DETERMINANTS OF COENZYME SPECIFICITY IN GLYCERALDEHYDE-3-PHOSPHATE DEHYDROGENASE - ROLE OF THE ACIDIC RESIDUE IN THE FINGERPRINT REGION OF THE NUCLEOTIDE-BINDING FOLD

On the basis of the three-dimensional structure of the glycolytic NAD- dependent glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and of sequ ence comparison with the photosynthetic NAD(P)-dependent GAPDH of the chloroplast, a series of mutants of GAPDH from Bacillus stearothermoph ilus have been constructed. The results deduced from kinetic and bindi ng studies suggest that the absence of activity of the wild-type GAPDH with NADP as a cofactor is the consequence of at least three factors: (1) steric hindrance, (2) electrostatic repulsion between the charged carboxyl group of Asp32 and the 2'PO4, and (3) structural determinant s at the subunit interface of the tetramer. The best value for k(cat)/ K(M) and K(D) for NADP was observed for the D32A-L187A-P188S mutant. T his triple mutation leads to a switch in favor of NADP specificity but with a k(cat)/K(M) ratio 50- and 80-fold less than that observed for the wild type with NAD and for the chloroplast GAPDH with NADP, respec tively. Substituting the invariant chloroplastic Thr33-Gly34-Gly35 for the B. stearothermophilus Leu33-Thr34-Asp35 residues on the double mu tant Ala187-Ser188 does not improve significantly the affinity for NAD P while substituting Ala32 for Asp32 on the double mutant does. Clearl y, other subtle adjustments in the adenosine subsite are needed to rec oncile the presence of the carboxylate group of Asp32 and the 2'-phosp hate of NADP. Kinetic studies indicate a change of the rate-limiting s tep for the mutants. This could be the consequence of an incomplete ap o-holo transition. The results taken all together suggest that it is d ifficult from a rationale to predict all the structural determinants t hat are implicated in determining specificity for a coenzyme. Comparat ive studies of the binding properties of various mutants suggest an es sential role of position 32 and more generally of the adenosine subsit e for revealing negative or positive cooperativity in GAPDH. Furthermo re, it is suggested that a correct positioning of the pyridinium ring is necessary but not sufficient for inducing cooperativity.