Proceedings of the National Academy of Sciences Vol. 68, No. 2, pp. 257-259, February 1971 The Comparative Specificity of Acid Proteinases IRENE M. VOYNICK AND JOSEPH S. FRUTON Kline Biology Tower, Yale University, New Haven, Cown. 06520 Communicated November 16, 1970 ABSTRACT Examination of the kinetic parameters for the site of attack by such a reagent (10). A reactive aspartyl the hydrolysis, by acid proteinases, of a single peptide bond residue has also been identified as part of the active site of (between p-nitro-L-phenylalanyl and L-phenylalanyl) in a series of oligopeptides has shown that secondary interac- penicillopepsin (6). Although further studies are needed, the tions are important factors in determining the catalytic available information suggests that in the action of pepsin efficiency. Comparison of the action of highly purified (and possibly of other acid proteinases) an imino-enzyme pepsinlike enzymes (Rhizopus proteinase, Mucor protein- intermediate is involved in the catalytic mechanism; it has ase, rennin) with that of swine pepsin A indicates signifi- cant differences among them, either in the binding of the been suggested that the active site contains a carboxylate substrate (as estimated by K.), or in the catalytic efficiency group acting as a nucleophile and another carboxyl group (as measured by kcat), or both. It may be concluded from (in its protonated form) acting as a proton donor (11, 12). these data that, in their action on oligopeptide substrates, Whereas the serine proteinases exhibit significant differ- the specificity of proteinases operating by a similar cata- ences in specificity with respect to the amino acid residue, lytic mechanism cannot be explained solely in terms of the in amino acid residues flanking the sensitive peptide bond; the substrate, which provides the carbonyl group of the sensi- in addition, the specificity includes significant contribu- tive peptide bond, the known acid proteinases appear to have tions from secondary interactions arising from comple- similar (although not identical) preference for hydrophobic mentary relations between parts of the substrate and of amino acid units flanking the sensitive bond (13). Thus, the the enzyme at a distance from the catalytic site. Data are also presented for the effect of urea (about 1 M) on the available data on the cleavage of the oxidized B chain of kinetic parameters of several acid proteinases; under the insulin by pepsin (14), rennin (3), and the acid proteinases of conditions of these studies, the binding of the substrate is Rhizopus (15) and Mucor (16) show, in all cases, a preferential affected to a much lesser degree than is the catalytic effi- attack at Leu-Tyr (15-16), Tyr-Leu (16-17), Phe-Phe (24-25), ciency. and Phe-Tyr (25-26), although differences have been noted Many of the known proteinases have been classified into in the apparent susceptibility of other peptide bonds. distinct types, according to the nature of the enzymic groups Despite their similarity of action on the B chain of insulin, identified with the catalytic process (1). The so-called serine however, several of the known acid proteinases have been proteinases, which include chymotrypsin, trypsin, elastase, reported to be relatively inactive toward simple synthetic and subtilisin, are characterized by the presence of a reactive substrates for crystalline swine pepsin A. Recent work has seryl residue and a histidyl residue at the catalytic site. In provided a large number of new pepsin substrates, some of the cysteine proteinases (papain, ficin, streptococcal pro- which are cleaved at a rapid rate, and conclusions have been teinase, etc.), a cysteinyl residue and a histidyl residue are drawn concerning the specificity of the action of this enzyme involved. These enzymes act optimally on proteins and on at peptide bonds (13). In particular, it has been found that synthetic peptide substrates at pH values in the range 5-9, with small synthetic substrates of the type A-X-Y-B, where and the available evidence is consistent with the intermediate X and Y are -amino acid residues flanking a sensitive peptide formation of an acyl-enzyme. In addition, there is a sizeable bond, pepsin prefers a phenylalanyl residue in the X position group of enzymes whose optimal action on protein substrates and a tryptophyl, tyrosyl, or phenylalanyl residue in the Y is in the pH range 2-5, and that have been termed "acid position; other hydrophobic amino acid residues also promote proteinases." The best-known member of this group is swine pepsin action, but to a much lesser degree (17). Moreover, it pepsin A (2); others are rennin (3), cathepsin D (4), and the has been found that the rate of pepsin action at the Phe-Phe acid proteinases of several molds, including Rhizopus chinensis [or Phe(NO2)-Phe] bond of substrates of the type A-Phe- (5), Penicillium janthinellum (penicillopepsin (6)), and AMucor Phe-B [or A-Phe(NO2)-Phe-B] is greatly influenced by the miehei (7). Besides pepsin, several of these enzymes have nature of the A and B groups on either side of the sensitive been prepared in crystalline form (3, 5, 6). dipeptidyl unit (18, 19). These effects have been interpreted A common feature of the acid proteinases appears to be as providing evidence for the view that the "secondary" in- their inhibition by diazo compounds (8, 9); in the case of teractions of the A and B groups of the substrate with com- pepsin, the carboxyl group of an aspartyl residue in the se- plementary enzymic groups relatively distant from the cata- quence Ile-Val-Asp-Thr-Gly-Thr-Ser has been identified as lytic site may alter the conformation of the enzyme in a manner that affects the efficiency of catalysis. It appeared likely, therefore, that the apparent discrepancy between the similar- Abbreviations: In addition to the usual abbreviations for L-amino ity of the action of the acid proteinases on the B chain of in- acid residues, the following have been used: Phe(NO2), p-nitro-I. sulin and the wide differences among them in the cleavage of phenylalany ; Z, benzyloxycarbonyl; OMe, methoxy. small synthetic substrates might be related to the effect of 257 Downloaded by guest on October 2, 2021 258 Biochemistry: Voynick and Fruton Proc. Nat. Acad. Sci. USA such secondary interactions. The present experiments were kcat/Km for the three substrates with B = OMe, Ala-OMe, undertaken to examine this possibility. and Ala-Ala-OMe were roughly similar to those obtained MATERIALS AND METHODS with pepsin, both kcat and Km being lower. One of the possible explanations for this difference in the kinetic parameters was obtained Swine pepsin (twice crystallized, lot PM693-7) observed with pepsin and the Rhizopus proteinase is that, in from the Worthington Biochemical Corp. When assayed in the latter case, nonproductive interaction makes a relatively the usual manner (20), with hemoglobin as the substrate, the larger contribution to the binding of the substrates (19, 23). proteinase activity of this enzyme preparation was 2700 That such nonproductive interaction may involve the ben- units/mg of protein. The acid proteinase of Rhizopus chinensis zyloxycarbonyl group is suggested by the finding that when (3 times crystallized) was obtained from the Miles Labora- A = Phe-Gly-His, the Km values for the two substrates tories, and had a specific activity of 870 units/mg of protein in tested with both pepsin and the Rhizopus proteinase are the the above assay (at pH 3.8, a similar activity was noted). same, although k0,t is much lower in the latter case. In con- The acid proteinase of Mucor meihei (7) was generously trast to the Rhizopus proteinase, the enzyme from Mucor provided by Dr. M. Ottesen. Crystalline rennin was prepared cleaves the substrates with A = Z-His much more slowly from commercial rennet powder according to the method of than does pepsin, the Km values being much higher; the effect Berridge (21); material generously provided by Dr. B. of the replacement of B = OMe by Ala-OMe on the value of Garnier was used for seeding, and greatly Foltmann and Dr. J. kcat/Km is similar, however, to the results for pepsin and the accelerated the preparation. The specific activity (milk- Rhizopus proteinase. With the Mucor proteinase, the re- clotting assay (22)) was 60 units/mg of protein. placement of A = Z-His by Phe-Gly-His exerts a markedly The synthesis of the oligopeptides used as substrates in unfavorable effect on kcat; thus the substrate with B = OMe the present study has been described previously, as has the was not cleaved at a measurable rate at the highest enzyme spectrophotometric method employed for the determination concentration (13 ,M) tested, and the substrate with B = of the initial rates of cleavage of the Phe(NO2)-Phe bond in Ala-Phe-OMe was cleaved at about 0.01% of the rate found such substrates (19). In all cases, the data were obtained at with pepsin. pH 4.0 (0.04 M formate buffer) and 370C, and accorded with The data in Table 1 for the action of rennin on pepsin sub- Michaelis-Menten kinetics over the range of substrate con- strates show that, of those tested, only those with B = Ala- centration (0.02-0.4 mM) employed; 5-8 runs were per- OMe, Ala-Ala-OMe, and Ala-Phe-OMe were cleaved at the formed for each determination of Km and of kcat (Vm per Phe(NO2)-Phe bond to a measurable extent under the con- molar equivalent of enzyme); the enzyme concentration ditions of these studies. With B = Ala-OMe, the value of (0.003-5.4 ,uM) was chosen to give reliable data for the initial Km was too high to permit an estimate of kcat to be made, and rate of hydrolysis.