Multiple roles of glutathione binding-site residues of glutathione s-transferase

This study was designed to characterize residues in the glutathione binding site of AdGSTD4-4 from the mosquito malaria vector Anopheles dirus. The data revealed that Leu33, His38 and His50 each play a role in enzyme catalysis and glutathione binding. The mutants of these three residues also displayed differences in hydrophobic substrate specificity, suggesting that changes in the active site conformation occurred. Differences in conformations was also suggested by protein stability changes. These results indicate that residues in the glutathione binding site are not only important in the catalytic function but also play a role in the structural integrity of the enzyme.     

Cooperative binding of gamma-glutamyl substrate to human glutathione synthetase.

Human glutathione synthetase is responsible for catalyzing the final step in glutathione biosynthesis. It is a homodimer with a monomer subunit MW of 52 kDa. Kinetic analysis reveals a departure from linearity of the Lineweaver-Burk double reciprocal plot for the binding of gamma-glutamyl substrate, indicating cooperative binding. The measured apparent K(m) values for gamma-glutamyl-alpha-aminobutyrate (an analog of gamma-glutamyl-alpha-aminobutyrate) are 63 and 164 microM, respectively. Neither ATP (K(m) of 248 microM) nor glycine (K(m) of 452 microM) exhibits such cooperative binding behavior. Although ATP is proposed to play a key role in the sequential binding of gamma-glutamyl substrate to the enzyme, the cooperative binding of the gamma-glutamyl substrate is not affected by alterations of ATP concentration. Quantitative analysis of the kinetic results for gamma-glutamyl substrate binding gives a Hill coefficient (h) of 0.75, indicating negative cooperativity. Our studies, for the first time, show that human glutathione synthetase is an allosteric enzyme with cooperative binding for gamma-glutamyl substrate.     

A crystallographic study of the glutathione binding site of glutathione reductase at 0.3-nm resolution

The binding of glutathione, some related molecules and two redox compounds to crystals of glutathione reductase has been investigated by X-ray crystallography at 0.3-nm resolution. Models for several bound ligands have been built and subjected to crystallographic refinement. The results clearly show the residues involved in glutathione binding as well as the geometry of the disulfide exchange. Glutathione-I is bound in a V-shaped conformation, while glutathione-II is extended. The zwitterionic glutamyl end of glutathione-II appears to be the most tightly bound part of the substrate. All glutathione conjugates and derivatives studied show binding dominated by the interactions at this site. In the reduced enzyme, glutathione-I forms a mixed disulfide intermediate with Cys58. Other structural changes are observed on reduction of the enzyme, and it is demonstrated that the carboxamidomethylated enzyme is a good model for the reduced species. Lipoate, a weak substrate of the enzyme, assumes a defined binding site where its disulfide is available for being attacked by Cys58-S gamma. A second region with affinity for a number of compounds has been found in a large cavity at the dimer interface of the enzyme. No functional role of this site is known.     

Novel kinetics of mammalian glutathione synthetase: characterization of gamma-glutamyl substrate cooperative binding

Glutathione (GSH) synthetase [L-gamma-glutamyl-L-cysteinyl:glycine ligase (ADP-forming), EC 6.3.2.3] catalyzes the final step in GSH biosynthesis. Mammalian glutathione synthetase is a homodimer with each subunit containing an active site. We report the detailed kinetic data for purified recombinant rat glutathione synthetase. It has the highest specific activity (11 micromol/min/mg) reported for any mammalian glutathione synthetase. The apparent K(m) values for ATP and glycine are 37 and 913 microM, respectively. The Lineweaver-Burk double reciprocal plot for gamma-glutamyl substrate binding revealed a departure from linearity indicating cooperative binding. Quantitative analysis of the kinetic results for gamma-glutamyl substrate binding gives a Hill coefficient (h) of 0. 576, which shows the negative cooperativity. Neither ATP, the other substrate involved in forming the enzyme-bound gamma-glutamyl phosphate intermediate, nor glycine, which attacks this intermediate to form GSH, exhibit any cooperativity. The cooperative binding of gamma-glutamyl substrate is not affected by ATP concentration. Thus, mammalian glutathione synthetase is an allosteric enzyme.     

Role of the charged groups of glutathione disulfide in the catalysis of glutathione reductase: crystallographic and kinetic studies with synthetic analogues

Six analogues of glutathione disulfide were synthesized. All of them involved the abolishment of charges, either by amidation of carboxylates or by removal of amino groups. Four of these analogues could be bound to crystalline oxidized glutathione reductase, and their binding modes could be established by X-ray analyses at 2.4-A resolution. All six analogues were catalytically processed; the kinetic parameters were determined. The two analogues that did not bind in the crystals had by far the poorest catalytic efficiencies. Kinetic parameters together with X-ray data show the influence of each charged group on binding and catalytic rate. Data analysis indicates that the enzyme avoids processing of incorrect substrates in two ways: First, it reduces their binding strengths and/or enforces displacement of catalytically important substrate parts. Furthermore, it forms a fragile cluster of bound substrate and catalytically competent residues, which is unbalanced by incorrect parts of the substrate such that catalysis is prevented. A scouting microcalorimetric study using glutathione disulfide yielded a binding enthalpy of -103 (+/- 10) kJ/mol at 25 degrees C and a heat capacity change of -8 (+/- 1) kJ.mol-1.K-1. The study showed that it is feasible to measure these parameters as a function of substrate modification.     

A novel selenocystine-beta-cyclodextrin conjugate that acts as a glutathione peroxidase mimic

A novel artificial glutathione peroxidase mimic consisting of a selenocystine-di-beta-cyclodextrin conjugate (selenium-bridged-6, 6'-amino-selenocystine-6,6'-deoxy-di-beta-cyclodextrin), in which selenocystine is bound to the primary side of beta-cyclodextrin through the two amino nitrogen groups of selenocystine, was synthesized. The glutathione peroxidase activities of the mimic-catalyzed reduction of H(2)O(2), tert-butylhydroperoxide, and cumene hydroperoxide by glutathione are 4.1, 2.11, and 5.82 units/micromol, respectively. The first activity was 82 and 4.2 times as much as that of selenocysteine and ebselen, respectively. Studies on the effect of substrate binding on the glutathione peroxidase activity suggest that it is important to consider substrate binding in designing glutathione peroxidase mimics. The detailed steady-state kinetic studies showed that the mimic-catalyzed reduction of H(2)O(2) by glutathione followed a ping-pong mechanism, which was similar to that of the native glutathione peroxidase.     

Truncated mutants of human thioredoxin reductase 1 do not exhibit glutathione reductase activity

The substrate spectrum of human thioredoxin reductase (hTrxR) is attributed to its C-terminal extension of 16 amino acids carrying a selenocysteine residue. The concept of an evolutionary link between thioredoxin reductase and glutathione reductase (GR) is presently discussed and supported by the fact that almost all residues at catalytic and substrate recognition sites are identical. Here, we addressed the question if a deletion of the C-terminal part of TrxR leads to recognition of glutathione disulfide (GSSG), the substrate of GR. We introduced mutations at the putative substrate binding site to enhance GSSG binding and turnover. However, none of these enzyme species accepted GSSG as substrate better than the full length cysteine mutant of TrxR, excluding a role of the C-terminal extension in preventing GSSG binding. Furthermore, we show that GSSG binding at the N-terminal active site of TrxR is electrostatically disfavoured.     

Inhibition of octopus glutathione transferase by Meisenheimer complex analog, S-(2,4,6-trinitrophenyl) glutathione

The tight binding of Meisenheimer intermediate with octopus digestive gland glutathione transferase was analyzed with 1,3,5-trinitrobenzene, which forms a trapped Meisenheimer complex with glutathione because there is no leaving group at the ipso carbon. By steady-state enzyme kinetic analysis, an inhibition constant of 1.89 ± 0.17 M was found for the transient formed, S-(2,4,6-trinitrophenyl) glutathione. The above inhibition constant is 407-fold smaller than the K _m value for the substrate (2,4-dinitrochlorobenzene). Thus, S-(2,4,6-trinitrophenyl) glutathione is considered to be a transition-state analog. The tight binding of this inhibitor to the enzyme provides an explanation for the involvement of the biological binding effect on the rate enhancement in the glutathione transferase-catalyzed S_NAr mechanism.     

Monobromobimane occupies a distinct xenobiotic substrate site in glutathione S-transferase pi

Monobromobimane (mBBr), functions as a substrate of porcine glutathione S-transferase pi (GST pi): The enzyme catalyzes the reaction of mBBr with glutathione. S-(Hydroxyethyl)bimane, a nonreactive analog of monobromobimane, acts as a competitive inhibitor with respect to mBBr as substrate but does not affect the reaction of GST pi with another substrate, 1-chloro-2,4-dinitrobenzene (CDNB). In the absence of glutathione, monobromobimane inactivates GST pi at pH 7.0 and 25 degrees C as assayed using mBBr as substrate, with a lesser effect on the enzyme's use of CDNB as substrate. These results indicate that the sites occupied by CDNB and mBBr are not identical. Inactivation is proportional to the incorporation of 2 moles of bimane/mole of subunit. Modification of GST pi with mBBr does not interfere with its binding of 8-anilino-1-naphthalene sulfonate, indicating that this hydrophobic site is not the target of monobromobimane. S-Methylglutathione and S-(hydroxyethyl)bimane each yield partial protection against inactivation and decrease reagent incorporation, while glutathionyl-bimane protects completely against inactivation. Peptide analysis after trypsin digestion indicates that mBBr modifies Cys45 and Cys99 equally. Modification of Cys45 is reduced in the presence of S-methylglutathione, indicating that this residue is at or near the glutathione binding region. In contrast, modification of Cys99 is reduced in the presence of S-(hydroxyethyl)bimane, suggesting that this residue is at or near the mBBr xenobiotic substrate binding site. Modification of Cys99 can best be understood by reaction with monobromobimane while it is bound to its xenobiotic substrate site in an alternate orientation. These results support the concept that glutathione S-transferase accomplishes its ability to react with a diversity of substrates in part by harboring distinct xenobiotic substrate sites.     

Large conformational changes in the catalytic cycle of glutathione synthase

Glutathione synthase catalyzes the final ATP-dependent step in glutathione biosynthesis, the formation of glutathione from gamma-glutamylcysteine and glycine. We have determined structures of yeast glutathione synthase in two forms: unbound (2.3 A resolution) and bound to its substrate gamma-glutamylcysteine, the ATP analog AMP-PNP, and two magnesium ions (1.8 A resolution). These structures reveal that upon substrate binding, large domain motions convert the enzyme from an open unliganded form to a closed conformation in which protein domains completely surround the substrate in the active site.     

Distinctions between the multiple cationic forms of rat liver glutathione S-transferase

Three cationic glutathione S-transferase forms isolated from rat liver were characterized as dimers that originated from different combinations of two subunit types, Ya and Yc. The cationic forms were purified using lysyl glutathione affinity matrices and were chromatographically resolved from anionic glutathione S-transferases that contain Yb subunits. The three classes of cationic transferase exhibited similar specific activities with 1-chloro-2,4-dinitrobenzene as a substrate, all forms cross-reacted with antibodies to glutathione S-transferase B, and all had comparable secondary structures and tryptophan fluorescence properties. In spite of those similarities, the Yc-containing forms were clearly distinguishable from Ya forms on the basis of characteristic differences in circular dichroic patterns associated with their aromatic side chains. All cationic transferases bound bilirubin with stoichiometric ratios of 1 mol/dimeric protein molecule, but discrete differences in mode of binding were ascribed to forms containing Ya subunits as compared to Yc dimers. Binding to Yc forms was of lower affinity and may be associated with the catalytic region of the protein since glutathione effectively displaced bilirubin from the Yc component.     

Functional mimicry of the active site of glutathione peroxidase by glutathione imprinted selenium-containing protein

For imitating the active site of antioxidant selenoenzyme glutathione peroxidase (GPx), an artificial enzyme selenosubtilisin was employed as a scaffold for reconstructing substrate glutathione (GSH) specific binding sites by a bioimprinting strategy. GSH was first covalently linked to selenosubtilisin to form a covalent complex GSH-selenosubtilisin through a Se-S bond, then the GSH molecule was used as a template to cast a complementary binding site for substrate GSH recognition. The bioimprinting procedure consists of unfolding the conformation of selenosubtilisin and fixing the new conformation of the complex GSH-selenosubtilisin. Thus a new specificity for naturally occurring GPx substrate GSH was obtained. This bioimprinting procedure facilitates the catalytic selenium moiety of the imprinted selenosubtilisin to match the reactive thiol group of GSH in the GSH binding site, which contributes to acceleration of the intramolecular catalysis. These imprinted selenium-containing proteins exhibited remarkable rate enhancement for the reduction of H2O2 by GSH. The average GPx activity was found to be 462 U/micromol, and it was approximately 100 times that for unimprinted selenosubtilisin. Compared with ebselen, a well-known GPx mimic, an activity enhancement of 500-fold was observed. Detailed steady-state kinetic studies demonstrated that the novel selenoenzyme followed a ping-pong mechanism similar to the naturally occurring GPx.     

Catalysis within the lipid bilayer-structure and mechanism of the MAPEG family of integral membrane proteins

Integral membrane enzymes of the MAPEG (membrane-associated proteins in eicosanoid and glutathione metabolism) family catalyze glutathione-dependent transformations of lipophilic substrates harvested from the lipid bilayer. Recent studies of members of this family have yielded extensive insights into the structural basis for their substrate binding and catalytic activity. Most informative are the structural studies of leukotriene C4 synthase, revealing a narrow hydrophobic substrate binding pocket allowing extensive recognition of the aliphatic chain of the LTA(4) substrate. A key feature of the pocket is a tryptophan residue that pins down the omega-end of the aliphatic chain into the active site. Since MAPEG members cannot utilize a hydrophobic effect for substrate binding, this novel mode of substrate recognition appears well suited for harvesting lipophilic substrates from the membrane. The binding mode also allows for the specific alignment of the substrate in the active site, positioning the C6 of the substrate for conjugation with glutathione. The glutathione is in turn bound in a polar pocket submerged into the protein core. Structure-based sequence alignments of human MAPEG members support the notion that the glutathione binding site is highly conserved among MAPEG enzymes and that they use a similar mechanism for glutathione activation.     

Thermodynamics of glutathione binding to the tyrosine 7 to phenylalanine mutant of glutathione S-transferase from Schistosoma japonicum

The binding of glutathione (GSH) to the tyrosine 7 to phenylalanine mutant of Schistosoma japonicum glutathione S-transferase (SjGST-Y7F) has been studied by isothermal titration calorimetry (ITC). At pH 6.5 and 25 °C this mutant shows a higher affinity for glutathione than wild type enzyme despite an almost complete loss of activity in the presence of 1-chloro-2,4-dinitrobenzene (CDNB) as second substrate. The enthalpy change upon binding of GSH is more negative for the mutant than for the wild type GST (SjGST). Changes in accessible solvent areas (ASA) have been calculated based on enthalpy and heat capacity changes. ASA values indicated the burial of apolar surfaces of protein and ligand upon binding. A more negative ΔC_p value has been obtained for the mutant enzyme, suggesting a more hydrophobic interaction, as may be expected from the change of a tyrosine residue to phenylalanine.     

Thermodynamic analysis of the binding of glutathione to glutathione S-transferase over a range of temperatures.

The binding properties of a glutathione S-transferase (EC 2.5.1.18) from Schistosoma japonicum to substrate glutathione (GSH) has been investigated by intrinsic fluorescence and isothermal titration calorimetry (ITC) at pH 6.5 over a temperature range of 15-30 degrees C. Calorimetric measurements in various buffer systems with different ionization heats suggest that protons are released during the binding of GSH at pH 6.5. We have also studied the effect of pH on the thermodynamics of GSH-GST interaction. The behaviour shown at different pHs indicates that at least three groups must participate in the exchange of protons. Fluorimetric and calorimetric measurements indicate that GSH binds to two sites in the dimer of 26-kDa glutathione S-transferase from Schistosoma japonicum (SjGST). On the other hand, noncooperativity for substrate binding to SjGST was detected over a temperature range of 15-30 degrees C. Among thermodynamic parameters, whereas DeltaG degrees remains practically invariant as a function of temperature, DeltaH and DeltaS degrees both decrease with an increase in temperature. While the binding is enthalpically favorable at all temperatures studied, at temperatures below 25 degrees C, DeltaG degrees is also favoured by entropic contributions. As the temperature increases, the entropic contributions progressively decrease, attaining a value of zero at 24.3 degrees C, and then becoming unfavorable. During this transition, the enthalpic contributions become progressively favorable, resulting in an enthalpy-entropy compensation. The temperature dependence of the enthalpy change yields the heat capacity change (DeltaCp degrees ) of -0.238 +/- 0.04 kcal per K per mol of GSH bound.     

Binding and protection of porphyrins by glutathione S-transferases of Zea mays L

Glutathione S-transferases (GSTs) are multi-functional enzymes, known to conjugate xenobiotics and degrade peroxides. Herein, we report on the potential of four Zea mays GST isoforms (Zm GST I-I, Zm GST I-II, Zm GST II-II and Zm GST III-III) to act as binding and protection proteins. These isoforms bind protoporphyrin IX (PPIX), mesoporphyrin, coproporphyrin, uroporphyrin and Mg-protoporpyhrin, but do not form a glutathione conjugate. The binding is non-covalent and inhibits GSTs enzymatic activity, dependent on the type of the porphyrin and GST isoform tested. I(50) values are in the range of 1 to 10 microM for PPIX, the inhibition by mesoporphyrin and Mg-protoporphyrin (Mg-PPIX) is two to five times less. The mode of binding is non-competitive for the hydrophobic substrate and competitive for glutathione. Binding affinities (K(D) values) of the GST isoforms are between 0.3 and 0.8 microM for coproporphyrin and about 2 microM for mesoporphyrin.Zm GST III-III prevents the nonenzymatic autoxidation of protoporphyrinogen to the phytotoxic PPIX. Zm GST II-II can reduce the oxidative degradation of hemin. This points to a specific ligand role of distinct GST isoforms to protect tetrapyrroles in the plant cell.     

Interaction of a glutathione S-conjugate with glutathione reductase. Kinetic and X-ray crystallographic studies

S-Conjugates of glutathione influence the glutathione/glutathione disulfide (GSH/GSSG) status of hepatocytes in at least two ways, namely by inhibition of GSSG transport into the bile [Akerboom et al. (1982) FEBS Lett. 140, 73-76] and by inhibition of the enzyme GSSG reductase (EC 1.6.4.2). The interaction of GSSG reductase with a well-studied conjugate, namely S-(2,4-dinitrophenyl)-glutathione and its electrophilic precursor 1-chloro-2,4-dinitrobenzene are described. For short exposures both compounds are reversible inhibitors of the enzyme, the Ki values being 30 microM and 22 microM respectively. After prolonged incubation, 1-chloro-2,4-dinitrobenzene blocks GSSG reductase irreversibly, which emphasizes the need for rapid conjugate formation in situ. As shown by X-ray crystallography the major binding site of S-(2,4-dinitrophenyl)-glutathione in GSSG reductase overlaps the binding site of the substrate, glutathione disulfide. However, the glutathione moiety of the conjugate does not bind in the same manner as either of the glutathiones in the disulfide.     

Structure-activity relationships of 4-hydroxyalkenals in the conjugation catalysed by mammalian glutathione transferases.

The substrate specificities of 15 cytosolic glutathione transferases from rat, mouse and man have been explored by use of a homologous series of 4-hydroxyalkenals, extending from 4-hydroxypentenal to 4-hydroxypentadecenal. Rat glutathione transferase 8-8 is exceptionally active with the whole range of 4-hydroxyalkenals, from C5 to C15. Rat transferase 1-1, although more than 10-fold less efficient than transferase 8-8, is the second most active transferase with the longest chain length substrates. Other enzyme forms showing high activities with these substrates are rat transferase 4-4 and human transferase mu. The specificity constants, kcat./Km, for the various enzymes have been determined with the 4-hydroxyalkenals. From these constants the incremental Gibbs free energy of binding to the enzyme has been calculated for the homologous substrates. The enzymes responded differently to changes in the length of the hydrocarbon side chain and could be divided into three groups. All glutathione transferases displayed increased binding energy in response to increased hydrophobicity of the substrate. For some of the enzymes, steric limitations of the active site appear to counteract the increase in binding strength afforded by increased chain length of the substrate. Comparison of the activities with 4-hydroxyalkenals and other activated alkenes provides information about the active-site properties of certain glutathione transferases. The results show that the ensemble of glutathione transferases in a given species may serve an important physiological role in the conjugation of the whole range of 4-hydroxyalkenals. In view of its high catalytic efficiency with all the homologues, rat glutathione transferase 8-8 appears to have evolved specifically to serve in the detoxication of these reactive compounds of oxidative metabolism.     

Kinetic analysis of the slow ionization of glutathione by microsomal glutathione transferase MGST1

An important aspect of the catalytic mechanism of microsomal glutathione transferase (MGST1) is the activation of the thiol of bound glutathione (GSH). GSH binding to MGST1 as measured by thiolate anion formation, proton release, and Meisenheimer complex formation is a slow process that can be described by a rapid binding step (K(GSH)d = 47 +/- 7 mM) of the peptide followed by slow deprotonation (k2 = 0.42 +/- 0.03 s(-1). Release of the GSH thiolate anion is very slow (apparent first-order rate k(-2) = 0.0006 +/- 0.00002 s(-)(1)) and thus explains the overall tight binding of GSH. It has been known for some time that the turnover (kcat) of MGST1 does not correlate well with the chemical reactivity of the electrophilic substrate. The steady-state kinetic parameters determined for GSH and 1-chloro-2,4-dinitrobenzene (CDNB) are consistent with thiolate anion formation (k2) being largely rate-determining in enzyme turnover (kcat = 0.26 +/- 0.07 s(-1). Thus, the chemical step of thiolate addition is not rate-limiting and can be studied as a burst of product formation on reaction of halo-nitroarene electrophiles with the E.GS- complex. The saturation behavior of the concentration dependence of the product burst with CDNB indicates that the reaction occurs in a two-step process that is characterized by rapid equilibrium binding ( = 0.53 +/- 0.08 mM) to the E.GS- complex and a relatively fast chemical reaction with the thiolate (k3 = 500 +/- 40 s(-1). In a series of substrate analogues, it is observed that log k3 is linearly related (rho value 3.5 +/- 0.3) to second substrate reactivity as described by Hammett sigma- values demonstrating a strong dependence on chemical reactivity that is similar to the nonenzymatic reaction (rho = 3.4). Microsomal glutathione transferase 1 displays the unusual property of being activated by sulfhydryl reagents. When the enzyme is activated by N-ethylmaleimide, the rate of thiolate anion formation is greatly enhanced, demonstrating for the first time the specific step that is activated. This result explains earlier observations that the enzyme is activated only with more reactive substrates. Taken together, the observations show that the kinetic mechanism of MGST1 can be described by slow GSH binding/thiolate formation followed by a chemical step that depends on the reactivity of the electrophilic substrate. As the chemical reactivity of the electrophile becomes lower the rate-determining step shifts from thiolate formation to the chemical reaction.     

Role of the glutamyl alpha-carboxylate of the substrate glutathione in the catalytic mechanism of human glutathione transferase A1-1.

The Glu alpha-carboxylate of glutathione contributes to the catalytic function of the glutathione transferases. The catalytic efficiency of human glutathione transferase A1-1 (GST A1-1) in the conjugation reaction with 1-chloro-2,4-dinitrobenzene is reduced 15 000-fold if the decarboxylated analogue of glutathione, dGSH (GABA-Cys-Gly), is used as an alternative thiol substrate. The decrease is partially due to an inability of the enzyme to promote ionization of dGSH. The pK(a) value of the thiol group of the natural substrate glutathione decreases from 9.2 to 6.7 upon binding to GST A1-1. However, the lack of the Glu alpha-carboxylate in dGSH raised the pK(a) value of the thiol in the enzymatic reaction to that of the nonenzymatic reaction. Furthermore, K(M)(dGSH) was 100-fold higher than K(M)(GSH). The active-site residue Thr68 forms a hydrogen bond to the Glu alpha-carboxylate of glutathione. Introduction of a carboxylate into GST A1-1 by a T68E mutation increased the catalytic efficiency with dGSH 10-fold and reduced the pK(a) value of the active site bound dGSH by approximately 1 pH unit. The altered pK(a) value is consistent with a catalytic mechanism where the carboxylate contributes to ionization of the glutathione thiol group. With Delta(5)-androstene-3,17-dione as substrate the efficiency of the enzyme is decreased 24 000-fold while with 4-nitrocinnamaldehyde (NCA) the decrease is less than 150-fold. In the latter reaction NCA accepts a proton and, unlike the other reactions studied, may not be dependent on the Glu alpha-carboxylate for deprotonation of the thiol group. An additional function of the Glu alpha-carboxylate may be productive orientation of glutathione within the active site.