PubMed1995
The active site base, Lys-258, of Escherichia coli aspartate aminotransferase (AATase) was mutated to a unique cysteine for chemical modification in a cysteine-free AATase [Quint; Gloss et al. (1992) Biochemistry 31, 32-39]. Two homolysine analogues (gamma-thiahomolysine and gamma-dithiohomolysine) and the carboxylate series [glutamate, (carboxymethyl)cysteine, and (carboxyethyl)cysteine] were prepared. The kcat and kcat/KM values for amino acids exhibited by the 258-homo-Lys AATases are ca. 1000-fold less than those of Quint. However, the kcat/KM values for keto acids are only reduced by 2.5-10-fold. The 258-side-chain carboxylates bind amino acids 1-4 orders of magnitude more tightly than Quint, but have first-order catalytic rate constants 5-7 orders of magnitude less than Quint. The 258-homo-Lys AATases stabilize the E.PLP vs the E.PMP species by approximately 3 kcal/mol relative to Quint; the 258-side-chain carboxylate variants stabilize E.PMP vs E.PLP by 5.8 kcal/mol relative to WT. The C alpha-2H kinetic isotope effects for the reactions of 258-homo-Lys and 258-side-chain carboxylate AATases demonstrate that transaldimination is equally or more rate-determining than C alpha proton abstraction for both sets of enzymes. The values of kcat/KM calculated for the 258-side-chain carboxylate variants from the Brønsted plot of Toney and Kirsch [(1989) Science 243, 1485-1488] are in close agreement with those observed. This "tethered" Brønsted analysis shows that the earlier reported poor reactivity of carboxylates in chemical rescue is due to electrostatic exclusion from the active site.
Aspartate Aminotransferases↗