EFFECTS OF GLN102ARG AND CYS97GLY MUTATIONS ON THE STRUCTURAL SPECIFICITY AND STEREOSPECIFICITY OF THE L-LACTATE DEHYDROGENASE FROM BACILLUS-STEAROTHERMOPHILUS

The L-lactate dehydrogenase of Bacillus stearothermophilus (BSLDH) is a thermostable enzyme with considerable potential for applications in asymmetric synthesis. An understanding of the factors controlling its structural specificity and stereospecificity is therefore of interest. In this paper the effects of Gln102 --> Arg and Cys97 --> Gly mutations have been evaluated. In a survey of thirteen 2-keto acids, the Q102R mutation was found to reduce the activity of BSLDH toward the reduction of RCOCOOH substrates with small or hydrophilic R groups without affecting its activity toward those with larger, hydrophobic R substituents. In addition, the mutants have a high affinity for C3- and C4-2-keto dicarboxylates. The extent of fructose 1,6-diphosphate activation of the mutant enzymes was similar to its effect on wild-type BSLDH. The mutants also retained the synthetically desirable thermostability. As a probe of the commitment of BSLDH to L stereospecificity, the Q102R mutation was introduced to allow the new 102R site to compete with Arg171 for binding of the COO- groups of the RCOCOOH substrates, which would reverse the normal RCOCOOH orientation at the active site and thereby open up the possibility of the formation of a D-2-hydroxy acid in place of the natural L product. However, L stereospecificity in 2-keto acid reduction was strictly retained by the Q102R mutants. This was confirmed by preparative-scale reductions of pyruvate and phenylpyruvate to give the corresponding L-hydroxy acids in enantiomerically pure form and by comparison of the kinetics of oxidation of L- and D-lactate and L- and D-phenyl lactate. No evidence for substrate activity for the D enantiomers of either of these was seen with WT or mutant enzymes. Some catalysis of D-malate oxidation by both WT and mutant BSLDH was observed, but the L enantiomer was still preferred to approximately the same degree in both cases. That the inability of BSLDH and its 102R mutants to catalyze D-2-hydroxy acid oxidations was not simply due to the failure of the D enantiomers to bind at the active site was established by a comparison of competitive inhibition constants for the above L- and D-hydroxy acids. CD spectroscopy showed that the Gln102 --> Arg mutations were not benign but induced significant structural perturbations. Electrostatic potential contours suggest that the structural changes are partly due to long-range interactions of the positive charge of the guanidinium group of Arg102 with several other residues that form an area of negative potential adjacent to the active site. The Cys97 --> Gly mutation, while inadvertent, was of interest because of the potential specificity effects arising from its location adjacent to the hinge of the flexible 98-110 loop. However, its effects on BSLDH specificity were minor.