'1 -“.‘..‘.‘. - to... .. circa-our. ".4 .-~;-- -“—--" .mm W ‘U *1 ‘W 1 J ,_ -_.::.- m 1.. ‘,..-..‘.a--¢J000~ —_.. -.-—.-.-. #- ABSTRACT PARTIAL PURIFICATION AND CHARACTERIZATION OF AN INTRACELLULAR PROTEASE FROM STREPTOCOCCUS DURANS By Donald L. Wallace An intracellular protease from Streptococcus durans was isolated, partially purified and characterized. Puri- fication entailed fractionation of the protein with 37 to 55% ammonium sulfate, heating at 80 C for 10 min and elution from hydroxylapatite gel with pH 7 sodium phos- phate buffer concentrations ranging from 0.1 to 0.2 M. A 67-fold purification was obtained with retention of 53% of the original activity. Purification of the protease was complicated by the presence of nucleic acids which could not be removed under the conditions employed without loss of proteolytic activity. Ribonuclease inactivated the protease but attempts to reactivate the protease by adding ribonucleic acid extracted from §4 durans were unsuccess- ful. Disc electrophoresis of the enzyme in polyacrylamide gel showed one protein zone with an adjoining nucleic acid zone. Proteolytic activity was determined by measuring the quantity of material soluble in 5% trichloroacetic acid (TCA) after the protease had acted for l min at 37 C on 0.5% casein suspended in 0.0“ M sodium phosphate buffer at Donald L. Wallace pH 7.5. The quantitation of material soluble in TCA was made by reading the change in absorbance at 280 mu on a spectrophotometer. The protease exhibited normal reaction kinetics with respect to substrate concentration, enzyme concentration and reaction rate. The proteolytic activity was affected by the ionic strength of the assay reaction mixture and an ionic strength of 0.1 was optimal. Proteolytic activity was optimal at pH 7.5 with secondary activity occurring at pH 5.5 to 6. Calcium (++), iron (++,+++), cobalt (++), nickel (++), copper (++), magnesium (++), zinc (++) and manganese (++) had no effect on the activity of the protease puri- fied 67-fold while mercury (++) inhibited the protease about 15%. However, treatment of the assay reaction mix- ture with 0.0A5 M EDTA resulted in partial loss of pro- teolytic activity. The loss of activity could be partially reversed by the addition of the metal ions mentioned above, except for iron (++,+++), copper (++) and mercury (++). These results were interpreted as an effect on the sub- strate and not an effect on the enzyme. The protease withstood heating at 97 to 99 C for 60 min at pH 6 and 7.5 but lost 21% of activity under the same conditions at pH 8.5. Casein and B-lactoglobulin were acted upon by the enzyme but there was no proteolytic activity detected on bovine serum albumin and hemoglobin or in the milk-clot test. Donald L. Wallace p-Chloromercuribenzoate was the most effective inhibi- tor of sulfhydryl groups followed by N-ethylmaleimide and iodoacetamide. Reducing agents and diisopropyl fluorophos- phate had no effect on the proteolytic activity. PARTIAL PURIFICATION AND CHARACTERIZATION OF AN INTRACELLULAR PROTEASE FROM STREPTOCOCCUS DURANS By. a ‘ ‘.. Donald L. Wallace A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science 1968 ' " -".' 5. to my wife and parents ACKNOWLEDGMENTS The author wishes to express his appreciation to his major professor, Dr. L. G. Harmon, for his continued interest and guidance throughout this study; to Dr. J. R. Brunner for his suggestions and for the use of his labora- tory facilities; to H. L. Sadoff, H. A. Lillevik and P. Markakis for serving as members of the guidance committee; and to Larry R. Beuchat and Francis H. Webster for their suggestions in editing this manuscript. Appreciation is extended to Dr. B. S. Schweigert, Chairman, Department of Food Science, for his interest in this program and to Michigan State University for the facilities which were provided. iii TABLE OF CONTENTS DEDICATION ACKNOWLEDGMENTS . . . . . . . . . . . LIST OF TABLES. LIST OF FIGURES INTRODUCTION LITERATURE REVIEW. Use of Streptococcus durans in Cheddar Cheese Manufacture Proteolysis by Cultures Used in Making Cheddar Cheese. . . . . . . . . . . . EXPERIMENTAL PROCEDURES. . . . . . . . . Production of Streptococcus durans Cells and Preparation of the Cell- free Extract. Preparation of Substrate . Standardized Assay for Proteolytic Activity Determination of Variables Involved in Assay . Substrate concentration Enzyme concentration Assay reaction time. pH studies. Assay temperature Buffer concentration Measurement of Protein Concentration . Partial Purification of an Intracellular Protease from S. durans . Experiments to Determine Additional .Charac- teristics of the Enzyme . . . . . Metal ions. . Ethylenediamine tetraacetic acid (EDTA) and metal ions. . . . . . . iv Page ii iii vi vii 1A l5 17 18 18 18 l9 l9 l9 19 2O 2O 22 22 22 q.- n . ‘ Thermal inactivation Reducing agents . . . Inhibitors. . . Other proteins as substrates. Analyses for nucleic acids Ribonuclease experiments Disc gel electrOphoresis RESULTS Variables Involved in Assay . Purification of an Intracellular Protease from S. durans EXperiments to Determine Additional Charac- teristics of the Protease DISCUSSION SUMMARY AND CONCLUSIONS. BIBLIOGRAPHY Page 24 24 2A 25 25 26 27 28 28 33 39 47 5A 58 Table LIST OF TABLES Page Amount of protein, specific activity and percentage yield obtained during the partial purification of an intracellular protease from S. durans assayed for l min at 37 c in o. si'caseifi'ln 0. on M phosphate buffer at pH 7. 5 . . . 38 Effect of various concentrations of EDTA on the activity of an intracellular protease from S. durans purified 67- fold and assayed for l min at 37 C in 0. 5% casein in 0. 0A M phosphate buffer at pH 7.5.. . . . . 40 Effect of various metal ions (0.01 M), EDTA (0.0A5 M) and the metal ions combined with EDTA in the reaction mixture on the activity of an intracellular protease from S. durans purified 67- fold and assayed for 1 min at 37 C in 0. 5% casein in 0. 0“ M phosphate buffer at pH 7.5.. . . . . A1 Effect of enzyme inhibitors on the activity of the intracellular protease from S. durans purified 67- fold and assayed for l min at 37 C in 0.5% casein in 0. 0A M phosphate buffer at pH 7.5. . . . . . AA vi 312"“ Figure LIST OF FIGURES Flow sheet showing the procedure for pro- ducing and harvesting S. durans and preparation of the cell- free extract Flow sheet showing the procedure for partial purification of an intracellular protease from S; durans. . . . . . . . Effect of substrate (casein) concentration on the activity of an intracellular pro- tease from S. durans purified 67- fold and assayed for l min at 37 C in 0. 04 M phos- phate buffer at pH 7. 5 . . . . . . Relationship between protease concentration and proteolytic activity in the CFE from S. durans assayed for l min at 37 C in 0.5% casein in 0. DA M phosphate buffer at pH 7 5 . . . . . . . Reaction rate of proteolysis by an intra— cellular protease from S. durans purified 67- fold and assayed in 0. 5% casein in 0. 04 M phosphate buffer at pH 7. 5 at 37 C Effect of pH on the activity of the CFE and an intracellular protease from S. durans puri— fied 67- fold and assayed for 1 min at 37 C in 0.5% casein in 0. 0A M phosphate buffer. Effect of pH on the activity of an intra- cellular protease from S. durans purified 67— fold and assayed for l min at 37 C in 0.5% casein in 0. 0A M phosphate buffer. Effect of temperature on the activity of an intracellular protease from S. durans puri- fied 67- fold and assayed for l min in 0.5% casein in 0. 0A M phosphate buffer at pH 7 5 Vii Page 16 23 29 3O 31 32 3A 35 Figure 10. ll. 12. Page Effect of sodium phosphate buffer concentra- tion on the activity of an intracellular protease from S. durans purified 67- fold and assayed for l min at 37 C in 0. 5% casein at pH 7. 5 . . . . . . . 36 Effect of sodium chloride concentration on the activity of an intracellular protease from S. durans purified 67- fold and assayed for l min at 37 C in 0.5% casein in the presence of the indicated sodium chloride concentrations at pH 7.5 . . . 37 Thermal stability of an intracellular pro- tease from S; durans purified 67-fold and heated at 97-99 C at the pH indicated and assayed for 1 min at 37 C in 0.5% casein in 0.04 M phosphate buffer. . . . . . A3 Disc electrophoresis in polyacrylamide gel of the intracellular protease from S. durans. Gel A is stained for detection of protein and Gel B is stained for detection of carbohydrate . . . . . . . . . A6 viii INTRODUCTION Cheesemaking and ripening involves the controlled growth of microorganisms and controlled activity of native, added, and microbial enzymes in the milk and curd system. Starter cultures, composed of organisms from several genera and species, are used to inoculate the cheese milk. These cultures proliferate in the mild and/ or the curd and are responsible for development of acid and volatile flavor components. The lactic streptococci and lactobacilli have a dominant influence on the char— acteristics of cheese. However, other organisms present in the microflora also influence the development of the body, texture, flavor and aroma during ripening of cheese. Streptococcus durans is a common organism in the microflora of milk and cheese. The high heat resistence and salt tolerance of this enteric streptococcus make it useful as a lactic starter culture organism in the development of short-time and continuous Cheesemaking pro- cesses. These processes involve higher than normal cook- ing temperatures (A3 to 5A.5 C) in the presence of salt. Cheese made by these processes are comparable in quality to normal cheese. The changes which take place in cheesemaking and ripening are largely the result of enzymatic activity. The enzymes are either naturally present, added to the milk or arise from microorganisms in the milk. Pro- teolysis contributes greatly to the flavor and body of the cheese. The production of proteases by the lactic streptococci and lactobacilli has been studied by a number of workers. However, no information was available on the production of proteases from S; durans. The pur- pose of this study was to isolate, purify and character- ize an intracellular protease from S; durans. The infor- mation will be beneficial in the future use of this organism or the enzymes extracted from this organism in cheesemaking. LITERATURE REVIEW The importance of the microbial flora in the manu- facture and curing of Cheddar cheese is well recognized. During Cheddar cheesemaking, a starter culture of lactic streptococci is added to the milk. These bacteria, commonly Streptococcus lactis, Streptococcus cremoris or both, utilize lactose to produce lactic acid which results in a lower pH in the cheese. A lower pH allows lacto- bacilli natively present in the milk to increase and con- tribute to further ripening of the cheese. Use of Streptococcus durans in Cheddar Cheese Manufacture The traditional Cheddar cheesemaking procedure calls for use of the lactic streptococci. However, Dahlberg and Kosikowsky (19A8) successfully used a selected strain of Streptococcus faecalis, an enteric streptococcus, to make Cheddar cheese. The cheese had more cheddar flavor of better quality and a more mellow and waxy body than Cheddar cheese made with a lactic starter. The authors' approach was to consider the characteristics desired of an organism in cheesemaking and curing and then isolate a bacterium Which met these requirements. The organism was required to be universally present in milk, survive pasteurization, produce lactic acid rapidly and use lactose as a source of energy. The bacterium had to grow in the 10 to A1 C range, produce little or no gas, be nonproteolytic and produce no objectionable flavors or odors. It should grow anaerobi— cally at pH 5 to 5.5 and at salt concentrations up to six per cent. In conjunction with this study, Kosikowsky and Dahlberg (19A8) reported that S; faecalis was able to grow and survive in Cheddar cheese in large numbers (A50 to 7A0 million) for a considerable time (120 days) when the cheese was ripened at 10 to 15.6 C. Clark and Reinbold (1966) studied the occurrence of enterococci in young Cheddar cheese. A total of 1,117 isolates of microorganisms frOm Al cheese samples were characterized and identified. Enterococci constituted 51.7%, or 578 of the total number of isolates. Sixty per cent of the enterococci belonged to the S; durans group, and the remaining A0% were S; faecalis or varieties of this organism. Mechanization of the cheesemaking process to reduce time and labor costs led to the development of short-time processes. Walter gt a1. (1953, l956a,b, 1957) developed a short-time, high—temperature process using thermoduric starters. A mixture of two starters was used; the first was a conventional mixture of lactic streptococci, and 2‘ the second was a heat— and salt-tolerant strain of S; durans. . A. ‘ "-“' ' ' a‘g ' ' A. o Hi azfilal‘ h 4"“:“1‘3 . 7%” r The latter was used because lactic streptococci were unable to grow or produce acid at temperatures above A0.6 C in four to six per cent salt concentrations. In this process, conventional Cheddar cheesemaking proce- dures were followed until the whey was drained off, and the curd was remixed with a small portion of whey. This “E mixture was heated to A6.l to A7.8 0, salt was added to a final concentration of four per cent and the whey was pressed out and drained. The making time was 3.5 hr from starter addition to completion of hooping. Subsequently, Walter 23 El: (1958) changed the process and used the routine procedure for making Cheddar cheese except for the steps between draining and milling. At this point, the curd and whey were pumped into a cloth—lined, per- forated curd-retaining device within a tank. The curd was allowed to mat for 2 hr without turning. This method was a simple means of mechanized draining and cheddaring, but it abandoned the main principles of the original short—time process. Czulak 23.3l- (195A) introduced short—time method known as the "New way" process for making Cheddar cheese. The process was based on the procedure suggested by Walter 33 a1. (1953) with the use of a two per cent lactic starter and a one per cent S; durans culture, but differed in that after cooking the curd at 37.8 0, half of the whey (.1l01. . . .....J¢.. . chi . . ....¢\| v“. I". w." Ans... A \ ‘was drained and the cooking temperature was raised to A2.2 to A3.3 C. The remaining whey was drained, and the curd was cheddared for one—half hour at A2.2 to A6.l C. The remainder of the manufacturing procedure was carried out as in the conventional process. The making time was 3 to 3.5 hr from addition of starter to completion of hooping. 43-, ' . Czulak and Hammond (19560) stated that fast—make cheese, fresh from the press, would have a pH of 5.1 to 5.3. The desired pH was achieved by using two per cent lactic starter and two per cent of an active thermoduric culture. Downs (1955) reported that an improved Cheddar cheese resulted when Streptococcus thermophilus was used in conjunction with S; durans to replace S, lactis. Czulak and Hammond (l956a) showed that the rate of acid production by S; durans and S; thermophilus at A5 C was comparable to S; lactis at 30 C. Streptococcus durans was preferred over S; thermophilus because of the latter's greater sensitivity to salt. Feagan (1956) reported S; thermophilus was unsatisfactory because the lag phase of the organism prevented early acid development in the whey. However, raising the first cooking temperature to A1.l C in A0 min allowed a more rapid acid development. Morris (1955) reported that experimental trials on the Czulak gt a1. (l95A) method produced cheese with a curdy body. He suggested that pH variations were responsi- ble and modified the method by allowing more time for acid development between the first and second cooking. Cook (1955) and Czulak and Hammond (1956b) reported that in commercial production the quality of the short- time Cheddar cheese compared favorably with cheese made by the traditional process. Proteolysis by Cultures Used in Making Cheddaerheese The lactic streptococci are described by Breed _t._l. (1957) as generally nonproteolytic. However, these bacteria do have proteolytic enzymes and carry out some proteolysis in milk or cheese. Sato and Ohmiya (1966) and Ohmiya and Sato (1967, l968a,b) found that lactic acid bacteria hydrolyzed a-casein but had little, if any, effect on B-casein. Jespersen (1966) compared the proteolytic activities of some lactic acid bacteria on skimmilk containing added calcium carbonate. Among the streptococci, Streptococcus diacetilactis was most proteolytic followed by S; lactis, S; cremoris, S; faecium and S; durans, respectively. The lactobacilli were, in general, more proteolytic than the streptococci. The importance of lactic acid bacteria in proteolysis during Cheddar cheese ripening has been established by Peterson gt_al. (19A8b), Ernstrom gt 31. (1958) and Emmons pt 31. (1962). Peterson pt 31. (19A8a) studied the 'protease activity in Cheddar cheese and found a primary pH Optimum at 5 and a secondary optimum at pH 7 to 8. Proteases active at pH 5 were activated by 0.015 M cysteine and were considered to have been intracellular bacterial proteases liberated by bacterial autolysis. Enzymes active at pH 7 to 8 were not affected by cysteine and were thought to have been extracellular proteases liberated by bacteria during the normal-life cycle. Baribo and Foster (1952) compared the characteris- tics of intracellular proteases from S; lactis, Lacto- bacillus casei and Micrococcus freudenreichii to natural milk proteases from one—year-old cheese. The cell-free extract from S; lactis was shown, upon heating to 60 C for 30 min, to contain both thermal-stable and thermal- labile proteases. The optimum temperature for activity was from A0 to A2 0. Two pH optima for activity were observed; one was a primary optimum at pH 7 and the other was a secondary optimum at pH 5 to 5.5. Lactobacillus casei contained thermal-stable and thermal-labile enzymes with an optimum at pH 7 and an optimum temperature of 37 C. Approximately the same was true of M; freudenreichii except that its proteases were all heat-labile. Pro- teases in the aqueous extract of cheese were stable when heated at 60 C for 30 min. These enzymes showed two temperature optima; one was above A5 C and the other was from 11 to 18 0. Two pH optima were apparent also, one 'being from pH 5 to 5.5 and the other near pH 7. The activity of all protease preparations was enhanced by reducing agents (approximately 0.01 M cysteine, sodium thioglycollate, sodium sulfite, ascorbic acid and potassium cyanide), but metal ions (approximately 0.01 M calcium, manganese, magnesium, cupric, zinc, cobalt, ferrous and ferric) had no effect. Collins and Nelson (19A9) and Vanderzant and Nelson (1953a) reported that S; lactis grown in skimmilk caused a substantial increase in the trichloroacetic acid (TCA)- soluble nitrogen during the first 2A hr, with a gradual increase throughout the remainder of the test period. In the work of Vanderzant and Nelson (1953a), controlling the pH at 7.0 during incubation considerably increased pro- teolysis. They found no proteolytic activity in the cell- free growth medium. Morgan and Nelson (1951) found marked increases in the concentration of ten amino acids in tungstic and lactic acid filtrates of skimmilk cultures of five S; lactis strains after a 1A-day incubation period. Lactic acid filtrates showed higher concentrations of amino acids than the tungstic acid filtrates. This was attributed to a greater proportion of low-molecular- weight protein degradation products. They suggested that these fractions from proteolysis by S; lactis might pro- vide the stimulus to the lactobacilli that are important to the later states of cheese ripening. 1 .' gflggdnwmm AV lO Amino acids and peptides present in protein-free fractions of milk before and after incubation with S; lactis were studied by Vanderzant and Nelson (l95Aa). At zero time, free alanine, glutamic acid, glycine, leucine, isoleucine and valine were present in the protein-free fraction of milk. These amino acids had increased in quantity after incubation for 2A hr. In addition, free lysine, phenylalanine, proline, serine, threonine and tyrosine were detected in the fraction. After 96 hr of incubation, further increases in the amounts of the above mentioned amino acids were noted except for glycine and valine. No aspartic acid was detected until after 96 hr of incubation. Peptides were present which contained all of the above mentioned amino acids except isoleucine, lysine, threonine and aspartic acid. Vanderzant and Nelson (1953b) investigated the characteristics of the intracellular proteolytic enzyme system of S; lactis. Skimmilk, casein and lactalbumin were used as substrates. Reducing agents in high con— centrations (0.01 to 0.1 M) increased proteolytic activ- ity. Metal ions (0.1, 0.01 and 0.001 M calcium, ferrous, ferric, manganese, magnesium, zinc, cupric and cobalt) had either no effect or were slightly inhibitory. The enzyme was most stable to heat at pH 7, but was completely inacti- vated for 2 min at 61.7 C. The enzyme was stable to storage at 2 C. ll Vanderzant and Nelson (l95Ab) reported the char- acteristics of some intracellular peptidases from S; lactis. The cell-free extract from S; lactis exhib- ited maximum peptidase activity at pH 7 to 8.5. Seven substrates were used, five dipeptides and two tripep- tides. When 0.33 M glycyl-L-leucine was used as the sub- strate, manganese ions increased activity and cupric, zinc and nickel ions were inhibitory. Magnesium and cobalt ions and cysteine had no effect. Peptidase activity was increased on 0.33 M DL-alanylglycine by cobalt and mag— nesium ions, inhibited by cupric, zinc and nickel ions and not affected by manganese ions and cysteine. Metal ions and cysteine were used in 0.01, 0.001 and 0.0001 M concentrations. Thermal stability was greatest at pH 7 and storage stability at 2 C was greatest at pH 6 to 9. Williamson 32 31. (196A) studied an extracellular protease from S; lactis. A l20-fold purification of the enzyme was obtained by fractionation with ammonium sul- fate, calcium phosphate gel chromatography and diethyl- aminoethyl cellulose. This protease exhibited extra- ordinary heat resistance by showing an increase in activity after being heated for 15 to 20 min at 98 C and losing only 32% of its activity after being heated for 60 min at 98 C. p-Chloromercuribenzoate (PCMB, 0.00A M) inhibited about 15% and sodium fluoride (0.005 M) inhib— ited about 26% of the protease activity. Diisopropyl l2 fluorophosphate showed no inhibition over a concentration range from 0.0000001 to 0.0001 M. The reducing agents, sodium thioglycollate (0.01 M) and sodium sulfite (0.01M), increased activity, while 0.1 M cysteine was required to increase activity. Cowman and Speck (1965a) reported that protease activity of S; lactis cells stored at 3 0 decreased with increased storage time until a relatively constant residual amount (12 to 16%) remained after six days. The loss of activity was most marked in the first and second days of storage at 3 C. Cowman and Speck (1965b) found similar results when S; lactis cells were stored at -20 and -196 0. However, the cells stored at -l96 C retained much higher residual protease activity. Cowman and Speck (1967) isolated and partially characterized the storage—labile proteolytic enzyme sys— tem of S; lactis is an effort to determine how it was inactivated by low-temperature storage. Proteolytic activity was optimal at pH 6 in cells which had not been stored at 3 C but also showed the greatest storage labil- ity at this pH when the cells were stored at 3 C. Gluta- thione restored the activity lost during storage. p-Hydroxymercuribenzoate (PHMB) had no effect on the pro- teolytic activity of the whole cells. Sonic disruption of the cells produced a soluble (intracellular) fraction and a particulate (cell-debris) fraction, both containing _a:F-=!!_y"7?;.b‘ emit yavj‘fi - “W“ " _‘j'- . ' . -, l3 protease activity. The soluble-fraction was stable to storage at 3 C, inhibited by PHMB and not activated by ferrous ions, magnesium ions or cysteine. The particulate- fraction was unstable to storage at 3 0, not inhibited by PHMB and activated by ferrous ions, magnesium ions and cysteine. However, after storage of the particulate fraction, magnesium ions were totally ineffective in reactivation; and only ferrous ions and cysteine would reactivate the fraction. Cowman gt_al. (1967) purified a storage-labile, membrane-bound protease from S; lactis. The enzyme showed maximum activity at pH 6, but minimum activity at this pH after storage. Gross structural alterations of the enzyme during storage were shown to occur by gel filtration and sedimentation velocity data. These data indicated that loss of activity during storage was due to aggregation, resulting in higher molecular weight forms (polymers). Activity was regained again upon disaggregation of the polymer forms to an active monomer. The PHMB derivative of the enzyme prevented aggregation during storage and, therefore, prevented the loss of activity. Ferrous ions, cysteine and glutathione reactivated the storage-inactivated enzyme. The importance of membrane protease in cellular function was apparent when glutathione restored the origi- nal acid-producing rate of stored whole cells when sub- cultured in milk. EXPERIMENTAL PROCEDURES Production of Streptococcus durans Cells and Preparation of the Cell-free Extract Streptococcus durans, as classified by Breed gt a1. (1957) was used as the source of intracellular protease for this study. The organism was activated from a frozen stock culture in litmus milk by three successive trans- fers into Trypticase Soy Broth (TSB) (Baltimore Biologi- cal Laboratory, Inc., Baltimore, Maryland). Thirty-five liters of TSB in a Fermacell Fermentor (New Brunswick Scientific Co., Inc., New Brunswick, New Jersey) were inoculated with 1% of an 18 hr culture of S; durans and incubated for 16 to 18 hr at 25 C. The cells were har- vested by continuous centrifugation at 35,000 x g at 0 C with a Sorvall RC-2 continuous, refrigerated centrifuge (Ivan Sorvall, Inc., Norwalk, Connecticut). The cells were pooled, suspended and washed in 0.0A M sodium phosphate buffer at pH 7.5 (standard buffer). Three additional washings were performed, using fresh buffer each time. The washed cells were suspended in standard buffer to a final volume of A50 ml. Twenty milliliter portions of the washed cells were disrupted by sonification in the presence of 7 g of glass 1A 15 'beads (0.1 to 0.2 mm in diameter). Sonification was con- tinued for 10 min at 0 to 10 C at an output of 8 to 10 amps with a Branson Model LS75 Sonifier (Branson Instruments, Inc., Stamford, Connecticut). The sonifier was set to deliver 35 watts per cm2 from the probe which was equal to a total of 56 watts since the tip of the probe had an area of 1.6 cm2. Cell debris was removed from the cell extract by centrifugation at 32,000 x g, washed twice with standard buffer and collected by centrifugation. The cell extract and the supernatants obtained from washing the debris were combined and diluted to 500 ml with standard buffer. This solution comprised the cell-free extract (CFE) and was stored at -20 C until needed. A schematic flow sheet for the growth and harvest of S; durans cells and preparation of the cell-free extract is shown in Fig. 1. Preparation of Substrate Casein was used as the substrate for measuring intracellular proteolytic activity from S; durans. Casein for this study was precipitated from fresh skimmilk at pH A.6 by addition of 1 N hydrochloric acid. The pre- cipitate was washed with copious amounts of water and then solublized by raising the pH to 7 with l N sodium hydroxide. Precipitation, washing and solublization were repeated once again and the casein was then lyophilized. 16 to 18 hr culture of S; durans in TSB at 25 C l Continuously centrifuged at 35,000 x g at 0-5 G \b S; durans cells Cell-frequrowth medium discarded Washed 3X in standard buffer - and centrifuged at 32,000 5 x g at 0-5 C for 30 min . h . I l ‘ 1. a Supernatant discarded Washed S; duranslcells i l Sonified for 10 min at 0-10 C and centrifuged at 32,000 x g at 0-5 C for A0 min 1 V V Cell debris Supernatant Washed 2X in standard buffer and centrifuged at 32,000 x g at 0—5 C for A0 min v V 6—‘——' Cell debris discarded Combined supernatants (Cell-free extract) Figure 1. Flow sheet showing the procedure for producing and harvesting S; durans, and preparation of the cell-free extract. 17 A 1% solution of casein was prepared in standard buffer and used for all assays. Standardized Assay for Proteolytic Activity The basis of the assay was the ability of the enzyme to hydrolyze casein and was determined by an increase in material soluble in 5% trichloroacetic acid (TCA). The increase in TCA-soluble material was quan- titated by measuring the increase in absorbance at 280 mu (AA280). The assay was performed by adding 1 ml of sub- strate to a test tube (reaction tube) which was placed in a 37 C waterbath for a 5 min preincubation period. Follow- ing preincubation, 1 m1 of appropriately diluted enzyme, preincubated in a separate tube, was added to the substrate. After the enzymatic reaction had proceeded for 1 min at 37 C, A m1 of 7.5% TCA were added to stop the reaction. A blank tube was prepared in the same manner as the reaction tube, except that TCA was added prior to addition of the enzyme. The blank and reaction tubes were allowed to stand at room temperature for 20 to 30 min to allow com- plete precipitation of the protein. Removal of the pre- cipitate was accomplished by centrifugation at full speed in an International clinical centrifuge (International Equipment 00., Boston, Massachusetts), followed by filtra- tion of the supernatant through Whatman Number A2 filter paper. The AA280 was read in a Beckman DB-G grating l8 Spectorphotometer (Beckman Instruments, Inc., Fullerton, California). All assays were made in duplicate, and at enzyme concentrations which produced a AA280 from 0 to 0.5. Activity was expressed in units, with one unit equal to 0.01 AA280 per min. Assays were also performed with other agents added such as reducing agents, metals, chelating agents, inhib- itors, ribonuclease and ribonucleic acid. In such cases, the same number of units of enzyme as used in the stand- ardized assay were added in a reduced volume (0.5 m1), and 0.5 m1 of the various agents were added. The volume of the reaction mixture used for assay was always stand- ardized to 2 m1. Determination of Variables Involved in Assay Substrate concentration. The effect of substrate (casein) concentration on the proteolytic activity was determined over a range from 0 to 7.5 mg of casein per 2 m1 of reaction mixture at pH 7.5. Standard buffer was used to dilute the substrate appropriately. Enzyme concentration. A range from 0.225 to 1.8 mg of CFE protein per ml of reaction mixture was used in performing assays. Standard buffer was used to dilute the CFE appropriately. l9 Assay reaction time. The initial reaction rate of the enzyme was determined by using reaction times ranging from 0 to A min in the standard assay. pH studies. Solutions of 1% casein in 0.0A M sodium phosphate buffer were made up at pH 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5 and 9. The CFE was adjusted to the prOper pH by dilution with buffer of appropriate pH and addition of 0.1 N sodium hydroxide or hydrochloric acid for final adjustment. Proteolytic activity was then determined at each pH mentioned above. Assay temperature. The CFE was assayed for pro- teolytic activity at 21, 25, 30, 37, A0, A5 and 50 C in order to obtain an optimum temperature for activity. Buffer concentration. Solutions of cell-free extract were prepared by dialyzing against sodium phos- phate buffers (pH 7.5) ranging from 0.025 to 0.055 M. Solutions of substrate were prepared in the same concen- trations of buffer. Proteolytic activity was determined at the various buffer concentrations. A similar experiment was performed using sodium chloride solution in place of sodium phosphate buffer. Sodium chloride solutions at pH 7.5 with concentrations ranging from 0.02 to 0.12 M were used for dialysis of the enzyme and substrate for 72 hr with agitation. 20 Measurement of Protein Concentration Protein concentrations were measured by the method of Lowry et a1. (1951) with certain modifications. Separate solutions of cupric sulfate and sodium tartrate were added to the sodium carbonate solution immediately before use and the phenol reagent was diluted 1:1 (v/v) with water. Bovine serum albumin (BSA) was used to pre— pare a standard curve over a concentration range from 0 to 500 pg per ml. Protein concentration was plotted against absorbance measured at 520 mu.. Partial Purification of an Intracellular Protease from_§; durans The first step in purification was to fractionate the CFE with ammonium sulfate. Five hundred milliliters of CFE at 0 to 2 C were slowly brought to 37% of satura- tion by addition of crystalline ammonium sulfate. After standing for 2 hr, the precipitate was removed by centri- fugation at 20,000 x g at 0 to 2 C for 30 min and dis— carded. The supernatant was then brought to 55% of sat- uration by addition of crystalline ammonium sulfate, allowed to stand for 2 hr and centrifuged at 20,000 x g for 20 min. The supernatant was discarded. ~The precipi- tate, which contained the enzymatic activity, was sus- pended in A00 m1 of standard buffer. 21 In the second step, the enzymatically active pre- paration from the preceding step was heated in a water- bath at 80 C for 10 min, cooled to 0 to 2 C and centri— fuged at 10,000 x g for 20 min to aggregate the precipitate into a pellet which was discarded. The supernatant (3A0 ml) was dialyzed with agitation against distilled, deionized water for 12 hr at A C. This dialyzed preparation was concentrated by perevaporation at room temperature to approximately one-third volume (110 ml) and dialyzed with agitation against standard buffer for A8 hr at A C. The third purification step involved the use of Bio- Gel HT (Bio-Rad Laboratories, Richmond, California) in batch technique to fractionate the enzyme preparation from the previous.step. Twenty-five milliliter portions of enzyme preparation were added to approximately 25 ml of hydrated gel (settled bed volume) in 0.0A M sodium phos- phate buffer (pH 7) at room temperature. The gel, with protein adsorbed, was removed by centrifugation at 5,000 x g for 10 min and the supernatant was discarded. Two hun- dred milliliters of 0.1 M sodium phosphate buffer at pH 7 were used to suspend and wash the gel. Two additional washings were performed, using fresh buffer for each wash- ing, and all supernatants were discarded. The gel was then suspended and washed twice in 200 ml of 0.2 M sodium phosphate buffer at pH 7 to elute the proteolytic enzyme 22 from the gel. These supernatants were combined and the A00 ml quantity was dialyzed with agitation for 12 hr at A C against distilled, deionized water. After dialysis, the enzyme was concentrated to A0 to 50 ml by perevapora- tion at room temperature and dialyzed with agitation against standard buffer for A8 hr at A C. This enzyme preparation was stored at —20 C until needed. The gel was prepared for re-use by removal of any remaining protein with 0.A M sodium phosphate buffer at pH 7 followed by equilibration with the same buffer. Proteolytic activity and protein concentration were determined at each purification step. A schematic flow sheet for the partial purification of an intracellular protease from S; durans is shown in Fig. 2. Experiments to Determine Additional Characteristics of the Enzyme Metal ions. Calcium (++), iron (++, +++), cobalt (++), nickel (++), copper (++), magnesium (++), zinc (++), manganese (++) and mercury (++) ions at a final concen- tration of 0.01 M were investigated to determine their effect on enzymatic activity. These ions were incor- porated into the assay by the addition of each separately to the substrate during preincubation. Ethylenediamine tetraacetic acid (EDTA) and metal ions. Standard assays were performed with enzyme and 23 Cell-free extract Ammonium sulfate added to 37% of saturation, the mixture allowed to stand 2 hr at 0-2 0 and centrifuged at 20,000 x g at 0-2 0 for 30 min Supernatant Precipitate discarded Ammonium sulfate added to 55% of saturation, the mixture allowed to stand 2 hr at 0-2 C and centrifuged at 20,000 x g at 0-2 C for 20 min v Supernatant Precipitate discarded Dissolved in standard buf- fer, heated at 80 C for 10 min and centrifuged at 10,000 x g at 0-2 C for 20 min t ' w Supernatant Precipitate discarded Dialyzed with distilled water for 12 hr, concentrated by perevaporation and dialyzed with standard buffer for A8 hr A Bio-Gel HT 1 l I 0.2 M phosphate eluant 0.1 M phosphate l eluant discarded Dialyzed with distilled water for 12 hr, concentrated by perevaporation and dialyzed with standard buffer for A8 hr c Protease purified 67‘fold Figure 2. Flow sheet showing the procedure for partial purification of an intracellular protase from S. durans. 2A substrate preparations which contained 0.0225, 0.0A5 and 0.09 M EDTA. In another experiment, each of the pre— viously mentioned metal ions (0.01 M) was tested for its effect on proteolytic activity in the presence of 0.0A5 M EDTA. Thermal inactivation. Twenty-seven milliliters of enzyme preparation were divided equally into nine frac- tions with 3 ml per fraction. The nine fractions were then subdivided into three groups. The three groups were adjusted to pH 6, 7.5 and 8.5, respectively, with small amounts of 1 N sodium hydroxide and 1 N hydrochloric acid. During heating at 97 to 99 C, one fraction from each group was removed at 0, 30 and 60 min and cooled in an ice bath. The groups at pH 6 and 8.5 were then adjusted to pH 7.5 in the same manner as described above. All fractions were assayed for proteolytic activity. Reducing agents. The effect of 0.000A5 and 0.0009 M concentrations of cysteine, sodium thioglycollate, ascorbic acid and glutathione on the proteolytic activity was determined. Reducing agents were incorporated into the assay by addition to the substrate during preincubation. Inhibitors. Enzyme inhibition was investigated by performing standard assays in the presence of 0.000078 and 0.000156 M PCMB (Aldrich Chemical Co., Inc., Milwaukee, Wisconsin), 0.0011 and 0.0022 M N—ethylmaleimide (Aldrich), 0.00113 and 0.00225 M iodoacetamide (Sigma Chemical Co., "I”: v -,-v- T 25 St. Louis, Missouri) and 0.000125 and 0.00025 diisopropyl fluorOphosphate (supplied by Dr. H. L. Sadoff, Department of Microbiology and Public Health, Michigan State Uni- versity, East Lansing). The respective inhibitors were incorporated into the assay by the addition of each to the substrate during preincubation. Otherpproteins as substrates. One per cent solu— tions of B-lactoglobulin, BSA and hemoglobin were pre- pared in standard buffer. The proteins were tested as substrate in place of casein in the standard assay. The milk-clotting method of Balls and and Hoover (1937) was also performed. Analyses for nucleic acids. The quantity of nucleic acids present in the enzyme preparation was measured by tests for phosphorus and pentose. Phosphorus was deter- mined by a colorimetric procedure adapted from the method by Sumner (19AA). The analysis was performed in duplicate. Dried protein (1 to 11 mg) was digested with 2.2 ml of 50% sulfuric acid. The digestion was carried out for 20 min on a sand bath heated by an electric heater maintained at 160 to 170 C. After cooling, five drOps of 30% hydro- gen peroxide were added to the protein digest, and the mixture was heated for 15 min. Then the digestion mixture was cooled, and five additional drops of 30% hydrogen peroxide were added. Digestion was continued for 1 hr. If the mixture was not clear at this point, additional 26 hydrogen peroxide was added and the digestion was con- tinued. After cooling, the mixture was transferred to a 50 ml volumetric flask; and 5 ml of a 6.6% ammonium molybdate solution and enough water were added to give a total volume of approximately A0 m1. Next, A ml of fresh acidified ferrous sulfate solution (5 g ferrous sulfate, 50 ml water and 1 ml of 7.5 N sulfuric acid) were added and mixed. The contents of the flask were made up to 50 ml with water and mixed. After standing for 30 min, the absorbance was read at 660 mu with a Beckman DB-G spectro- photometer. A standard curve was prepared covering the range of 0.0 to 0.31 mg of phosphorus with monobasic potassium phosphate. Pentoses present in the enzyme preparation were measured by the Dische modification (1953) of the orcinol reaction and quantitated from a standard curve prepared with ribonucleic acid from yeast covering a range from 0 to 120 ug. Ribonuclease experiments. Five milliliters of the enzyme preparation from S; durans were treated with 0.1 mg of protease-free ribonuclease (Calbiochem, Los Angeles, California) for l min. The ribonuclease was then removed from the reaction mixture by gel filtration on Bio-Gel P-60 gel (Bio-Rad) which had been equilibrated with standard buffer. Ribonucleic acid, which had been extracted from S; durans cells with phenol by the method .. .1*-'__.,i . “n..- ___w. 7.4 fix??? V W 27 of Asano (1965), was added to the reaction mixture in an attempt to regain the proteolytic activity. Controls for this experiment were a standard assay for original protease activity, an assay to detect any inactivation of the protease by ribonuclease, and an assay with ribo- nucleic acid and substrate but no enzyme to assure the absence of ribonuclease activity in the substrate. Disc gel electrophoresis. Disc gel electro- phoresis of the enzyme preparation was performed according to the method of Davis (l96A) except for one modification. The large pore gel was made by mixing one part small pore solution with one part sample which resulted in a 3.5% acrylamide gel. Eight-tenths milligram of enzyme dis- solved in 0.1 m1 of buffer was used for each duplicate gel. Electrophoresis was continued for 1 hr with an out- put of 5 milliamps per gel tube. One of the duplicate A gels was stained with amido black for the detection of protein, and the other gel was stained by the method of Clarke (196A) in which Schiff reagent was used for the detection of carbohydrate. RESULTS Variables Involved in Assay The effect of each variable involved in the assay on the activity of an intracellular protease from S; durans remained constant throughout purification. Data illus- trated in Figs. 3 through 10 show the effect of each vari- able considered in assaying the activity of the protease. The effect of concentration of the substrate on the pro- tease activity is indicated in Fig. 3. Under optimum assay conditions, 5 mg of casein per m1 of the reaction mixture provided excess substrate for the enzyme and was used in all standardized assays. The enzymatic activity was proportional to enzyme concentration within the range studied as shown in Fig. A. Figure 5 illustrates the initial reaction rate of the enzyme with determinations at half minute intervals, up to A min. A reaction time of l min gave a good indi- cation of initial reaction rate and was used in all experiments. The curves presented in Fig. 6 show the response from pH 5 to 9 for the crude CFE and for the protease 28 . ”17‘3“ ‘. AA280 29 O 2.5“ 2.0‘ 0.0 g 1 l l W e a —+— 4 O l 2 3 A 5 6 7 8 Substrate Concentration (mg/ml of Reaction Mixture) Figure 3.--Effect of substrate (casein) concentra- tion on the activity of an intracellular protease from S; durans purified 67-fold and assayed for 1 min at 37 C in 0.05 M phosphate buffer at pH 7.5. AA280 30 «r 0.0 ,l 11 5 O l 2 3 A Enzyme Concentration (mg/m1) Figure A.-¢Relationship between protease concen- tration and proteolytic activity in the CFE from S; durans assayed for 1 min at 37 C in 0.5% casein in 0.0A M phosphate buffer at pH 7.5. v. “q, w t 31 I'l‘ AA280 0 $ . + : 5 o l 2 3 A Time (min) Figure 5.--Reaction rate of proteolysis by an intra- cellular protease from S; durans purified 67-fold and assayed in 0.5% casein in 0.0A M phosphate buffer at pH 7.5 at 37 C. 32 AA280 C’CFE A.Purified 67—fold pH Figure 6.—-Effect of pH on the activity of the CFE and an intracellular protease from S; durans purified 67-fold and assayed for l min at 37 C in 0.5% casein in 0.0A M phosphate buffer. 33 purified 67-fold. Both preparations of protease gave an optimum at pH 7.5 to 7.6 and a shoulder of activity at pH 5.5 to 6.0. Depicted in Fig. 7 is the response from pH 7 to 8 for the protease purified 67-fold. The relationship between assay temperature and pro- teolytic activity is illustrated in Fig. 8. The proteo- 1ytic activity increased linearly as the temperature T: increased to 37 C. Above 37 C, the relationship became é " nonlinear; therefore, 37 C was used for all assays. 5 Data presented in Figs. 9 and 10 illustrate the r"! effects of ionic strength (u) of phosphate buffer and sodium chloride on the proteolytic activity. Sodium phosphate concentrations of 0.035 to 0.0A0 M (u = 0.09 to 0.11) produced maximum activity (Fig. 9) and were used in the standard assay. In a similar experiment, sodium phosphate buffer was replaced by sodium chloride (Fig. 10). The sodium chloride concentrations producing maximum activity were 0.08 to 0.10 M (u = 0.08 to 0.10). Purification of an Intracellular Protease from S; durans The protease from S; durans was purified approxi- mately 67—fold with 53% recovery of the original activity by the sequence of procedures reported in Table 1. Puri- fication procedures which were found unsatisfactory included alcohol fractionation, gel filtration chromato- graphy, DEAE-cellulose chromatography, manganese chloride AA28o 2. 2. 2. 2.0 % .L ‘r : : 7.0 7.2 7.A 7.6 7.8 8.0 pH Figure 7.--Effect of pH on the activity of an intracellular protease from S; durans purified 67-fold and assayed for l min at 37 C in 0.5% casein in 0.0A M phosphate buffer. 35 AA280 20 36 A6 50 Temperature (C) Figure 8.—-Effect of temperature on the activity of an intracellular protease from S. durans purified 67—fold and assayed for 1 min in 075% casein in 0.0A M phosphate buffer at pH 7.5. 1-2 : : . 0.025 0.035 0.045 0.055 Phosphate Buffer Concentration (M) J Figure 9.--Effect of sodium phosphate buffer concen- tration on the activity of an intracellular protease from S; durans purified 67-fold and assayed for l min at 37 C in 0.5% casein at pH 7.5. 37 1.50-t 1.25 4: 1.00l 530.7% AA2 O E 1 1 I 4'; O 0.08 0.16 0.2A 0.32 0.AO Sodium Chloride Concentration (M) Figure 10.--Effect of sodium chloride concentration on the activity of an intracellular protease from S; durans purified 67-fold and assayed for 1 min at 37 C in 0.5% casein in the presence of the indicated sodium chloride concentrations at pH 7.5. . 1.1.. .1. ' m.- - -;F5Ill, H.mm O.OO O.OHO.H O.O Omm OOH em HOOIOHO O.Oe O.HH m.Omm O.H OH: OOm :He OH pom O Omnsm Om Ompmmm H.HO O.H O.m: OH Om: OOH ammusm .HommAzsz 8 23 OOH O.H O.mm OH OOm OOm . maO a mE\mpH:: HE\wE HE\mpHQ: HE HOH>HOOa onHw COHmeHMHnsm oamHomam chpopm sz>Hpo< mEsHo> mHSBmOOHm .m.a ma pm pmHQSQImecomona z no.0 CH :Hmmmo nm.o CH 0 pm pm cHE H pom owmmmmm QOHSU .m Eopm ommmuOHQ MOHSHHmomHQCH cm mo soapOOHMHusa Hmfiunma on» wcflhso vmsfimpno,onHz mwmucmonma cam hpfi>fipom OHQHoon .cfimpona mo pcsoE< .H meme 39 precipitation, protamine sulfate precipitation and ultra- centrifugation in a sucrose gradient. Specific activity was calculated by divind units of activity per milliliter by milligrams of protein per milliliter. The degree of purification obtained with each procedure was calculated by dividing the specific activity (units/mg) at each step by the specific activity of the CFE. The percentage yield was that portion of the origi- nal total activity (units) retained after each purifica— tion step, and the total activity was obtained by multi- plying the volume (ml) and the activity (units/ml). Experiments to Determine Additional Characteristics of the Protease The data in Table 2 show the effect of a chelating agent, EDTA, on the proteolytic activity of the intra- cellular protease from S; durans. The protease activity was progressively inhibited by increasing concentrations of EDTA with complete inhibition occurring at a 0.09 M concentration of EDTA. The effect of metal ions on pro- teolytic activity in the presence and absence of 0.0A5 M EDTA is presented in Table 3. In the absence of EDTA, metal ions had no effect, except for 0.01 M mercury (++) which reduced the activity about 15%. In the presence of 0.0A5 M EDTA, all metal ions tested, except iron (++, +++), copper (++) and mercury (++), reactivated approxi- mately 53 to 61% of the activity destroyed by EDTA when A0 Table 2. Effect of various concentrations of EDTA on the activity of an intracellular protease from S; durans purified 67-fold and assayed for l min at 37 C in 0.5% casein in 0.0A M phosphate buffer at pH 7.5. EDTA concentration AA280 Percentage (M) inhibition 0.0 2.5 0 0.0225 1.7 32 0.0A5 0.6 76 0.09 0.0 100 Al Table 3. Effect of various metal ions (0.0A5 M) and the metal ions EDTA in the reaction mixture of an intracellular protease purified 67-fold and assayed (0.01 M), EDTA combined with on the activity from S. durans for l_min at 37 C in 0.5% casein in 0.0A M phosphate buffer at PH 7.5. Metal ion AA280 (0.01 M) NO EDTA 0.0A5 M EDTA Blank A.7 1.1 Calcium (++) A.6 3.3 Iron (++) A.6 * Iron (+++) A.6 * Cobalt (++) A.7 3.1 Nickel (++) A.5 3.0 Copper (++) A.6 * Magnesium (++) A.7 3.0 Zinc (++) A.6 3.0 Manganese (++) A.6 3.0 Mercury (++) A.0 0.0 * Erratic readings obtained A2 present without metallic ions. Iron and copper ions caused erratic readings, and mercury completely inhibited the protease in the presence of EDTA. The thermal stability of the intracellular pro- teolytic enzyme from S; durans is shown by data in Fig. 11. When heated at 97 to 99 C, the enzyme was stable at pH 6 and 7.5 for 60 min; but at pH 8.5, 10% of the activity was lost after 30 min, and 21% after 60 min. The reducing agents, cysteine, sodium thioglycollate, ascorbic acid and glutathione, had no effect on the pro- teolytic activity of an intracellular protease from §; durans. Data showing the effect of various concentra- tions of enzyme inhibitors on the S; durans protease is presented in Table A. The sulfhydryl inhibitors, PCMB, N—ethylmaleimide and iodoacetamide, were inhibitory to the enzyme while the esterase inhibitor, diisoprOpyl fluoro- phosphate, showed no effect on the enzyme. Substrate specificity of the enzyme was investigated on casein, B-lactoglobulin, BSA, hemoglobin and the milk clot test. The enzyme was specific for the milk proteins, casein and B-lactoglobulin, but showed no activity on BSA or hemoglobin or in the milk clot. Ribonuclease completely inactivated the purified pro- tease. However, all attempts to reactivate the protease with ribonucleic acid from S; durans cells were unsuccess— ful. Through the use of a control assay which contained AA28o OpH 6.0 APH 7.5 DpH 8.5 0 i +7 + if 0 15 30 A5 60 Heating Time (min) Figure ll.--Thermal stability of an intracellular protease from S; durans purified 67-fold and heated at 97-99 C at the pH indicated and assayed for 1 min at 37 C in 0.5% casein in 0.0A M phosphate bufferafl’pH'ZS. AA Table A. Effect of enzyme inhibitors on the activity of the intracellular protease from S; durans purified 67-fold and assayed for 1 min at 37 C on 0.5% casein in 0.0A M phosphate buffer at pH 7.5. Inhibitor and Percentage concenEration AA280 inhibition p-Chloromercuribenzoate None A.38 .000078 3.90 11.0 .000156 3.36 23.2 N-ethylmaleimide None e A.38 .0011 3.60 17.8 .0022 3.10 29.2 Iodoacetamide None 5.08 .00113 5.00 1.5 .00225 A.90 3.5 Diisopropyl fluorophosphate None 5.30 .000125 5.30 0.0 .00025 5.30 0.0 Y AS only ribonucleic acid and casein, the casein was shown to have no measurable ribonuclease activity. Disc electrophoresis of the S; durans protease in polyacrylamide is illustrated in Fig. 12. Disc gel A, which was stained with amido black for detection of pro- tein, shows one major zone for the enzyme protein. Adjoining the protein zone was a white zone which did not give a protein stain but reacted with the Schiff reagent used for the detection of carbohydrates. This is shown in disc gel B. On a dry weight basis, the phosphorus analysis showed 68.9% nucleic acids and the pentose analy- sis showed A9.3% nucleic acids in the enzyme preparation. A6 . {i A ‘ B Fig. 12. Disc electrophoresis in polyacrylamide gel of the intracellular protease from S. durans. Gel A is stained for detection of—protein and Gel B is stained for detection of carbohydrate. DISCUSSION In this study, an intracellular protease from §; durans was isolated and partially purified. Casein, the major protein in milk and most varieties of cheese, was used as a substrate to test various characteristics of the enzyme. The enzyme exhibited normal reaction kinetics with respect to substrate and enzyme concentra- tion and reaction rate. The initial reaction velocity of the assay decreased rapidly after 1 min and was probably due to inhibition of the enzyme by products of the enzy- matic reaction. Peterson gt 31. (19A8a), Baribo and Foster (1952) and Vanderzant and Nelson (1953b, 195Ab) reported an Optimum at pH 7.5 for proteases extracted from Cheddar cheese and other lactic streptococci. The protease from S; durans showed an optimum at pH 7.5 and some activity at pH 5.5 to 6 which are similar to the primary and secondary optima published for S; lactis by Baribo and Foster (1952). These results indicate either two enzymes or two ionizable forms of the enzyme. The ratio between the activity at pH 7.5 and the activity at pH 5.5 to 6 remained essentially constant during the purification I47 A8 procedure; and although this does not rule out the pre- sence of two enzymes, it does point to the possibility of two ionizable forms of the same enzyme. The Optimum-pH could be explained by ionizable groups at or near the active site which directly affect the reaction with the substrate. Ionizable groups remote from the active site but close enough to directly or indirectly affect the activity at pH 5.5 to 6 could cause the secondary activity. According to Webb (1963), the effect of ionic strength on enzyme kinetics may be broadly designated as (1) those producing changes in the solvent, (2) those directly affecting the interaction between the enzyme and the other components of the system, and (3) those indir- ectly affecting such interactions. The effects of ionic strength on the water serving as solvent were probably minor in the range of concentra- tions used in this study. The concentrations of electro- lyte were low and the ranges narrow so that variations in the characteristics of the solvent were small. Changes in the activity coefficients of the reaction components of the enzyme are responsible for directly affecting the interactions between charged groups of the protein. As the concentration of electrolyte increases, the ionic atmosphere surrounding each charged molecule and group is increased, and the electrical potential of each Y A9 ion is reduced. Therefore, the attraction between ions of opposite charge will decrease. Indirect effects of ionic strength on the interaction between components of the enzymatic reaction may be (1) changes in the prOperties of the enzyme or its reacting components, and (2) specific binding. A change in ionic strength will cause a change in ionization constants of the protein groups and thus could alter the pKa of the groups near the active site. The secondary effect would be the changing of the overall charge distribution and affinity between the enzyme and substrate. Specific ion binding means that it is possible that one or both of the electrolyte ions may react with either the enzyme or the other components of the system in a manner that is signifi- cantly more pronounced than simple electrostatic attrac- tion. An example might be a form of chelation of various metal ions. Although it was not the purpose of this study to investigate ionic strength in detail, the data showed an influence of ionic strength on the protease activity. The use of two electrolyte systems, sodium phosphate buffer and sodium chloride, partially ruled out specific ion binding. However, both electrolytes were sodium salts so that no conclusion can be drawn about the effect of the cation on the activity. The data gave no information about 50 whether the activity coefficients or ionization constants or both were responsible for the effect of ionic strength on the protease. Baribo and Foster (1952), Vanderzant and Nelson (1953b) and Cowman and Speck (1967) have reported intra— cellular proteases from S; lactis which were unaffected by metal ions. The S; durans protease was unaffected by all metal ions used except mercury (++). If the enzyme required a metal ion for activity, this requirement should have been apparent after purification. Results from the experiments with EDTA and metal ions indicated that the loss of proteolytic activity in the presence of EDTA was probably due to the effect of EDTA on casein rather than on the enzyme. Metal ions, except for iron (++, +++), c0pper (++) and mercury (++), restored part of the enzy— matic activity which had been destroyed by the addition of EDTA to the assay system. Casein is known to contain metal ions which are important to its structure. There- fore, EDTA may have altered the tertiary structure of the substrate by chelating the metal ions and, thereby making the substrate less susceptible to proteolysis. The addi- tion of metal ions would allow the casein to partially return to a configuration which was susceptible to pro- teolysis. Addition of too much metal caused precipitation so that complete reactivation probably was not possible. 51 The effects of iron (++, +++) and c0pper (++) could not be evaluated because erratic readings were obtained from controls which contained substrate and metal ions but no enzyme. Inhibition of the protease by mercury (++) was probably due to the formation of strong complexes between this heavy metal and the protein. The protease was stable to heating at 97 to 99 C for 60 min at pH 6 and 7.5; heating under the same conditions at pH 8.5 resulted in only 21% inactivation of the protease. Similar results were reported by Williamson gt a1. (196A) for an extracellular protease from S; lactis which was inactivated 32% by heating at 98 C for 60 min at pH 7.0. Koffler (1957) reviewed the heat stability of enzymes and other components of thermophilic bacteria and pointed out that the heat stability was probably due to inherent stability of the molecule or to the presence of protective factors. Either of these explanations could apply to S; durans because it is a thermoduric streptococcus and may produce an inherently stable protease. Also, the pro- tease preparation contained nucleic acids which may act as protective factors. Attempts were unsuccessful to reactivate the pro- tease with ribonucleic acid extracted from S; durans cells after inactivation with ribonuclease. This result is neither evidence for or against the requirement of nucleic acids for protease activity and gives no evidence about -c -Pflwmrv "535-” Fm“??? my 52 the nucleic acids as protective factors. However, if the nucleic acids act only to stabilize the protease to heat and most other conditions, it might be possible to remove the nucleic acids and retain proteolytic activity. Most of the proteases reported from lactic strep— tococci were either enhanced or inhibited by reducing agents (Baribo and Foster, 1952; Vanderzant and Nelson, 1953b, 195Ab; Williamson, gt al., 196A; Cowman and Speck, 1967). However, Cowman and Speck (1967) reported an intracellular protease from S; lactis that was not affected by reducing agents and was inhibited by PCMB. Similarly, in this study, S; durans protease was not affected by reducing agents and was inhibited by PCMB. Diisopropyl 1 fluorophosphate also had no effect on the activity, which indicated the enzyme was not an esterase. Inhibition of the protease by sulfhydryl reagents indicated these groups were involved at or near the active site. In a discussion of inhibitors, Webb (1966a,b) pointed out that the inhibition by mercurials is a selec-, tive reaction with sulfhydryl groups on enzymes. Mer- curials will complex most readily with free sulfhydryl groups followed by masked groups. On the other hand, N-ethylmaleimide reacts slowly with only free sulfhydryl groups and is highly selective for these groups. Fre- quently, protein sulfhydryl groups are resistent to iodoacetamide and only react with this compound after m I'D ‘1’ —" ' W . . ~ . ‘Q ‘ J... «a. v '_‘§"4- sw- b 53 "denaturation" of the protein. The results of experiments on inhibition of the intracellular protease from S; durans showed the PCMB was the most effective inhibitor followed by N—ethylmaleimide and iodoacetimide. The protease was specific for the milk proteins, casein and B—lactoglobulin. There was no measurable pro- teolysis of BSA and hemoglobin. Streptococcus durans is '? usually found in milk and cheese and this may explain the I'm specificity of the protease for milk proteins. Analyses for pentoses and phosphorus showed 50 to |_g 70% by weight, respectively, of nucleic acids in the enzyme preparation. The 20% difference between the two analyses could be partially explained by the incomplete removal of phosphate buffer causing a high phosphorus reading. How- ever, the results indicated a high percentage of nucleic acids in the enzyme preparation. The presence of nucleic acids was shown also by disc gel electrophoresis. A stain r for carbohydrate showed the nucleic acids to migrate with and immediately ahead of the protease. The protein showed high electrophoretic mobility because it moved nearly as fast as nucleic acids which are known to have a high mobility. However, if the nucleic acids were complexed with the protein, this would account for the high mobility of the protein. SUMMARY AND CONCLUSIONS An intracellular proteolytic enzyme from Strepto- coccus durans was isolated and partially purified. A 67-fold purification with 53% recovery of the original activity was obtained by ammonium sulfate fractionation, heating, and hydroxylapatite adsorption chromatography. Disc electrophoresis of the enzyme preparation in poly- acrylamide gel showed one protein zone and an adjoining zone of nucleic acids. Protease activity was determined on 0.5% casein in 0.0A M sodium phosphate buffer at pH 7.5 for l min at 37 C. The enzyme exhibited normal reaction kinetics with respect to substrate and enzyme concentration and reaction rate. The protease exhibited an optimum activity at pH 7.5 and a secondary activity at pH 5.5 to 6. Based on the constant ratio between the two optima during purification, the secondary optimum pH was probably due to a second active ionizable form of the same enzyme rather than a second enzyme. The protease activity was dependent on ionic strength of the assay reaction mixture with maximum activity occur- ring at an ionic strength of 0.1. The data gave very 5A 55 little indication whether ionic strength directly or indirectly affected the enzyme reaction components. Purification of the protease was complicated by the presence of nucleic acids. Attempts to remove the ribo- nucleic acid by ribonuclease resulted in the loss of pro— teolytic activity. The addition of ribonucleic acid extracted from S; durans cells did not reactivate the pro- tease. The nucleic acid appeared to stabilize the enzyme, but the relationship was not clear. The protease exhibited remarkable thermal stability. Heating at 97 to 99 C for 60 min at pH 6 and 7.5 resulted in no inactivation, and heating under the same conditions at pH 8.5 resulted in only 21% inactivation of the pro- tease. Streptococcus durans is a thermoduric organism which may have produced an inherently heat-stable enzyme or the nucleic acids present may have acted as protective factors which stabilized the enzyme to heat. Results from the experiments with metal ions indi- cated that no metal ions were required for protease activity. However, after an apparent inhibition of the protease by EDTA, most divalent metal cations reversed the inhibition. This was interpreted as an effect on the casein substrate rather than an effect on the enzyme. Sulfhydryl groups were shown to be involved at or near the active site because the protease was inhibited by three sulfhydryl reagents. p-Chloromercuribenzoate 56 was the most effective inhibitor followed by N-ethylmaleimide and iodoacetamide. Reducing agents and diisOprOpyl fluoro- phosphate had no effect on the proteolytic activity. O. . ..l w -—fl _. '1 'r m v .._1 ‘ “n: O- ‘ FJ' .- l P v a. on H. .. If BIBLIOGRAPHY Pee! .7 BIBLIOGRAPHY Asano, K. 1965. Size heterogeneity of T2 messenger RNA. J. Mol. Biol. 15:71-8A. Balls, A. K. and S. R. Hoover. 1937. The mild-clotting action of papain. J. Biol. Chem. 121:737-7A5. Baribo, L. E. and E. M. Foster. 1952. The intracellular proteinases of certain organisms from cheese and their relationship to the proteinases in cheese. J. Dairy Sci. 35:1A9-160. Breed, R. S., E. G. D. Murray and N. R. Smith. 1957. Bergey's Manual pf Determinative Bacteriology. 7th ed. Williams and Wilkins, Baltimore, Maryland. Clark, W. 8., Jr. and G. W. Reinbold. 1966. Enterococci in young Cheddar cheese. J. Dairy Sci. 39:121A-12l8. ' Clarke, J. T. l96A. Simplified "disc" (polyacrylamide gel) electrophoresis. Annals N. Y. Acad. Sci. 121:A28-A36. Collins, E. B. and F. E. Nelson. 19A9. The effect of Streptococcus lactis and coliform organisms on the soluble nitrogen of milk. J. Dairy Sci. 32:652—658. Cook, G. S. 1955. Practical results with the short-time cheese process. Australian J. Dairy Technol. 10:70-71. Cowman, R. A. and M. L. Speck. 1965a. Activity of lactic streptococci following storage at refrigeration tem- peratures. J. Dairy Sci. A§;1AAl-1AAA. Cowman, R. A. and M. L. Speck. 1965b. Ultralow temperature storage of lactic streptococci. J. Dairy Sci. 38: 1531-1532. Cowman, R. A. and M. L. Speck. 1967. Proteinase enzyme system of lactic streptococci. Appl. Microbiol. 15:851-856. 58 59 Cowman, R. A., H. E. Swaisgood and M. L. Speck. 1967. Proteinase enzyme system of lactic streptococci. 11. Role of membrane proteinase in cellular function. J. Bacteriol. 25:9A2-9A8. Czulak, J., L. A. Hammond and H. J. Meharry. 195A. A new method for Cheddar cheese manufacture. Australian Dairy Rev. 22:18, 20 and 22—23. Czulak, J. and L. A. Hammond. 1956a. The use of Streptococcug durans as a cheese starter. Proc. XIV Intern. Dairy Congr. 2:158—165. Czulak, J. and L. A. Hammond. l956b. The new method for Cheddar cheese manufacture in the li ht of commercial experience. Australian Dairy Rev. g_:11, l3 and 15—16. Czulak, J. and L. A. Hammond. 1956c. Acid development and pH in the new method of Cheddar cheese manufacture. Australian J. Dairy Technol. 11:165-166. Dahlberg, A. C. and F. V. Kosikowsky. 19A8. The develop- ment of flavor in American Cheddar cheese made from pasteurized milk with Streptococcug faecalis starter. J. Dairy Sci. 31:275-28A. Davis, B. J. 196A. Disc electrophoresis. II. Method and application to human serum proteins. Annals N. Y. Acad. Sci. l21:AOA—A27. Dische, Z. 1953. Qualitative and quantitative colorimetric determination of heptoses. J. Biol. Chem. 20A: 983-997. Downs, P. A. 1955. Preliminary studies of the Australian short-time method of making Cheddar cheese. J. Dairy Sci. 38:589. Emmons, D. B., W. A. McGugan, J. A. Elliott and Pamela M. Morse. 1962. Effect of strain of starter culture and of manufacturing procedure on bitterness and protein breakdown in Cheddar cheese. J. Dairy Sci. 55:332-3A2. Ernstrom, C. A., W. V. Price and A. M. Swamson. 1958. Effects of reducing rennet and adding calcium chloride on the manufacture and curing of Cheddar cheese. J. Dairy Sci. 51:61-69. 60 Feagan, J. T. 1956. A modification to the short method for Cheddar cheese manufacture. Australian J. Dairy Technol. 11:1A9-152. Jespersen, N. J. T. 1966. Proteolytic activity of some lactic acid bacteria. Proc. XVII Intern. Dairy Congr. Vol. D:A65-A70. Koffler, H. 1957. Protoplasmic differences between mesophiles and thermophiles. Bacteriol. Rev. gl;227-2A0. Kosikowsky, F. V. and A. C. Dahlberg. 19A8. The growth and survival of Streptococcus faecalis in pasteurized milk American Cheddar cheese. J. Dairy Sci. 31:285-292. Lowry, 0. H., N. J. Rosenbrough, A. L. Farr and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. Morgan, M. E. and F. E. Nelson. 1951. The distribution of certain amino acids in soluble fractions of milk cultures of Streptococcus lactis. J. Dairy Sci. 32:AA6-A56. Morris, T. A. 1955. Quensland experiments with the short process for Cheddar cheesemaking. Australian J. Dairy Technol. lQ:65-70., Ohmiya, K. and Y. Sato. 1967. Studies on the proteolytic action of dairy lactic acid bacteria. III. Changes of the casein treated with Streptococcus cremoris. Agr. Biol. Chem. 31:1318-1326. Ohmiya, K. and Y. Sato. l968a. Studies on the proteolytic action of dairy lactic acid bacteria. IV. Changes of the casein treated with Lactobacillus bulgaricus, Lactobacillus helveticus or Streptococcus lactis. Agr. Biol. Chem. 32:291-296. Ohmiya, K. and Y. Sato. l968b. Studies on the proteolytic action of dairy lactic acid bacteria. V. Changes of casein particle weight caused by lactic acid bacterial fermentation. Agr. Biol. Chem. 323297-302. Peterson, M. H., M. J. Johnson and W. V. Price. 19A8a. Determination of cheese proteinase. J. Dairy Sci. 3_1_:A7-5A. Peterson, J. H., M. J. Johnson and W. V. Price. l9A8b. Proteinase content of Cheddar cheese during making and ripening. J. Dairy Sci. 31:55—61. ( Y 61 Sato, Y. and K. Ohmiya. 1966. Study of protease of dairy bacteria. Agr. Biol. Chem. 30:731-7A1. Sumner, J. B. 19AA. A method for the colorimetric deter- mination of phosphorus. Science 100:Al3-A1A. Vanderzant, W. C. and F. E. Nelson. 1953a. Proteolysis by Streptococcus lactis grown in milk with and without controlled pH. J. Dairy Sci. 36;110A-1lll. Vanderzant, W. C. and F. E. Nelson. 1953b. Character- istics of an endocellular proteolytic enzyme system of Strgptococcus lactis. J. Dairy Sci. 36:1212-1222. Vanderzant, W. C. and F. E. Nelson. l95Aa. Amino acids and peptides in the protein-free fraction of milk before and after incubation with Streptococcus lactis. J. Dairy Sci. 31:790-79A. Vanderzant, W. C. and F. E. Nelson. 195Ab. Character- istics of some endocellular peptidases of Streptococcus lactis. J. Dairy Sci. 31:795-80A. Walter, H. E., A. M. Sadler and C. D. Mitchell. 1958. Labor-saving methods for making Cheddar cheese. J. Dairy Sci. 31:558-559. Walter, H. E., A. M. Sadler, J. P. Malkames and C. D. Mitchell. 1953. Short-time method for making American Cheddar cheese from pasteurized milk. Mimeogr. Report U. S. Dept. Agr. BDI-Inf-158. Walter, H. E., A. M. Sadler, J. P. Malkames and C. D. Mitchell. l956a. A simplified short-time method for making Cheddar cheese from pasteurized milk. Mimeogr. Report U. S. Dept. Agr. ARS-73—ll. Walter, H. E., A. M. Sadler, J. P. Malkames and C. D. Mitchell. 1956b. A simplified short—time method for making Cheddar cheese from pasteurized milk. Proc. XIV Intern. Dairy Congr. g 598-606. Walter, H. E., A. M. Sadler, J. P. Malkames and C. D. Mitchell. 1957. Method of manufacturing cheese. U. 8. Pat. 2,796,351. Webb, J. L. 1963. Enzyme and Metabolic Inhibitors. Vol. I. General prinprIes of Ifihibition. Academic Press, New York. 62 JWebb, J. L. 1966a. Enzyme and Metabolic Inhibitors. Vol. II. Malonate, analogs, dehydroacetate, sulfhydryl reagents, o-iodosobenzoate, mercurials. Academic Press, New York. , Webb, J. L. 1966b. Enzyme and Metabolic Inhibitors. Vol. III. Iodoacetate, maleate, N-ethylmaleimide, alloxan, quinones, arsenicals. Academic Press, New York. Williamson, W. T., S. B. Tove and M. L. Speck. l96A. Extracellular proteinase of Strgptococcus lactis. J. Bacteriol. 81:A9-53. 1 _. . . .0 'Iv‘o‘.¢;O‘COOL —a-.--¢c¢c¢x~ 'I.\“\"""'i,‘.nm¢1 . .--..—--~.¢laa¢r "AIAAIAAAAI