslational modifications discussed in this review is that the modified residues in each case are buried within the protein rather than exposed at the surface. Cofactor biosynthesis must precede catalytic function. Therefore, in the cases in which the posttranslational modification is self-processing, structural features of the protein must have initially evolved for some different function, which somehow created the environment for the posttranslational modifications that generated the protein-derived cofactor and a new catalytic function. It is also clearly evident that certain amino acid residues and bound metals at the active sites of these enzymes possess multiple functions: one in the posttranslational modification to create the protein-derived cofactor, and another in catalysis in the mature enzyme. In the cases where accessory enzymes are required to catalyze the posttranslational modifications, the nature of the substrate requires that catalysis occur somehow without direct contact between the active site of the modifying enzyme and the buried residues which are modified. The posttranslational modifications described in this review were discovered within the last 20 years, and it seems likely that more will be revealed in the coming years. The characterization of the mechanisms by which redox-active centers are generated by posttranslational modification raises interesting questions regarding the evolution of protein structures and functions, and is revealing unanticipated catalytic mechanisms for the formation of the protein-derived cofactors.

Abbreviations

TPQ	Topaquinone
LTQ	Lysine tyrosylquinone
TTQ	Tryptophan tryptophylquinone
CTQ	Cysteine tryptophylquinone
MADH	Methylamine dehydrogenase
QHNDH	Quinohemoprotein amine dehydrogenase

Acknowledgements

Work from the author’s laboratory was supported by NIH grant GM41574.

References

1. V. L. Davidson, Biochemistry, 2007, 46, 5283–5292.
2. N. M. Okeley and W. A. van der Donk, Chem. Biol., 2000, 7, R159–R171.
3. M. Mure, Acc. Chem. Res., 2004, 37, 131–139.
4. V. L. Davidson, Bioorg. Chem., 2005, 33, 159–170.
5. S. M. Janes, D. Mu, D. Wemmer, A. J. Smith, S. Kaur, D. Maltby, A. L. Burlingame and J. P. Klinman, Science, 1990, 248, 981–987.
6. S. X. Wang, M. Mure, K. F. Medzihradszky, A. L. Burlingame, D. E. Brown, D. M. Dooley, A. J. Smith, H. M. Kagan and J. P. Klinman, Science, 1996, 273, 1078–1084.
7. H. M. Kagan and P. C. Trackman, Am. J. Respir. Cell Mol. Biol., 1991, 5, 206–210.
8. N. Ito, S. E. Phillips, C. Stevens, Z. B. Ogel, M. J. McPherson, J. N. Keen, K. D. Yadav and P. F. Knowles, Nature, 1991, 350, 87–90.
9. Copper Amine Oxidases: Structures, catalytic mechanisms, and roles in pathophysiology, ed. G. Floris and B. Mondovi, CRC Press, Boca Raton, Florida, 2009.
10. A. Cona, G. Rea, R. Angelini, R. Federico and P. Tavladoraki, Trends Plant Sci., 2006, 11, 80–88.
11. T. T. Airenne, Y. Nymalm, H. Kidron, D. J. Smith, M. Pihlavisto, M. Salmi, S. Jalkanen, M. S. Johnson and T. A. Salminen, Protein Sci., 2005, 14, 1964–1974.
12. C. M. Stolen, R. Madanat, L. Marti, S. Kari, G. G. Yegutkin, H. Sariola, A. Zorzano and S. Jalkanen, FASEB J., 2004, 18, 702–704.
13. P. H. Yu, S. Wright, E. H. Fan, Z.-R. Lun and D. Gubisne-Harberle, Biochim. Biophys. Acta, Proteins Proteomics, 2003, 1647, 193–199.
14. M. Mure, S. A. Mills and J. P. Klinman, Biochemistry, 2002, 41, 9269–9278.
15. D. Mu, S. M. Janes, A. J. Smith, D. E. Brown, D. M. Dooley and J. P. Klinman, J. Biol. Chem., 1992, 267, 7979–7982.
16. R. Matsuzaki, T. Fukui, H. Sato, Y. Ozaki and K. Tanizawa, FEBS Lett., 1994, 351, 360–364.
17. D. Cai and J. P. Klinman, J. Biol. Chem., 1994, 269, 32039–32042.
18. B. J. Brazeau, B. J. Johnson and C. M. Wilmot, Arch. Biochem. Biophys., 2004, 428, 22–31.
19. M. C. Wilce, D. M. Dooley, H. C. Freeman, J. M. Guss, H. Matsunami, W. S. McIntire, C. E. Ruggiero, K. Tanizawa and H. Yamaguchi, Biochemistry, 1997, 36, 16116–16133.
20. M. Kim, T. Okajima, S. Kishishita, M. Yoshimura, A. Kawamori, K. Tanizawa and H. Yamaguchi, Nat. Struct. Biol., 2002, 9, 591–596.
21. J. E. Dove and J. P. Klinman, Adv. Protein Chem., 2001, 58, 141–174.
22. R. Li, J. P. Klinman and F. S. Mathews, Structure (London), 1998, 6, 293–307.
23. R. H. Moore, M. A. Spies, M. B. Culpepper, T. Murakawa, S. Hirota, T. Okajima, K. Tanizawa and M. Mure, J. Am. Chem. Soc., 2007, 129, 11524–11534.
24. Z. Chen, B. Schwartz, N. K. Williams, R. Li, J. P. Klinman and F. S. Mathews, Biochemistry, 2000, 39, 9709–9717.
25. Z. W. Chen, S. Datta, J. L. Dubois, J. P. Klinman and F. S. Mathews, Biochemistry, 2010, 49, 7393–7402.
26. H. M. Kagan and W. Li, J. Cell. Biochem., 2003, 88, 660–672.
27. J. T. Erler, K. L. Bennewith, M. Nicolau, N. Dornhofer, C. Kong, Q. T. Le, J. T. Chi, S. S. Jeffrey and A. J. Giaccia, Nature, 2006, 440, 1222–1226.
28. S. R. Vora, Y. Guo, D. N. Stephens, E. Salih, E. D. Vu, K. H. Kirsch, G. E. Sonenshein and P. C. Trackman, Biochemistry, 2010, 49, 2962–2972.
29. J. A. Bollinger, D. E. Brown and D. M. Dooley, Biochemistry, 2005, 44, 11708–11714.
30. M. M. Whittaker, P. J. Kersten, D. Cullen and J. W. Whittaker, J. Biol. Chem., 1999, 274, 36226–36232.
31. P. J. Kersten and T. K. Kirk, J. Bacteriol., 1987, 169, 2195–2201.
32. C. R. Simmons, Q. Liu, Q. Huang, Q. Hao, T. P. Begley, P. A. Karplus and M. H. Stipanuk, J. Biol. Chem., 2006, 281, 18723–18733.
33. J. E. Dominy, Jr., J. Hwang, S. Guo, L. L. Hirschberger, S. Zhang and M. H. Stipanuk, J. Biol. Chem., 2008, 283, 12188–12201.
34. M. M. Whittaker and J. W. Whittaker, J. Biol. Chem., 2003, 278, 22090–22101.
35. M. S. Rogers, A. J. Baron, M. J. McPherson, P. F. Knowles and D. M. Dooley, J. Am. Chem. Soc., 2000, 122, 990–991.
36. S. J. Firbank, M. S. Rogers, C. M. Wilmot, D. M. Dooley, M. A. Halcrow, P. F. Knowles, M. J. McPherson and S. E. Phillips, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 12932–12937.
37. J. W. Whittaker, Arch. Biochem. Biophys., 2005, 433, 227–239.
38. W. S. McIntire, D. E. Wemmer, A. Chistoserdov and M. E. Lidstrom, Science, 1991, 252, 817–824.
39. S. Datta, Y. Mori, K. Takagi, K. Kawaguchi, Z. W. Chen, T. Okajima, S. Kuroda, T. Ikeda, K. Kano, K. Tanizawa and F. S. Mathews, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 14268–14273.
40. A. Satoh, J. K. Kim, I. Miyahara, B. Devreese, I. Vandenberghe, A. Hacisalihoglu, T. Okajima, S. Kuroda, O. Adachi, J. A. Duine, J. Van Beeumen, K. Tanizawa and K. Hirotsu, J. Biol. Chem., 2002, 277, 2830–2834.
41. V. L. Davidson, Adv. Protein Chem., 2001, 58, 95–140.
42. G. R. Bishop, E. J. Valente, T. L. Whitehead, K. L. Brown, R. T. Hicks and V. L. Davidson, J. Am. Chem. Soc., 1996, 118, 12868–12869.
43. M. Husain and V. L. Davidson, J. Biol. Chem., 1985, 260, 14626–14629.
44. Y. L. Hyun and V. L. Davidson, Biochemistry, 1995, 34, 12249–12254.
45. G. R. Bishop, H. B. Brooks and V. L. Davidson, Biochemistry, 1996, 35, 8948–8954.
46. Z. Zhu and V. L. Davidson, Biochemistry, 1999, 38, 4862–4867.
47. C. M. Wilmot and V. L. Davidson, Curr. Opin. Chem. Biol., 2009, 13, 469–474.
48. C. J. van der Palen, W. N. Reijnders, S. de Vries, J. A. Duine and R. J. van Spanning, Antonie van Leeuwenhoek, 1997, 72, 219–228.
49. C. J. van der Palen, D. J. Slotboom, L. Jongejan, W. N. Reijnders, N. Harms, J. A. Duine and R. J. van Spanning, Eur. J. Biochem., 1995, 230, 860–871.
50. Y. Wang, M. E. Graichen, A. Liu, A. R. Pearson, C. W. Wilmot and V. L. Davidson, Biochemistry, 2003, 42, 7318–7325.
51. A. R. Pearson, T. de la Mora-Rey, M. E. Graichen, Y. Wang, L. H. Jones, S. Marimanikkupam, S. A. Aggar, P. A. Grimsrud, V. L. Davidson and C. W. Wilmot, Biochemistry, 2004, 43, 5494–5502.
52. Y. Wang, X. Li, L. H. Jones, A. R. Pearson, C. M. Wilmot and V. L. Davidson, J. Am. Chem. Soc., 2005, 127, 8258–8259.
53. X. Li, L. H. Jones, A. R. Pearson, C. M. Wilmot and V. L. Davidson, Biochemistry, 2006, 45, 13276–13283.
54. X. Li, R. Fu, S. Lee, C. Krebs, V. L. Davidson and A. Liu, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 8597–8600.
55. X. Li, M. Feng, Y. Wang, H. Tachikawa and V. L. Davidson, Biochemistry, 2006, 45, 821–828.
56. L. M. Jensen, R. Sanishvili, V. L. Davidson and C. M. Wilmot, Science, 2010, 327, 1392–1394.
57. N. Abu Tarboush, M. L. R. Jensen, M. Feng, H. Tachikawa, C. M. Wilmot and V. L. Davidson, 2010, submitted.
58. S. Lee, S. Shin, X. Li and V. L. Davidson, Biochemistry, 2009, 48, 2442–2447.
59. B. Meunier, S. P. de Visser and S. Shaik, Chem. Rev., 2004, 104, 3947–3980.
60. M. Sono, M. P. Roach, E. D. Coulter and J. H. Dawson, Chem. Rev., 1996, 96, 2841–2888.
61. S. Shin, S. Lee and V. L. Davidson, Biochemistry, 2009, 48, 10106–10112.
62. S. Shin, N. Abu Tarboush and V. L. Davidson, Biochemistry, 2010, 49, 5810–5816.
63. O. Adachi, T. Kubota, A. Hacisalihoglu, H. Toyama, E. Shinagawa, J. A. Duine and K. Matsushita, Biosci., Biotechnol., Biochem., 1998, 62, 469–478.
64. K. Takagi, K. Yamamoto, K. Kano and T. Ikeda, Eur. J. Biochem., 2001, 268, 470–476.
65. L. Chen, M. Doi, R. C. Durley, A. Y. Chistoserdov, M. E. Lidstrom, V. L. Davidson and F. S. Mathews, J. Mol. Biol., 1998, 276, 131–149.
66. P. A. Frey, A. D. Hegeman and F. J. Ruzicka, Crit. Rev. Biochem. Mol. Biol., 2008, 43, 63–88.
67. H. J. Sofia, G. Chen, B. G. Hetzler, J. F. Reyes-Spindola and N. E. Miller, Nucleic Acids Res., 2001, 29, 1097–1106.
68. K. Ono, T. Okajima, M. Tani, S. Kuroda, D. Sun, V. L. Davidson and K. Tanizawa, J. Biol. Chem., 2006, 281, 13672–13684.
69. V. H. Ng, J. S. Cox, A. O. Sousa, J. D. MacMicking and J. D. McKinney, Mol. Microbiol., 2004, 52, 1291–1302.
70. T. Bertrand, N. A. Eady, J. N. Jones, Jesmin, J. M. Nagy, B. Jamart-Gregoire, E. L. Raven and K. A. Brown, J. Biol. Chem., 2004, 279, 38991–38999.
71. Y. Yamada, T. Fujiwara, T. Sato, N. Igarashi and N. Tanaka, Nat. Struct. Biol., 2002, 9, 691–695.
72. X. Carpena, S. Loprasert, S. Mongkolsuk, J. Switala, P. C. Loewen and I. Fita, J. Mol. Biol., 2003, 327, 475–489.
73. R. A. Ghiladi, G. M. Knudsen, K. F. Medzihradszky and P. R. de Montellano, J. Biol. Chem., 2005, 280, 22651–22663.
74. J. Bravo, I. Fita, J. C. Ferrer, W. Ens, A. Hillar, J. Switala and P. C. Loewen, Protein Sci., 1997, 6, 1016–1023.
75. A. Diaz, E. Horjales, E. Rudino-Pinera, R. Arreola and W. Hansberg, J. Mol. Biol., 2004, 342, 971–985.
76. C. Ostermeier, A. Harrenga, U. Ermler and H. Michel, Proc. Natl. Acad. Sci. U. S. A., 1997, 94, 10547–10553.
77. S. Yoshikawa, K. Shinzawa-Itoh, R. Nakashima, R. Yaono, E. Yamashita, N. Inoue, M. Yao, M. J. Fei, C. P. Libeu, T. Mizushima, H. Yamaguchi, T. Tomizaki and T. Tsukihara, Science, 1998, 280, 1723–1729.
78. J. Hemp, D. E. Robinson, K. B. Ganesan, T. J. Martinez, N. L. Kelleher and R. B. Gennis, Biochemistry, 2006, 45, 15405–15410.
79. T. Klabunde, C. Eicken, J. C. Sacchettini and B. Krebs, Nat. Struct. Biol., 1998, 5, 1084–1090.
80. M. E. Cuff, K. I. Miller, K. E. van Holde and W. A. Hendrickson, J. Mol. Biol., 1998, 278, 855–870.

---

Footnote

† This article is part of a themed issue of Molecular BioSystems on Post-translational modifications.

---