Cellular regulation of glutamine synthetase activity in Escherichia coli

Abstract

Control of glutamine synthetase activity in E. coli is achieved by at least 4 separate mechanisms: (1) repression and depression of enzyme synthesis; (2) cumulative feedback inhibition by 8 end products of glutamine metabolism; (3) modulation of the kinetics, divalent ion specificity and feedback effector responses by enzymic adenylylation and deadenylylation of glutamine synthetase; (4) modulation of glutamine synthetase activity by variations in the ratios of ATP, Mn ++ and other nucleoside triphosphates. Hydrodynamic measurements and electron microscopic examination show that the enzyme has a molecular weight of 600,000±8% and is composed of 12 apparently identical subunits (M.W. 50,000) arranged in two hexagonal rings, one layered upside down on top of the other. With removal of divalent ions, the enzyme is converted to an inactive “relaxed” state that is susceptible to disaggregation by various treatments. Addition of Mn ++ to disaggregated subunits causes reaggregation and transient restoration of activity. Kinetic studies and direct binding measurements indicate that each subunit possesses separate binding sites for each of the 8 feedback inhibitors and 3 substrates. Two extreme forms of the enzyme exist; Enzyme I which contains little or no covalently bound AMP and Enzyme II which contains 12 equivalents of covalently bound AMP per mole (i.e., 1 per subunit). The glutamine synthetase activity of I specifically requires Mg ++, whereas II requires Mn ++. Under standard assay conditions, the γ-glutamyl transferase activity of I is about the same as that of II, but the glutamine synthetase activity of 1 is 3 times that of II. Histidine, tryptophan and CTP strongly inhibit both transferase and synthetase activities of II, but do not inhibit I. AMP inhibits the synthetase activity of I, but it activates I transferase activity; the converse is true with enzyme II. Alanine and glycine both inhibit I more strongly than II. An enzyme catalyzing the ATP-dependent conversion of I to II was partially purified from cell free extracts. This enzyme is activated by glutamine and is inhibited by glutamate and may be identical with the inactivating enzyme previously described by Mecke, Wulff and Holzer (23). Almost complete de-adenylylation of II and its conversion to a form similar to I is catalyzed by snake venom phosphodiesterase. There is evidence also for in vivo conversion of II to I. The in vivo production of 1 is strongly favored by growth on media containing limiting amounts of NH 4 + as the sole nitrogen source, whereas the production of II is favored when glutamate is the sole nitrogen source. However, depending upon the age of the culture and other conditions not yet identified, enzymes containing intermediate levels of covalently bound adenyl groups are produced. These intermediate forms exhibit marked differences in their kinetic behavior and in their responsiveness to various feedback inhibitors. Of particular interest is a form in which both biosynthetic and transferase activities are subject to cumulative feedback inhibition by the eight different end products of glutamine metabolism. A model is described to explain the cumulative effects of mixtures of inhibitors as well as the partial inhibition obtained with saturating concentrations of each.