The occurrence of proteolytic activity in ceU-free culture filtrates of some bacteria has given rise to many discussions on the question of whether the presence of proteins or protein derivatives is necessary for the protease formation. Haines (7) reviewed the literature on this subject up to 1931, and based on his own experimental results, concluded that proteases were formed in well grown cultures in simple synthetic media with an inorganic source of nitrogen provided that magnesium and calcium salts were present; the first mentioned would stimulate growth and the latter the protease formation (8). Gotini el a/. (4-6) studied the effect of calcium on the proteinases produced by several bacteria cultivated in nutrient agar and reached the conclusion that it was not essential for the protease formation but only for the function and stability of the enzyme. They supposed that the active form of the bacterial protease was a calcium-protein complex.In this paper evidence is presented (1) that Serratia marcescens proteinase was not found, at least in active form, in a synthetic medium with or without calcium provided that care was taken to avoid high concentrations of bacterial mass; and (2) that carefully washed bacteria from a proteolyticaUy inactive culture in the synthetic medium placed in an aqueous solution of gelatin showed activity after a short period of time. Materiels and MethodsSerratia n~rcescens.--Straln 9986 from the American Type Culture Collection. Synthet/c Med/a.mComposition described at the foot of Table I. Cu/ti~ion.mThe original strain was inoculated on plates of standard synthetic medium solidified with agar and incubated for 21 hours at 29°C. Then the surfaces were washed with sterile saline solution (0.125 ,~ NaC1), and the resulting bacterial suspension collected, centrifuged, and washed twice under sterile conditions. Finally the bacteria were suspended again and by suitable dilution, 10 s cells were inoculated * A preliminary report of cer~in phases of this work was presented at

The use of aqueous alkali for the titration of esterolytic activity when the esters are dissolved in alcoholic solutions, results in an error due to changes in the ionization of the buffer. This is corrected by titrating with alkali in the same solvent as the substrate.Alcohols and other substances which change the dielectric strength of water modify the rate of hydrolysis of BAEE 1 and TSAME by trypsin to an extent proportionate to their effect on the dielectric strength. The reaction rate increases with diminished dielectric strength and v/ce rersa. At low concentrations of substance there seems to be no specific effect other than that derived of the variation in dielectric strength. At higher concentrations, the enzyme might be denatured. In addition, it is probable that specific effects of each substance might intervene. The Coulombic and thermic energies of activation were calculated for the two esters in various solvents. The plot of the logarithm of rate constant rs. reciprocal of dielectric constant yields a straight line with positive slope. This behavior is similar to that of a non-enzymatic positive ion-dipole reaction. Trypsin reacts like a positive ion. The possible influence of the dielectric strength on the regulation of the equilibria involved in the interconversion of the various forms of trypsin in solution (active, inactive, denatured) is discussed. INTRODUCTIONSchwert and Eisenberg (19) hydrolyzed BAME and BAEE in different alcohols at various concentrations with the hope that by varying the solvent it might be possible to differentiate the two supposed steps of the enzymatic reaction (formation and activation of Michaelis complex). Although these investigators did not realize their objective, they did find that hydrolyses of these esters failed to follow zero order kinetics in the presence of alcohols. These solvents gave rise to an increase in the initial rate of hydrolysis to an extent i Throughout this paper the following abbreviations will be used: BAEE, benzoyl-T.-arginine ethyl ester. TSAME, p-toluenesulfonyl-T.-arginine methyl ester. BTEE, benzoyl-~tyrosine ethyl ester. BAME, benzoyl-~arginine methyl ester. ATEE, acetyl-L-tyrosine ethyl ester. 617

The kinetic effects resulting from changes in the medium ionic strength on reactions involving trypsin or a-chymotrypsin are different. The reaction rate increases continuously as the ionic strength increases with ot-chymotrypsin. With trypsin, the rate increases at low ionic strengths but as the ionic strength further increases a gradual inhibitory effect is observed. The effects produced by different salts of various valence types (from uni-univalent to unitrivalent or tri-univalent) are essentially the same, and they are a function of the square root of the ionic strength. The quantitative differences among the various salts may be accounted for on the basis of individual properties of the ions, such as the size of the hydrated ion, "association," etc. The effects of salts on the enzymic reactions described herein are amenable to the same electrostatic treatment applicable to non-enzymatic reactions. By applying Br6nstecl's basic kinetic concepts and the Debye-Hfickel law of electrolyte activity, it appears that the salt effects are mainly due to changes in the dissociation of ionizable groups. This appears to be a general method for analyzing the effect of inorganic ions on enzymic reactions. I N T R O D U C T I O NAlthough it is well known that inorganic ions can have profound non-specific effects on enzymic reactions there is no adequate explanation available. Trypsin and a-chymotrypsin appear particularly suitable for investigating the general phenomena of inorganic ion effects because pertinent information is available, including partial structures of their active centers, known ionizable groups, and the effects of some salts. J a n d o r f in 1950 (17) observed that magnesium sulfate and other salts increase markedly the activity of chymotrypsin zzo 3

The dielectric effects on trypsin and a-chymotrypsin activities have revealed that at pH 7.8 the active species of the former is the cation while that of the latter is the anion. The present study on the dielectric effects along the pH-activity curves shows that trypsin remains positive within the pH range of 5.5 to 8.5. Conversely, a-chymotrypsin is positive from pH 5.5 to 6.6, negative from 6.6 to about 8.1, and at pH 8.25 becomes positive again. The first point of inversion in charge sign shifts from 6.6 to 7.15 with the addition of 0.05 M phosphate buffer. The point of inversion does not seem to be modified significantly by changes in the substrate structure. At pH values near the point of inversion the plots of rate log vs. 100/D are broken lines formed by various straight portions, the slope of each varying progressively from a maximum positive to a maximum negative value. This suggests an effect of resonance possibly attributable to an imidazole group. As an attempt to explain the two observed points of sign inversion in a-chymotrypsin, the possibility is suggested that different enzyme configurations are disclosed by the combined action of pH and dielectric constant. On this theoretical basis, it is feasible that more than one isoionic point exists. I N T R O D U C T I O NIt was reported previously that the rate of ester hydrolysis by trypsin (5) or a-chymotrypsin (6) is modified by changes in the medium dielectric strength. By comparison with non-enzymatic reactions, it might be inferred from the behavior of these two enzymes that at p H 7.8 the active f o r m of trypsin is the cation and that of ot-chymotrypsin is the anion. These observations are in harmony with two earlier findings : (a) Northrop (16) deduced, by D o n n a n equilibrium considerations, that trypsin behaves like a univalent positive