The dependence of the substrate specificity on the conformation of crystalline glutamate dehydrogenase.

Glutamate dehydrogenase (L-glutamate-NAD(P) oxidoreductase EC 1.4.1.3) from bovine liver catalyzes the reversible oxidative deamination of various monocarboxylic amino acids, as well as of n-glutamate (1, 2). Several years ago, we reported that estrogenic steroids both inhibited the oxidation of L-glutamate (3) and stimulated the oxidation of L-alanine (4). Conversely, adenosine diphosphate was found to stimulate the glutamate dehydrogenase reaction (5) (and prevent its inhibit’ion by the steroids (3)), and to inhibit the alanine dehydrogenase reaction (4). The glutamate and alanine activities are reciprocally affected under other experimental conditions as well. It was, therefore, concluded that different conformations of the enzyme catalyzed each of the two different reactions (4). The equilibrium between the forms was thought to be influenced by regulator molecules such as the steroids or adenosine diphosphate. At that time, it was known that the enzyme could undergo a reversible concentration-dependent aggregabion reaction (6). This prompted Frieden to propose (5) that only the most highly aggregated form of the enzyme was active for the glutamate dehydrogenase reaction while the disaggregated molecule (.s~~,~ 12.5) was enzymically inactive. Under our experimental conditions, reagents which favored disaggregation also inhibited the glutamate dehydrogenase and stimulated the alanine dehydrogenase (4). Conversely, reagents which favored enzyme aggregation stimulated the glutamate and inhibited the alanine reaction. We, therefore, concurred with the Frieden proposal that the glutamate-active species was the polymeric form, and we went further to assign the alanine-active species to the monomer (4). Subsequent work has shown that the monomeric form of the protein catalyzes the glutamate reaction (7-Q). Therefore, the state of aggregation of the enzyme does not itself appear to influence enzyme activity. Still, the correlation between activity and state of aggregation remains to be explained, and, accordingly it was suggested that three conformations of the protein in rapid equilibrium could account for the data (8,Q).