br Discussion In the present paper it was

Discussion In the present paper it was shown for the first time that the conjugation of ethacrynic C527 and glutathione catalyzed by GSTP1-1 stereospecifically forms one of the diastereoisomers of the glutathione conjugate (EASG). Chemical conjugation results in formation of a mixture of both diastereoisomers with a slight, but not significant preference for diastereoisomer B (ratio A:B=46:54). From the human GSTs of the alpha class, only GSTA1-1 appeared to be stereoselective and to form diastereoisomer A preferentially. The fact that only GST A1-1 seems to be stereoselective in the conjugation of ethacrynic acid is quite remarkable as GSTA1-1 and GSTA2-2 only differ in 11 amino acids, three of which are in the hydrophobic binding site (as reviewed in [17]). Human GSTM1a-1a has a very low specific activity [2], so it was impossible to test its EA conjugation to an extent significantly above the rate of the chemical background reaction in this study. Furthermore, Ploemen et al. [2]already showed that rat GST3-3 was not stereoselective towards EA conjugation. The GSTP1-1 mutant, C47S, was demonstrated to be stereoselective towards EA conjugation in the same way as GSTP1-1, so it can be concluded that the cysteine 47 residue does not influence the stereoselectivity of the enzyme. This result is in line with previous conclusions that the amino acid does not influence the catalytic mechanism [18]. Since the crystallization of the GST isoenzymes, numerous studies have been performed, characterizing the three-dimensional structures of GST isoenzymes complexed with substrates and/or GSH (conjugates) 19, 20, 21, 22, 23, 24. For both GSTA1-1 [23]and GSTP1-1 [24]a three-dimensional structure of the enzyme in complex with EA and an EASG diastereoisomer has been described and deposited in the Brookhaven PDB databank. However, for both these GST-EASG complexes arbitrarily only one of both EASG diastereoisomers is depicted in complex with the enzyme. The experimental electron density is reported to be not sufficiently well defined to be conclusive as to whether the EASG bound to the enzyme can be identified as the R or S diastereoisomer. As a consequence the authors have apparently arbitrarily added only one of the two diastereoisomers to their pdb file. However, it is interesting to notice that in the pdb file of GSTA1-1 the R-form is included whereas the GSTP1-1 pdb file contains the S-form. This is remarkable taking into account the results of the present study showing that both isoenzymes actually form the same diastereoisomer. For the glutathione conjugation of 7β,8α-dihydroxy-9α,10α-oxy-7,8,9,10-tetrahydrobenzo[a]pyrene, Hu et al. [25]suggest that the enantioselectivity of GSTP1-1 may be predicted by the structure of the active site. For the glutathione conjugation of ethacrynic acid, such a prediction on the basis of a crystal structure will be more difficult. This becomes clear when looking in detail at the possible mechanisms for glutathione conjugation of ethacrynic acid as a means to find a clue to whether the R or the S isomer is the product to be expected. The diastereoisomeric products are formed by the protonation of C9 of the ethacrynic acid-glutathione intermediate, generating a chiral center, in principle after the initial attack of the sulfur atom from glutathione (Fig. 2). It is very interesting to notice that only one diastereoisomer is actually formed as the proton is usually assumed to arise from the solvent or from a pool of protons that readily exchanges with the solvent [9]; the enzyme thus determines the direction of this protonation. As is shown in Fig. 1B, an intermediate compound is formed during the formation of the actual conjugates, which is most likely the enol tautomer of the glutathione conjugate. The fact that this intermediate is only detected in the chemical conjugation reaction implies that in the enzyme catalyzed reaction the protonation of the C9 takes place in the active site of the enzyme.