Proteolysis as a probe of ligand-associated conformational changes in rat carbamyl phosphate synthetase I.

Elastase, V8 protease, subtilisin, trypsin, and chymotrypsin all cleaved the 1462-residue polypeptide of rat carbamyl phosphate synthetase I in segment C 160-180 residues from the COOH-end. Its activator N-acetylglutamate (AcGlu) increased the rate of cleavage approximately ninefold, presumably by binding preferentially to the conformation in which C is exposed. ATP/Mg2+ prevented proteolysis both +/- AcGlu. Kd,app for AcGlu (66 microM) and ATP (4.2 microM with AcGlu and 5 mM Mg2+) was estimated from the pseudo-first-order rate constants for inactivation caused by cleavage with elastase at C. Chymotrypsin and trypsin also hydrolyzed the enzyme, independent of AcGlu, at site D within less than 20 residues of the COOH-end. D was protected by ATP only in the presence of AcGlu and K+, and enzyme hydrolyzed exclusively at D had greater than 30-fold higher Km's for AcGlu and ATP. Digestion by trypsin at a third site (B) approximately 530 residues upstream from C appeared to occur subsequent to hydrolysis at C. Slow cleavage by elastase at an additional site (A) to give 360- and 1100-residue peptides was unaffected by AcGlu and ATP, and caused only modest loss of activity. These peptides were isolated by chromatography on DEAE-cellulose. Assignment of the smaller one to the NH2-end on the basis of its cysteine content places site A in the junction between the segments homologous to the small glutaminase and large synthetase subunits of Escherichia coli carbamyl phosphate synthetase II. Neither peptide alone was active; maximal regain of activity (approximately 25%) occurred on combining them in equimolar proportions. The sizes of the peptides produced by further digestion of the site A digest gave the approximate locations of the other sites. Sites A (Ala-417) and B (Arg-787) have recently been identified by NH2-terminal sequencing (S. G. Powers-Lee and K. Corina (1986) J. Biol. Chem. 261, 15349-15352). Reasons for the low value of KAcGlu,app are examined, and protection by ATP is discussed in relation to previous models for the conformational equilibria of the enzyme.