Probing the active site of glyoxalase I from human erythrocytes by use of the strong reversible inhibitor S-p-bromobenzylglutathione and metal substitutions

Molecular enzymology of the glyoxalase system

The glyoxalase system catalyzes the conversion of 2-oxoaldehydes into the corresponding 2-hydroxyacids. This biotransformation involves two separate enzymes, glyoxalase I and glyoxalase II, which bring about two consecutive reactions involving the thiol-containing tripeptide glutathione as a cofactor. The physiologically most important substrate methylglyoxal is converted by glyoxalase I into S-D-lactoyl-glutathione in the first reaction. Subsequently, glyoxalase II catalyzes the hydrolysis of this thiolester into D-lactic acid and free glutathione. The structures of both enzymes have been obtained via molecular cloning, heterologous expression, and X-ray diffraction analysis. Glyoxalase I and glyoxalase II are metalloenzymes and zinc plays an essential role in their diverse catalytic mechanisms. Both enzymes appear linked to a variety of pathological conditions, but further investigations are required to clarify the different physiological aspects of the glyoxalase system.

Characterization of the glyoxalases of the malarial parasite Plasmodium falciparum and comparison with their human counterparts

The glyoxalase system consisting of glyoxalase I (GloI) and glyoxalase II (GloII) constitutes a glutathione-dependent intracellular pathway converting toxic 2-oxoaldehydes, such as methylglyoxal, to the corresponding 2-hydroxyacids. Here we describe a complete glyoxalase system in the malarial parasitePlasmodium falciparum. The biochemical, kinetic and structural properties of cytosolic GloI (cGloI) and two GloIIs (cytosolic GloII named cGloII, and tGloII preceded by a targeting sequence) were directly compared with the respective isofunctional host enzymes. cGloI and cGloII exhibit lowerKmvalues and higher catalytic efficiencies (kcat/Km) than the human counterparts, pointing to the importance of the system in malarial parasites. A Tyr185Phe mutant of cGloII shows a 2.5-fold increase inKm, proving the contribution of Tyr185 to substrate binding. Molecular models suggest very similar active sites/metal binding sites of parasite and host cell enzymes. However, a fourth protein, which has highest similarities to GloI, was found to be unique for malarial parasites; it is likely to act in the apicoplast, and has as yet undefined substrate specificity. Various S-(N-hydroxy-N-arylcarbamoyl)glutathiones tested asP. falciparumGlo inhibitors were active in the lower nanomolar range. The Glo system ofPlasmodiumwill be further evaluated as a target for the development of antimalarial drugs.

Characterization of glyoxalase I (E. coli)-inhibitor interactions by electrospray time-of-flight mass spectrometry and enzyme kinetic analysis

Potential inhibitors of the enzyme glyoxalase I from Escherichia coli have been evaluated using a combination of electrospray mass spectrometry and conventional kinetic analysis. An 11-membered library of potential inhibitors included a glutathione analogue resembling the transition-state intermediate in the glyoxalase I catalysis, several alkyl-glutathione, and one flavonoid. The E. coli glyoxalase I quaternary structure was found to be predominantly dimeric, as is the homologous human glyoxalase I. Binding studies by electrospray revealed that inhibitors bind exclusively to the dimeric form of glyoxalase I. Two specific binding sites were observed per dimer. The transition-state analogue was found to have the highest binding affinity, followed by a newly identified inhibitor; S-(2-[3-(hexyloxy)benzoyl]-vinyl)glutathione. Kinetic analysis confirmed that the order of affinity established by mass spectrometry could be correlated to inhibitory effects on the enzymatic reaction. This study shows that selective inhibitors may exist for the E. coli homologue of the glyoxalase I enzyme.

The human glutathione transferase P1-1 specific inhibitor TER 117 designed for overcoming cytostatic-drug resistance is also a strong inhibitor of glyoxalase I

gamma-L-Glutamyl-S-(benzyl)-L-cysteinyl-R-(-)-phenylglycine (TER 117) has previously been developed for selective inhibition of human glutathione S-transferase P1-1 (GST P1-1) based on the postulated contribution of this isoenzyme to the development of drug resistance in cancer cells. In the present investigation, the inhibitory effect of TER 117 on the human glyoxalase system was studied. Although designed as an inhibitor specific for GST P1-1, TER 117 also competitively inhibits glyoxalase I (K(I) = 0.56 microM). In contrast, no inhibition of glyoxalase II was detected. Reduced glyoxalase activity is expected to raise intracellular levels of toxic 2-oxoaldehydes otherwise eliminated by glyoxalase I. The resulting toxicity would accompany the potentiation of cytostatic drugs, caused by inhibition of the detoxication effected by GST P1-1. TER 117 was designed for efficient inhibition of the most abundant form GST P1-1/Ile105. Therefore, the inhibitory effect of TER 117 on a second allelic variant GST P1-1/Val105 was also studied. TER 117 was shown to competitively inhibit both GST P1-1 variants. The apparent K(I) values at glutathione concentrations relevant to the intracellular milieu were in the micromolar range for both enzyme forms. Extrapolation to free enzyme produced K(I) values of approximately 0.1 microM for both isoenzymes, reflecting the high affinity of GST P1-1 for the inhibitor. Thus, the allelic variation in position 105 of GST P1-1 does not affect the inhibitory potency of TER 117. The inhibitory effects of TER 117 on GST P1-1 and glyoxalase I activities may act in synergy in the cell and improve the effectiveness of chemotherapy.

Glyoxalase I is a novel nitric-oxide-responsive protein

To clarify the molecular mechanisms of nitric oxide (NO) signalling, we examined the NO-responsive proteins in cultured human endothelial cells by two-dimensional (2D) PAGE. Levels of two proteins [NO-responsive proteins (NORPs)] with different pI values responded to NO donors. One NORP (pI 5.2) appeared in response to NO, whereas another (pI 5.0) disappeared. These proteins were identified as a native form and a modified form of human glyoxalase I (Glox I; EC 4. 4.1.5) by peptide mapping, microsequencing and correlation between the activity and the isoelectric shift. Glox I lost activity in response to NO, and all NO donors tested inhibited its activity in a dose-dependent manner. Activity and normal electrophoretic mobility were restored by dithiothreitol and by the removal of sources of NO from the culture medium. Glox I was selectively inactivated by NO; compounds that induce oxidative stress (H(2)O(2), paraquat and arsenite) failed to inhibit this enzyme. Our results suggest that NO oxidatively modifies Glox I and reversibly inhibits the enzyme's activity. The inactivation of Glox I by NO was more effective than that of glyceraldehyde-3-phosphate dehydrogenase (G3PDH), another NO-sensitive enzyme. Thus Glox I seems to be a novel NO-responsive protein that is more sensitive to NO than G3PDH.

Involvement of an active-site Zn2+ ligand in the catalytic mechanism of human glyoxalase I

The Zn2+ ligands glutamate 99 and glutamate 172 in the active site of human glyoxalase I were replaced, each in turn, by glutamines by site-directed mutagenesis to elucidate their potential significance for the catalytic properties of the enzyme. To compensate for the loss of the charged amino acid residue, another of the metal ligands, glutamine 33, was simultaneously mutated into glutamate. The double mutants and the single mutants Q33E, E99Q, and E172Q were expressed in Escherichia coli, purified on an S-hexylglutathione matrix, and characterized. Metal analysis demonstrated that mutant Q33E/E172Q contained 1.0 mol of zinc/mol of enzyme subunit, whereas mutant Q33E/E99Q contained only 0.3 mol of zinc/mol of subunit. No catalytic activity could be detected with the double mutant Q33E/E172Q (<10(-8) of the wild-type activity). The second double mutant Q33E/E99Q had 1.5% of the specific activity of the wild-type enzyme, whereas the values for mutants Q33E and E99Q were 1.3 and 0. 1%, respectively; the E172Q mutant had less than 10(-5) times the specific activity of the wild-type. The crystal structure of the catalytically inactive double mutant Q33E/E172Q demonstrated that Zn2+ was bound without any gross changes or perturbations. The results suggest that the metal ligand glutamate 172 is directly involved in the catalytic mechanism of the enzyme, presumably serving as the base that abstracts a proton from the hemithioacetal substrate.

Overexpression of glyoxalase-I in bovine endothelial cells inhibits intracellular advanced glycation endproduct formation and prevents hyperglycemia-induced increases in macromolecular endocytosis

Methylglyoxal (MG), a dicarbonyl compound produced by the fragmentation of triose phosphates, forms advanced glycation endproducts (AGEs) in vitro. Glyoxalase-I catalyzes the conversion of MG to S-D-lactoylglutathione, which in turn is converted to D-lactate by glyoxalase-II. To evaluate directly the effect of glyoxalase-I activity on intracellular AGE formation, GM7373 endothelial cells that stably express human glyoxalase-I were generated. Glyoxalase-I activity in these cells was increased 28-fold compared to neo-transfected control cells (21.80+/-0.1 vs. 0. 76+/-0.02 micromol/min/mg protein, n = 3, P < 0.001). In neo-transfected cells, 30 mM glucose incubation increased MG and D-lactate concentration approximately twofold above 5 MM (35.5+/-5.8 vs. 19.6+/-1.6, P < 0.02, n = 3, and 21.0+/-1.3 vs. 10.0+/-1.2 pmol/ 10(6) cells, n = 3, P < 0.001, respectively). In contrast, in glyoxalase-I-transfected cells, 30 mM glucose incubation did not increase MG concentration at all, while increasing the enzymatic product D-lactate by > 10-fold (18.9+/-3.2 vs. 18.4+/- 5.8, n = 3, P = NS, and 107.1+/-9.0 vs. 9.4+/-0 pmol/10(6) cells, n = 3, P < 0.001, respectively). After exposure to 30 mM glucose, intracellular AGE formation in neo cells was increased 13.6-fold (2.58+/-0.15 vs. 0.19+/-0.03 total absorbance units, n = 3, P < 0.001). Concomitant with increased intracellular AGEs, macromolecular endocytosis by these cells was increased 2.2-fold. Overexpression of glyoxalase-I completely prevented both hyperglycemia-induced AGE formation and increased macromolecular endocytosis.

Crystal structure of human glyoxalase I--evidence for gene duplication and 3D domain swapping

The zinc metalloenzyme glyoxalase I catalyses the glutathione-dependent inactivation of toxic methylglyoxal. The structure of the dimeric human enzyme in complex with S-benzyl-glutathione has been determined by multiple isomorphous replacement (MIR) and refined at 2.2 A resolution. Each monomer consists of two domains. Despite only low sequence homology between them, these domains are structurally equivalent and appear to have arisen by a gene duplication. On the other hand, there is no structural homology to the 'glutathione binding domain' found in other glutathione-linked proteins. 3D domain swapping of the N- and C-terminal domains has resulted in the active site being situated in the dimer interface, with the inhibitor and essential zinc ion interacting with side chains from both subunits. Two structurally equivalent residues from each domain contribute to a square pyramidal coordination of the zinc ion, rarely seen in zinc enzymes. Comparison of glyoxalase I with other known structures shows the enzyme to belong to a new structural family which includes the Fe2+-dependent dihydroxybiphenyl dioxygenase and the bleomycin resistance protein. This structural family appears to allow members to form with or without domain swapping.

Antitumor activity of S-(p-bromobenzyl)glutathione diesters in vitro: a structure-activity study

S-(p-Bromobenzyl)glutathione is a competitive inhibitor of human glyoxalase I which is part of the cytosolic glyoxalase system. It may be delivered into the cystosol of cells by diesterification wherein it is deesterified by cytosolic nonspecific esterases. S-(p-Bromobenzyl)glutathione diesters had antitumor activity in vitro and in vivo. The inhibition of human leukemia 60 cell growth in vitro by a series of alkyl and cycloalkyl diesters of S-(p-bromobenzyl)glutathione was investigated. For n-alkyl diesters, the n-propyl diester was the most potent derivative with a median growth inhibitory concentration GC50 value of 7.77 +/- 0.01 microM (N = 18). The most potent derivative was S-(p-bromobenzyl)glutathione cyclopentyl diester which had a GC50 value of 4.23 +/- 0.01 microM (N = 21) and also had antitumor activity in vivo.

Optimized heterologous expression of the human zinc enzyme glyoxalase I

DNA coding for human glyoxalase I was isolated from a HeLa cell cDNA library by means of PCR. The deduced amino acid sequence differs form previously isolated sequences in that a glutamic acid replaces an alanine in position 111. This variant cDNA may represent the more acidic isoform of glyoxalase I originally identified at the protein level. An expression clone was constructed for high-level production of glyoxalase I in Escherichia coli. For optimal yield of the recombinant protein, silent random mutations were introduced in the cDNA coding region. Antisera against human glyoxalase I were used to select a high-level expression clone. This clone afforded 60 mg of purified enzyme per litre of culture medium. Addition of a zinc salt to the culture medium was essential to obtain an active enzyme and a stoicheiometric metal content. The functional characterization of the recombinant enzyme included determination of kinetic constants for methylglyoxal, phenylglyoxal and p-phenylphenylglyoxal, as well as inhibition studies. The kinetic properties of recombinant glyoxalase I were indistinguishable from those of the enzyme purified from human tissues.

A simplified method for the purification of human red blood cell glyoxalase. I. Characteristics, immunoblotting, and inhibitor studies

Glyoxalase I (EC 4.4.1.5) was purified from human red blood cells by a simplified method using S-hexylglutathione affinity chromatography with a modified concentration gradient of S-hexylglutathione for elution. The pure protein had a specific activity of 1830 U/mg of protein, where the overall yield was 9%. The pure protein had a molecular mass of 46,000 D, comprised of two subunits of 23,000 D each, and an isoelectric point value of 5.1. The KM value for methylglyoxal-glutathione hemithioacetal was 192 +/- 8 microM and the kcat value was 10.9 +/- 0.2 x 10(4) min-1 (N = 15). The glyoxalase I inhibitor S-p-bromobenzylglutathione had a Ki value of 0.16 +/- 0.04 microM and S-p-nitrobenzoxycarbonylglutathione, previously thought to inhibit only glyoxalase II, also inhibited glyoxalase I with a Ki value of 3.12 +/- 0.88 microM. Reduced glutathione was a weak competitive inhibitor of glyoxalase I with a Ki value of 18 +/- 8 mM. The polyclonal antibodies were raised to the purified enzyme and were found to react specifically with glyoxalase I antigen by immunoblotting. This procedure gave a protein of high purity with simple low pressure chromatographic techniques with a moderate but adequate yield for small-scale preparations.

The formation of methylglyoxal from triose phosphates. Investigation using a specific assay for methylglyoxal

In Krebs-Ringer phosphate buffer, the rate of formation of methylglyoxal from glycerone phosphate and glyceraldehyde 3-phosphate was first order with respect to the triose phosphate with rates constant values of 1.94 +/- 0.02 x 10(-5) s-1 (n = 18) and 1.54 +/- 0.02 x 10(-4) s-1 (n = 18) at 37 degrees C, respectively. The rate of formation of methylglyoxal from glycerone phosphate and glyceraldehyde 3-phosphate in the presence of red blood cell lysate was not significantly different from the non-enzymatic value (P > 0.05). Methylglyoxal formation from glycerone phosphate was increased in the presence of triose phosphate isomerase but this may be due to the faster non-enzymatic formation from the glyceraldehyde 3-phosphate isomerisation product. For red blood cells in vitro, the predicted non-enzymatic rate of formation of methylglyoxal from glycerone phosphate and glyceraldehyde 3-phosphate may account for the metabolic flux through the glyoxalase system. The reactivity of glycerone phosphate and glyceraldehyde 3-phosphate towards the non-enzymatic formation of methylglyoxal under physiological conditions suggests that methylglyoxal formation is unavoidable from the Embden-Meyerhof pathway.

Chemical modification of tyrosine residues in glyoxalase I from yeast and human erythrocytes

1. Yeast glyoxalase I was inactivated by N-acetylimidazole and tetranitromethane, the latter process following pK app 7.31 and irreversibly producing a protein with a spectrum typical of 3-nitrotyrosine. 2. For yeast glyoxalase I, amino-acid analysis and protection studies with S-(p-bromobenzyl)glutathione, a competitive inhibitor, indicated two classes of tetranitromethane-reactive tyrosine residues, fast- and slow-reacting, with the latter class containing the crucial tyrosine(s). 3. Human erythrocyte glyoxalase I was inactivated by tetranitromethane in fast and slow processes, protection studies in this case indicating the important tyrosine(s) as fast-reacting.

Inhibition of mammalian glyoxalase I (lactoylglutathione lyase) by N-acylated S-blocked glutathione derivatives as a probe for the role of the N-site of glutathione in glyoxalase I mechanism

A series of twelve S-blocked and N,S-blocked glutathione derivatives has been studied as inhibitors of glyoxalase I [R)-S-lactoylglutathione methylglyoxal-lyase (isomerising), EC 4.4.1.5) from human erythrocytes. A number of new N,S-blocked glutathiones have been synthesised. Inhibition at pH 7.0, 25 degrees C was linear-competitive in all cases and the Ki values were interpreted in terms of the absence of a specific binding interaction for the N-site of the inhibitor and the absence of coupling between binding processes at N- and S-sites (the regions around the NH2 and HS groups, respectively, of GSH analogues bound to enzyme). These observations are in strong contrast to previous results with the yeast enzyme. Some Ki values were measured for yeast glyoxalase I. A special binding interaction of the phenyl groups with enzyme from both species was found for glutathione derivatives with N-acyl groups of structure -NH X CO X X X Y X Ph but not for -NH X COPh, where X and Y were variously -CH2-, -NH- and -O-. Studies were made of the range of stability of human erythrocyte glyoxalase I to pH. The pH profiles for the Ki values of S-p-bromobenzyl)glutathione and N-acetyl-S-(p-bromobenzyl)glutathione indicated no pH dependence for the latter and little, if any, for the former inhibitor. The mean Ki over the pH range 5-8.5 for S-(p-bromobenzyl)glutathione was 1.21 +/- 0.37 microM and for N-acetyl-S-(p-bromobenzyl)glutathione in the same pH range, Ki decreased from 1.45 +/- 0.26 microM to 0.88 +/- 0.11 M.

Methylglyoxal inhibits the translation of natural and chemically decapped mRNAs

Methylglyoxal was a weak inhibitor of translation in the reticulocyte-lysate cell-free system and it did not display cap-dependent inhibition. A similar inhibition was obtained in a wheat-germ cell-free system that displayed extensive cap-dependent inhibition with the cap analogue 7-methylguanosine phosphate. These results show that the chemical reaction of methylglyoxal with 7-methylguanosine is not the mechanism for the inhibition of protein synthesis by methylglyoxal and that methylglyoxal is a weak general inhibitor of translation.

Fluorescence and nuclear relaxation enhancement studies of the binding of glutathione derivatives to manganese-reconstituted glyoxalase I from human erythrocytes. A model for the catalytic mechanism of the enzyme involving a hydrated metal ion

The apoenzyme of glyoxalase I (EC 4.4.1.5) from human erythrocytes was prepared by removal of Zn2+ with ethylenediaminetetraacetic acid (EDTA). Methanol was used as a stabilizing agent. Extended dialysis was required to remove EDTA from the resulting solution of apoenzyme. Reconstitution with Mn2+ was followed by measuring enzyme activity, electron paramagnetic resonance of free Mn2+ ions, and nuclear magnetic resonance of water protons. The holoenzyme contained two Mn2+ per protein dimer and had approximately 50% of the catalytic activity of the native enzyme. The binding of the cosubstrate glutathione (gamma-L-glutamyl-L-cysteinylglycine), the product S-D-lactoyl-glutathione, and the competitive inhibitor S-(p-bromo-benzyl)glutathione was monitored by the quenching of the intrinsic tryptophan fluorescence and by the proton relaxation enhancement of water bound to Mn2+ in the active site of the enzyme. The dissociation constants were 1.1 mM, 0.42 mM, and 0.54 microM for glutathione, S-D-lactoylglutathione, and S-(p-bromobenzyl)glutathione, respectively. The temperature and frequency dependences of the longitudinal and transverse paramagnetic relaxation rates, 1/T1p and 1/T2p, were studied for water. The results were analyzed in terms of correlation and exchange times. In addition proton and deuteron relaxation rates were measured in parallel at two different magnetic fields. Good agreement between the two approaches of analysis was noticed. The data show that two water molecules are bound in the first coordination sphere of Mn2+ in the active site of glyoxalase I. When S-(p-bromobenzyl)glutathione or S-D-lactoylglutathione is bound to the enzyme, only one exchangeable water molecule could be detected, indicating occlusion of the second water molecule. An enediol mechanism involving the metal-bound water is proposed for the catalysis effected by glyoxalase I.