The proteolytic enzymes of the K-1 strain of Streptomyces griseus obtained from a commercial preparation (Pronase). II. The activity of a serine enzyme in 6 M guanidinium chloride.

Abstract A homogeneous serine protease, homologous with bovine chymotrypsin, purified from a commercial preparation (Pronase) of the K-1 strain of Streptomyces griseus demonstrated marked stability in the presence of 6 m guanidinium chloride. In this solvent the enzyme was active against N-α-acetyl-l-tyrosine ethyl ester showing a slight increase in Km but a significant decrease in Vmax as compared to results of studies in the absence of denaturant. The rate of hydrolysis of casein was enhanced in the presence of 6 m guanidinium chloride. Ovalbumin, which was minimally hydrolyzed in the absence of denaturant was extensively hydrolyzed by the enzyme in the presence of denaturant. These results probably reflect the selective unfolding of the substrate proteins by denaturant. The enzyme revealed no difference in circular dichroism spectra in the presence or absence of 6 m guanidinium chloride. Sodium ethylenediaminetetraacetate in the presence of denaturant caused a sustained irreversible loss of activity associated with a change in the circular dichroism spectrum. In the absence of guanidine, minimal effects of the chelating agent were seen. These results suggest the probable stabilizing effect of a cation. We believe this is the first demonstration of conservation of enzymatic activity in the presence of 6 m guanidinium chloride. This protease may be useful for structural studies of polypeptides which are, in the native state, resistant to proteolysis.

A homogeneous serine protease, homologous with bovine chymotrypsin, purified from a commercial preparation (Pronase) of the K-l strain of Sfrepfomyces griseus demonstrated marked stability in the presence of 6 M guanidinium chloride. In this solvent the enzyme was active against N-ol-acetyl-Ltyrosine ethyl ester showing a slight increase in K, but a significant decrease in Vlnax as compared to results of studies in the absence of denaturant. The rate of hydrolysis of casein was enhanced in the presence of 6 M guanidinium chloride.
Ovalbumin, which was minimally hydrolyzed in the absence of denaturant was extensively hydrolyzed by the enzyme in the presence of denaturant. These results probably reflect the selective unfolding of the substrate proteins by denaturant.
The enzyme revealed no difference in circular dichroism spectra in the presence or absence of 6 M guanidinium chloride. Sodium ethylenediaminetetraacetate in the presence of denaturant caused a sustained irreversible loss of activity associated with a change in the circular dichroism spectrum.
In the absence of guanidine, minimal effects of the chelating agent were seen. These results suggest the probable stabilizing effect of a cation. We believe this is the first demonstration of conservation of enzymatic activity in the presence of 6 M guanidinium chloride. This protease may be useful for structural studies of polypeptides which are, in the native state, resistant to proteolysis.
;\ companion report describes the facilitated purification to homogeneity of serine proteases present in a commercial preparation, I'ronase, obtained from the K-l strain of Streplolnyces griseus (1). In the course of studying some of the properties of that enzyme with the greatest affinity for carbosymethylcellulose, the stability of the enzyme in the presence of some denaturants was analyzed. This enzyme with an estimated molecular weight of 17,500 is also an esterase with activity against N-o-acety-L-tyrosine ethyl ester. It had been noted during the course of analytical studies for purity by acrylamide gel electrophoresis at pH 4.3 that this protease demonstrated curious staining properties. Treatment of the gels with either Amido schwarz or Coomassie blue gave the usual coloration of a single protein band.
However, after a period of several days following the application of Amido schwarz or 2 to 3 weeks following the application of Coomassie blue, each gel was almost completely decolorized.
This occurred in an acidic solution where the enzyme is completely inactive. Restaining yielded no further dye retention.
This suggested that the protease had diffused from the gel. A general inquiry revealed that this characteristic had not been previously observed. Therefore, the consideration arose that this enzyme might be unusually resistant to drli;rturntion.

MATERIALS
AY\iD METHODS Pronase (grade B) was obtained from Calbiochem ; several lots (numbers 900053, 000130, and 000333) were used for these studies.
The serine proteases were purified as previously described (I).
Only the enzyme with the greatest affinity for carboxymethylcellulose was utilized in these studies. It was found to be homogeneous by chromatography through carbosymethylcellulose, by gel filtration, and by acrylamide gel electrophoresis.
Ac After cooling the solution in an ice bath, 1) volumes of cold diethyl ether was added (7). The filtered precipitate was dried under vacuum in a desiccator.
The activity against Ac-Tyr-OEt was determined by previously described techniques (8). The kinetic constants toward Xc-Tyr-OEt were determined at substrate concentrations of 5 rnM to 30 InM and enzyme concentrations of 0.015 to 0.03 mg per ml.
The assay medium in the kinetic studies contained 3% diosane to permit the complete solubility of the ester substrate at the higher concentrations.
Titrations were performed at pH 8.0 at 25" with 0.049 M NaOH.
The extent of casein or ovalbumin hydrolysis by the enzyme was determined after incubation at 37" by measuring the amount of supernatant absorbance at 280 nm after precipitation with trichloroacetic acid. Protein substrates in guamdme stood at room temperature for 1 hour to allow time for unfolding before the addition of the protease. Aliquots (1 ml each) of incubation mixtures were utilized for each assay. In order to reduce the differential effect of guanidinium chloride on the assays, the following procedure was done.
To those aliquots without guanidine, 1 ml of 6 M denaturant was added; to those with guanidine, 1 ml of water was added. Immediately thereafter 2 ml of 10% trichloroacetic acid was added to each sample. Each experiment was terminated when no further increase in absorbance at 280 nm was noted or when no material could be precipitated after addition of the trichloroacetic acid solution. In the latter case the optical densities of the soluble peptides gave values close to those espected for the complete solubilization of the parent proteins.
Measurements of circular dichroism were performed on a Cary 60 spectropolarimeter with CD attachment. The instrument was standardized with an aqueous solution of d-lo-camphorsulfonic acid (K and K Laboratories, Plainview, New York, batch 4829), giving [0] = 7260 + 165 deg cm2 per decimole at 290 nm.
Spectra were obtained on solutions in the protein con-centration range of 1 to 2 mg per ml at 25". Determinations of protein concentration were made from the absorbance for which the correspondence is Ej?,,, = 8.1 at 280 nm. This value was determined by the standard technique with the Folin phenol reagent with bovine serum albumin as a standard (9). Concentrated enzyme samples were first prepared in 4 mM Tris (pH 8.0) containing 4 mM CaCL. Solutions of guanidinium chloride (6.3 M) or urea (8.4 M) in the same buffer were added to separate aliquots of protein solution for a final guanidine concentration of 6.0 M and a final urea concentration of 8.0 M. Fig. 1 demonstrates the results of a study of the stability of the enzyme in 6 M guanidinium chloride.

RESULTS
The incubations were carried out at the pH of maximal activity for this enzyme. Retention of activity was noted and increasing calcium ion concentrations appeared to protect against slight loss of esterase activity.
Therefore, these results could mean either the unusual retention of the native state in this denaturant or refolding of completely denatured enzyme upon dilution in the guanidinefree assay medium.
The significant finding was the effect of sodium EDTA.
With excess chelating agent there appeared a moderately rapid loss of activity. The loss of activity was not reversible if excess calcium chloride was added to an aliquot with a l-hour incubation before assays against Ac-Tyr-OEt; nor was the activity restored when an aliquot was diluted into a guandine-free solution containing excess calcium ion at pH 8.0. These findings suggested that in the absence of EDTA the enzyme retained conformational stability in the denaturant. It was assumed without further evidence that calcium ion is the only cation involved in the stabilization process. Very slight inhibition of this enzyme by EDTA was noted in the absence of denaturant.
A separate experiment was carried out under conditions similar to that listed in Fig. 1   The decrease in maximal velocity probably represents diminished activity of most of the enzyme molecules rather than maximal activity of a small fraction of the molecules, because maximal activity returned after dilution of the enzyme into guanidine-free solutions.
It would be unlikely that this represented extensive reversible renaturation of most of the molecules for one would expect unfolded species to be susceptible to rapid autolysis. Fig. 2 demonstrates the effect of the enzyme on casein in 6 M guanidinium chloride.
The rate of proteolysis is apparently greater in the presence than in the absence of denaturant.
No difference is noted in the extent of proteolysis. This rate enhancement is probably not due to a general salt effect.
In the presence of 2 x KC1 the casein solution became more opaque during the reaction with the Pronase enzyme and a diminished apparent rate of hydrolysis was noted. If the velocity of peptide bond cleavage was reduced in 6 M guanidine to a degree similar to the reduction of e&erase rate, it follows that the increase in rate of hydrolysis must reflect a much greater accessibility of substrate to enzyme in denaturant.
.L similar effect of guanidine on the hydrolysis of casein by streptococcal proteinase has been reported (10). Fig. 3 demonstrates the effect of the enzyme on ovalbumin. By 2 hours the extent of apparent proteolysis in denaturant is 35 times greater than that achieved in the absence of guanidine.
In fact minimal apparent digestion is seen without denaturant. Crude Pronase, a misture of many proteases, hydrolyzes native ovalbumin extensively (II), an observation differing from that of the present study with the purified single component.
In corltrast to the study with casein, a protein considered to be ramdomly coiled in the absence of denaturant, the proteolysis of ovalbumin necessarily first required the unfolding of substrate. Fig. 4 depicts the circular dichroism of native and denatured enzyme.
In Tris buffer, 8 M urea or 6 M guanidinium chloride solution the spectrum of the native enzyme is unchanged. This supports the view that the diminished e&erase activity in guanidine does not reflect gross denaturation of a large fraction of the enzyme molecules.
However, undetected minor conformational changes could account for the difference in the activity against Ac-T-r-OEt.
Enzyme activit'y decreases over a Y-hour period when EDT.4 is added to the guanidilGum chloride solution. This loss of enzyme activity corresponds in part to changes seen in the CD spectrum with the 215 nm minimum becoming less pronounced by 10 to 15%.
However, CD spectral changes continue to occur subsequent to the diminution in enzyme activity. Over in the enzyme aliquot. The total volume in each assay was 3.04 ml. a 24.hour period the CT) spectrum of the enzyme in the guanidinium chloride and EDTA solution converts to the dashed curve in Fig. 4. This study excludes the remote possibility that this protein has a random conformation in the native state, a condition that would be highly unlikely for a proteolytic enzyme. This stability is presumably dependent upon some cation since sodium EDTA causes moderately rapid irreversible denaturation.
It has been suggested that 6 to 8 M guanidinium chloride may not be sufficient to effect a conformational change for exceptionally stable proteins (12). This consideration is apparently borne out by the present example. It has been noted that block copolymers containing either polyleucine or polyphenylalanine retain helical stability in 7.2 M guanidine (13).
In the present study, the CD curve of the native enzyme demonstrates very little evidence of helical structure. The shoulder at 230 nm is not particularly sensitive to denaturitlg conditions, a feature suggesting that this band belongs primarily to :t nonhelical contribution to the optical activity. Therefore, other structural features must be considered to explain the present example of stability.
The following meager facts are known about the structure of this protein.
Like bovine chymotrypsin there is the Asp-Ser-Gly sequence around the reactive serine residue (14) Five to 6 half-cystine residues are found and since preliminary studies reveal IIO free sulfhydryl groups, these are probably fully incorporated into disulfide bonds. There could not be a very large number of these bonds to account for the marked stability.
Several studies have been done on the stability of chymotrypsin in urea and guanidinium chloride (17)(18)(19). It was demonstrated that the bovine enzyme was stable at low concentrations of either denaturant.
However suggests that the enzyme is structurally less rigid in denaturant but that the susceptibility to proteolysis is only slightly increased.
If autolysis is primarily an intermolecular event this differential effect between chelating agent and another protease molecule would not be surprising.
For, a small molecule such as EDT.4 in contrast to a protein molecule might easily be accommodated by t,he interstices of a slightly unfolded enzyme.
It is to be noted that calcium ion was present at a concentration of at least 5 in&f at every step of purification of this protease.
Preliminary studies in this laboratory of the Prollase enzyme have not revealed marked thermal stability. The property of significant retention of the native state of this enzyme in urea or guanidine may be of use in the structural studies of some polypeptides.
Certain problems may arise \vith the standard procedures for the hydrolysis of proteins by specific endopeptidases. Initial preparation of the substrate usually includes some form of denaturation since native proteins are generally resistant to proteolysis.
These techniques for irreversible denaturation may result in the cleavage of certain covalent bonds such as the oxidation of disulfide bonds. In addition, proteolysis of the denatured protein in many cases is incomplete leaving an ii,soluble residue.
The use of this enzyme with denaturant should in principle by-pass these two complications. The specificity of this protease is under investigation to determine its applicability in sequence studies.
An importalIt, consideratioll will be the determination of conservation of specificity in denaturant.