Inhibition by cyclopropylamine of the quinoprotein methylamine dehydrogenase is mechanism-based and causes covalent cross-linking of alpha and beta subunits

Cyclopropylamine acted as a mechanism-based inhibitor of the quinoprotein methylamine dehydrogenase from Paracoccus denitrificans. The protein-bound quinone cofactor of this enzyme was rapidly reduced by addition of a stoichiometric amount of cyclopropylamine, but this compound did not serve as a substrate for the enzyme in the steady-state kinetic assay. Time-dependent inactivation of the enzyme by cyclopropylamine was observed only in the presence of a reoxidant. Saturation behavior was observed, and values of KI of 3.9 microM and K(inact) of 1.7 min-1 were determined. Enzyme inactivation was irreversible, as no restoration of activity was evident after gel filtration of methylamine dehydrogenase which had been incubated with cyclopropylamine in the presence of a reoxidant. The inactivated enzyme exhibited an altered absorption spectrum. Electrophoretic analysis of inactivated methylamine dehydrogenase indicated that covalent cross-linking of the alpha and beta subunits of this alpha 2 beta 2 oligomeric enzyme had occurred and that the quinone cofactor had been modified. A mechanism for this inhibition is proposed which is based upon the data presented and is consistent with the available structural information on methylamine dehydrogenase.     

Factors affecting the production of pyrroloquinoline quinone by the methylotrophic bacterium W3A1

Two variants of the methylotrophic bacterium W3A1, designated W3A1-S (slimy) and W3A1-NS (nonslimy), were compared with respect to their ability to grow in batch culture on the C1 substrates methylamine, methanol, and trimethylamine. Substrate utilization, cell density, pH, cellular and soluble polysaccharide production, and concentrations of the enzymes methylamine dehydrogenase, trimethylamine dehydrogenase, and methanol dehydrogenase produced were measured as a function of growth. The ability of the two bacterial variants to excrete the redox cofactor pyrroloquinoline quinone into the growth medium was also investigated. The two variants were similar with respect to all properties measured, except that W3A1-S produced significantly more capsular polysaccharides than variant W3A1-NS. Pyrroloquinoline quinone was excreted when either variant was grown on any of the C1 substrates investigated but was maximally produced when the methylamine concentration was 0.45% (wt/vol). This cofactor is excreted only as bacterial growth enters the stationary phase, a time when the levels of trimethylamine dehydrogenase and the quinoproteins methanol dehydrogenase and methylamine dehydrogenase begin to decline. It is not known whether the pyrroloquinoline quinone found in the medium is made de novo for excretion, derived from the quinoprotein pool, or both. Pyrroloquinoline quinone excretion has been observed with other methylotrophs, but this is the first instance where the excretion was observed with substrates other than methanol.     

Factors affecting the production of pyrroloquinoline quinone by the methylotrophic bacterium W3A1.

Two variants of the methylotrophic bacterium W3A1, designated W3A1-S (slimy) and W3A1-NS (nonslimy), were compared with respect to their ability to grow in batch culture on the C1 substrates methylamine, methanol, and trimethylamine. Substrate utilization, cell density, pH, cellular and soluble polysaccharide production, and concentrations of the enzymes methylamine dehydrogenase, trimethylamine dehydrogenase, and methanol dehydrogenase produced were measured as a function of growth. The ability of the two bacterial variants to excrete the redox cofactor pyrroloquinoline quinone into the growth medium was also investigated. The two variants were similar with respect to all properties measured, except that W3A1-S produced significantly more capsular polysaccharides than variant W3A1-NS. Pyrroloquinoline quinone was excreted when either variant was grown on any of the C1 substrates investigated but was maximally produced when the methylamine concentration was 0.45% (wt/vol). This cofactor is excreted only as bacterial growth enters the stationary phase, a time when the levels of trimethylamine dehydrogenase and the quinoproteins methanol dehydrogenase and methylamine dehydrogenase begin to decline. It is not known whether the pyrroloquinoline quinone found in the medium is made de novo for excretion, derived from the quinoprotein pool, or both. Pyrroloquinoline quinone excretion has been observed with other methylotrophs, but this is the first instance where the excretion was observed with substrates other than methanol.     

Structural comparison of crystal and solution states of the 138 kDa complex of methylamine dehydrogenase and amicyanin from Paracoccus versutus

Methylamine can be used as the sole carbon source of certain methylotrophic bacteria. Methylamine dehydrogenase catalyzes the conversion of methylamine into formaldehyde and donates electrons to the electron transfer protein amicyanin. The crystal structure of the complex of methylamine dehydrogenase and amicyanin from Paracoccus versutus has been determined, and the rate of electron transfer from the tryptophan tryptophylquinone cofactor of methylamine dehydrogenase to the copper ion of amicyanin in solution has been determined. In the presence of monovalent ions, the rate of electron transfer from the methylamine-reduced TTQ is much higher than in their absence. In general, the kinetics are similar to those observed for the system from Paracoccus denitrificans. The complex in solution has been studied using nuclear magnetic resonance. Signals of perdeuterated, (15)N-enriched amicyanin bound to methylamine dehydrogenase are observed. Chemical shift perturbation analysis indicates that the dissociation rate constant is approximately 250 s(-1) and that amicyanin assumes a well-defined position in the complex in solution. The most affected residues are in the interface observed in the crystal structure, whereas smaller chemical shift changes extend to deep inside the protein. These perturbations can be correlated to small differences in the hydrogen bond network observed in the crystal structures of free and bound amicyanin. This study indicates that chemical shift changes can be used as reliable indicators of subtle structural changes even in a complex larger than 100 kDa.     

Steady-state kinetic analysis for the reaction of ammonium and alkylammonium ions with methylamine dehydrogenase from bacterium W3A1

The steady-state kinetic mechanism for the reaction of n-alkylamines and phenazine ethosulfate (PES) or phenazine methosulfate (PMS) with methylamine dehydrogenase from bacterium W3A1 is found to be of the ping-pong type. This conclusion is based on the observations that 1/v versus 1/[methylamine] or 1/[butylamine] plots, at various constant concentrations of an oxidizing substrate, and 1/v versus 1/[PES] or 1/[PMS] plots, at various constant concentrations of a reducing substrate, are parallel. Additionally, the values of kcat/Km for four n-alkylamines are identical when PES is the oxidizing substrate, as were the kcat/Km values for four reoxidizing substrates when methylamine was the reducing substrate. Last, analysis of steady-state kinetic data obtained when methylamine and propylamine are presented to the enzyme simultaneously and PES and PMS are used simultaneously also supports the involvement of a ping-pong mechanism. The enzymic reaction with either methylamine or PES is dependent on the ionic strength, and the data indicate that each interacts with an anionic site on methylamine dehydrogenase. The presence of ammonium ion at low concentration activates the enzyme, but at high concentration this ion is a competitive inhibitor in the reaction involving methylamine and the enzyme. A complete steady-state mechanism describing these ammonia effects is presented and is discussed in light of the nature of the pyrroloquinoline quinone cofactor covalently bound to the enzyme.     

Crystallographic investigations of the tryptophan-derived cofactor in the quinoprotein methylamine dehydrogenase

A model of tryptophan tryptophylquinone (TTQ), recently proposed by McIntire et al. (Science (1991) 252, 817-824) to be the prosthetic group of the quinoprotein methylamine dehydrogenase, has been compared with electron density maps of this dehydrogenase from Thiobacillus versutus and Paracoccus denitrificans. The comparison shows that the TTQ model can be neatly accommodated, providing strong supportive evidence that TTQ is indeed the cofactor for this group of quinoproteins.     

Cofactor-directed inactivation by nucleophilic amines of the quinoprotein methylamine dehydrogenase from Paracoccus denitrificans.

Phenylhydrazine, semicarbazide, aminoguanidine, hydrazine, and hydroxylamine each irreversibly inactivated methylamine dehydrogenase from Paracoccus denitrificans and caused changes in the absorbance spectrum of the protein-bound tryptophan tryptophylquinone [TTQ] prosthetic group. Different spectral perturbations were observed on reaction with each of these inactivators. In each case a stoichiometry of 2 mol per mol of enzyme (1:1 per cofactor) was required to observe complete modification of the absorbance spectrum. Identical changes were observed in the presence and absence of oxygen. The reactions of hydrazine and hydroxylamine were very rapid, with stoichiometric inactivation occurring in less than 30 s. Inactivation by phenylhydrazine and semicarbazide exhibited apparent bimolecular kinetics and second order rate constants for inactivation, respectively, of 25 min-1 mM-1 and 39 min-1 mM-1. In contrast, inactivation by aminoguanidine exhibited saturation behavior and kinetic parameters of KI = 2.5 mM and kinact = 0.5 min-1 were obtained. Ammonium salts did not inactivate the enzyme, but were reversible competitive inhibitors with respect to methylamine. A Ki of 20 mM was obtained for ammonium chloride. A mechanism for the reactions of these compounds with the TTQ cofactor of methylamine dehydrogenase is proposed, and the relationship of these data to the mechanisms of interaction of these compounds with o-quinones and other quinoproteins which possess TTQ and other quinone cofactors is discussed.     

Active site aspartate residues are critical for tryptophan tryptophylquinone biogenesis in methylamine dehydrogenase

The biosynthesis of methylamine dehydrogenase (MADH) requires formation of six intrasubunit disulfide bonds, incorporation of two oxygens into residue betaTrp57 and covalent cross-linking of betaTrp57 to betaTrp108 to form the protein-derived cofactor tryptophan tryptophylquinone (TTQ). Residues betaAsp76 and betaAsp32 are located in close proximity to the quinone oxygens of TTQ in the enzyme active site. These residues are structurally conserved in quinohemoprotein amine dehydrogenase, which possesses a cysteine tryptophylquinone cofactor. Relatively conservative betaD76N and betaD32N mutations resulted in very low levels of MADH expression. Analysis of the isolated proteins by mass spectrometry revealed that each mutation affected TTQ biogenesis. betaD76N MADH possessed the six disulfides but had no oxygen incorporated into betaTrp57 and was completely inactive. The betaD32N MADH preparation contained a major species with six disulfides but no oxygen incorporated into betaTrp57 and a minor species with both oxygens incorporated, which was active. The steady-state kinetic parameters for the betaD32N mutant were significantly altered by the mutation and exhibited a 1000-fold increase in the Km value for methylamine. These results have allowed us to more clearly define the sequence of events that lead to TTQ biogenesis and to define novel roles for aspartate residues in the biogenesis of a protein-derived cofactor.     

pH-dependent semiquinone formation by methylamine dehydrogenase from Paracoccus denitrificans. Evidence for intermolecular electron transfer between quinone cofactors

The quinonoid confactors of Paracoccus denitrificans methylamine dehydrogenase exhibited a pH-dependent redistribution of electrons from the 50% reduced plus 50% oxidized to the 100% semiquinone redox form. This phenomenon was only observed at pH values greater than 7.5. The semiquinone was not readily reduced by addition of methylamine, consistent with the view that this substrate donates two electrons at a time to each cofactor during catalysis. Once formed at pH 9.0, no change in redox state from 100% semiquinone was observed when the pH was shifted to 7.5, suggesting that the requirement of high pH was for formation and not stability of the semiquinone. The rate of semiquinone formation exhibited a first-order dependence on the concentration of methylamine dehydrogenase, indicating that this phenomenon was a bimolecular process involving intermolecular electron transfer between reduced and oxidized cofactors. The rate of semiquinone formation decreased with decreasing ionic strength, suggesting a role for hydrophobic interactions in facilitating electron transfer between methylamine dehydrogenase molecules. Methylamine dehydrogenase was covalently modified with norleucine methyl ester in the presence of 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide (EDC). This modification did not affect the catalytic activity of the enzyme but greatly inhibited the intermolecular redistribution of electrons at high pH. This modification also prevented subsequent cross-linking by EDC of the large subunit of methylamine dehydrogenase to amicyanin, the natural electron acceptor for this enzyme.(ABSTRACT TRUNCATED AT 250 WORDS)     

Resonance Raman spectroscopy of methylamine dehydrogenase from bacterium W3A1

Resonance Raman spectroscopy has been used to probe the structure of the covalently bound quinone cofactor in methylamine dehydrogenase from the bacterium W3A1. Spectra were obtained on the phenylhydrazine and 2-pyridylhydrazine derivatives of the native enzyme, on the quinone-containing subunit labeled with phenylhydrazine, and on an active-site peptide also labeled with phenylhydrazine. Comparisons of these spectra to the corresponding spectra of copper-containing amine oxidase derivatives indicate that the quinones in these two classes of quinoproteins are not identical. The resonance Raman spectra of the native enzyme and small subunit have also been measured. 16O/18O exchange permitted the carbonyl modes of the quinone to be identified in the resonance Raman spectrum of oxidized methylamine dehydrogenase: a band at 1614 cm-1, together with a shoulder at 1630 cm-1, are assigned as modes containing substantial C = O stretching character. D2O/H2O exchange has pronounced effects on the resonance Raman spectrum of the oxidized enzyme, suggesting that the quinone may have numerous hydrogen bonds to the protein or that it is unusually sensitive to the local environment. Resonance Raman spectra of the isolated small subunit, and its phenylhydrazine derivative, are considerably different from the corresponding spectra of the intact protein. An attractive explanation for these observations is that the quinone cofactor in methylamine dehydrogenase from W3A1 is located at the interface between the large and small subunits, as found for the enzyme from Thiobacillus versutus [Vellieux, F. M. D., Huitema, F., Groendijk, H., Kalk, K. H., Frank, J. Jzn., Jongejan, J. A., & Duine, J. A. (1989) EMBO J. 8, 2171-2178].(ABSTRACT TRUNCATED AT 250 WORDS)     

Apparent oxygen-dependent inhibition by superoxide dismutase of the quinoprotein methanol dehydrogenase.

Methanol dehydrogenase activity, when assayed with phenazine ethosulfate (PES) as an electron acceptor, was inhibited by superoxide dismutase (SOD) and by Mn2+ only under aerobic conditions. Catalase, formate, and other divalent cations did not inhibit the enzyme. The enzyme also exhibited significantly higher levels of activity when assayed with PES under anaerobic conditions relative to aerobic conditions. The oxygen- and superoxide-dependent effects on methanol dehydrogenase were not observed when either Wurster's Blue or cytochrome c-55li was used as an electron acceptor. Another quinoprotein, methylamine dehydrogenase, which possesses tryptophan tryptophylquinone (TTQ) rather than pyrroloquinoline quinone (PQQ) as a prosthetic group, was not inhibited by SOD or Mn2+ when assayed with PES as an electron acceptor. Spectroscopic analysis of methanol dehydrogenase provided no evidence for any oxygen- or superoxide-dependent changes in the redox state of the enzyme-bound PQQ cofactor of methanol dehydrogenase. To explain these data, a model is presented in which this cofactor reacts reversibly with oxygen and superoxide, and in which oxygen is able to compete with PES as an electron acceptor for the reduced species.     

Oxidation of C1 compounds by Pseudomonas sp. MS

Pseudomonas sp. MS is capable of growth on a number of compounds containing only C₁ groups. They include trimethylsulphonium salts, methylamine, dimethylamine and trimethylamine. Although formaldehyde and formate will not support growth they are rapidly oxidized by intact cells. Methanol neither supports growth nor is oxidized. A particulate fraction of the cell oxidizes methylamine to carbon dioxide in the absence of any external electron acceptor. Formaldehyde and formate are more slowly oxidized to carbon dioxide by the particulate fraction, although they do not appear to be free intermediates in the oxidation of methylamine. Soluble NAD-linked formaldehyde dehydrogenase and formate dehydrogenase are also present. The particulate methylamine oxidase is induced by growth on methylamine, dimethylamine and trimethylamine, whereas the soluble formaldehyde dehydrogenase and formate dehydrogenase are induced by trimethylsulphonium nitrate as well as the aforementioned amines.     

The role of PQQ and quinoproteins in methylotrophic bacteria

Abstract Quinoprotein dehydrogenases play a non-exclusive role in the dissimilation of C₁ compounds. Methanol and methylamine oxidation occur by covalent catalysis while the reduction equivalents are transferred to the respiratory chain in one-electron steps. Cytochrome c _L is an excellent electron acceptor for methanol dehydrogenase at pH 7.0 and a bad one at pH 9.0. Efficient methanol oxidation (with NH₃ as activator) occurs at pH 9.0, but (due to the failure of NH₃) not at pH 7.0. Since stimulation occurred at the latter condition with a compound prepared from Hyphomicrobium X, most probably methanol oxidation in vivo requires the presence of a natural activator. The finding of pro-PQQ in methylamine dehydrogenase implicates that certain quinoproteins may have a modified tyrosine as cofactor. This type of quinoprotein is involved in assimilation routes which also occur in methylotrophs. l -Tyrosine and l -glutamate are the precursors of PQQ biosynthesis. Free intermediates in the route of biosynthesis have not been found. Most probably the whole process occurs on a protein matrix. In view of the significant amounts found in their culture fluid, methylotrophic bacteria seem particularly well suited for the fermentative production of PQQ.     

An unexpected role for the active site base in cofactor orientation and flexibility in the copper amine oxidase from Hansenula polymorpha.

The role of the active site aspartate base in the aminotransferase mechanism of the copper amine oxidase from the yeast Hansenula polymorpha has been probed by site-directed mutagenesis. The D319E mutant catalyzes the oxidation of methylamine and phenethylamine, but not that of benzylamine. kcat/Km for methylamine is found to be 80-fold reduced compared to that of the wild type. Viscosogen and substrate and solvent deuteration have no effect on this parameter for D319E, which is suggestive of limitation of kcat/Km by a conformational change. This conformational change is proposed to be the movement of the cofactor into a productive orientation upon the binding of substrate. In the absence of substrate, a flipped cofactor orientation is likely, on the basis of resonance Raman evidence that the C5 carbonyl of the cofactor is less solvent accessible than the C3 hydrogen. kcat for D319E methylamine oxidase is reduced 200-fold compared to that of the wild type and is unaffected by substrate deuteration, but displays a substantial solvent isotope effect. A 428 nm absorbance is evident under conditions of saturating methylamine and oxygen with D319E. The D319N mutant is observed to produce a similar absorbance at 430 nm when treated with ammonia despite the fact that this mutant has no amine oxidase activity. Resonance Raman spectroscopy indicates the formation of a covalent ammonia adduct and identifies it as the deprotonated iminoquinone. In contrast, when the D319E mutant is reacted with ammonia, it gives predominantly a 340-350 nm species. This absorbance is ascribed to a localization of the cofactor oxyanion induced by binding of the cation at the active site and not to covalent adduct formation. Resonance Raman spectroscopic examination of the steady state species of D319E methylamine oxidation, in combination with the kinetic data, indicates that the 428 nm species is the deprotonated iminoquinone produced upon reoxidation of the reduced cofactor. A model is proposed in which a central role of the active site base is to position the free cofactor and several enzyme intermediates for optimal activity.     

Three-dimensional structure of the quinoprotein methylamine dehydrogenase from Paracoccus denitrificans determined by molecular replacement at 2.8 A resolution.

The three-dimensional structure of the quinoprotein methylamine dehydrogenase from Paracoccus dentrificans (PD-MADH) has been determined at 2.8 A resolution by the molecular replacement method combined with map averaging procedures, using data collected from an area detector. The structure of methylamine dehydrogenase from Thio-bacillus versutus, which contains an "X-ray" sequence, was used as the starting search model. MADH consists of 2 heavy (H) and 2 light (L) subunits related by a molecular 2-fold axis. The H subunit is folded into seven four-stranded beta segments, forming a disk-shaped structure, arranged with pseudo-7-fold symmetry. A 31-residue elongated tail exists at the N-terminus of the H subunit in MADH from T. versutus but is partially digested in this crystal form of MADH from P. denitrificans, leaving the H subunit about 18 residues shorter. Each L subunit contains 127 residues arranged into 10 beta-strands connected by turns. The active site of the enzyme is located in the L subunit and is accessible via a hydrophobic channel between the H and L subunits. The redox cofactor of MADH, tryptophan tryptophylquinone is highly unusual. It is formed from two covalently linked tryptophan side chains at positions 57 and 107 of the L subunit, one of which contains an orthoquinone.     

Redox properties of quinohemoprotein amine dehydrogenase from Paracoccus denitrificans

Paracoccus denitrificans produces two primary enzymes for the amine oxidation, tryptophan-tryptophylquinone (TTQ)-containing methylamine dehydrogenase (MADH) and quinohemoprotein amine dehydrogenase (QH-AmDH). QH-AmDH has a novel cofactor, cysteine tryptophylquinone (CTQ) and two hemes c. In this work, the redox potentials of three redox centers in QH-AmDH were determined by a mediator-assisted continuous-flow column electrolytic spectroelectrochemical technique. Kinetics of the electron transfer from QH-AmDH to three kinds of metalloproteins, amicyanin, cytochrome c(550), and horse heart cytochrome c were examined on the basis of the theory of mediated-bioelectrocatalysis. All these metalloproteins work as a good electron acceptor of QH-AmDH and donate the electron to the terminal oxidase of P. denitrificans, which was revealed by reconstitution of the respiratory chain. These properties are in marked contrast with those of MADH, which shows high specificity to amicyanin. These electron transfer kinetics are discussed in terms of thermodynamics and structural property.     

Characterization of the tryptophan-derived quinone cofactor of methylamine dehydrogenase by resonance Raman spectroscopy

The resonance Raman (RR) spectrum of oxidized methylamine dehydrogenase (MADHOX) exhibits a set of C-H, C-C, C = C, and C = O vibrational modes between 900 and 1700 cm-1 that are characteristic of the quinone moiety of the tryptophan tryptophlyquinone (TTQ) cofactor. The close similarity of the RR spectra for MADHs from Paracoccus denitrificans (Pd), Thiobacillus versutus (Tv), and bacterium W3A1 proves that the same cofactor is present in all three proteins. The MADHs from Pd and Tv have a v(C = O) mode at approximately 1625 cm-1 that shifts approximately 20 cm-1 upon 18O substitution of one of the carbonyl oxygens and is assigned to the in-phase symmetric stretch of the two C = O groups. The semiquinone form of Pd MADH has its own characteristic RR spectrum with altered peak frequencies and intensities as well as a decrease in the total number of peaks. The hydroxide and ammonia adducts of MADHOX produce RR spectra similar to that of the semiquinone. The spectral changes in all three cases are interpreted as being due to reduced conjugation of the cofactor. The ammonia adduct is formulated as a carbinolamine, a likely intermediate in the enzymatic mechanism. In contrast, formation of the electron-transfer complex between amicyanin and MADHOX has no effect on the vibrational frequencies (and, hence, structure) of either the MADH quinone or the amicyanin blue copper site. The behavior of the TTQ cofactors of Pd and Tv MADHs are very similar to one another and somewhat different from W3A1 MADH, particularly with regard to adduct formation and ability to undergo isotope exchange with solvent. These differences are ascribed to the cofactor environments within the proteins rather than to the structure of the cofactor itself.     

A spectrophotometric method for determining the oxidative deamination of methylamine by the amine oxidases

Previously published studies on the oxidative deamination of methylamine by the amine oxidases have determined the formation of radioactively labeled formaldehyde from [(14)C]methylamine. The present work describes a coupled spectrophotometric assay, using formaldehyde dehydrogenase, for the continuous determination of the oxidative deamination of methylamine by semicarbazide-sensitive amine oxidase (SSAO) and its potential use for determining methylamine concentrations in plasma. In this assay, the formaldehyde produced by methylamine deamination is further oxidized to formate, with the reduction of NAD(+), by formaldehyde dehydrogenase. The NADH generated is monitored continuously at 340 nm. Interference from the presence of a rotenone-insensitive NADH oxidase activity in crude tissue homogenates and microsomal fractions can be minimized by pretreating samples with Triton X-100 or substituting NAD(+) by APAD(+) in the coupled assay. This relatively inexpensive and reproducible assay procedure avoids the use of radioactively labeled material.     

Understanding quinone cofactor biogenesis in methylamine dehydrogenase through novel cofactor generation

Cofactors made from constitutive amino acids in proteins are now known to be relatively common. A number of these involve the generation of quinone cofactors, such as topaquinone in the copper-containing amine oxidases, and lysine tyrosylquinone in lysyl oxidase. The biogenesis of the quinone cofactor tryptophan tryptophylquinone (TTQ) in methylamine dehydrogenase (MADH) involves the post-translational modification of two constitutive Trp residues (Trp(beta)(57) and Trp(beta)(108) in Paracoccus denitrificans MADH). The modifications for generating TTQ are the addition of two oxygens to the indole ring of Trp(beta)(57) and the formation of a covalent cross-link between Cepsilon3 of Trp(beta)(57) and Cdelta1 of Trp(beta)(108). The order in which these events occur is unknown. To investigate the role Trp(beta)(108) may play in this process, this residue was mutated to both a His (betaW108H) and a Cys (betaW108C) residue. For each mutant, the majority of the protein that was isolated was inactive and exhibited weaker subunit-subunit interactions than native MADH. Analysis by mass spectrometry suggested that the inactive protein was a biosynthetic intermediate with only one oxygen atom incorporated into Trp(beta)(57) and no cross-link with residue beta108. However, in each mutant preparation, a small percentage of the mutant enzyme was active and appears to possess a functional tryptophylquinone cofactor. In the case of betaW108C, this cofactor may be identical to cysteine tryptophylquinone, recently described in the bacterial quinohemoprotein amine dehydrogenase. In betaW108H, the active cofactor is presumably a histidine tryptophylquinone, which has not been previously described, and represents the synthesis of a novel quinone protein cofactor.