Half-of-the-sites reactivity of bovine serum amine oxidase. Reactivity and chemical identity of the second site

Maize polyamine oxidase in the presence of spermine/spermidine induces the apoptosis of LoVo human colon adenocarcinoma cells

Amine oxidases, which contribute to the regulation of polyamine levels, catalyze the oxidative deamination of polyamines to generate H2O2 and aldehyde(s). In this study, and at least to the best of our knowledge, maize polyamine oxidase (ZmPAO) was used for the first time with the aim of identifying a novel strategy for cancer therapy. The cytotoxicity and the mechanisms of cell death induced by the enzymatic oxidation products of polyamine generated by ZmPAO were investigated. Exogenous spermine and ZmPAO treatment decreased cell viability in a spermine dose‑ and time‑dependent manner, particularly, the viability of the multidrug‑resistant (MDR) colon adenocarcinoma cells, LoVo DX, when compared with drug‑sensitive ones (LoVo WT). Further analyses revealed that H2O2 derived from spermine was mainly responsible for the cytotoxicity. Flow cytometric analysis revealed that treatment with ZmPAO and spermine increased the apoptotic population of LoVo WT and LoVo DX cells. In addition, we found that treatment with ZmPAO and spermine markedly reduced mitochondrial membrane potential in the LoVo DX cells, in agreement with the results of cell viability and apoptosis assays. Transmission electron microscopic observations supported the involvement of mitochondrial depolarization in the apoptotic process. Therefore, the dysregulation of polyamine metabolism in tumor cells may be a potential therapeutic target. In addition, the development of MDR tumor cells is recognized as a major obstacle in cancer therapy. Therefore, the design of a novel therapeutic strategy based on the use of this combination may be taken into account, making this approach attractive mainly in treating MDR cancer patients.

Probing the molecular mechanisms in copper amine oxidases by generating heterodimers

For some homodimeric copper amine oxidases (CuAO), there is suggestive evidence of differential activity at the two active sites implying potential cooperativity between the two monomers. To examine this phenomenon for the Arthrobacter globiformis CuAO (AGAO), we purified a heterodimeric form of the enzyme for comparison with the homodimer. The heterodimer comprises an active wild-type monomer and an inactive monomer in which an active-site tyrosine is mutated to phenylalanine (Y382F). This mutation prevents the formation of the trihydroxyphenylalanine quinone (TPQ) cofactor. A pETDuet vector and a dual fusion tag strategy was used to purify heterodimers (WT/Y382F) from homodimers. Purity was confirmed by western blot and native PAGE analyses. Spectral and kinetic studies support the view that whether there are one or two functional monomers in the dimer, the properties of each functional monomer are the same, thus indicating no communication between the active sites in this bacterial enzyme.

The role of protein crystallography in defining the mechanisms of biogenesis and catalysis in copper amine oxidase

Copper amine oxidases (CAOs) are a ubiquitous group of enzymes that catalyze the conversion of primary amines to aldehydes coupled to the reduction of O₂ to H₂O₂. These enzymes utilize a wide range of substrates from methylamine to polypeptides. Changes in CAO activity are correlated with a variety of human diseases, including diabetes mellitus, Alzheimer’s disease, and inflammatory disorders. CAOs contain a cofactor, 2,4,5-trihydroxyphenylalanine quinone (TPQ), that is required for catalytic activity and synthesized through the post-translational modification of a tyrosine residue within the CAO polypeptide. TPQ generation is a self-processing event only requiring the addition of oxygen and Cu(II) to the apoCAO. Thus, the CAO active site supports two very different reactions: TPQ synthesis, and the two electron oxidation of primary amines. Crystal structures are available from bacterial through to human sources, and have given insight into substrate preference, stereospecificity, and structural changes during biogenesis and catalysis. In particular both these processes have been studied in crystallo through the addition of native substrates. These latter studies enable intermediates during physiological turnover to be directly visualized, and demonstrate the power of this relatively recent development in protein crystallography.

Kinetic and structural analysis of substrate specificity in two copper amine oxidases from Hansenula polymorpha

The structural underpinnings of enzyme substrate specificity are investigated in a pair of copper amine oxidases (CAOs) from Hansenula polymorpha (HPAO-1 and HPAO-2). The X-ray crystal structure (to 2.0 A resolution) and steady state kinetic data of the second copper amine oxidase (HPAO-2) are presented for comparison to those of HPAO-1. Despite 34% sequence identity and superimposable active site residues implicated in catalysis, the enzymes vary considerably in their substrate entry channel. The previously studied CAO, HPAO-1, has a narrow substrate channel. In contrast, HPAO-2 has a wide funnel-shaped substrate channel, which also contains a side chamber. In addition, there are a number of amino acid changes within the channels of HPAO-2 and HPAO-1 that may sterically impact the ability of substrates to form covalent Schiff base catalytic intermediates and to initiate chemistry. These differences can partially explain the greatly different substrate specificities as characterized by k(cat)/K(m) value differences. In HPAO-1, the k(cat)/K(m) for methylamine is 330-fold greater than for benzylamine, whereas in HPAO-2, it is benzylamine that is the better substrate by 750-fold. In HPAO-2, an inflated (D)k(cat)/K(m)(methylamine) in relation to (D)k(cat)/K(m)(benzylamine) indicates that proton abstraction has been impeded more than substrate release. In HPAO-1, (D)k(cat)/K(m)(S) changes little with the slow substrate and indicates a similar increase in the energy barriers that control both substrate binding and subsequent catalysis. In neither case is k(cat)/K(m) for the second substrate, O(2), significantly altered. These results reinforce the modular nature of the active sites of CAOs and show that multiple factors contribute to substrate specificity and catalytic efficiency. In HPAO-1, the enzyme with the smaller substrate binding pocket, both initial substrate binding and proton loss are affected by an increase in substrate size, while in HPAO-2, the enzyme with the larger substrate binding pocket, the rate of proton loss is differentially affected when a phenyl substituent in the substrate is reduced to the size of a methyl group.

Potential anticancer application of polyamine oxidation products formed by amine oxidase: a new therapeutic approach

The polyamines spermine, spermidine and putrescine are ubiquitous cell components. These molecules are substrates of a class of enzymes that includes monoamine oxidases, diamine oxidases, polyamine oxidases and copper-containing amine oxidases. Amine oxidases are important because they contribute to regulate levels of mono- and polyamines. In tumors, polyamines and amine oxidases are increased as compared to normal tissues. Cytotoxicity induced by bovine serum amine oxidase (BSAO) and spermine is attributed to H(2)O(2) and aldehydes produced by the reaction. This study demonstrated that multidrug-resistant (MDR) cancer cells (colon adenocarcinoma and melanoma) are significantly more sensitive than the corresponding wild-type (WT) ones to H(2)O(2) and aldehydes, the products of BSAO-catalyzed oxidation of spermine. Transmission electron microscopy (TEM) observations showed major ultrastructural alterations of the mitochondria. These were more pronounced in MDR than in WT cells. Increasing the incubation temperature from 37 to 42 degrees Celsius enhances cytotoxicity in cells exposed to spermine metabolites. The combination BSAO/spermine prevents tumor growth, particularly well if the enzyme has been conjugated to a biocompatible hydrogel polymers. Since both wild-type and MDR cancer cells after pre-treatment with MDL 72527, a lysosomotropic compound, are sensitized to subsequent exposure to BSAO/spermine, it is conceivable that combined treatment with a lysosomotropic compound and BSAO/spermine would be effective against tumor cells. It is of interest to search for such novel compounds, which might be promising for application in a therapeutic setting.

Hydrazine and amphetamine binding to amine oxidases: old drugs with new prospects

Tranylcypromine (TCP), an amphetamine, is a reversible inhibitor of copper-containing amine oxidases. We have solved the structure of the complex of TCP with the amine oxidase from E. coli (ECAO) and shown that only the (+)-enantiomer of TCP binds. Kinetic studies on 2-phenylethylamine and TCP binding to wild-type ECAO and mutational variants fully support the model in which binding of the protonated amine is the first step in the catalytic cycle. Hydrazines are irreversible inhibitors of copper-containing amine oxidases. Binding of hydrazines leads to an adduct ("Adduct 1") with a chromophore at 430 nm which converts at higher pH to another adduct ("Adduct 2") with a chromophore at 520 nm. We have determined the structures of Adduct 1 and 2 for 2-hydrazinopyridine reacted with ECAO. It has been found that Adduct 1 corresponds to the hydrazone and azo tautomers whilst Adduct 2 corresponds to the azo tautomer coordinated to the active site copper. The implications of these results in developing more specific drugs are discussed.

Amine oxidases in apoptosis and cancer

Amine oxidases, the major enzymes of biogenic amines metabolism, are considered to be biological regulators, especially for cell growth and differentiation. A primary involvement of amine oxidases in cancer growth inhibition and progression, especially by means of aldehydes, H(2)O(2) and other reactive oxygen species, the amine oxidase-mediated products of biogenic amines oxidation, has been demonstrated. Amine oxidases are involved in cancer growth inhibition because of the higher content in tumour cells of biogenic amines in comparison to normal cells. The cytotoxic effect can be explained by a damage to cell membranes and/or nuclei or, indirectly, through modulation of membrane permeability transition and therefore apoptosis. The oxidation products of biogenic amines appears to be also carcinogenic, while acrolein, produced from the oxidation of spermine and spermidine, should be a key compound both carcinogenic and cytotoxic. The cancer inhibition/promotion effect of amine oxidases could be explained by taking into consideration the full pattern of the enzyme content of the cell. The balance of amine oxidases and antioxidant enzymes appear to be a crucial point for cancer inhibition or progression. A long lasting imbalance of these enzymes appears to be carcinogenic, while, for a short time, amine oxidases are cytotoxic for cancer cells.

Crystal Structure of Amine Oxidase from Bovine Serum

Copper-containing amine oxidase extracted from bovine serum (BSAO) was crystallized and its three-dimensional structure at 2.37A resolution is described. The biological unit of BSAO is a homodimer, formed by two monomers related to each other by a non-crystallographic 2-fold axis. Each monomer is composed of three domains, similar to those of other amine oxidases from lower species. The two monomers are structurally equivalent, despite some minor differences at the two active sites. A large funnel allows access of substrates to the active-site; another cavity, accessible to the solvent, is also present between the two monomers; this second cavity could allow the entrance of molecular oxygen necessary for the oxidative reaction. Some sugar residues, bound to Asn, were still present and visible in the electron density map, in spite of the exhaustive deglycosylation necessary to grow the crystals. The comparison of the BSAO structure with those of other resolved AO structures shows strong dissimilarities in the architecture and charge distribution of the cavities leading to the active-site, possibly explaining the differences in substrate specificity.

Is the catalytic mechanism of bacteria, plant, and mammal copper-TPQ amine oxidases identical?

This short review is mostly concerned with the work carried out in Rome on the copper amine oxidase from bovine serum (BSAO). The first target was the copper oxidation state and its relationship with the organic cofactor. It was found that copper is not reduced on reaction with amines under anaerobic conditions or along the catalytic cycle and that it is not within bonding distance of the quinone cofactor. The copper stability in the oxidised state was supported by BSAO ability to oxidise benzylhydrazine, a slow substrate, in the presence of N,N-diethyldithiocarbamate (DDC) and by the substantial catalytic activity of Co(2+)-substituted BSAO. Parallel work established that only one subunit of the dimeric enzyme readily binds reagents of the carbonyl group. Flexible hydrazides with a long aromatic tail were found to be highly specific inhibitors, suggesting the presence of an extended hydrophobic region at the catalytic site. A study by stopped-flow transient spectroscopy and steady state kinetics led to the formulation of a simplified, yet complete and consistent, catalytic mechanism for BSAO that was compared with that available for lentil seedling amine oxidase (LSAO). As in other copper amine oxidases, BSAO is inactivated by H(2)O(2) produced in the catalytic reaction, while the cofactor is stabilised in its reduced state. A conserved tyrosine hydrogen-bonded to the cofactor might be oxidised.

Biochemical and toxicological evaluation of agent-cofactor reactivity as a mechanism of action for osteolathyrism

In vitro reactivity for each of four osteolathyrogens with a model compound for the lysyl oxidase (LO) cofactor was evaluated and coupled with mixture toxicity testing to evaluate agent-cofactor reactivity as a potential mechanism of action for osteolathyrism. Reactivity of the model cofactor (mLTQ: 4-butylamino-5-methyl-o-quinone), with each of two ureides, semicarbazide (SC) and thiosemicarbazide (TSC), and each of two aminonitriles, aminoacetonitrile (AAN) and beta-aminopropionitrile (betaAPN), was assessed using UV-vis spectrophotometry; both in the absence and presence of Cu(II)-bipyridine (bipy) complex. Two sets of mixture toxicity experiments were conducted using a frog embryo assay that assessed the incidence of osteolathyrism in the notochord of tadpoles after 96-h exposure. The resulting concentration-response curves for each set were evaluated (chi(2) goodness-of-fit test) against theoretical curves for two combined effects models: dose-addition and independence, to determine the combined effect of each osteolathyrogen combination. The agents SC, TSC and AAN each showed rapid, irreversible reactivity with mLTQ, both in the absence and presence of Cu(II)-bipy complex, as indicated by bleaching of the mLTQ peak (504 nm) and formation of an adduct at 350 nm. betaAPN showed no apparent reactivity in the absence of prolonged incubation with mLTQ, whether Cu(II)-bipy complex was present or not. After prolonged incubation (24-144 h) a new peak formed at 350 nm, suggesting that betaAPN reacts weakly with the cofactor, but in a manner different from the other agents examined. The toxicity tests indicated a dose-additive combined effect for the SC:TSC, AAN:SC and AAN:SC:TSC mixtures (0.1<P<0.9; with slope and EC(50) additivity quotient values between 0.93-1.14), suggesting these agents induce osteolathyrism via a common mechanism. The betaAPN:SC and betaAPN: SC:TSC mixtures showed a greater-than dose-additive combined effect (P<0.001; slope and EC(50) additivity quotient values from 0.74 to 0.89), suggesting a different or additional mechanism of osteolathyrism for betaAPN. Taken together the results indicate that SC, TSC and AAN induce osteolathyrism primarily, if not completely, by binding to the LO cofactor, thereby inhibiting proper connective tissue fiber cross-linking. While betaAPN also has the potential to react with the cofactor, the nature of the reactivity observed suggests that betaAPN-cofactor binding is, at best, a secondary mechanism for induction of osteolathyrism in vivo.

Probing the catalytic mechanism of Escherichia coli amine oxidase using mutational variants and a reversible inhibitor as a substrate analogue

Copper amine oxidases are homodimeric enzymes containing one Cu(2+) ion and one 2,4,5-trihydroxyphenylalanine quinone (TPQ) per monomer. Previous studies with the copper amine oxidase from Escherichia coli (ECAO) have elucidated the structure of the active site and established the importance in catalysis of an active-site base, Asp-383. To explore the early interactions of substrate with enzyme, we have used tranylcypromine (TCP), a fully reversible competitive inhibitor, with wild-type ECAO and with the active-site base variants D383E and D383N. The formation of an adduct, analogous to the substrate Schiff base, between TCP and the TPQ cofactor in the active site of wild-type ECAO and in the D383E and D383N variants has been investigated over the pH range 5.5-9.4. For the wild-type enzyme, the plot of the binding constant for adduct formation (K(b)) against pH is bell-shaped, indicating two pK(a)s of 5.8 and approximately 8, consistent with the preferred reaction partners being the unprotonated active-site base and the protonated TCP. For the D383N variant, the reaction pathway involving unprotonated base and protonated TCP cannot occur, and binding must follow a less favoured pathway with unprotonated TCP as reactant. Surprisingly, for the D383E variant, the K(b) versus pH behaviour is qualitatively similar to that of D383N, supporting a reaction pathway involving unprotonated TCP. The TCP binding data are consistent with substrate binding data for the wild type and the D383E variant using steady-state kinetics. The results provide strong support for a protonated amine being the preferred substrate for the wild-type enzyme, and emphasize the importance of the active-site base, Asp-383, in the primary binding event.

Amine oxidase from lentil seedlings: energetic domains and effect of temperature on activity

Copper/TPQ amine oxidases from mammalian and plant sources have shown many differences in substrate specificity and molecular properties. In this work the activity of lentil seedling amine oxidase was followed at various temperatures in 100 mM potassium phosphate buffer, pH 7, using benzylamine as substrate. The discontinuous Arrhenius plot of lentil amine oxidase showed two distinct phases with a jump between them. Thermal denaturation of the enzyme, using differential scanning calorimetry under the same experimental conditions, showed a transition at the same temperature ranges in the absence of substrate, indicating the occurrence of conformational changes, with an enthalpy change of about 175.9 kJ/mole. The temperature-induced changes of the activity of lentil amine oxidase are compared with those of bovine serum amine oxidase (taken from the literature).

The oxidation and reduction reactions of bovine serum amine oxidase. A kinetic study

The presteady-state and steady-state kinetics of bovine serum amine oxidase (BSAO) were analyzed by stopped-flow transient spectroscopy. A simplified model of the catalytic cycle was found to describe the experimental data and the rate constants of the individual steps were used to calculate Michaelis parameters that agree with the direct determinations. In spite of many studies on selected reactions from the catalytic cycle, this is amongst the first efforts to provide a comprehensive kinetic description of the reactions of BSAO, whose results can be compared with the steady-state parameters. The reoxidation reaction by dioxygen is more complex than previously thought, in agreement with a recent report [Su, Q. & Klinman, J.P. (1998) Biochemistry 37, 12513-12525], and occurs in at least two steps whose rate constants, previously undetermined, have been measured. The reaction of the oxidized enzyme with the amine substrate is poorly determined in this type of experiment, thus irreversible combination with aromatic hydrazine inhibitors was used as a model system, demonstrating that the mechanism and rate constants of their reaction is fully compatible with an accurate description of the catalytic cycle with the physiological substrate. These results constitute a simplified, yet complete and consistent, description of the catalytic cycle and offer an interesting comparison with those obtained on plant amine oxidases; two steps of the catalytic cycle are significantly slower in BSAO than in pea seedling or lentil seedling amine oxidases, namely the reoxidation and the trans-iminative proton abstraction occurring in the enzyme-substrate complex. The former difference is rationalized as being due to the low to zero concentration of the semiquinolamine-radical intermediate, while the latter is less easily interpreted.

Modulation of bovine serum amine oxidase activity by hydrogen peroxide

Bovine serum amine oxidase (BSAO), reduced by excess amine under limited turnover conditions, was over 80% inactivated by H(2)O(2) upon oxygen exhaustion. The UV-Vis spectrum and the reduced reactivity with carbonyl reagents showed that the cofactor topaquinone (TPQ) was stabilized in reduced form. The protein large M(r) (170 kDa) prevented the identification of modified residues by amino acid analyses. Minor changes of the Cu(2+) EPR signal and the formation of a radical at g = 2.001, with intensity a few percent of that of the Cu(2+) signal, unaffected by a temperature increase, suggest that Cu(2+)-bound histidines were not oxidized and the radical was not the Cu(+)-semiquinolamine in equilibrium with Cu(2+)-aminoquinol. It may derive from the modification of a conserved residue in proximity of the active site, possibly the tyrosine at hydrogen-bonding distance of TPQ C-4 ionized hydroxyl. The inactivation reaction appears to be a general feature of copper-containing amine oxidases. It may be part of an autoregulatory process in vivo, possibly relevant to cell adhesion and redox signaling.

Purification and characterization of membrane-bound semicarbazide-sensitive amine oxidase (SSAO) from bovine lung

Semicarbazide-sensitive amine oxidase (SSAO) has been purified from bovine lung microsomes in a form which is catalytically active and stable to storage. The enzyme, an integral membrane protein, was solubilized with Triton X-100 and purification was achieved, in the presence of detergent, by chromatography with Cibacron Blue 3GA-agarose, hydroxylapatite, Lens culinaris-agarose, Resource Q-FPLC and gel filtration on Superdex 200 HR-FPLC. This is the first reported procedure for the extensive purification of a membrane-bound SSAO. The purified enzyme had an apparent Mr of 400000 but exhibited microheterogeneity with SDS/PAGE and isoelectric focusing, probably as a result of its glycoprotein nature. It behaved as a tetramer with subunits with apparent Mr values of 100. Antibodies raised towards the purified enzyme cross-reacted with the enzymes from human lung and bovine plasma. Redox-cycling staining and reaction with carbonyl reagents were consistent with the presence of a quinone cofactor, possibly topa quinone. The enzyme was also shown to contain two mol of Cu/mol of enzyme and removal of half of this bound copper resulted essentially in complete inhibition of enzyme activity. In contrast to the reported behaviour of the SSAO enzymes from plasma, the bovine lung enzyme was relatively insensitive to inhibition by cyanide, copper-chelating agents and amiloride. The specificity of the bovine lung enzyme was also narrower than reported for soluble SSAO. It catalysed the oxidative deamination of benzylamine, methylamine, 2-phenylethylamine and histamine but had no significant activity towards dopamine, 5-hydroxytryptamine, tryptamine or tyramine.

Reconstitution of Cu2+-depleted bovine serum amine oxidase with Co2+

Two different Cu2+-depleted derivatives of bovine serum amine oxidase (BSAO) have recently been prepared, which contain about 0.5 mol/dimer of phenylhydrazine-reactive topa quinone (TPQ) cofactor and, depending on the reagents used, about 0.2 or 0.7 residual Cu2+/dimer [Agostinelli, De Matteis, Sinibaldi, Mondovi and Morpurgo (1997) Biochem. J. 324, 497-501]. The benzylamine oxidase activity of both derivatives was <5% and increased up to approximately 20% on incorporation of Co2+, irrespective of the residual Cu2+ content, which was unaffected by the treatment according to atomic absorption and ESR spectroscopy. The residual Cu2+ ions appeared to be distributed one per dimer and to be bound to inactive subunits, whereas Co2+ was bound to active subunits. The change in the active site had an appreciable influence on the kinetic behaviour. With several amines, the kinetic parameters, Km and kc, measured for Co2+-BSAO were different from those for native BSAO. This excludes the possibility that the catalytic activity was due to residual Cu2+. Furthermore, Co2+ restored to nearly native level the intensity of the TPQ 480 nm band and the reactions with phenylhydrazine or benzylhydrazine, which had been slowed down or abolished, respectively, in Cu2+-depleted samples. The CD spectrum, measured for the derivative with low Cu2+ content, was compatible with Co2+ binding to the copper site. The amine oxidase activity of the Co2+ derivative, which cannot form a semiquinone radical as an intermediate of the catalytic reaction, strongly suggests that the Cu+-semiquinone is not an obligatory intermediate of BSAO catalytic pathway.

Reactions of the oxidized organic cofactor in copper-depleted bovine serum amine oxidase

A novel copper-depleted bovine serum amine oxidase (BSAO), in which about half the molecules contained the organic cofactor in the oxidized form, was prepared by adding a reductant in anaerobic conditions to the cyanide-reacted protein. The CuI-semiquinone formed in these conditions reoxidizes after the removal of copper. The inactive derivative was reduced by benzylamine at approx. 1/1000 the rate of BSAO. The pseudo-first-order reaction was preceded by the formation of a protein-benzylamine complex with dissociation constant, Kd, of 4.9+/-0.5 mM, similar to the Km of BSAO (2.2 mM). Also the reactions with phenylhydrazine and benzohydrazide were considerably slower than in holo-BSAO, whereas the reactions with p-pyridine-2-ylphenylacetohydrazide, containing a longer aromatic tail, and semicarbazide, lacking an aromatic moiety, were less severely affected. Removal of copper had no effect on the optical spectra of BSAO and of most adducts, containing the cofactor in quinol form, showing that copper is bound to neither the oxidized nor the reduced cofactor. Benzylhydrazine did not produce optical effects but was tightly bound, as inferred from its inhibitory effect on reaction with other molecules. Substrate and inhibitors might bind a hydrophobic pocket at some distance from the quinone, probably near the protein surface, with their affinity depending on the hydrophobic character and pKa. The binding, which is not greatly influenced by copper removal, probably induces a copper-dependent change of conformation, 'opening' a pathway to the active site buried in the protein interior.

Crystal structure of a eukaryotic (pea seedling) copper-containing amine oxidase at 2.2 A resolution

BACKGROUND

Copper-containing amine oxidases catalyze the oxidative deamination of primary amines to aldehydes, in a reaction that requires free radicals. These enzymes are important in many biological processes, including cell differentiation and growth, would healing, detoxification and signalling. The catalytic reaction requires a redox cofactor, topa quinone (TPQ), which is derived by post-translational modification of an invariant tyrosine residue. Both the biogenesis of the TPQ cofactor and the reaction catalyzed by the enzyme require the presence of a copper atom at the active site. The crystal structure of a prokaryotic copper amine oxidase from E. coli (ECAO) has recently been reported.

RESULTS

The first structure of a eukaryotic (pea seedling) amine oxidase (PSAO) has been solved and refined at 2.2 A resolution. The crystallographic phases were derived from a single phosphotungstic acid derivative. The positions of the tungsten atoms in the W12 clusters were obtained by molecular replacement using E. coli amine oxidase as a search model. The methodology avoided bias from the search model, and provides an essentially independent view of a eukaryotic amine oxidase. The PSAO molecule is a homodimer; each subunit has three domains. The active site of each subunit lies near an edge of the beta-sandwich of the largest domain, but is not accessible from the solvent. The essential active-site copper atom is coordinated by three histidine side chains and two water molecules in an approximately square-pyramidal arrangement. All the atoms of the TPQ cofactor are unambiguously defined, the shortest distance to the copper atom being approximately 6 A.

CONCLUSIONS

There is considerable structural homology between PSAO and ECAO. A combination of evidence from both structures indicates that the TPQ side chain is sufficiently flexible to permit the aromatic grouf to rotate about the Cbeta-Cgamma bond, and to move between bonding and non-bonding positions with respect to the Cu atom. Conformational flexibility is also required at the surface of the molecule to allow the substrates access to the active site, which is inaccessible to solvent, as expected for an enzyme that uses radical chemistry.