The carboxypeptidase inhibitor from Russet Burbank potatoes was subjected to a variety of chemical modifications and their effects on inhibitory activity toward carboxypeptidases A and B were determined. The importance of the alpha carboxylate of glycine-39 to the enzyme-inhibitor interaction was demonstrated by the observation that a derivative in which all four carboxyls were modified was inactive whereas a derivative in which only the beta carboxylates of aspartic acid residues 5, 16, and 17 were masked retained full inhibitory activity. In addition to these three aspartic acid residues, lysine residues 10 and 13, histidine residues 3 and 15, and arginine-32 were modified and residues 1-5 removed with little effect on inhibitory activity. Tryptophan residues 22 and 28 did not react with 2-hydroxy-5-nitrobenzyl bromide or o-nitrophenylsulfenyl chloride, and thus are presumed to be buried in the interior of the inhibitor molecule. Although tyrosine-37 was acetylated without affecting binding characteristics, both carboxypeptidases A and B protected against deacetylation by hydroxylamine. These studies indicate that the carboxyl terminal region of the inhibitor is in contact with enzyme in the complex. The parallel effects of modifications on inhibitory activity toward carboxypeptidases A and B support previous evidence that both enzymes utilize the same binding site on the inhibitor [C. A. Ryan (1971), Biochem. Biophys. Res. Commun. 44, 1265].

The mechanism of action of a carboxypeptidase inhibitor from potatoes has been probed by studying its interaction with derivatives of carboxypeptidase A containing modified residues at the active site. Arsanilazocarboxypeptidase A, a derivative containing a chromophore attached to tyrosine 248, exhibits a circular dichroism spectrum which is sensitive to the presence of ligands at the active site (Kagan, H.M., and Vallee, B.L. (1969), Biochemistry 8, 4223). Since the spectral change attending binding of the carboxypeptidase inhibitor to arsanilazocarboxypeptidase A is similar to that produced by small substrates and inhibitors, the enzyme-inhibitor interaction also involves the enzyme active site. Catalytic activity is not required for inhibitor binding. Complexes of the inhibitor with apocarboxypeptidase A anc carboxypeptidase A which was inactivated by treatment with the affinity label, N-bromoacetyl-N-methyl-L-phenylalanine, are demonstrated by gel filtration experiments. Morever, competitive binding studies reveal that the latter derivative, in which the binding pocket is presumably blocked by reagent, binds inhibitor nearly as strongly as does the native enzyme, and differences in free energy of association being only 0.4 kcal/mol of a total binding energy of - 11 kcal/mol. A model is proposed to account for both the tight binding of inhibitor to the N-bromoacetyl-N-methyl-L-phenylalanine derivative and the involvement of the active site of arsanilazocarboxypeptidase A. It is suggested that the inhibitor fits into a shallow depression at the active site of the enzyme but does not penetrate into the binding pocket.