Denaturation and reactivation of dimeric human glutathione reductase--an assay for folding inhibitors.

Human glutathione reductase (GR; which catalyzes the reaction NADPH + GSSG + H+ --> 2 GSH + NADP+) is an obligatory FAD-containing homodimer of known geometry. Native human GR, a potential target of antimalarial and cytostatic agents, cannot be dissociated by dilution or by means of subunit-interface mimetics, similarly to well-studied viral dimeric proteins. However, ab initio folding and/or dimerization of human GR can be inhibited by point mutations or by peptides corresponding to subunit-interface areas, for example synthetic peptide P11, which represents the intersubunit-contact helix H11. The structure of this peptide, which might assist inhibitor design, was solved by high-resolution NMR spectroscopy. Residues 440-453, were found to be alpha helical in the isolated peptide. To quantitate the efficacy of inhibitors such as P11, we developed the following unfolding/reactivation assay. The effects of various guanidine hydrochloride (Gdn/HCl) concentrations were studied by analytical ultracentrifugation. It was shown that human GR denatured by greater than 3 M Gdn/HCl is monomeric and free of FAD. Circular-dichroism experiments at 223 nm indicated a half-life of approximately 20 s at 20 degrees C for the unfolding process. To optimize the reactivation yield, four parameters [protein concentration (x) in the range 0.3-10 microg/ml, cofactor supplementation, temperature (y: 0-32 degrees C), and time (0-72 h)] were varied systematically, and a reactivation score z was given to each constellation of parameters. This type of analysis might be useful to optimize refolding and activation yields for other proteins. For human GR, the highest recovery was found not to occur at one of the corners of the x,y plane, but close to its center. Consequently, the optimal assay conditions for folding and dimerization inhibitors are as follows. The enzyme (at 300 microg/ml) is denatured by 5 M guanidine hydrochloride/5 mM dithiothreitol, then reactivated by dilution to 1 microg/ml at pH 6.9 and 20 degrees C. In the absence of inhibitors, this procedure leads to 70% of the control activity within 8 h. Peptides representing the upper subunit interface (for instance residues 436-478) of human GR were found to inhibit refolding with EC50% values in the micromolar range, whereas fragments from other regions of the protein had no influence on this process. For peptide P11, the EC50% value was 20 microM. In conclusion, hGR, enzyme with a tight intersubunit contact area of 21 nm2, appears to be suitable for studying protein folding, dimerization, and prosthetic-group complexation in the absence and presence of compounds that inhibit these processes. There is a shortage, at least for oligomeric enzymes of eukaryotes, of published systematic studies on protein (re)activation.