Use of transferred nuclear-Overhauser-effect spectroscopy to measure the bound conformation of a disulphide-replaced analogue of glutathione disulphide as an inhibitor of yeast glutathione reductase.

Abstract

The analogue of glutathione disulphide (GSSG) in which the disulphide bridge of GSSG is replaced by -CH2-S- was synthesised from L-cystathionine using t-butoxycarbonyl and t-butyl ester protection with triethylsilane-promoted deprotection. This analogue (GCSG) was found to be a linear, competitive inhibitor of yeast glutathione reductase (Ki value 981 microM at pH 7.0), a very poor substrate and not to act as an irreversible inhibitor of glutathione reductase. The weak binding of GCSG to glutathione reductase permitted the use of transferred nuclear Overhauser effect spectroscopy (TRNOESY) to investigate the bound conformation of GCSG in its complex with glutathione reductase. The solution structure of free GCSG was investigated by NMR spectroscopy using a range of NMR techniques. The TRNOESY experiment allowed a range of conformations to be determined for the central bridge region (containing the -CH2-S- replacement) of GCSG bound to yeast glutathione reductase. Using the nuclear Overhauser effect constraints thus derived, in combination with molecular graphics and energy minimisation based on the known crystal coordinates of glutathione disulphide (GSSG) bound to human erythrocyte glutathione reductase, allowed an explanation of the lack of substrate activity of GCSG, its inactivity as a suicide inactivator and its relatively weak binding in terms of the enforced mislocation of the -CH2-S- bridge with respect to the catalytic residues (relative to GSSG). Thus, the simple replacement of -S- by -CH2-, common in medicinal chemistry, can lead to poor receptor binding if the replacement occurs in a central, rather than peripheral, part of the ligand under modification.