\J PURIFICATiON Am muttamumu Gr STAPHYLG-wAGULA-Sfi Think, huh; flown 9f 5311.9.» ‘ MiCHIGAN STATE ‘UZNfiIEasiTY‘ land 1.5%}?! Jr, ; 13963? - . This is to certify that the thesis entitled Pfirification and Characterization of Staphylocoagulase presented by Zeno Zolli, Jr. has been accepted towards fulfillment of the requirements for Ph.D. Microbiology and Public Health degree in (fiizrv4£:;h(14fi24vut4;2(5 Major professor Duel5 February 1963 0-169 LIBRARY Michigan State University ABSTRACT PURIFICATION AND CHARACTERIZATION OF STAPHYLOCOAGULASE by Zeno Zolli, Jr. The antigenic relationship between coagulases from some twenty phage propagating strains of the International-Blair series of staphylococci was studied. The coagulases were prepared by acid and ethanol precipitation according to methods of Tager (1948) and recent modifications of Blobel et al. (1960). Inconclusive results from these materials stimulated a research program to develop procedures for the isolation and purification of coagulase. Progress in this quest was gauged by a series of serological and chemical characterizations. Inhibition tests using early preparations with a puri- fication factor of 21 to 79X showed that maximal anticoagu- lase activities occurred in homologous antigen-antisera systems. Although there was a low degree of cross-inhibition among the strains, no consistent correlation was observed corresponding to phage group. This fact suggested possible Zeno Zolli, Jr. strain specificity of coagulase. Further elucidation of this phenomenon using gel diffusion patterns proved diffi- cult to interpret because of the high number of precipitin bands. Several approaches were investigated for the increased purification of coagulase from Staphylococcus aureus 70. In the first, coagulase was separated by concentration in the syneretic fluid from the clot complex. This method resulted in a partially purified preparation with a 300- fold purification factor. Another method yielding a par— tially purified preparation (approximately BOO-fold) employed only gel filtration (Sephadex G-200) which per- mitted separation of proteins by differences in molecular weight. The final procedure achieved extreme purification of this clotting material by using three cycles of dialysis in ethanol-water mixtures under controlled conditions according to modifications of methods of Cohn et a1. (1946) and Pillemer et a1. (1948) followed by molecular sieving through a column of Sephadex G-200. By manipulation of» five variables (pH, ionic strength, temperature, protein and ethanol concentration), the final preparation showed an Zeno Zolli, Jr. approximate 3700—fold increase in activity per mg protein. The final procedure for the separation and isolation of coagulase, evolved from a series of detailed experiments, was composed of four steps as follows: (a) The first fraction (Cg-I-P) was precipitated from six 200 ml aliquots of cell-free broth filtrate by dialysis against a sodium acetate buffer(pH 3.8 and ionic strength 0.1) containing 10% ethanol (v/v) at -4C. Each precipitate was then redissolved in 50 ml of 0.1 N sodium acetate solution. (b) Secondly, each solution of Cg-I—P was dialyzed against a sodium acetate buffer (pH 5.2 and ionic strength 0.05) containing 10% ethanol (v/v)at -4C, and the resulting precipitates (Cg-II-P) were each redis- solved in 5 ml of 0.1 N sodium acetate solution. (c) The solution of Cg-II-P was then dialyzed against a phosphate buffer (pH 6.1 and ionic strength 0.1) containing 10% ethanol (v/v) at —4C; in this step, the supernatant fluid (Cg-III-S) was retained while the precipitate was discarded. (d) In the final step, the solution of fraction Cg- III-S was concentrated (10X) prior to molecular sieving Zeno Zolli, Jr. through a column of Sephadex G—200. The final active fraction was designated as Cg-IV-C. The successfully isolated coagulase was serologically and chemically characterized. Using gel diffusion tech- niques, the use of the fourth fraction as antigen against anti-fraction III serum produced one precipitin band. A more critical aspect of this test employing the fourth fraction against anti-fraction I serum produced a second but weak band. Additional confirmation of purity was evi- denced by the appearance of a single peak using cellulose acetate paper electrophoresis. Progressive elimination of carbohydrate, deoxyribonuclease, lipase and phosphatase was observed through the four stages of purification. Temperature stability studies showed that increasing purity corresponded to decreasing heat stability. PURIFICATION AND CHARACTERIZATION OF STAPHYLOCOAGULASE by Zeno Zolli, Jr. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1963 3— 2 5’ '7 <1" 5/ > I - J’ 'I,‘ h. ACKNOWLEDGEMENT The author wishes to thank Dr. Charles L. San Clemente for his stimulating discussions and many helpful sugges- tions during the course of this study. His willingness to assist students, regardless of imposition on his own time, was deeply appreciated. ii TABLE OF CONTENTS ACKNOWLEDGEMENT . . . . . . . . . LIST OF TABLES . . . . . . . . . LIST OF FIGURES . . . . . . . . . INTRODUCTION . . . . . . . . . . REVIEW OF LITERATURE . . . . . . Coagulase and Pathogenicity Antigenicity of Staphylocoagulase Serological Specificity of Coagulase Purification of Coagulase MATERIALS AND METHODS . . . . . . Cultures and Their Maintenance Concentration and Purification Partial purification by acid precipitation according to Tager (1948) and Blobel e1 of Coagulase and ethanol methods of a1. (1960) Partial purification by separation from clot complex Partial purification by gel filtration (Sephadex alone) Extreme purification using ethanol-water mixtures under controlled conditions, and Sephadex Characterization of the Various Fractions Specialized procedures Electrophoretic analysis Production of anticoagulase Serological techniques Coagulase'inhibition test Gel diffusion methods iii PAGE ii vi ix WOW-boobs)!“ 18 18 19 19 20 21 21 22 22 22 24 26 26 28 PAGE Assay methods 29 Coagulase . 29 Phosphatase 30 Total protein 31 Nitrogen 32 Phosphorus 33 Miscellaneous tests 33 Estimation of carbohydrate 33 Estimation of lipase 34 Estimation of deoxyribonuclease 35 Effect of pH 35 Effect of inhibitors 36 Temperature stability 36 RESULTS . . . . . . . . . . . . . . . . . . . . . . . 37 Evaluation of Various Cultural Conditions for Optimal Production of Coagulase 37 Effect of incubation temperature 37 Effect of shaking 38 Effect of addition of trace minerals to medium 40 Concentration and Purification of Coagulase 40 Partial purification by acid and ethanol precipitation 40 Partial purification by separation from clot complex 43 Partial purification by gel filtration (Sephadex alone) 44 Extreme purification using ethanol-water mixtures under controlled conditions, and Sephadex 47 iv Preliminary experiments Summary of final procedure Characterization of the Various Fractions Results of specialized procedures Electrophoretic analysis Serological studies Results of assays Phosphatase Nitrogen and phosphorus Results of miscellaneous tests Estimation of carbohydrate Estimation of lipase Estimation of deoxyribonuclease Effect of pH Effect of inhibitors Temperature stability DISCUSSION . . . . . . . . . . . . . . . . . SUMMARY . . . . . . . . . . . . . . . . . BIBLIOGRAPHY . . . . . . . . . . . . . . . . PAGE 47 so 63 63 63 63 68 68 71 74 74 74 75 75 7s 76 82 89 91 TABLE LIST OF TABLES Effect of shaking and the addition of trace minerals to brain heart infusion upon the production of coagulase by Staphylococcus aureus 70 . . . . . . . . . . . . . . . . . . Purification factors of coagulases prepared by acid and ethanol precipitation . . . . . . Separation of coagulase from phosphatase by concentration in the syneretic fluid from the clot complex . . . . . . . . . . . . . . . . . Experimental proof for the retention of coagulase in the clotting complex . . . . . . Precipitation of coagulase fraction Cg-I-P using various concentrations of ethanol at fixed conditions of temperature, protein concentration, pH and ionic strength of acetate buffer . . . . . . . . . . . . . . . . Precipitation of coagulase fraction Cg-I-P using an acetate buffer of various ionic strengths at fixed conditions of temperature, protein concentration, pH and ethanol . . . . Precipitation of coagulase fraction Cg-I-P at two strengths of ethanol using different pH values of acetate buffer at fixed conditions of temperature, protein concentration and ionic strength . . . . . . . . . . . . . . . A set of optimal conditions for the precipita- tion of coagulase fraction Cg—II-P in ethanol- water mixtures at constant temperature and protein concentration but at various pH values and ionic strengths of acetate . . . . . . . . vi PAGE 41 42 45 46 52 53 54 55 TABLE 10. ll. 12. 13. 14. The set of conditions for the solubility of coagulase fraction Cg-III-S in ethanol-water mixtures varying in pH but at a fixed tempera- ture, protein concentration and ionic strength of phosphate buffer . . . . . . . . . . . . Summary of the purification factors and per cent recovery of successive fractions of coagulase . . . . . . . . . . . . . . . . . Cross-neutralization of constant amounts (depending upon purity of preparation) of the coagulases from staphylococcal strains of the International-Blair groups by their respective antisera O O O O O O O I O O O O O O O O O 0 Amounts of residual phosphatase and its eventual elimination during the purification of coagulase as measured in the precipitates and corresponding supernatant fluids of successive fractions obtained in three cycles of ethanol-water mixtures under congrolled conditions followed by passage through Sephadex‘ G-200. Proof of separability required meas- urement of phosphatase at two pH values . . Phosphorus and nitrogen determinations of each successive fraction of coagulase obtained from three cycles of ethanol-water mixtures under controlled conditions followed by molecular sieving through Sephadex G—200 . . . . . . . The progressive elimination of carbohydrate, lipase, deoxyribonuclease and phosphatase as measured in the precipitates and corresponding supernatant fluids of successive fractions obtained from three cycles of ethanol-water mixtures under controlled conditions followed by molecular sieving through Sephadex G-200 vii PAGE 56 62 66 73 77 78 TABLE PAGE 15. Effect of trypsin, sodium fluoride and EDTA on the activity of successive fractions of increasingly purified coagulase . . . . . . . . 80 viii FIGURE LIST OF FIGURES PAGE Effect of temperature on coagulase production by Staphylococcus aureus 70 in brain heart infusion . . . . . . . . . . . . . . . . . . . 39 Separation by gel filtration (Sephadex G-200) of a coagulase preparation obtained by preci- pitation of a cell-free filtrate at pH 3.8 . . . 48 Preparation of extremely purified coagulase Cg-IV—C by molecular sieving of Cg-III-S through a column of Sephadex G-200. The eluant was 0.1 N sodium acetate (pH 7.2) . . . . . . . 61 Patterns of the first fraction (A) of coagulase, Cg-I—P as contrasted with the final fraction (B) of coagulase Cg—IV—C indicating electro- phoretic homogeneity on cellulose acetate strips using barbital buffer at pH 8.6 and ionic strength 0.07 . . . . . . . . . . . . . . . . . 64 Precipitation bands of the gel diffusion tech- nique indicating the progressive purity of each of the four fractions of coagulase. Wells A (top) contain antisera prepared in rabbits against the third fraction Cg-III-S. wells B (bottom) contain antisera prepared in rabbits against the first fraction of coagulase, Cg-I-P. 69 Separation of phosphatase from coagulase by molecular sieving the first fraction (20 mg protein per ml of sample) of coagulase Cg—I-P through a column of Sephadex G-200. The eluant was 0.1 N sodium acetate (pH 7.2). Phosphatase activity was measured at both pH 5.6 and 7.2 . . 72 Effect of pH on the clotting activity of extremely purified coagulase fraction Cg-IV—C. . 79 Effect of temperature at various intervals upon the stability of successive fractions of purified coagulase . . . . . . . . . . . . . . . 81 ix INTRODUCTION The exact role of coagulase in the pathogenic process of Staphylococcus aureus has been a matter of conjecture for over half a century. The means by which this enzyme-like material may enhance virulence is unknown. Although other cellular constituents have been incriminated as factors of pathogenicity, coagulase is demonstrated most consistently and readily among the virulent strains of this organism. The development of procedures for the purification of coagulase has made possible more critical studies of its antigenicity and mode of action. Barber and Wildy (1958) and Blobel et a1. (1960) showed a relationship between the bacteriophage groups of staphylococcal strains and the anti- genic specificity of free coagulase produced by correspond- ing strains. Significant findings along these lines could conceivably lead to simpler diagnostic procedures for staphylococcal strain identification. This report is concerned with a study of the possible relationship between phage type and antigenic specificity of coagulase. The inconclusive results obtained with partially purified coagulase indicated the need for a preparation of greater purity than the one employed. 1 Methods were developed to separate and purify coagulase to a high degree. Antigenic and chemical properties of the purified coagulase were determined. REVIEW OF LITERATURE Coagulase and Pathogenicity Since the discovery by Loeb (1903) that S, aureus possesses the ability to clot goose plasma, persistent attempts have been made to determine the exact relationship of this clotting material to the virulence of g, aureus and the role it plays in pathogenesis. The earliest observa- tion that a correlation may exist was reported by Much (1908). Among the reports which followed and agreed with this view were those of Daranyi (1926), Cruickshank (1937), Smith et a1. (1947) and Blair (1958). Other factors which have been reported to be associ- ated with virulence were phosphatase (Rangam and Katdare, 1954), urease (Fusillo and Jaffurs, 1955), mannitol fermen- tation (Schaub and Merrit, 1960) and alpha hemolysin (Brown, 1960). Jeffries (1961) attempted to correlate coagulase with several activities which included deoxyribonuclease, mannite fermentation, pigment production, reduction of phenyltetrazolium, and growth on tellurite glycine agar. Some 494 strains of staphylococci isolated from routine specimens in the laboratory of the Detroit Receiving Hospi- tal were used in the study. Deoxyribonuclease, mannite 3 fermentation or pigment production correlated to a high degree with coagulase production and could be used as pre- liminary tests for potential pathogenicity among the staphylococci. Among the organisms tested, deoxyribonu- clease activity correlated with coagulase production to the highest degree. Citations for other tests for the pathoqenicity of §, aureus include those of Blair (1939, 1958) and Angyal (1961) who both reported data supporting the reliability of coagulase production, and suggested its use as sole criterion. Antigenicity of Staphylocoagulase The correlation of coagulase production to the viru- lence of S, aureus led to investigations of the possibility that antibodies specific for coagulase could confer pro- tection for g, aureus. Initial attempts to elicit antibodies specific for coagulase by culture supernatant fluid proved fruitless (Gross, 1931; Walston, 1935; Smith and Hale, 1944). How- ever, Lominski and Roberts (1946) showed that coagulase inhibition by human sera exhibited "inhibitory substances r which have antibody characteristics, but cannot yet be accepted as such." It was not until Rammelkamp (1948) noted an increase in anticoagulase titers of acute and con- valescent phase sera of patients infected with staph- ylococci that an intensive study of this phenomenon was instigated. Subsequent studies by Rammelkamp et a1. (1950) showed anticoagulase titers in monkeys injected with cell- free coagulase preparations. Tager (1948) developed a method for the partial purification of coagulase and paved the way for additional studies involving the experimental production of antibodies to staphylocoagulase (Tager and Hales, 1948). These experiments showed that coagulase was antigenic in rabbits. Exposure to a partially purified preparation necessitated intensive and prolonged injection regimes as well as combination with alpha—lysin. Aluminum phosphate as an adjuvant with purified coagu- lase was used by Duthie and Lorenz (1952). A suspension of washed aluminum phosphate (6 mg/ml) adsorbed up to 3.5 mg coagulase per mg of aluminum phosphate from a 1% coagulase preparation at pH 4.8. Intramuscular or subcutaneous inoc- ulations of rabbits with this material consistently yielded higher anticoagulase levels than coagulase alone or in con- junction with alpha lysin. Barber and Wildy (1958) combined potassium aluminum sulfate with coagulase at pH 6.8 for four or more weekly injections in rabbits. Blobel et a1. (1960) obtained antibodies from rabbits against alcohol precipitated coagulase. Potassium aluminum sulfate was used as an adjuvant in the manner described by Barber and Wildy (1958). Six intramuscular injections were administered at weekly intervals and consisted of a total dosage of 12 mg of 1yophilized preparation per animal. An adjuvant was used which consisted of mineral oil and an emulsifying agent (mannide monooleate). The electrophor- etically purified preparation of coagulase was dissolved in physiological saline solution (pH 7.0) and mixed with the oil and emulsifying agent in a ratio of 10:9:1 respec- tively. The presence of either adjuvant doubled the anti- coagulase titer (1:180). Serological Specificity of Coagulase Rammelkamp et a1. (1950) observed that coagulase may be serologically distinct. Using antisera produced against coagulases derived from four different strains of staphylo- cocci, Duthie (1952) and Duthie and Lorenz (1952) found four antigenically distinct types of coagulases and suggested the existence of others. In spite of a low degree of cross inhibition, the majority of coagulases were inhibited by only a single antiserum. A suggestion by Smith (1954) was followed with an attempt by Barber and Wildy (1958) to determine whether a correlation does in fact exist between coagulase antigenic specificity and bacteriophage type. They chose strains which were representative of three main bacteriophage groups (strain M, phage type 52/80 - group I; strain W120, phage types 3B/3C/55 - group II; strain Newman, phage types 7/47/53/54/73/75/77 - group III). Antisera were prepared by injection of alum precipitated partially puri- fied coagulase into young adult rabbits. A close relation- ship between bacteriophage groups of staphylococcal strains and the antigenic specificity of free coagulase was found from corresponding strains. Blobel and Berman (1960) reported experiments designed to obtain data concerning the serologic specificity of coagulase. By using rabbit anticoagulase sera against alcohol precipitated coagulases representing phage groups I, II, III, IV and Misc., each coagulase was neutralized to a greater degree by its homologous antiserum. This indicated a relationship between bacteriophage group and antigenic specificity, but a low level of cross neutrali- zation did occur among all pairs. The difference in neu- tralization titers did not appear to be sufficiently dis- tinct to justify the use of coagulase cross inhibition tests to classify staphylococci for epidemiological inves— tigations. Using a limited number of organisms, similar results were obtained by Zen—Yoji et a1. (1961). More conclusive results may have been obtained by using an extremely purified coagulase preparation. Purification of Coagulase The determination of the role of coagulase in the pathogenesis of staphylococci spurred early attempts to purify this material. Walston (1935) and Fisher (1936) were among the first to use alcohol to precipitate crude coagulase from broth cultures of S, aureus. Lominski (1944) separated supernatant fluid from 12 hr broth cultures of S, aureus and found that coagulase passed through Chamber- land L3 filters. Similarly, Smith and Hale (1944) found coagulase to be readily filterable through gradacol membranes of suitable porosity. In addition, they reported coagulase thermostable and particulate, or associated with uniform particles. Rammelkamp et a1. (1950) obtained coag- ulase preparations for serological investigations by growth of S, aureus in albumin enriched broth for 3 days at 37C. Gerheim et a1. (1947) prepared crude coagulase extracts by alcohol precipitation. The cells were removed from the broth of 48 hr cultures of S, aureus by filtration or centrifugation. Ethanol (95%) was added to the cell-free filtrate in a 3:1 ratio respectively at 0C. The yield from 850 ml of broth culture filtrate was approximately 1.5 g of a "greyish-brown powder" which was completely soluble as a 1% solution in citrate-borate buffer (pH 7.55). No inactivation of the "powder" occurred when stored at 5C for 8 months. Solutions remained stable for 10 days at room temperature. Tager (1948) was one of the earliest workers to purify coagulase. An inoculum was prepared by growing S, aureus #104 in brain heart infusion for 3-6 hr at 37C. A 10-15% inoculum of this culture was transferred to a 1 liter flask containing 175 m1 of the same broth. Then flasks were incubated for 4—6 days at 37C. The cells were removed from 10 15-20 liters of broth culture with a Sharples super— centrifuge. The cell—free supernatant fluid was acidified to pH 3.8 - 4.0 by the addition of 4 N HCl. This mixture was allowed to stand at 4C for 12—24 hr and the precipi- tate collected. This precipitate was then washed several times with phosphate buffer solution (pH 3.8 - 4.0), resus- pended in M/15 phosphate buffer solution (pH 8.2) and the insoluble residues discarded. Reprecipitation was accom- plished by reacidification to pH 6.5 followed by the addi- tion of 3 vol ethanol (95%) and storage at —5C. The pre- cipitate was washed, dissolved in phosphate buffer solu- tion (pH 8.2) and insoluble debris removed. Ammonium sul- fate (8-12% saturation) was added and the resulting precip- itate discarded. Ethanol and ammonium sulfate fractiona- tion were repeated several times. Results showed a 300- 400 fold increase in purification and indicated that (a) coagulase was most stable at pH 4.5 - 7.0; (b) crude coag— ulase was thermostable and resisted autoclaving at 120C but that the partially purified material was more heat 1a— bile; (c) purified coagulase was non-dialyzable; and (d) coagulase was protein in nature. Drummond and Tager (1959) showed that coagulase preparations possessed 0.5% 11 carbohydrate plus esterase activity. When 20 mg samples were used, they obtained a coagulase titer of 1:112,640 in 18 hr. Boake (1956) modified Tager's (1948) technique by eliminating the addition of ammonium sulfate and relied only on alcohol and pH adjustments for purification. From 10 liters of culture supernatant fluid, 2 g of concentrated, partially purified coagulase were recovered. Taking advantage of the thermostability of crude coag- ulase preparations, Walker et a1. (1948) cultivated S, aureus in beef heart tryptic digest broth and autoclaved the culture filtrates for 20 min at 120C in order to remove certain interfering proteins by coagulation. Then the supernatant fluid was acidified to pH 4.0 with HCl, cooled to OC and the precipitate was allowed to settle overnight. The precipitate was washed three times with 0.1 original volume of filtrate as follows: (1) cold sodium acetate buffer solution at pH 4.0 and ionic strength 0.1 (2) cold sodium acetate buffer solution (pH 4.0, ionic strength 0.01) (3) cold distilled water. Finally, the precipitate was adjusted to pH 7.5 and the insoluble debris discarded. This final product was stored in vacuum dried form. Yields 12 varied between 25 mg and 60 mg per 100 ml of original supernatant fluid. Duthie and Lorenz (1952) used cadmium sulfate to par- tially purify coagulase. Two liter flasks containing 400 m1 nutrient broth were seeded with S, aureus strain Newman (NTCT 8178) and incubated at 37C for 9 hr without shaking. An additional 3 hr of cultivation was conducted with shak- ing. It was shown that shaken cultures contained four to five times more coagulase activity than non-shaken cultures did. Bacteria were removed from the broth culture by mix- ing with Filter-Cel and passage through filter paper. After cooling to 4C, cadmium sulfate was added to a final concentration of 0.5% (w/v) and the pH adjusted to 5.8. The precipitate, which resulted from overnight storage at 4C, was dissolved in l N HCl solution and adjusted to pH 2.0. Dialysis against tap water removed the cadmium at this pH and the material was stored in 1yophilized form. The resulting preparation contained 6000 MCD (minimal clot- ting doses) per mg. Using cultures grown in 200 ml quanti- ties of peptone yeast extract media and incubated in shal- low layers in Roux bottles for 1 to 3 days at 37C, Barber and Wildy (1958) obtained coagulase preparations which were 13 then purified with cadmium sulfate according to methods of Duthie and Lorenz (1952). Duthie and Haughton (1958) used casein hydrolysate as a growth medium and continuous shaking to obtain optimal coag- ulase production. After removal of the cells from the broth by filtration through EMchner funnels containing Hyflo Super—Cel, cadmium sulfate was added to the culture free fluid and allowed to stand fOr 16 hr at 4C. A water slurry was then made of the precipitate and HCl was added to a pH of 2.0. This solution was dialyzed for 48 hr and ammonium sulfate added at a concentration of 67% saturation. Reprecipitation with ammonium sulfate was performed. Murray and Gohdes (1959) also used cadmium sulfate to purify coagulase. Cultures of S, aureus were grown in tryptose broth for 72 hr at 37C. After removal of the cells, ammonium sulfate was added to the supernatant fluid (33%1w/v) and allowed to stand at 4C for 24 hr. The precip- itate was harvested by centrifugation, dispersed in sodium citrate solution (3.2% w/v) and adjusted to pH 5.0. Cadmium sulfate was added and the precipitate discarded. Coagulase was then precipitated from the supernatant fluid 14 by the addition of ammonium sulfate (33% w/v) and subse— quently solubilized with a 3.2% solution of sodium citrate. Further purification was accomplished by stepwise elutions from a cadmium sulfate chromatOgraphic column with increas- ing concentrations of phosphate buffer. Using S, aureus grown in heart infusion for 5 days on a shaker at 37C, Blobel et a1. (1960) partially purified coag- ulase by acid and ethanol precipitation. Addition of ammo- nium sulfate proved useless since even at 5% saturation considerable amounts of contaminating proteins were precip- itated. Additional purification was accomplished by starch block electrophoresis (pH 8.4, 200 v, 20-30 ma) for 16 hr. Migration of active coagulase was toward the anode. Although an overall purification factor of 387.5 was achieved and the final preparation contained high activity per unit nitrogen, electropherograms revealed a hetero- geneous mixture. The use of anion exchange chromatography produced a slight increase in purification; and, recovery of active coagulase was less than 50%. Using high voltage column electrophoresis, a contaminating egg yolk factor was separated from purified coagulase preparations by Blobel et a1. (1961). 15 Using methods similar to those of Blobel et a1. (1960), Inniss and San Clemente (1961, 1962) were unable to sep- arate phosphatase from coagulase activity. Results obtained by anion exchange chromatographic studies, in which gradient elutions were made from DEAE-cellulose columns with 0.01 M tris buffer solution, showed that maximal coagulase and phosphatase activity occurred in the same fraction. Subse- quent experiments using starch block electrophoresis (pH 8.6, 200 v, 2 ma) gave similar results. A small degree of separation by electrophoresis did occur when a discontinu- ous buffering system was used. According to Cohn et al. (1946, 1950), the separation of components from blood plasma by fractional precipitation has the advantage that the material which remains insoluble is protected from various chemical and enzymatic changes which may rapidly occur in solution. By use of variations in ionic strength, pH, temperature, protein and alcohol concentration, they were able to separate several proteins successfully from plasma. Using this same technique, Pillemer et a1. (1948) crystallized tetanal toxin. Their procedures involved the use of methanol, since toxin denaturation by ethanol was encountered. 16 Initial attempts to utilize these procedures to purify coagulase were made by Tagerand Lodge (1951). Cell-free culture filtrate was acidified to pH 5.2 with 5 M acetic acid solution and cold methanol was added to a final con- centration of 20%. Maintenance of the temperature below 1C for 18 hr resulted in a precipitate which was separated and then redissolved in 0.15 M sodium acetate solution (pH 7.4). This step was followed by adjustment to pH 5.4 and addition of cold methyl alcohol (final concentration of 17%). Finally, a third precipitation was carried out at pH 5.3 with 0.075 M sodium acetate, 20% methyl alcohol and -5C. An increase of approximately 125-fold in purifica- tion and 50% recovery of coagulase was found. Further attempts to utilize alcohol precipitation for purification of coagulase were made by Blobel (1959). After precipitating coagulase from cell-free broth fil— trate at pH 3.8, the material was suspended in M/15 phos- phate buffer solution (pH 7.4) and methanol added to final concentrations of 25%, 50% and 75%. Temperatures were maintained at below —10C for 16 hr. No precipitation occurred at alcohol concentrations of less than 25% under these conditions. When methanol was used in final 17 concentrations of 50%, approximately 85% of coagulase activity was lost. The best yield with only 28% loss of total coagulase activity was found at alcohol concentra— tions of 75%° Similarly designed experiments indicated that optimal coagulase precipitation occurred when the final concentration of ethanol was 70%. MATE RIALS AND METHODS Cultures and Their Maintenance Twenty phage propagating strains of S, aureus of the International-Blair series (Blair and Carr, 1953 and 1960) were used in these studies. These organisms were desig- nated as follows: Group I — phage types 52A/79, 80; Group II — phage types 3A, 3B, 3C, 55; Group III - phage types 187, 53, 83(VA4), 73, 6, 77, 71, 47, 54, 75, 7, 70; Group IV - phage type 42D; Group Miscellaneous - phage type 81. Stock cultures were maintained on brain heart infusion agarl slants at 4C and were transferred approxi- mately every two months. To eliminate strain variation with respect to loss of coagulase production, periodic transfers were made onto brain heart infusion agar plates containing 10% human plasma. Typical colonies around which the largest zone of fibrin occurred were chosen for restocking. 1Difco Laboratories, Inc., Detroit, Michigan. 18 19 Concentration and Purification of Coagulase Several methods were used to concentrate and purify coagulase. In initial investigations, coagulase was partially purified according to methods of Tager (1947) and recent modifications of Blobel et a1. (1960). Experimental results showed the need for a preparation of increased purity. Thus, several other techniques were developed in an attempt to purify this clotting material to a high degree. Partial purification by acid and ethanol precipitation according to methods of Tager (1948) and Blobel et a1. (1960) Twenty strains of S, aureus representing phage groups I, II, III, IV and Miscellaneous were used as sources of coagulase for partial purification. Three hundred ml of culture in log phase growth were added to 6 liter Florence flasks which contained 3 liters of brain heart infusion. These flasks were uncubated for 12 hr at 37C on a rotary shaker (ca 150 cycles/min). The cells were removed by use of a Servall continuous-flow superspeed centrifugel (Model Ivan Sorvall, Inc., Norwalk,Connecticut. 20 KSB-l) at a flow rate of 50 ml/min. Using 4 N HCl, the supernatant fluid was acidified to pH 3.8 and allowed to stand at 4C for 18 hr. The precipitate was collected, 50 ml distilled water added and the solution adjusted to pH 7.2 with 0.066 M disodium phosphate buffer. After stir- ring, the insoluble residues were removed by centrifuga- tion at 12,000 x g for 20 min. To the filtrate, ethanol (95%X was added slowly to a final concentration of 70% (v/v). Precipitation was allowed to occurr at -20C for 18 hr. The precipitate was removed at -20C, 50 ml distilled water and enough of 0.066 M potassium dihydrogen phosphate buffer solution were added to adjust the pH to 7.4. Three cycles of ethanol precipitation were repeated before con- centration of the product by dialysis overnight against polyvinylpyrrolidonel and subsequent lyophilization. Partial purification by_separation from clot complex Elek (1959) suggested that the conversion of fibrino- gen to fibrin by coagulase may involve the formation of a complex with substances in the clot. This observation prompted a series of investigations to determine whether a leford Laboratories, Redwood City, California. 21 complex did indeed occur and whether this system would yield purified coagulase. Partial purification by gel filtration (Sephadex alone) The use of particular dextran gels for separation of amino acids, peptides and proteins has been described by Porath (1960). These methods using Sephadex1 G-200 were adapted and modified to the purification of coagulase. Extreme purification using ethanol-water mixtures under controlled conditions, and Sgphadex The separation of extremely purified coagulase from cell-free supernatant fluid was accomplished in four steps, three of which employed modifications of procedures of Cohn et a1. (1946) and Pillemer et a1. (1948). The fourth step utilized gel filtration (Sephadex G-200) according to modifications of methods of Porath (1960). l . , . PharmaCia Fine Chemicals, Uppsala, Sweden. 22 Characterization of the Various Fractions Specializedyprocedures Electrophoretic analysis. With partially purified material, starch block electrophoresis was used as an attempt to separate coagulase from phosphatase activity. Procedures according to Blobel et a1. (1960) and modified by Inniss (1961) were used. Approximately 500 g of insol- uble potato starch were washed 2-3 times with 1 liter amounts of 0.02 N potassium hydroxide solution followed by washing with distilled water and then the desired buffer. To mold the starch block, a thick slurry was prepared and added to a plastic template (2 x 24 cm). After hardening, a cross-wise well was excised from the starch block and the sample in the desired concentration was placed in the well. The plastic template was subsequently arranged in the Universal apparatus1 in a manner similar to that used for cellulose acetate paper electrophoresis. Migration of components was allowed in an electrical field. The material was subjected to 200 v and 2 ma current for 16 hr at 4C. 1Shandon Scientific Co., London, England. 23 The starch block was segmented at 1 cm intervals and the contents eluted with barbital buffer solution (pH 8.6). With extremely purified coagulase preparations, small scale electrophoresis using cellulose acetate strips was performed as described by Smith (1960). Oxoid strips (2.5 x 12 cm) were impregnated by floating them on the surface of the desired buffer solution. This technique eliminated opaque spots (entrapped air) which occurred when the strips were rapidly submerged. After the strips were removed from the buffer solution and lightly blotted with filter paper, they were applied across the bridge gap of a Universal apparatus. Using a micropipette and straight edge ruler, in order to obtain narrow zones of origin, 0.01 ml sample was applied to the strip. The time interval varied with the conditions, for example, when barbitone buffer (pH 8.6, ionic strength 0.07) and a current of 0.4 ma/cm were used, optimal separation was achieved in 2 hr. Immediately after removal of the strips from the electrophetic apparatus, they were "fixed" by immersion into 5% trichloracetic acid for 20 min. Protein components were stained by either 0.2% 1 Ponceau S in 3% aqueous trichloracetic acid, or 0.002% 1Allied Chemical Corp., New York, New York. 24 Nigrosinl in 2% acetic acid solution. Staining was more rapid with Ponceau S and gave satisfactory results in 10-20 min. With Nigrosin, overnight immersion of the strips was necessary. After staining, the strips were transferred to a washing solution (5% aqueous acetic acid) until a color- less background appeared. The strips were dried by placing them between two pieces of paper towel and pressing them with glass. An instrument consisting of a double beam recording and integrating reflectance densitometer (Chrom- oscanz) was used to determine the relative concentration of electrical components on the strip. Production of anticoagulase. In our earlier studies employing acid and ethanol precipitated coagulase prepara- tions, either mature Dutch Belt rabbits (2-4 lb) or New Zealand White rabbits (4-6 1b) were used for production of anticoagulase serum. Freund3 adjuvant (1:1) was used as an antibody enhancing agent in all coagulase solutions. For proper emulsification, the addition of adjuvant to sample lAllied Chemical Corp., New York, New York. 2Joyce, Loebl and Co., Gateshead-on—Tyne, England. 3 . . . . . Difco Laboratories, DetrOit, Michigan. 25 was done in small increments while carefully drawing and withdrawing the mixture with a one ml pipette; this tech- nique resulted in a stable emulsion which did not separate upon standing overnight at 4C. The rabbits were inocula- ted subcutaneously by multiple injections in the subscapu- 1ar region at weekly intervals for a period of 5 weeks. For this particular route of administration, 5 mg of coag- ulase mixture was injected each week for a total of 25 mg. Ten days after the last injection, the rabbits were care- fully restrained on a board and blood was removed by cardiac puncture using a 10 ml syringe with attached 2 in. needle (20 gauge). With extremely purified coagulase (3700X) preparations, only mature Dutch Belt rabbits (2-4 lb) were used. An antibody enhancing agent (Freund adjuvant) was used as previously described. However, this material was injected according to techniques of Leskowitz and Waksman (1960) via the foot pad route instead of the subscapular region. In this case, the total inoculation consisting of 2 mg of extremely purified antigen contained in 0.8 m1 of a 0.1 N sodium acetate solution and Freund adjuvant mixture (1:1) was given in 0.2 ml quantities per foot of each rabbit. 26 The rabbits were then bled approximately 10 days after this initial injection. Obviously, since this route required only one injection, some 5 weeks were saved over the sub- scapular route. Adequate quantities of blood (50 ml) were obtained from the marginal ear vein while keeping the appendage warm with a special, electrically heated test tube. Serological techniques. The coagulase inhibition test was used to determine the highest dilution of anti-serum which completely inhibited the clotting activity of coagu- lase. To a series of tubes containing a minimal amount of coagulase sufficient in 0.2 ml to yield a 4+ clot there was added 0.2 m1 of serially diluted serum. The tubes were incubated in a water bath at 37C for 30 minutes. Then, all tubes received 0.5 ml of human plasma (diluted 1:5 in 0.85% NaCl solution), mixed well, and incubated at 37C for three more hours. In some instances, reinforcement of this sys- tem with rabbit serum was followed by an increase in coag- ulase activity. This enhancement was caused by an adequate level of coagulase reacting factor (CRF). To eliminate the masking of coagulase antibodies, normal serum was used in 27 control tubes. In place of human plasma as the source of fibrinogen, an alternate method involved the use of 1.5% (w/v) human fibrinogen plus 1% normal rabbit serum. In our earlier studies, gel diffusion evaluation of coagulase preparations was made according to modifications of methods of Ouchterlony (1958). A good quality agar was dissolved in water at a concentration of 1%. Following the addition of calcium chloride and removal of the preci- pitates by filtration through glass wool, the agar was cut into small cubes and washed for several days in running tap water prior to final washing in distilled water. The agar was melted and to it was added sodium chloride (0.85%). Antibiotic glass petri dishes (90 mm diameter) were used because their flat surfaces were ideal for layering a thin film of this agar. Perfectly flat bottoms are essential for the maintenance of constant agar depth across the plate. Before addition of the agar, six strips of filter paper (Whatman l) measuring approximately 1.5 x 0.5 in. were folded over both ends of the lip of the bottom plate so that part of the strip extended along the bottom and toward the center of the plate and part folded upward and out over the lip. A stainless steel wire bent to form a 28 circle was placed inside of the bottom plate to hold the strips in place. After sterilization of the petri plates in a hot air oven (300F) for 3 hr, 30 ml of sterile agar were poured into each plate and allowed to cool slowly. Before use, the plates were stored at 4C overnight. A central well and six circumferential wells were cut so that distances between each well were equal (10 mm). The anti- gen was placed in the central well and the serum in the peripheral wells, precipitation zones were observed after 5 days incubation at room temperature. A gel diffusion method (Murty, 1960) was used with the extremely purified coagulase preparations. Difco ionagar at a final concentration of 1% was added to phosphate buf- fered physiological saline solution (pH 7.4). Ten m1 quantities of this agar were layered on standard glass lan- tern slides (3 1/4 x 4 in.) and allowed to harden at 4C for 2 hr. After cutting wells to the desired pattern, antigen and serum samples were added. The glass slides were stored at room temperature under humid conditions for 3-5 days. With this system, the development time of precipitin zones was reduced from 5—6 days to 48 hr. After immersing the slides into 0.85% NaCl solution for 4 hr, they were washed 29 in distilled water overnight and the agar dried to a thin film by use of a warm air heater. The zones of precipita- tion were stained according to methods described by Crowle (1961). The slides containing the dried agar film were immersed into stain solution (thiazine red R, 10%; amido- swarz 10B, 0.1%; light green SF, 0.1%; acetic acid, 2%; mercuric chloride, 0.1%; to a desired volume with distilled water) for 5 minutes. The slides were then washed in 5% acetic acid solution to remove excess non-specific stain. Assay methods Coagulase. Coagulase activity was assayed by methods similar to those of Tager and Hales (1948). To the first of a series of non-etched, scrupulously clean test tubes (13 x 100 mm) containing 1 ml 0.85% NaCl solution, 1 ml of coagulase solution was added and serially diluted two—fold. Each tube then received 1 ml diluted human plasma. The plasma was prepared by adding 4 parts of 0.15 M NaCl solu- tion and enough thimerosal to give a final concentration of 1:5000. After thorough mixing, the tubes were incubated in a 37C bath and clot formation was observed at 3 and 24 hr. Reciprocals of the highest dilution of coagulase 30 solution in which any visible clot occurred were recorded as "coagulase reciprocal titers." Importance of early initial readings was essential to detect those samples con- taining fibrin lysing substances with consequent dissolu- tion of the clot. Phosphatase. Modifications of methods of Pelczar et a1. (1956), Barnes and Morris (1957) and Inniss and San Clemente (1962) were used to determine quantitatively any residual phosphatase activity in representative samples from successive fractions of increasingly purified coagu- lase. A 0.4% solution of disodium p-nitrophenylphosphate was used as the stock substrate. P-nitrophenol standard was prepared so that 1 ml contained a concentration of 0.2 uM. Acid phosphatase was measured in citrate buffer (pH 5.6) while alkaline phosphatase was determined in glycine buffer (pH 10.4). Tris buffer (pH 7.2) was also used. To 1.4 ml of the desired buffer solution contained in a test tube (20 x 150 mm), 1 m1 of sample and 0.6 ml stock substrate solution were added. After incubation in a 37C bath for 30 min, 1 ml of l N NaOH solution was added to each tube to stop the reaction and to develop the color of the liberated p—nitrophenol. Optical density was 31 determined with a Bausch and Lomb Spectronic 20 spectro- photometer using a filter with a wavelength of 425 mu. Enzyme activity was determined from a calibrated curve in which optical density was plotted against uM of p—nitro— phenol. Total protein. Quantitative determination of total protein was made with techniques established by Lowry et a1. (1951). Bovine serum albumin1 was prepared so that values on the standard curve ranged from 7.8 to 500 ug per test tube. One ml samples were added to tubes contain- ing 1 ml of 1 N NaOH solution and allowed to stand at room temperature for 10 min. To each tube there were added 1.5 m1 triple distilled water and 5 m1 reagent A (1 ml of 2% copper sulfate 5H 0 and 1 m1 of 2.7% sodium potassium 2 tartrate 4H 0 added to 98 ml of 2% sodium carbonate). 2 After standing at room temperature for 15 min, each tube received 0.5 m1 Folin-Ciocalteu reagent, carefully mixed, and was incubated an additional 35 min at the same tempera- ture. A wavelength of 660 mu on a Bausch and Lomb Spec- tronic 20 spectrophotometer was used to determine the lNutritional Biochemicals Corp., Cleveland, Ohio. 32 optical density. Total protein per ml of sample was deter- mined from a calibrated chart currently prepared. Nitrogen. The micro-Kjeldahl method was used to deter- mine protein nitrogen as described by Kabat and Mayer (1961). To a 10 m1 micro-Kjeldahl flask containing 1 m1 of sample there was added 1 ml of a sulfuric acid solution. The sul- furic acid solution was prepared by adding 0.2 g copper sulfate to 20 m1 of concentrated H2804. Approximately 0.25 g of potassium sulfate and some spherical glass beads were added to each digestion flask and allowed to boil. Diges- tion was carried out by boiling on a gas heated apparatusl until all particles disappeared and the solution became clear. Boiling was continued for an additional one-half hour. Distillations were made with a micro—Kjeldahl dis- tillation apparatus. The ammonia was distilled into a boric acid-methyl red indicator solution. This indicator solution was prepared by the addition of 2 m1 methyl red solution (saturated methyl red in 50% ethyl alcohol) to 100 m1 saturated boric acid solution, and adding 100 ml dis- tilled water. The distillate was titrated with N/70 HCl lPrecision Scientific Co., Chicago, Illinois. 33 and the endpoint recorded when the sample agreed in color intensity with that of the distilled blank. Phosphorus. The Fiske and Subbarow method (1925) was used for the colorimetric determination of phosphorus. A series of test tubes receiving 1 ml sample and 4 ml of 10% trichloracetic acid were allowed to set at room tempera- ture for 10 min. After centrifugation and decantation, the filtrate received 0.5 ml molybdate reagent (2.5% ammonium molybdate in 3 N sulfuric acid), 0.5 ml aminonaphtholsul- fonic acid powder in 195 m1 of 15% sodium bisulfite and 5 m1 of 20% sodium sulfite), and 2 ml distilled water. The tubes were thoroughly mixed and stored in the dark for 10 min. The optical density was measured at a wavelength of 660 mu using a Bausch and Lomb Spectronic 20 spectrophoto- meter. Phosphorus was determined from a calibrated curve prepared with known amounts of standard phosphate solution (1 m1 = 0.004 mg P). Miscellaneous tests Estimation of carbohydrate. The Molisch test according to the method of Gunsalus (1959) was used for the qualita- tive detection of carbohydrate. One-half m1 of solution 34 was placed into a small test tube. Two drops of 5% alco- holic a~naphthol solution were added and the contents thor- oughly mixed. One ml of concentrated sulfuric acid was added slowly down the side of the tube forming a layer beneath the aqueous phase. A pink band at the interface indicated a positive carbohydrate reaction. Estimation of lipase. According to Tietz et a1. (1959), lipase activity was satisfactorily estimated when olive oil was used as a substrate. The oil substrate was an emulsion of 93 m1 of a solution containing 0.2 g sodium benzoate and 7 g gum arabic and 93 ml olive oil. Three m1 of this substrate were added to a test tube containing 1 m1 sample, 2.5 ml water and 1 ml buffer (0.02 M tris, pH 8.0). The tube was mixed thoroughly before incubating in a 37C bath for 14 hr. After adding 3 m1 of ethanol (95%), the entire contents were transferred to a 50 ml beaker. A con- trol tube was similarly treated. The contents of both beakers were then titrated electrometrically with N/20 NaOH to pH 10. By calculating the difference between the control and sample value, the units of lipase activity were recorded as the number of m1 of N/20 NaOH used to neutralize the liberated free fatty acids. 35 Estimation of deoxyribonuclease. DNase test agar1 was used to detect deoxyribonuclease activity of each fraction obtained during the process of coagulase purification. A thin layer of turbid agar was placed into the base of a sterile pertri dish. Shallow wells similar to those used for the gel diffusion technique were then cut into the agar at desired intervals. Several drops of sample were added to each well and diffusion allowed to occur at room tempera— ture for 6-8 hr. After flooding the plates with N/l HCl, positive reactions were recognized by clear zones around the well. Effect of pH. The effect of pH was studied for the various fractions of purified coagulase. Addition of appropriate buffer to the coagulase reaction system resulted in the desired pH value. Coagulase activity was measured using a two—fold serial dilution technique as previously described but in this case we adjusted the pH of the system to 5.0, 6.0, 6.8, 7.0, 7.2, 8.0 and 9.0. A citrate buffer was used to obtain values of pH 5.0 and 6.0. Phosphate buffer was used for ranges between pH 6.8 and 7.2, while tris buffer was used for pH values of 8.0 and 9.0. 1Difco Laboratories, Detroit, Michigan. 36 Effect of inhibitors. Each of the increasingly puri- fied coagulase fractions was exposed to trypsin, ethyl- enediaminetetraacetate (EDTA) and sodium fluoride to deter- mine their effect on the clotting activity of coagulase. To 2 ml of each coagulase fraction was added 1 ml of a highly active (1:250) trypsin preparation1 to a final con- centration of 40 ug/ml. Final concentrations of 0.025 M were used with sodium fluoride and EDTA. After incubation of these mixtures in a 37C bath for 1 hr, coagulase activ- ities were measured using a two-fold serial dilution sys- tem as previously described. Temperature stability. Two m1 aliquots of each coag- ulase sample in 0.1 N sodium acetate were placed into sep- arate test tubes and incubated at 56C, 37C and 25C. Each of the fractions was standardized to obtain approximately equivalent initial coagulase titers. Samples were tested for coagulase at various intervals over a 2 day period. Difco Laboratories, Detroit, Michigan. RESULTS Evaluation of Various Cultural Conditions for Optimal Production of Coagulase During the development of procedures to purify coagu- lase frustration was caused by continually limited yields. Initial enhancement of coagulase activity in the culture filtrate was attempted to overcome this problem. Recently, Solomon (1962) developed a synthetic growth medium for S, aureus Which was not as effective as brain heart infusion in the production of coagulase. In addition to confirming the observations of Tyrrell et a1. (1957) that a biphasic growth medium enhanced bacterial growth 2-30 times over a nonbiphasic system, Inniss (1961) also found no significant change in coagulase production by S, aureus 70. A series of investigations was thus conducted to determine whether temperatures of incubation, increased aeration, or addition of various additives to the medium would affect coagulase production. Effect 9: incubation temperature When incubation using brain heart infusion was carried out on a rotary shaker at various temperatures, maximal 37 38 coagulase was produced at 12 hr and 37C (FIG. 1). A slight delay and reduction in peak activity was noted at 25C. No coagulase appeared at the lowest temperature. Fukui et a1. (1960) reported that initial storage of Pasteurella pestis at 5C for 24 hr followed by incubation at 37C for 6 hr increased the elaboration of certain anti— gens. No significant increase was noted in viable or total cell count over that which occurred during normal incubation at 37C. Experiments were conducted to determine whether a similar system would be beneficial for coagulase production by staphylococci. Primary incubation at 4C was carried out for 24, 48, 72 and 96 hr followed by incubation at 37C for 2, 8, 12 and 24 hr. Neither increase in viable count nor evidence of coagulase activity were observed for periods up to 96 hr at 4C. Additional incubation of these cultures at 37C showed no further increase in coagulase activity over those which were not initially incubated in the cold. Effect 2: shaking Duthie and Haughton (1958) showed that shake cultures gave higher and more rapid coagulase titers than did non- shake cultures. To substantiate these studies coagulase activity of shake versus non-shake cultures were compared 39 2048— I28 — a .1 l, 2.- '{f 7. 32— fin ":5 {.5 :7 .5... {ii-1 Coagulase Reciprocal Titer per ml Filtrate l l I in i 4’ 8 I2 I6 20 24 28 32 36 Time(hours) FIG. 1. Effect of temperature on coagulase production by Staphylococcus aureus 70 in brain heart infusion. 40 over a 5 day incubation period. When S, aureus 70 was incubated on a rotary shaker at 37C, maximal titers were obtained at 12 hr. However, within 48 hr continued shaking resulted in a sharp decrease in coagulase and little or no activity remaining at 72 hr. In contrast, highest coagulase activity under non-shake conditions occurred in 4 days fol- lowed by a slight decrease the next day (TABLE 1). Effect of addition of trace minerals §2_the medium Tager and Hales (1948) suggested that the addition of trace elements to brain heart infusion increased coagulase production. The composition of this mineral "cocktail," as Tager called it, is indicated on TABLE 1. When coagulase titers were compared between brain heart infusion alone and brain heart infusion plus 1 ml of the mineral solution per liter of medium, no enhancement of activity occurred (TABLE 1). Concentration and Purification of Coagulase Partial purification Sy_gglg and ethanol precipitation The early attempts to purify coagulase included methods originally developed by Tager (1948) and more recently 41 TABLE 1. Effect of shaking and the addition of trace minerals to brain heart infusion upon the pro- duction of coagulase by Staphylococcus aureug 70 Coagulase Reciprocal Titers Time (Hr) Shake an-shake BHI BHI+Minerals* BHI BHI+Minerals* 12 1024 1024 128 128 18 512 256 128 128 24 512 256 512 512 48 16 16 1024 1024 72 0 2048 2048 96 0 4096 4096 120 0 2048 2048 *Composition of trace mineral solution (g/liter). m1 of this solution was added to a liter of brain heart infusion. MnSO 4 H20 Boric acid CuSO Molybdic acid 4 FeCl3 ZnSO KI 4 SHZO 6H20 0.03 0.06 0.04 0.02 0.25 0.20 0.10 One 42 TABLE 2. Purification factors of coagulases prepared by acid and ethanol precipitation Phage Phage Purification Phage Phage Purification Group Type Factor Group Type Factor I 52A/79 26.6 III 83(VA4) 21.6 80 66.2 73 65.0 II 3A 33.3 6 42.4 3B 58.3 77 42.4 3C 58.3 71 50.3 55 35.0 47 26.6 III 7 42.0 54 50.2 70 56.0 75 56.0 187 42.6 IV 42D 79.0 53 28.7 Misc. 81 36.6 43 modified by Blobel et a1. (1960). Twenty strains of S, aureus were used representing phage groups I, II, III, IV and Miscellaneous. The inadequate degree of purification of each coagulase is indicated in TABLE 2. The purifica- tion factor was determined by dividing the coagulase re- ciprocal titer per mg protein of the purified material by the coagulase reciprocal titer per mg protein of the original culture filtrate. Although the purification fac- tors tend to vary among the different strains of staphy- lococci, an average of these results (46.8) agrees closely with a factor of 48.4 obtained by Blobel et a1. (1960). Partial purification 9y separation from clot complex Since Elek (1959) suggested that the conversion of fibrinogen to fibrin by coagulase may involve the formation of a complex with substances in the clot, we devised a series of experiments to determine whether this system would yield purified coagulase. we added enough partially purified coagulase to plasma to obtain a 2+ clot (approxi- mately 25% of the total volume) after incubation at 37C for 10 min. The clot was removed by low speed centrifugation and washed several times with 0.85% NaCl solution. After 44 carefully disrupting the clot and removing the fibrin debris by centrifugation, the syneretic fluid was tested for pro- tein, phosphatase and coagulase activity. Results (TABLE 3) indicate an 8-fold increase in coagulase purification. This purification factor was similar to that obtained by Blobel et a1. (1960) who used starch block electrophoresis for further purification of the acid and ethanol precipitated coagulase preparation. Furthermore, to determine the extent to which available coagulase would become trapped within the clot, trials were conducted in which 5 mg of partially purified material were added to 4.9 ml of physiological saline solution. To this mixture was then added 0.1 ml normal rabbit serum (source of coagulase reacting factor) and purified human fibrinogen. By adding the aforementioned three substances at various stages and removal of the clot when formed, it appeared that the clot consisted of fibrin, coagulase and coagulase re- acting factor (TABLE 4). Partial purification Sy gel filtration (Sephadex alone) Separation of coagulase from undesirable protein was attempted using Sephadex G—200 chromatography. After 1300 45 TABLE 3. Separation of coagulase from phosphatase by concentration in the syneretic fluid from the clot complex Reciprocal of Coagulase Residual Fraction Coagulase Titer Purification Phosphatase per Mg Protein Factor Activitya Ib 5,120 - - c II 194 0.04 0.30 d III 40,960 8.00 0.02 auM p-nitrophenol liberated per ml. bPartially purified coagulase from Staphylococcus aureus 70. CFree fluid in which clot was immersed. Recovered syneretic fluid. 46 TABLE 4. Experimental proof for the retention of coagulase in the clotting complex Coagulase Reacting Factor (CRF) (0.1 ml Normal Rabbit Serum) + Acid and Ethanol Precipitated Coagulase + Purified Human Fibrinogen l FORMATION OF CLOT l FI' ' l , Remove Immersing Fluid Clot + Fibrinogen NO CLOT FORMATION + “ii FORMATION OF CLOT .l . ImmerSing Fluid Remove + Clot CRF + NO CLOT FORMATION + Fibrinogen + NO CLOT FORMATION + Coagulase l FORMATION OF CLOT 47 ml of cooled cell-free supernatant fluid were adjusted to pH 3.8 with 4 N HCl, the resulting precipitate was removed and dissolved in 50 ml of 0.1 N sodium acetate. Concen- tration to 5 ml was accomplished by evaporation at 4 C through dialysis tubing. Application of 2 ml (20 mg/ml) of this material to a column containing Sephadex G-200 and subsequent collection of 5 ml aliquots resulted in the removal of about 75% of the inert protein present in the original sample (FIG. 2). Extreme purification using ethanol-water mixtures under controlled conditions, and Sephadex Preliminary experiments. Modifications of methods of Cohn et a1. (1946) and Pillemer et a1. (1948) were used to purify coagulase. These techniques involving rigid control of pH, ionic strength, temperature, protein and alcohol concentration have been successfully used by the above workers for the separation of various components of blood plasma as well as purification of several toxins. It was deemed feasable to use a similar approach for the isolation and purification of coagulase. The following preliminary experiments were devised to arrive at conditions for the optimal recovery of coagulase: 48 2400 o—-— — Coagulase 560~ 0— Protein l l l l l l l 480 — l| l l 400— | l l l l 240— ug Protein per ml of Fraction l60~ 1 0 IO 20 3O 4O 50 Fraction Collected (tube number) FIG. 2. Separation by gel filtration (Sephadex G-200) of a coagulase preparation obtained by precipitation of a cell-free filtrate at pH 3.8. Coagulase Reciprocal Titers per ml 49 1. Fraction Cg-I-P. As a starting point, it seemed desirable to obtain maximal coagulase by precipitating it from the cell-free broth filtrate. Since Tager (1948) found that most of the active material precipitated at pH 3.8, we repeated this value. In addition, we used an acetate buffer at a constant ionic strength of 0.1 and various con- centrations of ethanol between 0 and 60% (v/v). A tempera- ture of -4C was used throughout these trials except where noted. TABLE 5 indicates that optimal activity was found in the precipitate that appeared at an ethanol concentration of 10% (v/v). Fixing the pH at 3.8, the ethanol concentration at 10% (v/v) and the ionic strength of the acetate buffer at various values from 0.01 to 0.1, optimal coagulase activity was found in the precipitate obtained at 0.1 ionic strength (TABLE 6). Furthermore, when ionic strength was held constant at 0.1 and two different ethanol concentrations were used for a series of pH values, some paired, ideal precipitating conditions prevailed at pH 3.8 (TABLE 7). 2. Fraction Cg—II—P. To obtain this fraction, satis- factory results were obtained by first redissolving fraction 50 Cg-I-P in 0.1 N sodium acetate followed by reprecipitation at favorable conditions. Ionic strength and pH of acetate buffer and ethanol concentrations were individually varied to determine optimal conditions. Maximal activity was ob- tained at pH 5.2, ionic strength 0.05, ethanol concentration 10% (v/v) and protein concentration of 3.1 mg/ml (TABLE 8). 3. Fraction Cg—III-S. In the preparation of the third fraction, it was considered appropriate to remove inert protein by precipitation and keeping the active coagulase in solution. Anticipating the need for higher pH values, a change was made to a phosphate buffer system. TABLE 9 represents the set of conditions for the solubility of coagulase fraction Cg-III—S in ethanol—water mixtures vary- ing in pH but at a fixed temperature, protein concentration and ionic strength of the phosphate buffer. Summary g£_fina1 procedure. As a result of the fore- going series of detailed experiments, we arrived at exact conditions for the separation of coagulase to a high degree of purity. TABLE 10 represents a summary of the purifica- tion factors and per cent recovery of successive fractions showing an apparent similar degree of purification (3,793X) for both fractions Cg-III-S and Cg-IV-C. Although these 51 results might indicate that the Sephadex step was not es— sential, one must remember that the immunological tests are far more critical and sensitive than the protein or coagu- lase determinations. Subsequent studies revealed that the third fraction contained three immunological components which were reduced to one precipitin band by the additional use of Sephadex to remove residual but interfering protein. The following protocol, therefore, represents a summary of the final procedure for the isolation of coagulase in ethanol-water mixtures under carefully controlled conditions followed by molecular sieving through a column of Sephadex G-200. 1. Culture conditiogg. Brain heart infusion cultures (10 ml) in logarithmic phase of growth were used as inocula for 100 m1 of the same infusion contained in a group of 1 liter Erlenmeyer flasks. Following incubation on a rotary shaker (ca 150 cycles/min) at 37C for 7 hr, a final inocula- tion with the contents of one Erlenmeyer flask was made into a 6 liter Florence flask containing 1200 m1 broth. Additional incubation was then allowed for either 12 hr on a shaker or 96 hr without shaking. The organisms were then 52 TABLE 5. Precipitation of coagulase fraction Cg-I-P using various concentrations of ethanol at . . . a . fixed conditions of temperature, protein . b . . concentration, pH and ionic strength of acetate buffer Ionic % Ethanol (95%) Coagulase PH c Strength (v/v) Activity 3.8 0.1 0 1,000 10 5,300 20 2,000 40 900 60 400 aTemperature of bath was -4C. Protein concentration of 12 mg/ml. c . . . . Coagulase rec1proca1 titer per mg protein in precipitate. 53 TABLE 6. Precipitation of coagulase fraction Cg-I-P using an acetate buffer of various ionic strengths . . . a . at fixed conditions of temperature, protein . b concentration, pH and ethanol Ionic % Ethanol (95%) Coagulase PH c Strength (v/v) Activity 3.8 0.01 10 1,200 0.02 1,200 0.04 1,100 0.06 1,000 0.08 3,600 0.10 5,500 a Temperature of bath was -4C. Protein concentration of 12 mg/ml. CCoagulase reciprocal titer per mg protein in precipitate. 54 TABLE 7. Precipitation of coagulase fraction Cg-I-P at two strengths of ethanol using different pH values of acetate buffer at fixed conditions of a . . b . . temperature, protein concentration and ionic strength Ionic % Ethanol (95%) Coagulase pH Strength (v/v) Activityc 3.4 0.1 10 3,600 3.6 10 3,300 3.6 20 3,200 3.8 10 5,300 3.8 20 2,000 10 4,400 4.0 20 1,000 10 3,300 4.6 10 3,300 5.0 10 2,400 5.4 10 No ppt 5.8 10 No ppt 6.2 10 No ppt 10 No ppt 8.0 10 No ppt a Temperature of bath was -4C. Protein concentration of 12 mg/ml. cCoagulase reciprocal titer per mg protein in precipitate. 55 TABLE 8. A set of optimal conditions for the precipita- tion of coagulase fraction Cg-II-P in ethanol- water mixtures at constant temperaturea and protein concentrationb but at various pH values and ionic strengths of acetate Ionic % Ethanol (95%) Coagulase pH Strength (v/v) Activityc 5.2 0.01 10 10,700 0.01 20 4,200 0.05 10 99,300 0.05 20 17,400 5.4 0.01 10 46,800 0.01 20 26,400 0.03 10 11,700 0.03 20 15,200 0.05 10 49,600 0.05 20 37,000 0.06 10 26,000 0.06 20 45,000 5.6 0.01 10 24,000 0.01 20 41,900 0.05 10 25,600 0.05 20 21,500 aTemperature of bath was -4C. bProtein concentration of 3.1 mg/ml. c . . . . Coagulase reCiprocal titer per mg protein in precipitate. 56 TABLE 9. The set of conditions for the solubility of coagulase fraction Cg-III-S in ethanol-water mixtures varying in pH but at a fixed tempera- ture,a protein concentrationb and ionic strength of phosphate buffer Ionic % Ethanol (95%) Coagulase PH .4. c Strength (v/v) ActiVity 5.5 0.1 10 1,400 20 900 6.1 10 218,000 20 600 6.5 10 20,000 20 4,000 7.1 10 20,000 20 10,000 7.5 10 19,000 20 8,000 7.9 10 19,000 20 11,000 aTemperature of bath was -4C. Protein concentration of 0.7 mg/ml. cCoagulase reciprocal titer per mg protein in the supernatant fluid. 57 removed by use of a Servall continuous—flow superspeed centrifuge at a flow rate of 50 ml/min. 2. Separation g§_fraction Cg—I-P. Two liters of ace- tate buffer solution (pH 3.8, ionic strength 0.1, 10% ethanol v/v) were placed into each of six heavy walled cylindrical glass jars (15 x 20 cm) and transferred to a cold bath (-4C). Since a suitable water bath was not available for studies with temperatures below the freezing range, and since similar sized commercial laboratory coolers were economically prohibitive, a unique innovation using a "wet-type" Coca-Cola cooler merits mention here. By manipulation of the thermostat and addition of alcohol to the water bath, a constant temperature below zero centigrade was obtained. After the temperature of the buffer solution was equilibrated with that of the surrounding medium, 200 ml aliquots of cell-free culture filtrate, transferred to each of 6 bags prepared from cellulose dialyzing tubing, were immersed into each of the six jars containing buffer solution. With constant stirring, dialysis was allowed to proceed for 12 hr. At this time, a freshly prepared 2 liter batch of buffer solution was added to each jar replacing that already present and dialysis was allowed to occur for 58 an additional 12 hr. The precipitate accumulating in the dialyzing bag was removed by use of a refrigerated centri- fuge (International PR-l, 2000 rpm/60 min at -4C) and re- hydrated to 50 ml with 0.1 N sodium acetate solution. After thorough mixing, insoluble residues were removed by centri- fugation at 2000 rpm for 30 min and discarded. 3. Separation 9; fraction CgeII—P. For this step, 2-1iter quantities of acetate buffer solution (pH 5.2, ionic strength 0.05, 10% ethanol v/v) were placed into jars and treated in the manner described in the preceding step. Each of the 50 m1 aliquots obtained as fraction Cg-I-P was placed into six dialyzing bags and immersed individually into six jars containing buffer solution. Following dialysis for 24 hr, the precipitate was removed by centrifugation in a Servall superspeed centrifuge (15,000 x g) for 10 min and dissolved in 5 ml 0.1 N sodium acetate solution. 4. Separation 9; fraction Cg-III-S. In contrast to recovery of the active coagulase in the precipitate as done in fractions Cg-I-P and Cg-II-P, conditions were arranged so that coagulase for this fraction was retained in solution and inactive material was precipitated. Since six 5 ml samples were obtained from the preceding step, only one jar 59 containing 2 liters of phosphate buffer (pH 6.1, ionic strength 0.1, 10% ethanol v/v) was necessary. Again, after equilibration of the buffer to -4C, each of the samples was placed into individual dialysis bags and immersed into the buffer solution and stirred. Dialyzing time, temperature and centrifugation speed were similar to those of the pre- ceding step. HOwever, the precipitate was now discarded. The supernatant fluid was recovered since it contained maximal coagulase activity per unit protein. 5. Separation 9: fraction QgeIV-C. Five grams of dry Sephadex powder (G-200) were added to 250 ml of 0.15 M sodium chloride solution and allowed to swell. Henceforth, all operations were carried out at 4C. After transfer of this mixture to a glass chromatographic tube (2.5 x 45 cm), enough material was allowed to settle and pack overnight to achieve a height of 35 cm. To prevent the loss of the salt solution and subsequent dehydration of the gel, a rubber tube was clamped to the outlet of the column. A disc of filter paper whose diameter measured slightly less than that of the bore of the column tube was carefully placed on the surface of the bed. Above the disc 20 ml of 0.1 N sodium acetate (pH 7.2) were carefully pipetted and an 60 equal volume was allowed to drain from the column. Next, 2 m1 of fraction Cg-III-S (concentrated to approximately 1 mg per ml of protein) were dissolved in the same buffer solution and added to the column in the same manner. To avoid disturbance of the surface of the gel column and subsequent skewing of the components to be separated, the filter disc was absolutely essential. After the sample was started through the column, approximately 20 ml of eluant was added again. This was followed by the application of a continuous, regulated flow (20 ml/hr) of eluant from a reservoir which fed into the top of the column by capillary tubing through a rubber stopper. Mechanical collection of 5 m1 samples was accomplished by one of two available in- struments (Preparative Fraction CollectorF'Model No. 85000; Rotary Fraction Collector,2 Model No. 1205A). To reduce the possibility of bacterial contamination of the polysac- charide gel, merthiolate solution (1:20,000) was allowed to flow through the column between sample runs. FIG. 3 shows the concomitant protein and coagulase peaks by molecular sieving of Cg-III—S through a column of Sephadex G-200. lCalifornia Corp. for Biochem. Res., Los Angeles, Calif. 2 . . . . . Research SpeCialties Co., Richmond, California. 61 -5000 1 4000 12000 -|OOO “I o— Coagulase 3O - l \ O——— Protein ' a II I _ l I E II ¢ 3 to I 5 20— ’ l 2 ’ l e ’ ‘ CL " I‘ U i P I I IO- I I I I I I I I . I l . IO 20 3O 4O Fraction Collected (tube number) FIG. 3. Preparation of extremely purified coagulase, Cg-IV-C, by molecular sieving of Cg-III-S through a column of Sephadex G-200. (pH 7.2). The eluant was 0.1 N sodium acetate Reciprocal Coagulase Titers per ml TABLE 10. Summary of the purification factors and per cent recovery of successive fractions of coagulase Coagulase Purification % Fraction a Activity Factor Recovery Filtrate 58 - - Cg-I-P 2,600 45 61 Cg-II-P 92,000 1,586 24 Cg-III-S 220,000 3,793 12 Cg-IV-C 220,000 3,793 12 aCoagulase reciprocal titer per mg protein (average value of at least 5 separate trials). 63 Characterization of the Various Fractions Results g§_§pecia1ized procedures Electrophoretic analysis. Evaluation of protein homogeneity was obtained by use of paper electrophoresis using cellulose acetate strips. Duplicate strips were prepared to detect the coagulase component. One strip was stained while the other was cut into small squares (2 x 2 mm) and immersed into 0.5 m1 of 0.1 N sodium ace- tate solution. After elution of the component, the pieces of oxoid strip were removed with tweezers. To this solu- tion was then added 0.5 ml of 0.85% NaCl solution followed by measurement of coagulase activity with a two-fold serial dilution system as previously described. For com- parative purposes, electrophoretic patterns of coagulase fractions Cg-I-P and Cg-IV-C are indicated in FIG. 4. The extremely purified preparation (Cg-IV—C) showed only one peak while the relatively crude fraction (Cg-I-P) showed multiple components. Serological studies. Suggestions by Barber and Wildy (1958) indicated a possible antigenic relationship between 64 H (+) (‘l (+) FIG. 4. Patterns of the first fraction (A) of coagulase, Cg-I-P, as contrasted with the final fraction (B) of coagulase, Cg—IV-C, indicating elec— trophoretic homogeneity on cellulose acetate strips using barbital buffer at pH 8.6 and ionic strength 0.07. 65 coagulases within major phage groups of S. aureus and prompted an extensive series of studies to confirm or deny these results. For these early studies, we isolated coagulase from each strain of the International-Blair series of staphylococci by acid and alcohol precipitation according to modifications of techniques of Tager (1948). Rabbits were used for serological studies, the material being injected subcutaneously along with Freund adjuvant. The sera from two rabbits used for each sample were pooled and employed in the coagulase inhibition and agar diffu- sion tests to determine antigenic specificity. These experiments were arranged to detect any possible cross- inhibition as well as cross—precipitation reactions. Re- sults (TABLE 11) indicate that highest coagulase inhibition occurred in the homologous antigen—antiserum system. Al- though low cross-inhibition titers occurred in hetero— 1ogous combinations, no definite relationships could be noted corresponding to phage groups. Therefore, these results may indicate that coagulases are strain specific. Using agar diffusion techniques, further confirmation of this phenomenon proved exceptionally difficult to interpret because of the appearance of a high number of precipitin zones . (56 owuHOQOHQno nu“: uoHo +2 I nausea yuan gown: vane am venous poem-Louovonn on» .qldeHn .oonaauooo can. no acacia coca-Louovoun acouucoo u vouuaanusoc yuan sums: coHusawv Moons“: any «0 aH¢009aHoon «cowouaou consummc OO O OH OH O O O OH OH OH O OH O O OH OH OH O O HO OOH: OH mu O OH O O O O OH OH O O O OH O OH O OH OH OH: >H O OH In O OH O OH OH OH OH O O O O O O O O O H O O O mm OH O OH OH O O O O O O O O O O O OH O O O O mu OH OH O OH OH O OH O O O O OH O O OO OH O O OH OH mu OH OH OH O OH OH O O O O O O OH HO O O O O O OH mm OH OH OH O O O O O O O OH O HH OH O OH O O O OH mm OH O OH O O O O O O O O HH OH OH OH OH OH OH OH OH mum OH O OH OH OH OH OH OH OH OH O OH O O OH OH O OH OH OH mm OH O OH OH O O O OH O HH O OH OH OH OH OH OH OH OH O ma O O O O OH O OH O :(> O O O OH OH O OH 3O O OH OH mm O O O O O OH O HO O O O O O O O O OH O O O an O O O O O O HOH HHH OH O O O O O OH O O OH O O O an O O OH O O OO O O O O OH O OH O O O O O O OH an OH O OH O OH OH OH OH O O OH O O OH OH OH O O OH O: mum O O OH OH O O OH O O OH O O O OH O O OH O O O mm OH O HO HOH OO OH OH muooauou awonw up unsouu uHonqucoHuvcnoucn one uo madcap» HooooooHasaeua noun moouasu-Oo any we AcoHuogonoun uo >uwusa can: unavconovv «acacia acouucoo no camunumHvuusocuquLo HH mHm