University of Southern California Dissertations and Theses

Studies of trypsin-binding ɑ₂ macroglobulin of human plasma

(USC Thesis Other)

Studies of trypsin-binding ɑ₂ macroglobulin of human plasma

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content STUDIES OP TRYPSIN-BINDING *2 MACROGLOBULIN OP HUMAN PLASMA by Sally Mayfield Howard A Dissertation Presented to the PACULTY OP THE GRADUATE SCHOOL UNIVERSITY OP SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OP PHILOSOPHY (Biochemistry) June 1966 UNIVERSITY O F S O U T H E R N CA LIFO RN IA T H E G R A D U A TE SC H O O L U N IV E R S IT Y PA R K L O S A N G E L E S . C A L IF O R N IA 0 0 0 0 7 This dissertation, written by Sally Mayfield Howard under the direction of h&T„-..Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R O F P H I L O S O P H Y Dean Date. June 1966 DISSERTATION COMMITTEE f I Chairman ACKNOWLEDGEMENTS The report of this work would be Incomplete without a sincere expression of appreciation for the help given by Dr. John Mehl. my adviser. I also wish to thank Dr. Donald Visser and Dr. Ronald Brown for their cooperation and assistance as mem bers of my Dissertation Committee, and Dr. Milton Heinrich for his encouragement and support as my adviser during the early part of my graduate studies. Grateful acknowledgement 1b made to the National Institutes of Health for a three-year Pre- doctoral Fellowship, and to the Department of Bio chemistry, University of Southern California, for support from U. S. Public Health Service Training Grant funds for one year. Special thanks goes to Mr. Russell Saunders and Mr. John DeGroot for their excellent assistance. --S. M. H. ii TABLE OP CONTENTS ACKNOWLEDGEMENTS Page ii LIST OP TABLES v LIST OP ILLUSTRATIONS vi LIST OP ABBREVIATIONS viii Chapter I. INTRODUCTION AND HISTORICAL BACKGROUND 1 II. EXPERIMENTAL PLAN 11 III. MATERIALS AND METHODS lb BAPNA Assay Standardization of Trypsin Activity Activity Units Purification Procedure Analytical Ultracentrifuge Starch Gel Electrophoresis Polyacrylamide Gel Disc Electrophoresis Esterase Assay pH-stat Assays Casein Assay Purification Procedure Physical Properties of Purified CtgM Zone Electrophoresis Sedimentation and Diffusion Ultraviolet Absorptivity Specific Trypsin-Binding Activity Spec t ro pho t ome t ry Dissociation IV. RESULTS iii Chapter Page Kinetics of Free Trypsin Inhibition of Trypsin by c*pM Kinetics of Trypsin- c*2M Complex V. DISCUSSION............................... 77 VI. SUMMARY..................... 91 LIST OF REFERENCES............................... 96 lv LIST OP TABLES Table Page 1• Purification of Trypsin-binding dLp Macroglo bulin from Cohn III-O Fraction................35 2. TAME Esterase Activity during Purification • . . Mf 3. Hate Modification by Reaction Products at High Substrate Concentration........................65 b . Effects of Some Modifiers on the Rates of Hydrolysis of TAME and BABE by Trypsin • • • . 69 v LIST OP ILLUSTRATIONS Figure Page 1 • Summary of Purification Procedure for Trypsin- binding from Cohn III-O Fraction • • • • 21 2, Concentration Dependence of the Sedimentation Coefficient of Purified d 2M............ . . 2 * + 3. Disc Electrophoresis of Lipid-poor Protein from Cohn III-O Fraction and PEG Fractions . • • • 36 *+. DEAE-cellulose Column Chromatography of PEG Precipitate............................. IfO 5. Sedimentation of d^M at Intermediate Stages of Purification................ l+l 6. Vertical Starch Gel Electrophoresis of d ^ at Various Stages of Purification. ....... k-2 7. Sedimentation of Purified dgM at Neutral pH, and After Dialysis at pH l+.O in Acetate ... *+6 8. Ultraviolet Absorption Spectrum of Purified d.2M.................................... 50 9. Difference Spectrum Developed by Trypsin on Interaction with at^ . . ......... 52 10. Sedimentation of Purified doM at pH 3.0 and 9.5....................f ................ 5»f 11. Eadle Plot of Trypsin-catalyzed Hydrolysis of TAME.................................... 58 12. Effects of TAME and TA on the Rate of Hydroly sis of TAME............................ 60 13. Inhibition of the Hydrolysis of TAME by STI . . 62 1*+. Trypsin-catalyzed Hydrolysis of BAEE at pH 6.0 and 8.0............................. 63 vi Figure Page 15. Inhibition of the Tryptic Hydrolysis of Low Concentrations of BAEE by TA • ......... 67 16. pH-Rate Profiles of Trypsin-catalyzed Hydroly sis of TAME and BAEE ................ . 70 17* d oM Inhibition of Tryptic Hydrolysis of TAME and BAEE............................... 72 18. Catalysis of TAME Hydrolysis by Free Trypsin and Trypsin- Complex............ 7 * t vii LIST OP ABBREVIATIONS cl 2* * 0I 2 ®acroglobulin BAPNA benzoyl-DL-arginlne £-nitroanilide STI soybean trypsin inhibitor TAME p-toluene sulfonyl-L-arginine methyl ester TA p-toluene eulfonyl-L-arginine BAEE benzoyl-L-arginine ethyl ester BA benzoyl-L-arginine PEG polyethylene glycol Till CHAPTER I INTRODUCTION AND HISTORICAL BACKGROUND It has been known for some time that a small fraction of- the protein of normal human serum consists of globulins of high molecular weight, approaching one million. In the earliest systematic investigation of serum in the analytical ultracentrifuge, Mutzenbecher (1) found that there is a component with sedimentation coefficient of 17-20 S, and later studies by many' different workers con firmed this observation. More recently, the high-molecular* weight proteins of plasma have been analyzed by a combina tion of centrifugation and conventional electrophoretic techniques used for classification of serum proteins. In barbital buffer of pH 8.6 and ionic strength 0.1', either moving boundaxy or filter paper electrophoresis separates serum into at least five major components (2). In 19*+7, Oncley et al. (3) isolated rapidly- sedimentlng globulins from the Cohn III-O Fraction of human plasma. Sedimentation coefficients were 16-20 S, and one 20 S fraction was obtained that appeared to be about 80% homogeneous. These proteins moved in the zone 1 electrophoretically, but in contrast with the £ lipopro tein in the same III-O material, contained little or no lipid. They were later referred to as " lipid-poor euglobulins" (*f). In 195*+ Brown and associates (5) reported isola tion, from the Cohn III-O Fraction, of a glycoprotein hav ing an electrophoretic mobility in the ot2 region and a sedimentation coefficient of l*+.6 S for a 2% protein solu tion. As the sedimentation coefficient was not corrected for concentration, it later was thought that this might be the same protein as the 19 S ^ 2 glycoprotein characterized by Schultze and associates (6) in the course of classifica tion of a number of proteins from human plasma. Schultze's group later purified and characterized this protein more extensively, calling it olg macro globulin, and made it available commercially (7). In 1955 Brattsten fractionated whole serum by con tinuous zone electrophoresis and examined the fractions in the analytical ultracentrifuge (8). He found sedimentation coefficients above 17 S in the 7 and fractions only. Shortly thereafter, Wallenlus et al. (9) concentrated the heavy globulins of serum by repeated centrifugation, then examined this material electrophoretically, finding peaks in the 7 and 6L 2 regions. Hence, the rapidly-sedimenting (17-20 S) globulins of human plasma are now considered to Tall into two distinct classes, ct ^ and P' macro globulins. As the distinction between adjacent electrophoretic zones is somewhat arbitrary, the A lipid-poor euglobulin frac tion of Oncley et al. (3) can undoubtedly be considered to belong to the ^ 2 class. The Y maoroglobullns have been much more exten sively studied than cl2 macro globulin, due to interest in their role as antibodies and the elevation of this fraction in Waldenstrom's macroglobulinemla. Immunologic ally, as well as electro phoretically, the normal Y macroglobulins are closely related to pathological Waldenstrom macroglo bulins and to normal 7 S Y globulins, but appear to be unrelated to ^ 2 macro globulin (10). 2 macroglobulin does not appear to be an antibody (11), and until quite recently no particular physiological activity had been associated with this protein. Marked elevation of dg macroglobulin is often found in the sera of nephrosis patients (10). In 1962 Haverback et al. (12) reported that a human serum protein migrating in the & 2 region is able to combine with trypsin and chymotxypsln in such a way that enzymatic activity is retained but is no longer subject to inhibition by protein inhibitors such as soybean trypsin inhibitor. Subsequently, Mehl et al. (13) isolated <*2 macroglobulin both from plasma and from fraction III-O and showed that the trypsin-binding activity described by Haverback is found in this fraction. The large size difference in the macro globulin and trypsin, and the fact that each molecule seems to bind a limited number of tryp sin molecules, accounts for the relatively slight altera tion of electrophoretic mobility of “acroglobulin when trypsin is bound, even though free trypsin would migrate in the opposite direction at pH 8,6 due to its extremely high isoelectric point. The complex is less active than free trypsin, so 0(2 macroglobulin may be considered an inhibi tor, but further inhibition by soybean trypsin inhibitor is prevented. These workers also found that the binding activ ity of olg macroglobulin is destroyed by ammonium ion; hence, any fractionation method employing ammonium sulfate, such as that used by Schultze*s group to obtain highly purified ot-g macroglobulin (7), would not be expected to yield active protein. Oncley and associates (1*0, who developed the method for subfractionation of the Cohn II + III Fraction, noted that the III-O Fraction contains materials with plas- mln Inhibitor and antithrombin activities. This III-O Fraction consists of a large amount of 0 lipoprotein together with a number of other proteins, and these workers did not attempt to isolate the proteolytic inhibitors. Jacobsson reported in 1953 that the trypsin inhibitors of 5 human plasma migrate as ol^ and 0*2 globulins (15). He further found that the dg fraction also inhibits plasmin, and that this activity Increases In nephrosis serum (1 6 ), Steinee and Mehl (17) investigated the trypsin-binding activity of nephrotic sera in which macroglobulin was elevated. They found a linear correlation between amount of macro globulin, as measured from the ultracentrifuge pattern, and trypsin-binding activity, determined by addi tion of excess soybean trypsin inhibitor to a mixture of the serum and trypsin and measurement of the tryptic activ ity remaining* Schultze and associates have partially purified at2 plasmin inhibitor and identified it, by an immunological method, with their highly purified <*2 macroglobulin (1 8). Thus it appears that the plasma protein designated as otg macroglobulin on the basis of its electrophoretic and ultracentrifugal characteristics is able to bind vari ous proteolytic enzymes in a specific manner and modify their activity* Although investigations of trypsin-binding protein, 6L2 plasmin inhibitor, and at least part of the 0I2 trypsin inhibitor of human plasma are probably con cerned with the same protein, the preparations used for these studies have been less homogeneous than the highly purified d2 macroglobulin obtained some years ago in Schultze's laboratory without benefit of any type of activity assays and, in fact, by a method vhich would have probably destroyed any enzyme-binding activity. It is evident that purification of o*2 macroglobu- lin by other methods, in which binding activity can be assayed at each step of the procedure, is essential to any further study of the interaction of this protein with the proteolytic enzymes. Its rather unusual behavior as both an inhibitor and anti-inhibitor of trypsin is of consider able interest, and the measurement of the extent to which tiypsin is protected against soybean trypsin inhibitor affords a convenient assay method for following the purifi- j- cation of the trypsin-binding activity. The detection of the reaction between trypsin and oi 2 macroglobulin has been dependent upon some measurement of the catalytic activity of trypsin. The mechanism of catalysis and the interpretation of kinetic parameters for hydrolysis of various substrates has, consequently, also been a concern of the present investigation. Trypsin catalyzes the hydrolysis of peptide bonds in proteins involving the carboxyl group of arginine or lysine (19)* The discovery that amides (20) and esters (21) of arginine and lysine are also substrates for trypsin gave great impetus to Investigations of the kinetics and mechanism of trypsin catalysis. It is one of a number of hydrolytic enzymes containing an "active serine" residue which is very reactive toward diisopropyl fluorophosphate and, when blocked by phosphorylation with this reagent, is enzymatically inactive (22). Evidence that hydrolysis by trypsin and chymotrypsin proceeds by (1) acylation of this active serine residue by the substrate, and (2) deacylatlon of the acyl-enzyme intermediate catalyzed by an uncharged histidine residue in the protein to give produot and free enzyme, has been recently reviewed by Bender and K4zdy (23). At the present time this reaction sequence is generally, but not unanimously, accepted as describing the hydrolysis of ester and amide substrates by trypsin and chymotrypsin, Gutfreund and Sturtevant (2*0 have shown that the usual Michaelis-Menten scheme can be extended to apply to such a case. If the reaction is written: kg ko E + S (ES) ----- *(ESf ) ---- * E + P* a rate equation can be derived by steady-state treatment, and the data plotted according to any of the usual methods to obtain constants analogous to, but more complex than, the Michael is-Menten Km and Vm. Because is dependent on enzyme concentration, the constant kca- £ is often used; kca^ . is equal to Vm divided by E, the molar concentration of enzyme. This convention will be followed. Modified Michaelis-Menten schemes have been used successfully for trypsin- and chymotrypsin-eatalyzed hydrolysis of a number of different substrates, and kinetic parameters determined (23). In general, the affinities of trypsin and chymotrypsin for ester substrates are higher than for the corresponding amides, and the rates of hydrolysis of esters are faster (21, 25, 26, 27)• It has been suggested that the acylation step is rate-limiting for amide, but not ester, hydrolysis (28, 29)* Acylation seems to be at least as fast as deacylation, and usually faster, with ester substrates. In 1963 Trowbridge et al. (30) reported deviation from simple Michaelis-Menten kinetics in the case of tryptic hydrolysis of £-toluenesulfonyl-L-arginine methyl ester (TAME), and showed that substrate activation could account for this behavior. This observation has been con firmed by other workers (31> 32), and probably accounts for discrepancies in values of K^i and Vm reported from earlier kinetic analyses of this system (33> 3^, 35K Substrate inhibition has not been reported for any of the ester or amide substrates of trypsin, but it is known that in some cases substrate analogues can be inhibi tory. For example, inhibition by the reaction product benzoylarginine has been reported for trypsin-catalyzed hydrolysis of the amide, but not the esters, of this amino acid derivative (36, 37)• Another useful trypsin substrate, benzoyl arginine p-nitroanilide (BAPNA), is currently avail able only as the LL mixture; Erlanger et al. (29) have found that only the L-isomer is hydrolyzed, the D-isomer being a competitive inhibitor. Trowbridge et al. (30) showed that, although both optical Isomers of TAME can be hydrolyzed by trypsin, the L-isomer is a much better sub strate and presence of the L-isomer leads to apparent com petitive inhibition. Xnagaml (38) has found that alkyl- ammonium ions, which can be considered as analogues of the side chain of lysine, are competitive inhibitors of trypsin-catalyzed hydrolysis of benzoylarginine ethyl ester (BAEE). There are a number of proteins from diverse sources which are specifically inhibitory to trypsin (39)• The best known are those from soy beans, lima beans, and pan creas; ovomucoid from chicken eggs; and inhibitor from plasma. They appear to be different proteins with molecu lar weights generally in the range 10,000 to 50,000. All have quite high affinity for trypsin, and the inhibition is stoichiometric, although in some cases slight enzymatic activity is retained in the presence of excess inhibitor. In the neutral to slightly alkaline pH range where trypsin is most active, interaction with all except pancreatic inhibitor is quite rapid. With most of these inhibitors a 1*1 complex can be Isolated and crystallized and shown to 10 be equal in molecular weight to the sum of the molecular weights of trypsin and the inhibitor, and it is essentially devoid of either proteolytic or inhibitory activity. Until quite recently it was thought that these complexes are addition compounds and that interaction does not involve any changes in covalent structure of either component, Finkenstadt and Laskowski (M-0) have reported that combina tion of trypsin with soybean trypsin inhibitor (STI), measured at acid pH by release of hydrogen ions, is accom panied by an overshoot which is consistent with the forma tion of an intermediate. This intermediate seems to be an altered form of STI which will dissociate from trypsin at acid pH, No overshoot occurs on subsequent interaction with added trypsin at a higher pH, and the altered inhibi tor is fully inhibitory. These workers believe that STI, and possibly other inhibitors, sore therefore substrates for limited proteolysis and after cleavage of a peptide are covalently bonded to the active site of trypsin at neutral pH, Such a mechanism might have some relation to both the inhibitory and anti-inhibitory activities of macro- globulin (otgM). CHAPTER II EXPERIMENTAL PLAN Trypsin-binding < Xg macroglobulin ( ka<3 been partially purified in this laboratory (1 3 » *H) from both plasma and the Cohn III-O Fraction, and some of its proper ties had been investigated. For further study of its phys ical and enzymatic properties a more highly purified proteir^ prepared in relatively large quantity by a standardized procedure, was desirable. Therefore, one aim of this study was the refinement of methods previously used in isolating trypsin-binding *^2M» 6 1 2 1 ( 3 development of further steps which would give a good yield of product of higher specific activity and in greater quantity than was heretofore avail able. It was hoped that the final product would be a homogeneous protein which could be characterized in some detail by physical methods, but this was not the case. Some possible reasons for the difficulty in achieving com plete homogeneity are discussed in a later chapter. Some of the physical characteristics of the purest preparation available were measured, giving some indication of its size and the minimum number of trypsin-binding sites. 11 12 Such data, although somewhat provisional since the protein is not entirely pure, are useful for any continuation of the studies described here, as well as for comparison with results from other laboratories where several different approaches are being used in ^studying proteins of quite similar properties. The discovery that c*2M appears to dissociate into discrete subunits at acid pH led to some preliminary studies of changes in sedimentation behavior and activity under various conditions. A more complete investigation is necessary before conclusions can be drawn about the rela tionship between the subunits and binding activity, for example. Some investigation of the nature of the trypsin- otgM complex, in which trypsin is partially inhibited but is no longer subject to further inhibition by STI, was another aim of this study. This rather unusual behavior was evident in the standard BAPNA assay used previously, but in order to compare the kinetic properties of free and complexed trypsin, a pH-stat assay method was introduced which proved much more adaptable to studies involving variation of substrate concentration, pH, etc. Initial studies of hydrolysis of the ester substrates TAME and BABE by free trypsin revealed some interesting activation and inhibition effects by these substrates and their reaction 13 products which extended the information about these systems previously reported in the literature* These effects were investigated in some detail, both because of their Inherent interest and because understanding of the kinetic behavior of free trypsin was necessaiy before comparison could be made with the kinetics of the trypsin- ^2^ complex. The binding activity of otgM was considered from two points of view. First, its inhibitory properties when added to trypsin in varying amounts were investigated, and second, the kinetics of its complex with trypsin were studied as though the complex were a separate enzyme. Results of these studies gave some further insight into the nature of the association of this interesting protein with trypsin. CHAPTER III MATERIALS AND METHODS BAPNA Assay The assay method used to follow the purification of trypsin-binding activity was adapted from the trypsin assay of Erlanger et al. (29)* Worthington trypsin, 3 X crystal lized, was dissolved in 0,0025 N HCl-0,01 M CaCl2 at an approximate concentration of 1 mg/ml, and this stock solu tion stored at 0-5°• Assay buffer was 0.17 M Tris HC1- 0.01 M CaCl2, pH 7.7. Throughout this work Tris buffers were used, nearly always in the hydrochloride foim; here after they will be designated simply Tris. Trypsin stock was diluted with assay buffer to a concentration of 50/tg<fal, and 0.3 n i l (15 /ug trypsin) was added to the a*2M sample in the same buffer and preincubated 15 minutes at 37° • Soy bean trypsin inhibitor (STI) (Worthington, 3 X crystal lized) was dissolved in assay buffer at a concentration of 1 mg/ml for stock solution, and diluted to 100 ^sg/ml as needed. To each assay tube 0.2 ml (20 /tg STI) was added, and preincubation continued for an additional 15 minutes. Substrate was a supersaturated aqueous solution of 15 benzoyl-DL-arginine £-nitroanilide (BAPNA) (Mann Research Laboratories) at a concentration of 1 mg/ml, dissolved by heating at 70°, cooled, filtered, and stored in the dark at room temperature. To each preincubated sample tube was added 1,0 ml of substrate solution. After exactly 15 minutes at 37°* 1 ml of 5% phosphotungstic acid in 1 M acetate buffer, pH *f.5, was added, to stop the reaction and precipitate the protein. Tubes were allowed to stand for about 3 0 minutes, then centrifuged and filtered and the absorbancy of the clear filtrates read at 383 mM- In a Beckman Model B spectrophotometer. Each assay included a tiypsin standard containing no sample or STI; this gave the absorbancy at 383 &M- for BAPNA hydrolysis by 15 A8 of free trypsin. The blank contained both trypsin and STI, as well as substrate, and allowed correction for nonenzymatic hydrolysis of substrate. In the assay volume of 3 DL-BAPNA concentration was 0.77 mM. Erlanger et al. have shown that the D-BAPNA present is not hydrolyzed and is a competitive inhibitor, and that the Kjq for L-BAPNA is 0.939 mM (at pH 8.15 and 15°)* However, under the assay conditions described here the extent of hydrolysis was proportional to time during the entire incubation period, so could be used as a measure of trypsin activity. The stoichiometric weight ratio of trypsin to STI is 1,0*1.2, and free trypsin was totally inhibited by this ratio of STI, in the BAPNA assay. Therefore the only trypsin activity remaining was due to trypsin which was bound to the sample and thus protected from STI. To deter mine the activity of bound trypsin, relative to free trypsin, trypsin was preincubated with graduated amounts of ot2M, before addition of substrate. Activity decrease was proportional to amount of added at low concentrations, but leveled off to about 65% that of free trypsin at the equivalence point. If excess STI was added to trypsin which had been preincubated with graduated amounts of activity was zero for free trypsin, increased with increas ing aHd, as before, leveled off at about 65% the activity of uninhibited free trypsin. Standardization of Trypsin Activity Trypsin concentration can be determined by weight of crystalline enzyme used, or spectrophotometrically using the optical factor 0.651 (^2) to convert *280 mj u > However, all trypsin solutions contain some inactive pro tein and the concentration of active enzyme actually present can be measured by titration with STI, which is available as a much more uniform and stable crystalline protein. Graduated amounts of STI (the optical factor for STI is 1.1) (*t3) were added to a standard trypsin concentra tion in the BAPNA assay, and the weight of STI required to 17 just inhibit a given weight of trypsin was determined from a plot of STI added vs. A.^83 ma * Since STI combines only with active trypsin, and the molar ratio is 1*1, the weight of active trypsin thus determined (taking the molecular weight of STI as 20,000 and of trypsin, 2*f,000) could be compared, with the total weight of trypsin measured spectro- photometrically. Trypsin solutions used in this work con tained 60# to 70# active enzyme. Except as otherwise noted, however, trypsin concentrations are expressed on the basis of weight of protein, uncorrected for Inactive enzyme. Activity Units Binding activity is expressed as m-S of trypsin activity per ml of solution, and is a measure of residual trypsin activity not inhibited by STI. As noted above, it is lower than the activity of an equivalent amount of free txypsin due to approximately 35# inhibition of trypsin by under the standard conditions of this assay. Specific activity is fig of trypsin activity per ml of otgM solution, divided by the absorbancy of the cX^m sample at 280 m/ir • Throughout the purification, approxi mate protein concentration was measured spectrophotometri- cally, by assuming that the A2gQ is roughly equal to a protein concentration of 1 mgAil. Purification Procedure The starting material is the Cohn III-O fraction of human plasma, obtained from Hyland Laboratories, which comes as a frozen paste. To avoid denaturation by the ethanol present, it was placed in dialysis tubing without thawing, and dialyzed against partly frozen buffer (0,01 M Tris, pH 7,6), The ethanol-free III-O was dialyzed against 1.65 M NaCl in 0.01 M Tris buffer, pH 7.6, then centrifuged at 29,000 rpm for 22 hours in the Spinco #2>0 preparative rotor, to float the £ lipoprotein which con stitutes about 60# by weight of the proteins present. 0^ 2M was present in the clear gummy pellet and viscous blue liquid in the bottom of the tube. It was removed by slic ing the tube just below the interface, about one inch from the bottom. This procedure was adapted from Brown et al. (5). The crude lipid-poor protein sedimented from the III-O Fraction, still containing a high concentration of NaCl, was mixed with a solution of polyethylene glycol (PEG) of molecular weight 6,000-7,500 in 0.05 M acetate buffer, pH 5.5, to give a final protein concentration of 2-3# and final PEG concentration of 6#. After standing overnight in the coldroom, the precipitate was centrifuged down and dissolved in 0.02 M Tris-0.06 M NaCl, pH 7,7, at about 10# protein concentration. It was then dialyzed 19 against the same buffer to remove excess salt and PEG, A 50 ml sample of the protein solution was applied to a DEAE-cellulose column (a four-foot length of 3”inch- diameter lyrex pipe, packed with 600-700 grams (dry weight) of DEAE-cellulose, that had been pretreated with HC1 and NaOH) which had been equilibrated with 0,02 M Tris-0.06 M NaCl, pH 7.7, and the column was washed with the same buffer until all unadsorbed protein (Peak I) was eluted. The flow rate was M-50 or 900 ml per hour, and one-hour or one-half-hour fractions were collected. The operation of this column and fraction collector have been described (U-l, M+). A linear concentration gradient was then applied, from 0.06 M to 0.12 M NaCl, in the same Tris buffer; total gradient volume was six or eight liters. The smooth peak (Peak II) eluted by this gradient was subdivided into IIA and IIB, having higher and lower specific activities, respectively. Elution was continued with 0.02 M Tris-0.12 M NaCl until the eluate was nearly free from protein and binding activity. NaCl concentration was then increased to 0.5 M and Peak III was eluted, containing essentially all the remaining protein. The IIA material was concentrated by pressure filtration to about 1% protein, and equilibrated with 0.01 M acetate-0.05 M NaCl, pH 5.^, by dialysis. The dialyzate was diluted stepwise with distilled water, over a period of about one week, to a final concentration of 0.001 M acetate- 20 0,005 M NaCl, and the dialysis bag containing the protein allowed to remain in this low-ionic-strength buffer for several days, at 0-5°* ^he precipitate was removed by centrifugation in a rotor precooled to the same temperature. To the supernatant was added a weight of bentonite equal to approximately three times the weight of protein present; the mixture was allowed to stand at room temperature, with occasional stirring, for one-half hour or longer, and the bentonite then removed by centrifugation and the superna tant clarified by filtration, if necessary. The purification procedure is summarized in Pig, 1, Analytical Ultracentrifuge Sedimentation velocity runs were done in a Spinco Model E analytical ultracentrifuge, with schlieren optics, in *t° sector cells; the partition cell used had a fixed, perforated Kel-P partition. Except for the experiments to investigate pH and salt effects on sedimentation coeffi cient, the buffer used was sodium phosphate, pH 7,0-7,2, ionic strength 0,15, and relative viscosity 1,0. Rotor speed was 56,100 rpm, and pictures were usually taken at eight-minute intervals after reaching full speed; the 17 S peak is veiy near the bottom of the cell after **0 minutes. Protein solutions were equilibrated with the appro priate buffers by dialysis; for some experiments, pH was adjusted by addition of HC1 or NaOH. Relative viscosities COHN III-O FRACTION 21 top lipoproteiru disc&rd dialyze out EtOH density to 1.06 centrifuge slice tubes bottom lipld-poor protein S.A, = 18 add PEG centrifuge S.A. * 28 S.A. = 2 discard dissolve in buffer DEAE-cellulose column elute with 0.02 M Tris, pH 7.7, containing: - ■ i i i 0.06 M NaCl 0.06— 0.12 M NaCl , 0.5 M NaCl Peak I S.A. = 6 discard 1 Peak II i | Peak' III S.A. = 13 1 discard IIA S.A. = 5b iIb S.A. = bl i i i ppt ---- S.A. - bO discard rec hro ma tograph concentrate to 1% protein dialyze vs, 0.01 H acetate- 0.05 M NaCl, pH 5A dilute dialyzate 10-fold centrifuge . low-ionic-strength super S.A. * 63 add 3 x wt. bentonite centrifuge ppt disoard super S.A. * 78 PURIFIED ctgM Fig. 1.— Summary of purification procedure for trypsin-binding o^M from Cohn III-O Fraction. 22 of solvents other than the standard phosphate buffer were measured in an Ostwald viscometer, in order to correct the sedimentation coefficients to son & U |W For calculation of sedimentation and diffusion coefficients, enlargements of the schlieren patterns photo graphed during the runs were made, to ten or fifteen times the actual cell height, either photographically or by manual tracing of the projected image; exact enlargement factors were determined from enlarged photographs of a ruled plate. Sedimentation coefficients (s) were calculated from 1 d In x the formula s * — — — — — , where u > is the rotor speed O o 2 dt in radians/sec, x is the distance of the boundary (schlieren peak) from the center of rotation, in cm, and t is time in seconds. The Slope of a plot of In x vs. t (values from different exposures made during a single run), divided by t * > , gives s directly (in Svedberg units, S). Since all runs in the present study were done at or very near 20°, the only temperature correction needed was the ratio of the viscosity of water at the experimental temperature to that o at 20 . Correction to was made by multiplying the experimental s by the viscosity of the solvent used, rela tive to water. Diffusion constants (D) were calculated from the sedimentation velocity data by measuring peak areas with a planimeter and peak heights directly on the enlarged 23 photographs. A plot of the square of the second moment This method is quite accurate to homogeneous materials, but in the present case the peak was asymmetric and boundary spreading was due to inhomogeneity as well as to diffusion. A rough estimate of D for the major component was made by drawing a leading edge to exclude part of the shoulder and give a peak which was symmetrical; this area was then measured and the above formula applied. The concentration dependence of the sedimentation coefficient was determined by measuring sedimentation coefficients of purified cXgM (about 0*5% protein) and of four different dilutions, down to 1/8 the original concen tration. Absorbancies at 280 ma of the various dilutions agreed with dilution factors within - 2$. s was plotted against concentration, (Pig. 2), and extrapolated to zero concentration by the method of least squares. All experi mental points were within - 0*6% of the theoretical values which were determined by the least-squares line. Starch Gel Electrophoresis Starch gel electrophoresis was done both horizon tally, in narrow trays (dimensions 25.0 cm x 2.0 cm x 0.6 cm) with a single sample (applied on filter paper) run in each tray, and vertically, with liquid samples applied against time has a slope equal to 2D. 2»f 18- c n 02 0.3 0. 1 04 03 PERCENT PROTEIN Pig, 2.— Concentration dependence of the sedimentation coefficient of purified Sample was bentonite supernatant in phosphate buffer; points are sedimentation coefficients calculated as described in text, from five different runs at the protein concentrations indicated. The schlieren pattern of the 0.53^ solution is shown in Pig. 7A. in preformed slots and up to ten samples run on a single slab* Connaught Starch-hydrolyzed was used, and in both methods the discontinuous buffer system described by Poulik (*f5) was employed* The starch was prepared in 0.08 M Tris- 0.005 M citrate, pH 8.5* cooled, and the samples applied. The electrolyte was 0.025 M borate buffer, pH 8.6. Horizon tal runs were done at room temperature, with a constant current of 3 mA per tray, and up to four trays could be run simultaneously. For vertical electrophoresis, the Buchler apparatus was used, the starch concentration was increased slightly to give a less fragile gel, and runs were done in the coldroom (5°) for about 1 6 hours at 150 volts. Gel slabs were removed from the molds, sliced longitudinally with a steel wire, and stained *+5 seconds with Amido Schwarz Black. They were then rinsed several times with methanol:water:acetic acid (50 : 50:10) and left overnight or longer in the same solvent to destain the back ground. In some cases serum was run simultaneously as a standard, but usually the purpose was comparison of differ ent fractions obtained at various stages of purification, and unfractionated material served as "standard." Polyacrylamide Gel Disc Electrophoresis The simplified Clarke (**6) method of disc electro phoresis was followed. Reagent solutions were: (a) 30 g acrylamide 1 g N,N*-methylenebisacrylamide 1 2 3 ml water 26 (b) 0.28$ (v/v) Nl-tetramethylenediamine (TEMED) solu tion (aqueous) (c) 0.1*t# (w/v) ammonium persulfate solution (aqueous) (d) 29#0 g glycine 6,0 g Tris 9 8O ml water A mixture was made of 2 parts (a), 1 part (b), b parts (c), 1 part (d), and immediately poured into glass tubes (7.7 cm long x 6 mm i.d., mounted vertically in rubber serum stoppers) to a height of 5.9 cm, overlayered carefully with water to form a horizontal interface, and allowed to polymerize for about one hour. The water was removed and the tubes mounted vertically with the upper ends extending through hollow rubber stoppers into the upper electrode compartment. Samples in 5% sucrose were applied directly to the gel surface; sample volume was 0.20 to 0.25 ml, containing 25 to 50 /xg protein (for a puri fied fraction) or larger amounts, up to M-00 ^ig (for serum). Electrolyte was then layered carefully over the samples by means of a thin glass capillary or a hypodermic needle, and the tubes plugged loosely with cotton. Electrolyte (a 10- fold dilution of a stock solution containing 29*0 g glycine, 6.0 g Tris, 5.0 ml 1 N HC1, 975 ml water) was placed in both compartments and a few drops of bromphenol blue solu tion added to the upper compartment as a "tracking dye." A current of 1 mA per tube was applied until the dye front appeared in the gel, then increased to ^ mA per tube and 27 electrophoresis continued until the dye reached the bottom of the tube, or longer if desired, Gels were loosened from the sides of the tubes with a thin wire, removed, and stained for one hour with 1% Amido Schwarz Black in 35# methanol-8# acetic acid. Prior to electrophoretic destaining, they were rinsed several times, and leached overnight or longer, with 8# acetic acid, Destaining tubes were 7,5 cm lengths of I^-rex tub ing (7 mm l,d,), with lower ends constricted by drawing out in a"flame. Gels were placed in these tubes, and mounted vertically in the apparatus as before, with 8# acetic acid in both electrode compartments, A current of 8 mA pet tube was applied until the free dye had moved completely out of the gels. Up to eight tubes could be conveniently handled in a single run, and samples were usually run in duplicate, with serum or unfractionated material as a standard. Esterase Assay Esterase activity of the protein solutions was measured by the pH-stat assay method described in detail in another section, A known concentration of TAME was added to an appropriate dilution of the protein in the reaction vessel, and the volume of standard base added to maintain constant pH was plotted as a function of time. Rather large amounts of protein were required because TAME 28 esterase activity of solutions was much lower than that of trypsin, but the method was useful for comparing relative esterase activities of partially purified frac tions, Activity is expressed as micromoles of TAME hydro lyzed per minute, per ml of reaction volume, by one ml of protein sample, and specific activity calculated by divid ing by the A2qq of the sample. Trypsin assays were usually done at the same time, under the same conditions, so that esterase activity of samples could be expressed as a per cent of the activity of an equal weight of trypsin, uH-stat Assays A Radiometer automatic titrator (Model TTT1) was used, equipped with a recorder and motor-driven syringe (0.5 ml capacity) for delivering standard base into the reaction vessel. Reaction volumes could be varied from ml to 750 ml by using different electrodes (Radiometer) and vessels. Radiometer vessels, thermostated by a coil of copper tubing through which water was circulated from a constant-temperature bath, were used for volumes up to 10 ml. Large-volume (200 and 750 ml) reactions were run in beakers, covered with a piece of lucite with holes drilled to admit electrodes and a thermometer, and sealed with tape. It was not necessary to thermostat large volumes for reactions at 25°; provided reagents were at this temperature initially, variation was less than 0.5° 29 during the assay* Reactions were stirred magnetically, and were ran under nitrogen to exclude carbon dioxide. Standard base was prepared by diluting saturated NaOH solution to approximately 0*1 N and titrating with potassium biphthalate. For use, it was diluted (usually to 0.02 N) with boiled water and placed in the reservoir of the titrator, with a tube of soda lime attached to the air inlet. The reaction solvent was 0.0*+ M KC1-0.01 M CaClg* unbuffered. Substrates were £-1oluenesulfony 1 -L-arginine methyl ester (TAME), obtained from Calbiochem or from Mann Research Laboratories, and benzoyl-L-arginine ethyl ester (BAEE), from Mann. Sources of other reagents used were: £-1oluenesulfonyl-L-arginine (TA), from Mann and from K & K Laboratories; benzoyl-L-arginine (BA), from Calbiochem. Stock solutions of reagents were made in the reaction sol vent. Acidic substrate solutions were not neutralized, to avoid hydrolysis on storage; pH of other reagent solutions was usually adjusted. When high concentrations of BA and TA were required, these were dissolved by heating, cooled quickly, and added carefully to the reaction mixture as supersaturated solutions. For the assay, substrate (and product or other analogue, when used) was added to the appropriate volume of solvent in the vessel, the pH adjusted as necessary, and 30 the solution stirred under nitrogen with the machine turned on until no further base was being added, or, when very- high substrate concentrations were used, when the “blank" rate reached a constant value. Enzyme was then added with a syringe to start the reaction. Stock solution of trypsin was 1 mg/ml in 0.2 M KC1- 0.05 M CaClg; this was diluted, usually to 10 jig.ml for TAME assays and 25 pg/ml for BAEE, with 0.005 M Tris-0.01 M CaCl2> pH 7.9> at least 30 minutes prior to use. Stand ards run at intervals during the day showed that the activ ity of the trypsin usually did not vary over a period of several hours. Fresh dilutions were made daily, and there was slight variation of standard rates from day to day, so when assays for the same experiment were done on different days, rates were corrected to a single standard (either an average or an arbitrarily selected standard). Reactions usually did not start smoothly but the rate became constant within a few seconds so that in most cases an unequivocal straight line could be drawn on the chart and the initial rate calculated from its slope • For the lowest substrate concentrations V /umoles of substrate was used, and the entire course of the reaction occupied only b0% of the chart width. In these cases the Initial rate was calculated as follows: tangents were drawn to the curve at various points, and the rates at these points calculated from their slopes. The concentration of unreacted substrate at each point was known from the ini tial concentration and the extent of reaction. A plot of the velocity, v, vs. v/[S] (where [S] is the molar concen tration of substrate) was made and initial rate determined by extrapolation to initial substrate concentration. This treatment assumes that Michaelis-Menten kinetics are followed, which turns out to be the case for low substrate concentrations (see "Results"). When STI was used, the crystalline protein was dissolved in reaction solvent and the pH adjusted to 8. It was then mixed with trypsin in the desired ratio and allowed to stand for a few minutes before use. For assays using bound trypsin (trypsin complexed with <X2M) the usual procedure included preincubation of trypsin and in a concentrated solution, buffered at the appropriate pH, and addition of a measured amount o-f this solution to the reaction vessel, to start the assay. In order to determine whether the extent of binding of oCgM to trypsin is concentration-dependent, in some assays the bound trypsin was diluted to the final assay volume and allowed to stand for various periods before starting the reaction by addition of substrate. An attempt was made also to determine the time required for binding, by adding trypsin and ctfgM separately to large volumes of solvent, 32 and starting the reactions with substrate after various time intervals; these experiments were inconclusive but demonstrated that binding is not instantaneous and a pre incubation period is necessary. The OCgM solution used for all the pH-stat studies of bound trypsin was a low-ionic-strength supernatant, which was the purest preparation available at that time. All assays were done in duplicate, and the average initial rates of substrate hydrolysis in moles liter 1 sec ^ were divided by molar concentration of trypsin (using spectrophotometric measurement of concentration and taking the molecular weight to be 2^,000) to give v0/[E0] in sec""1. Substrate concentrations were calculated from the weight of crystalline reagent used in making stock solu tions; occasionally this was checked by allowing the reac tion to go to completion, and the agreement was usually within Casein Assay The method of Bundy and Mehl (*f7) was used. Sub strate solution was 3 grams of vitamin-free casein (Pfanstiel) dissolved in 100 ml of 0.17 M Tris buffer, pH 7.6. Graduated amounts of *2** (fraction IIA from DEAE-cellulose column) were preincubated with trypsin in the same buffer for one-half hour at 37°» each tube con tained 10 ^ug trypsin in a total volume of 2 ml. One ml of 33 substrate solution was added, and after incubation at 37° for 20 minutes the reaction was stopped with 6 ml of 2.5# trichloroacetic acid. The suspensions were allowed to stand for one and one-half hours, then centrifuged, fil tered, and the absorbancy of the clear filtrates measured at 280 mji. Blanks contained CXgM at the various concentra tions used, and substrate; trypsin was added after the tri chloroacetic acid. CHAPTER IV RESULTS Purification Procedure Five kg of frozen III-O paste, after dialysis and centrifugation to remove lipoprotein, yielded about one liter of a 15% solution of lipid-poor protein. Figures in Table 1 are based on a 5o-ml aliquot of this, the amount that could be handled in a single column run, and are aver ages of results of a number of different runs. Slightly better than four-fold purification was achieved, with about 50% recovery of activity. In the original method of removing lipoprotein by flotation in a high-density salt solution (5), only the pellet was recovered for further purification of For large amounts of-III-O it seemed more efficient to use a lower speed and a shorter time (insufficient to sediment the macroglobulin completely) and to recover the entire lipid-free fraction, including slower-sedimenting proteins which could be removed by other methods. The disc electro phoresis pattern of this fraction (Fig. 3) shows that albumin is present, as well as ceruloplasmin, which gives a 3*f 35 TABLE I PURIFICATION OF TRYPSIN-BINDING MACROGLOBULIN FROM COHN III-O FRACTION Step Fraction S.A.a Total mg Protein ■ . Recovered Total Apparent pg Trypsin Bound0 Centrifugation of III-O Lipid-poor Protein 18 7250 130,500 PEG precipita tion Supernatant Precipitate 2 28 • • • * M -520 126,900 DEAE-cellulose column chroma tography I IIA 6 5*f • • • • 1270 6 8 ,500 IIB *fl • • • • III 13 • • • • IIB rechromato graphed IIA' 5>f 200 1 0 ,800 Total (IIA + IIA*) 1*1-70 79,300 Low ionic strength pre cipitation Supernatant Precipitate 63 bo 1270 * • • • 80,000 Bentonite treatment Supernatant (Purified 78 760 59,^00 ug tiypsin activity per mg protein; standard BAPNA assay. 280 mp ‘ times volume in ml. s S.A. times total mg protein. A B C Fig* 3*— Disc electrophoresis of lipid-poor protein from Cohn III-O Frac tion, before and after PEG precipitation. Samples were dialyzed vs. 0.02 M Tris- 0.06 M NaCl, pH 7.7, then diluted 10- fold with 10% sucrose. Final protein concentrations were about 0.5 mg/ml; 0.2 ml applied to each gel. A, lipid- poor protein; B, PEG precipitate; C, PEG supernatant. The diffuse band near bottom of A and C is albumin; otoM is the dark band about 0.5 cm from top in A and B. 37 deep greenish-blue color to the solution, and a number of unidentified proteins. Due to the large volume of material to be handled, removal of some inactive proteins by a batch procedure, such as fractional precipitation, was considered desirable at this stage. Use of ammonium sulfate is undesirable because ammonia, and primary amines, destroy the trypsin- binding activity (13). A two-step salt fractionation using phosphate was developed, in which trypsin-binding activity * precipitated at pH 5«5 between 1.5 and 2.0 M phosphate. It was later found to be simpler and more convenient to use PEG, of molecular weight about 6,000 (*+8), as a precipitant in a one-step procedure, for the entire batch of crude pro tein solution. At 6$ PEG concentration, about 60$ of the protein precipitated as a gummy mass, and contained essen tially all of the trypsin-binding activity, as determined by assay. Comparison of analytical ultracentrifuge schlie ren patterns of supernatant and precipitate shows that all of the macroglobulin (17-19 S) was precipitated under the conditions used, but the precipitate still contained a number of other proteins. Pig. 3 shows the disc electro phoresis patterns. PEG was present in the precipitate, but the concentration was low enough so that dialysis removed it readily. High concentrations of PEG interfere with the tiypsin assay, giving a finely-divided precipitate when 38 phosphotungstic acid is added and thus making it difficult to exactly quantitate the recovery of activity in the supernatant. PEG also absorbs in the ultraviolet, but this was negligible compared to the protein absorbancy of both supernatant and precipitate at this stage. The EEAE-cellulose column method described is essentially the same as that used for preparation of cy-gM from plasma (*fl). In an attempt to improve the yield, pur ity, and column capacity, samples of partially purified 0*2^ were chromatographed on small EEAE-cellulose columns under various conditions of pH and ionic strength. Condi tions tried included salt gradients (0.06 to 0.12 or 0.18 M NaCl) at pH 8.6 and 6.7, and pH gradients from pH 7.6 to 5.0 using 0.02 and 0.05 M imidazole buffer contain ing 0.06 and 0.03 M NaCl. None of these showed any notice able improvement over the previous method. The column chromatography step was quite useful, giving a two-fold or better purification, but was time-consuming for large amounts because the capacity was only about 1 mg protein per gram (dry weight) of EEAE-cellulose, Each column run, including washing with 0.5 M NaCl until the eluate was free from protein and re-equilibrating with starting buffer, required 2 1/2 to 3 days, and usually five runs could be made without repacking the column. Unless the column was overloaded, the unadsorbed protein (Peak I) contained no activity or macroglobulin. Peak IIIj which was also discarded, had specific activity of 10 to 15, but about b-5% of the protein present was macroglobulin-jwlth a sedimentation coefficient over 18 S. It seems likely that this material consists of a mixture of trypsin-binding oi^^M and an inactive faster-sedimenting component. IIB contained substantial activity, and frac tions from several columns were pooled, concentrated, and rechromatographed to improve the overall yield (Table l). Pig. b shows a typical column run. The major schlieren peak In the ultracentrifuge pattern of IIA is macroglobulin, having a sedimentation coefficient of 16.8 S (uncorrected for protein concentra tion), but the pattern is noticeably skewed with a 19 S "shoulder" (Pig. 5). There is also a distinct 6.5 S peak and smaller amounts of other slowly-sedimenting material. Starch gel electrophoresis of IIA shows at least three bands (Pig. 6), and five bands can be seen on polyacryl amide gel disc electrophoresis. Dialysis of IIA at low ionic strength and acid pH precipitated the 19 S material primarily, as seen from the ultracentrifuge pattern (Pig. 5), though some of the 6.5 S and 17 S proteins also precipitated. The specific esterase activity of the precipitate was about twenty times that of the supernatant, indicating that this activity is probably l+ o 04 r^t -02 o -40 « 08- W 30 ( S I 20 0.4 30 20 1 0 LITERS ELUTED Fig. l+.— DEAE-cellulose column chromatogra phy of PEG precipitate. Sample containing 1+.M+ grams protein was applied to column, and eluted with 0.02 M Tris buffer, pH 7.7, containing NaCl at the concentrations indicated (inset). Void volume was b liters. Protein concentration (left scale) - - - Activity, determined by BAPNA assay (right scale) Pig. 5.— Sedimentation of olgM at intermediate stages of purification, showing fractionation of heavy components. SampleB run in phosphate buffer. Times (after reaching full speed) at which exposures were made, and phase plate angles, are given. See text for units of specific activity (S.A.).V A-Column Peak IIA: Trypsin-binding S.A.,!& esterase S.A., 0.039; 21 min,^0; 28 min^O®; 35 min,35°; ^2 min,35°. B-Low-ionic-strength ppt* Trypsin-binding S.A.,l K)sesterase S.A., 0.126; 16 min, 20 min, 26 min, 29 min, all at ^5° angle. C-Low-ionic-strength supers Trypsin-binding S.A,,63; esterase S.A., 0.0066; 16 min,60°» overexposed; 29 min,60°; 39 min,If Jr. h2 A B C D E Pig. 6.— Vertical starch gel electrophoresis of c* gM at various stages of purification. pH 8.6, sam ple volumes about 0.1 ml. Sample A280 mu Apparent jag trypsin ^ bound, per ml sample A Low-ionic-strength ppt. 15,6 336 B Low-ionic-strength ppt. 12.9 280 C Bentonite super, **.6 356 D Low-ionic-strength super. 5*5 317 E Peak IIA, LEAE-cellulose 9.2 28lf **3 associated with the 19 S material. It is not known whether the trypsin-binding protein itself will catalyze ester hydrolysis; if so, this activity is very slight. Specific esterase activity appears to Increase at successive steps of purification until the low ionic strength precipitation removes most of the 19 S material (Tahle 2). Trypsin- binding specific activity of the precipitate was lower than in the supernatant. On starch gel electrophoresis, the major component of the precipitate moves faster than that of the supernatant (Pig. 6). The low Ionic strength precipitation step was rather difficult to control. The amount of precipitation was sensitive to slight changes in temperature, pH, and ionic strength. The solubility characteristics of the pro teins present at this stage appeared to be quite similar, and they tended to co-precipitate initially, then gradually solubility equilibrium was reached over a period of several days. Several methods were tried for the removal of the 6.5 S material and other minor slowly-sedimenting contami nants from the low-ionic-strength supernatant. Gel filtra tion on either polyacrylamide gel (P-200 and P-300) or Sephadex G-200 was effective, and so was fractionation with Rivanol at pH 5.^, but the bentonite treatment described was equally effective and far simpler. M+ TABLE 2 TAME ESTERASE ACTIVITY HJRING FURIPICATION Sample Specific esterase activity PEG precipitate 0.029 DEAE-cellulose column Peak IIA 0.039 Low ionic strength supernatant 0.0066 Low ionic strength precipitate 0 . 1 2 6 Trypsin 76 A ^umoles/ml of TAME hydrolyzed, per min., per mg protein at pH 8.0, temp. 25° * TAME concentration 25.0 mM. ^5 The hentonite supernatant had a specific activity near 8 0 , and represented the highest degree of purification which was achieved. In the following discussion of its physical properties, bentonite supernatant will be referred to as purified ot^M. Physical Properties of Purified ot . 2M Zone Electrophoresis On starch gel electrophoresis, purified otgM has two bands, the less intense one moving slightly faster (Pig. 6 ). Both are in the region. An additional band is seen on disc electrophoresis. Sedimentation and Diffusion Purified oL^i shows a single sharp peak in the ultracentrifuge (Pig. 7), with S20,w » 17.3 S (Pig. 2). It i« still slightly skewed, however, indicating incomplete removal of the 1 9 S material, so this sedimentation coeffi cient is probably an average value for the two * * overlapping" components and therefore slightly higher than that of the pure 17 S protein would be. Due to the correlation between increase of specific activity and increasing homogeneity of the 17 S peak, it seems certain that trypsin-binding activ ity is associated predominantly, if not exclusively, with this protein. Although it is not possible in such a case to determine with any accuracy the degree of purity of the J B r J L -A - J j *1 ' A Fig. 7.— Sedimentation of purified ot-oM at neutral pH, and after dialysis at pH W.O in low and high acetate, snowing dissociation due to H+ and acetate. Times given are minutes after reaching full speed; phase plate angle 6 0° throughout, r /2 * ionic strength; * ) rel = relative vis cosity. A-In phosphate buffer, pH 7.1; F /2, 0.15s ? r-el* 1 . 0 0 . 8,16,2^,32, and * t 0 min. B-3h 0.01 M acetate-0.0985 M KC1, pH lf.0; /V2, 0.1; 9 ^ , l.Q. 6,1^,22,30, and 38 min. C-In 0.67 M acetate, pH lf.0; F/2> 0.1; ^reijl.075. 8,16,2^,32, and ^0 min. -r *+7 major component, a symmetrical peak (constructed as described under "Methods’ *) Includes about 80$ of the total area under the curve. The diffusion constant calculated - 7 2 - 1 from this symmetrical peak is 1 , 7 x 1 0 cm sec , which may be taken as a rough estimate, at best, of D for the major 17 S component. Using these values for s and D, and assuming that the partial specific volume, like that for most proteins, lies between 0,70 and 0 .7 5 , the molecular weight would be in the range 8 0 0 , 0 0 0 to 1 , 0 0 0 , 0 0 0 . Ultraviolet Absorptivity Purified was dialyzed exhaustively against 0,01 M Tris-0.01 M NaCl, pH 7,78. The absorbancy of a 10- fold dilution was measured in a Beckman BU spectrophoto meter at wavelengths from 277“280 m^i. Measured volumes ( 1 5 ml each) of protein solution and dialyzate were placed in dried weighing bottles and dried to constant weight in vaouo, over P2®5 » 6 0°, Mettler balance was used for weighing. Weight of buffer salt was 31 mg (theor,, 30.2 mg) and of protein, 6 3 mg. The absorptivity, a(l#, 1 cm) of trypsin-binding oi- 2* * was found to be 8.7> at 280 nyi, Thus, protein concen trations of relatively pure solutions can be determined by multiplying the absorbancy at 2 8 0 nyi by the factor 1 . 1 5 . Specific Trvoain-Binding Activity The specific activity of the purest preparations was near 80. This of course represents apparent trypsin activity, which is about 0 . 6 5 that of free trypsin in the standard BAFNA assay. The trypsin concentration was measured spectrophotometrically and required correction for the 3 0 to U-0% inactive protein present, as determined by titration with STI. When these corrections are made, and the assumed a(l^,l cm,2 8 0 mp) of 1 0 . 0 on which specific activities of crude preparations are based is replaced by the experimentally determined value of 8 , 7 , the combining ratio of trypsin to ol2M, weight, is about 1 . 0 to l^.^. Therefore one mole of trypsin, of molecular weight 2^,000, combines with about 3^6,000 grams of ol2M. Consequently, there must be at least two trypsin-binding sites on the ol 2M molecule. It is possible that there are more, since the best preparations do not appear to be entirely pure, and there was some loss of activity at various stages of purification due to time lags while experimental methods were being developed. Preparations were stored at 0-5° and, where prolonged storage was necessary, were sterilized by passage through a Seitz filter. There was gradual loss of activity, over a period of several months, in most cases. Lyophilization could also be used but the protein did not all redissolve, and that which did usually had a lower specific activity than the original solution. Spectrophotometry of Solutions Ultraviolet and visible spectra of purified < = * - 2* * samples, and difference spectra of mixtures of ol 2^ and trypsin, vs. the same solutions in separate cuvettes, were measured with a Cary recording spectrophotometer. Spectra of two different <*2M solutions in 0.01 M Tris-0.01 M NaCl, pH 7.78, at protein concentrations of 1*75 mg/ml, were recorded, with buffer in the reference cell each time. The spectrum of purified ^ 2^ showed no absorption in the visi ble range. Ultraviolet spectra of the two different preparations were identical. Pig. 8 shows that A max is at 278 n p j . , and there is a shoulder at 2 8 3 n j p . which probably is due to tryptophan. For the difference spectra, two pairs of 1-cm cells in series were read against each other, one pair containing a mixture of equivalent amounts (determined by assay) of trypsin and o l _ 2M "the reference pair containing the same amounts of the two proteins in separate cuvettes. Trypsin was dissolved in 0.0025 N HC1-0.01 M CaCl2 at a concentra tion of 1 0 mg/ml, and 0 . 0 5 ml was mixed with 3 « 0 ml of <i2M solution (2.63 mg/ml) in 0.01 H Tris-0.01 M NaCl, pH 7.78. The other cell of this pair contained -0.05 ml of the HC1 solution in 3*0 ml of buffer. One of the reference cells contained oL2M in Tris-NaCl with HC1 added, the other con tained trypsin diluted with Tris-NaCl, so that the solvent 5o i . o o CO 0.5 248 268 288 308 328 WAVELENGTH (Mji) Pig. 8 .— Ultraviolet absorption spectrum of purified o<2M. Sample: bentonite supernatant, 1 . 7 5 mg/ml in 0.01 M Tris-0.01 M NaCl, pH 7.78. Opti cal path was 1 . 0 cm. composition was the same in all four cells, and total amounts of protein were identical in the two pairs. The spectrum of trypsin alone, at the concentration used for the difference spectra, was also recorded. The difference spectrum of oi 2^“' t ryP8in complex, vs. the separate protein solutions, shows some rather poorly-defined relative maxima and minima in the range from 276-290 mp. (Pig. 9)* spectrum of trypsin and oi 2M ijl separate cells in series was not recorded, but the total absorbancy expected from various relative concentrations can be calculated, and it is found that the irregularities in the difference spectrum cannot be due to slight experi mental variations in relative concentrations of the two proteins. The difference spectrum shows relative maxima at 2 7 6-2 7 8, 28*+, and 2 9 0 - 2 9 1 mp, and minima at 2 8 0 and 2 8 6- 287 i e j u . These features are probably due to perturbations of the absorption of both tyrosyl and tryptophyl residues in.either or both proteins. Dissociation Some interesting changes in the ultracentrifuge pattern and trypsin-binding activity of purified otgM were produced by changing the pH, and by use of other buffers than the standard phosphate. Dialysis against acetate or citrate buffer at pH below 5 caused the appearance of an 11 S peak in the ultracentrifuge pattern; the 17 S peak was 52 0. 10- 005 < < 3 - 0.05 -0.10 258 298 278 308 WAVELENGTH (M p) Fig* 9*— Difference spectrum developed by trypsin on interaction with a" * : 7.7. Reference* Purified ( X p M , 2*o3 mg/ml in 0.01 M Tris-0.01 M NaCl, in 1-cfi cell; second cell, in series, contained trypsin, 0.17 mg/ml in same buffer. Sample: oLpM end trypsin mixed, final concentrations as above. Second cell contained buffer. 5 3 still present but was diminished in size* The same changes occurred on direct adjustment of the pH with HC1. Further, when two samples of were dialyzed against buffers of identical pH (*f.O) and ionic strength (0.1) but one con taining 0.67 M acetate and the other 0.01 M acetate with KC1 added to adjust the ionic strength, the ultracentrifuge pattern of the "high-acetate" sample had an 11 S peak which was significantly larger in area than that of the ”low- acetate" sample (Fig. 7). Therefore, acetate, in addition to H+ concentration, appears to promote formation of an 11 S peak at the expense of the 17 S material. A decrease of this magnitude in sedimentation coefficient could be due to a change in size and shape of the protein molecule (swelling or unfolding) but, in such a case, the diffusion coefficient would also decrease. The 11 S material actually diffuses more rapidly; in fact, cal culation of s and D for both peaks in the schlieren pattern of the 0 . 6 7 M acetate-treated shows that the molecular weight of the faster-sedimenting material must be about twice that of the slower. The most likely explanation for the appearance of the 11 S peak is that, on acidification, the 17 S material partially dissociates into identical (in sedimentation behavior) subunits. The extent of dissocia tion was a function of H+ (and acetate) concentration, but a pH as low as 3.0 did not produce complete dissociation (Fig. 10). Pig, 10,— Sedimentation of purified ^ 2^ P® 3*0 9»5, immediately after pH adjustment and 2 k hours later, Por all samples, times of exposures after reaching full speed, and phase plate angles, were: 8 and lb min, 60°; 2m-, 32, and *f0 min, ^-5 . Ionic strengths were adjusted with NaCl to about 0.15. S.A. of solution before pH adjustment was 60, A-Purif, + 1 N NaOH to pH 9*5; S.A, = 55> protein concentration O.^l*. B-Purif. oioM + 0,5 H HC1 to pH 3.0; S.A, = 0, protein concentration 0.33$* C-Same sample as B, 2*f hours later; S.A, = 0. D-Same sample as A, 2k hours later; S.A, = 55. ¥ Returning the acidified protein to neutral or slightly alkaline pH by dialysis resulted in partial reassociation; that is, the area of the 17 S peak increased at the expense of the 11 S, Rapid neutralization with NaOH did not appear to accomplish this. Loss of activity at acid pH was also partly rever sible, if the pH did not go too low. At pH 3.0, all activ ity was lost, irreversibly. At pH *+.0 about 25% of the original activity remained, and this increased to slightly over 50% if the protein remained at neutral or slightly alkaline pH overnight. Since the assays were done at pH 7.7, and required preincubation with trypsin at this pH, the values obtained immediately for the acid-treated pro tein probably included some reactivation which occurred before substrate was added. There does not appear to be any direct correlation between reversible activity loss and dissociation, because; (1) at pH 3*0, more than half of the protein remained in the 17 S peak but was totally inactive (Pig. 10); (2 ) rapid neutralization of which had been at pH b.Q did not diminish the size of the 11 S peak in the ultracentrifuge pattern, but did restore the activity to at least twice that of the acidified material. Use of a partition cell made it possible to sedi ment the 17 S boundary below the partition, leaving only 56 11 S subunits above where they could be recovered and assayed* This material was found to be 50 to 60% as active as the original, but of course this does not necessarily mean that the subunits themselves were active; they may have recombined under the assay conditions. o * . 2* * which had been treated with 0.67 M acetate, pH .0, then returned to pH 7*7 by dialysis, showed an additional faster-moving band on starch gel electrophoresis, beside the two bands already present before acidification. In the ultracentrifuge pattern of the same sample, an 11 S peak was still present. The corresponding Mlow-acetateM sample, after neutralization, had a much smaller 11 S peak and a much fainter fast band. Results of a few experiments to investigate changes in o<- 2*4 at alkaline pH may also be mentioned. If the ori ginal (non-acidified) protein solution was adjusted to pH 9*5 with NaOH, the activity remained high (over 90% of the original) for several days* However, there was a progres sive change in sedimentation behavior. An hour after pH adjustment the pattern was quite similar to the original, but 2* + hours later the peak had become very broad and skewed, suggesting possible aggregation (Pig. 10). Similar changes in the 17 S peak were also found in the pattern of HCl-treated and rapidly neutralized solution, although the pH was never above 7.2. Kinetics of free Trypsin Initial experiments to determine suitable condi tions for assays using TAME as substrate revealed that deviations from Michaelis-Menten kinetics are quite evident in the substrate concentration range usually employed for these assays. Consequently, some information about the kinetics of hydrolysis of TAME and BAEE by free trypsin was necessary before any comparison with the behavior of tryp sin bound to could be made. Dependence of initial reaction rate on substrate concentration for trypsin-catalyzed hydrolysis of TAME, at pH 8.0 and 25 - 0.5°» was determined for TAME concentra tions from 6.7 x 10 ^ M to 0.15 M, using volumes from 10 ml to 750 ml. Assays with the same substrate concentration in different volumes demonstrated that the rates are independ ent of assay volume so no error was introduced by changing the volume. The Eadie plot (according to the Michaelis- Menten equation in the form rSfr= keat' n“ ! r s ] ’ is shown in Pig. 11; the deviation from Michaelis-Menten kinetics is very marked at substrate concentrations above 0.2 mM TAME. Apparent constants can be obtained from the data for lower substrate concentrations, and are in good agreement with values reported in the literature by workers 58 1 5 0 , 1 0 0 I o UJ C O ' • ' UJ 90- 20 2 5 * / [ £ . ] [ 8 ] * I 0 " 9 ( M - 'S E C -1 ) Pig. 11.— Eadie plot of trypsin-catalyzed hydrolysis of TAME, showing substrate activation. TAME concentrations were 1,3 x 1 0"5 M to 2.5 x 1 0”^ M; assayed at pH 8.0, temperature 25 - 0.5°. - - - - extrapolation to.determine kcat = 62.5 see”;* * 1.06 x 10 5 M who have used a sufficiently wide range of substrate con- centrations (30). The Kj, of 1,1 x 1 0 M and kca^ of 62.5 sec- 1 can be taken to represent the affinity of substrate for the hydrolytic site, and the maximum rate when this site is saturated with substrate, if activation did not occur. The extent of activation can be expressed as a ratio of the observed rate, at any substrate concentration, to a theoretical rate for that concentration if the system followed Michaelis-Menten kinetics. For example, the observed rate is 1.3 times kca^ . for 2.5 mM TAME, a concen tration over 2 0 0 times the 1^. Substrate activation explains the strong dependence of initial rate on substrate concentration at concentra tions well above the K^, where the enzyme would be expected to be saturated with substrate and therefore relatively insensitive to such variations; but it does not account for the observed failure of rates to decrease during the course of the reaction as substrate is depleted. It was postu lated that the product, TA, might also serve as activator, and evidence that this does occur is shown in Fig. 12 (*+9). An arbitrary "standard” TAME concentration of 2.5 mM was used, and to this were added various amounts of TAME and TA. It can be seen that the rate for the standard sub strate concentration is approximately doubled by addition of a 1 2-fold excess of either substrate or product. 6 0 1 8 0 1 60 1 4 0 i o UJ to 120 100 80 60 80 60 40 20 mM TA M E OR TAM E + p-TO SYLARGININE Pig. 12.— Comparison of the effects of TAME and mixtures of TAME and TA at equal total concentrations on the rate of hydrolysis of TAME. • TAME alone; 0 2.5 mM TAME plus TA to the indi cated total concentration; 015 mM TA plus TAME to the indicated total concentration. 61 In order to rule out the possibility that this activation involves an increase in the amount of active trypsin, the titration of active trypsin with STI was done at two different substrate concentrations and also with TA added* Pig. 13 shows that the amount of STI required for 1 0 0^ inhibition is the same in all cases, and therefore the amount of active trypsin is unchanged even though the rates are higher with product or additional substrate. Similar kinetic studies were done with trypsin and BAEE, from 6*7 )iM to 2.0 mM, at both pH 6.0 and 8.0. The results are plotted according to another form of the Michaelis-Menten equation, L§,TLPq] = [s]. i , vo ^cat ^cat in Pig. I1 *. Apparent kinetic constants were obtained from an expanded plot of the low-concentration region, and the straight lines in the higher-concentration plot calculated from these values. Since the reciprocal of the rate is plotted, substrate activation is indicated by downward curvature at high substrate concentrations. In changing from pH 8.0 to 6.0 Kjq increases, kca^ . decreases, and the extent of activation increases. Por example, at 2.0 mM BAEE, the observed rate is 1.09 times k^^ at pH 8.0, and 1.2*t times k^-^ at pH 6.0. The results thus far are compatible with the idea 62 100 80 60 i o UJ C O 40 o UJ > 20 0. 8 0.6 0.4 0.2 GM S T l/G M TR Y P S IN Pig. 13*— Inhibition of the hydrol. ^s of TAME by STI with a sub strate concentration of 2.5 mM ( • ), a substrate concentration of 13*75 mM (O), or 2.5 mM TAME and TA to bring the total concentration to 13*75 mM ( □). -10 0 1 0 20 30 40 [BAEE] (MM) 02 0.5 1 . 0 [BAEE] ( mM) 2.0 Fig. ^.--Trypsin-catalyzed hydrolysis of BAEE at pH 6.0 ( •) and 8.0 ( o ; temperature 2 5 ± 0 .5®* other conditions given in text. Low concentration region used for calculation of con- S't&U'tS pH 6 * Km = 8 x 10“® M, kcat * 5.^ sec"1 pH 8t 5 « 3 X 10“ 6 M, kca^ * 12.6 sec" 1 The lines defined by these constants plotted on a different scale to show deviation of points at high [S], due to activation. 6k that activation involves binding of a second molecule of substrate (or product) by tiypein, that the activating molecule is not hydrolyzed, and that the dissociation con stant of the second molecule is much higher than the Since TA is an effective activator of trypsin- catalyzed TAME hydrolysis, it might be expected to have the same effect on BAEE hydrolysis, but instead, it is inhibi tory, Several different BAEE concentrations in the milli- molar range were assayed with and without addition of equal or greater concentrations of TA. The degree of inhibition was dependent on both substrate and inhibitor concentra tions , but the usual plots to determine type of inhibition are of little value because substrate activation occurs at the BAEE concentrations used, Benzoylarginine was also considered as a possible activator at high concentrations of both TAME and BAEE, At pH 8,0, the concentrations tested had no significant effect on BAEE rates, but increased the rate of hydrolysis of TAME, although less effectively than similar concentra tions of TA, Table 3 gives examples of the changes in trypsin-catalyzed hydrolysis rates of high concentrations of BAEE and TAME, due to addition of various concentrations of the two products. In order to determine whether there is competitive inhibition by the products at the hydrolytic site of 6 5 TABLE 3 RATE MODIFICATION BY REACTION PRODUCTS AT HIGH SUBSTRATE CONCENTRATION Substrate Modifier Concentra tion (mM) vo (sec“^) % Change CB0) none lk,k •RAtra BA 12.5 Ik,9 + 3.5 2,5 mM BA 27.5 lV 0 TA 3.75 1 0 . 0 - 30.5 TA 2 0 . 0 6 . 0 -l*f 0 none 6 8 BA 1 0 . 0 75 + 1 0 TAME BA 25.0 77 + 13 2*5 mM TA 1 0 . 0 91 + 31 * TA 2 0 . 0 117 + 72 TA 25.0 1 2* f + 8 2 66 trypsin, rate studies were done at low substrate concentra tions (between 10 ^ and 10 * * M) to avoid as much as possi ble the effects of substrate activation* BAEE hydrolysis was readily inhibited by TA, and slightly by BA, at pH 8*0. TAME hydrolysis was inhibited by BA, but inhibition by TA was barely detectable, at pH 8.0. At pH 6.5, TA was slightly more inhibitory to low concentrations of TAME. In most cases these studies were done by holding the substrate concentration, [S], constant, and varying [I], the inhibi tor concentration. [I] was usually between 1 and 10 mM. According to Dixon (50), for either competitive or non competitive inhibition, a plot of 1 /v vs. [I] will be straight; the point of intersection of two lines for differ ent [S] distinguishes between these types. In noncompeti tive inhibition, intersection is on the [I] axis; for the competitive case the intersection is above the axis, at an [I] corresponding to the negative of the K j . (enzyme- inhibitor dissociation constant). Pig. 15 shows this type of plot for inhibition of BAEE hydrolysis by TA, at pH 8.0. Although the lines curve somewhat, competitive inhibition is suggested, with a near 1 mM. TA also inhibits BAPNA, and the determined in the standard BAPNA assay was some what higher, about 7 mM. The of BA and trypsin was determined, using both TAME and BAEE as substrates, in the pH-stat assay at pH 67 0.12 0. 10- o L L l in > ° 0. 0 6 -2 Pig. 15.— Inhibition of the tryptic hydroly sis of low concentrations of BAEE by TA, Dixon plot. Assays at pH 8.0, temperature 25 - 0.5°* o 2 x 10 £ M BAEE • 8 x 10*' M BAEE - - — Kj^ ■ 1.5 mM 68 8,0, and using BAPNA at pH 7.7 in the standard assay; values ranged from 2 to 6 mM. Because of the rather com plex kinetics of all these systems, and the curvature of the plots, there is considerable uncertainty in K^s, but they seem to be between 1 and 7 mM for both BA and TA. It can be concluded that the affinity of the hydrolytic site of tiypsin for these reaction products is probably much less than for the respective substrates. " A summary of activation and inhibition by these substrates and products is shown in Table *f, with an indi cation of the magnitude of the effect in each case. It can be seen that rate modification by the reaction products is dependent on the substrate used, as well as on the modifier. Rates of trypsin-catalyzed hydrolysis of TAME and BAEE, at substrate concentration 5.0 x 10”^ M, and tempera ture 25 - 0.5°, were measured at several other pH's; the pH-rate profiles are shown in Pig. 1 6 * Inhibition of Trypsin by otPM It has been pointed out that, In the standard BAPNA assay, ot^M is inhibitory; the decrease in initial rate is proportional to the concentration of until the equiva lence point is reached, then in the presence of excess the rate levels off at about 6 5# the activity of free tryp sin. Similar studies of the inhibition of trypsin by were made by the pH-stat method. In this experiment, 69 TABLE h EFFECTS OF SOME MODIFIERS ON THE RATES OF HYDROLYSIS OF TAME AND BAEE BY TRYPSIN TAME, high concentration TAME, low concentration Activated by TAME++ Activated by TA++ Activated by BA+ Inhibited Inhibited by TA- by BA— BAEE, high concentration BAEE, low c one e nt ra t i on Activated by BAEE+ Inhibited by TA- Inhibited by TA— No effect by BA O Inhibited by BA- 70 O u i i / i ■3 6.0 70 8.0 pH Pig. 16.— pH-Rate profiles of trypsin-oatalyaed hydrolysis of TAME and BAEE. Reaction volume 200 ml, solvent 0.0*t M KOI-0.01 M CaCl2, temp. 25 ± 0.5° •TAME, o,BAEE, both at concentration of 5.0 x 1 0 " 5 h. 50 40 90 20 71 graduated concentrations of cx-^M were mixed with trypsin (concentrations of trypsin were 10 ;ug/ml for TAME assays, and bO jug/ml for BAEE assays), preincubated, then to start the reaction 1 ml of the mixture was added to the vessel containing, substrate in the reaction solvent. No signifi cant difference in rates was observed if the preincubated trypsin- £* 2M was added first to the vessel of solvent, then after an interval of 1 0 minutes the reaction started by adding substrate, Fig, 17 shows the results; the combining ratio of trypsin and ot2M> represented by the break in the curve, is independent of type and amount of substrate, and of reaction volume, and is also in close agreement with that determined for the same solution by the standard BAPNA assay. However, the rate of hydrolysis catalyzed by trypsin- ol2M complex, relative to free trypsin, varies con siderably with the substrate used, and its concentration. When trypsin was preincubated with increasing amounts of and casein was used as substrate, the rate decreased as a function of concentration and leveled off at about lb% of the activity of free trypsin. The °*2M used for these assays was from an earlier stage of purifi cation than that used with TAME and BAEE, so the combining ratio with trypsin was different. Kinetics of Trypsin- Q*gM Complex The kinetic properties of the trypsin- o*2M complex 72 60- i o Ui J2 40 20 pLITER S a z M PER }iG T R Y P S IN Fig. 17*— inhibition of tryptic hydrolysis or TAME and BAEE. otpM solution was low-ionic-strength supernatant at concentration of 6 . 3 mg/ml, mixed with trypsin in the indicated pro portions and preincubated at pH 8.0 before assaying. O 2.0 mM TAME; # 0.1 mM TAME; £7 0.1 mM BAEE. Rates with excess 7 3$ of free trypsin rate with 2 mM TAME, 33$ with 0.1 mM TAME, 6 8$ with BAEE. were investigated by the pH-stat assay method. Trypsin was preincubated with a slight excess of solution, and this complex added to the required concentration of TAME, to start the reaction. As before, temperature was 25 * 0,5°» and pH was 8,0, Tig, 18 shows the results; data for free trypsin hydrolysis of TAME are included for compari son, In order to include the concentration range where activation is noticeable, the scale used makes the points for lower concentrations very crowded. An expanded plot of these lower points was used to determine constants, by the method shown in Fig. l*t for BAEE, Both the and kcat are altered when trypsin is bound to ^ 2^* substrate acti vation is greater with the complex than with free trypsin. For example, at a TAME concentration of 5«0 mM, the rate with free trypsin is 1 , ^ 3 times kcat, and with the complex, 2 , 3 8 times kca- £ . Similar studies were made of BAEE hydrolysis by tiypsin- o^M complex, and as with TAME, Km was higher and kCat lower for the complex than for free trypsin. There is considerable uncertainty in data for these BAEE assays, however, because duplicate rates did not agree closely, so the numerical values of kca^ = 1 1 , 5 sec”^ and = -5 3 x 10 M for the trypsin- crtgM-BAEE system are approxi mate, It seems likely that the difficulties may have been due to adverse effects on the glass electrode by the high 7*f y I L I w > X o x >° ’ -a UJ C O 25 20 [TAME] ( mM) Pig. 18.— Catalysis of TAME hydrolysis by free trypsin and trypsin- complex. Assayed at pH 8.0, temperature 25 - 0.5°. O free tiypsin,- - - theor., a 1 . 0 6 x 1 0~^ M kcat = 62.5 sec • complex, _____ theor., » 2 . 7 5 x 1 0”?M kcat = 29.5 sec- 1 Theoretical lines are based on values of Km and kcat at ' * - ow substrate concentrations. Deviation or points from these lines shows substrate activa tion. protein concentrations required for these assays. BAEE assays require about four times as much trypsin as TAME assays, for equal rates, and the trypsin complex contains about fifteen times as much protein, by weight, as free tiypsin. It was hoped that the pH dependence of binding could be investigated by determination of the trypsin- equivalence point from TAME assays, at various pH*s. These experiments would require fairly low TAME concentrations, and hence large volumes, so that inhibition would be at least 50%; tiypsin and o^M should be mixed and allowed to equilibrate at the final assay volume and pH. However, even at pH 8 the assays done in this way gave poor results. When tiypsin and were mixed in 200 ml volume and the assay started immediately, the rate decreased sharply dur ing the assay due to progressive inhibition, as expected. If the mixture was allowed to stand for 1 0 minutes before adding substrate, initial rates were measurable but when plotted vs. cx^M added, indicated that binding was still incomplete. Longer preincubation periods were tried but, probably due to difficulty in maintaining uniform condi tions, -the results were quite variable. Thus, this method did not prove feasible for measuring trypsin- oc^tl binding as a function of pH, or the kinetics of the binding reac tion. 76 It can be concluded, then, from the pH-stat studies of free tiypsin and the tiypsin- complex, that inhibi tion of tiypsin by is of a "mixed" type, since both Km and kca^ are changed. The combining ratio of trypsin and ^ 2m is independent of type and concentration of sub strate, so substrate does not appear to displace o<2M from the enzyme. Complex formation is relatively slow, requir ing at least several minutes. Due to alteration of the kinetic behavior of tiypsin on binding to appar ent extent of inhibition of trypsin by excess varies with type and concentration of substrate. If substrate activation involves an additional binding site on trypsin, this site is not blocked by because substrate activa tion occurs with the complex. CHAPTER V DISCUSSION When trypsin-binding activity of the M of human plasma is purified about *fO-fold by Cohn fractionation followed by the additional steps described here, the prod uct does not appear to be a homogeneous protein, but the increase in specific activity at successive steps of the procedure is accompanied by an increase in the relative concentration of a component having a sedimentation coeffi cient near 17 S, and a decrease in concentration of a slightly faster-sedimenting component. Both of these com ponents move in the < ^ 2 zone electrophoretically, and the 1 relationship between them is not known. If there is more than one °*2^> this might explain why the oigM isolated by Schnitzel group (7) by a method consisting primarily of ammonium sulfate fractionation has somewhat different physical properties (s20,w s 19*6 S, a[l^,l cm,280 djp. ] = 6.9) from those reported here (8 2 0, w = S, a[l$, 1 cm, 280 n y u . ] 9 8.7) and by Steinbuch and associates (SgQ w ■ 16.9 S) (51) for an oCgM which inhibits proteolytic enzymes. Steinbuch*s is isolated from plasma by a procedure 77 involving Rivanol binding, and it, too, shows a slightly asymmetric major peak in the analytical ultracentrifuge, as well as having inhibitory properties which appear very simi lar to those described here and in a report by Lanchantin (52), Methods aimed at Isolation of an activity (ability to interact with proteolytic enzymes), then, yield a pro tein with a sedimentation coefficient near 17 S. The faster-sedimenting material which is discarded due to its relative or -total lack of this activity is possibly the 19 S ot^M purified by Schultze. However, antiserum to the purified 19 S is reported to interact quantitatively with 2 plasmin inhibitor isolated by Rivanol fractiona tion (18), indicating that the active protein may be immunologically identical with the 19 S OfgM. Further, Ganrot and Laurell (53) have recently found that the °*2 fraction of human plasma from a single individual contains components with a range of slightly differing electropho retic mobilities, all of which show identical specificity toward highly purified antibody to an O^M isolated by an ammonium sulfate method similar to that of Schultze, Thus, the information currently available on human °t2** would seem to indicate that it may be homogeneous immunologically but heterogeneous in electrophoretic and sedimentation characteristics and in binding activity. It is possible that forms of d 2m differing some what in chemical composition (amino acid or carbohydrate _ 79 content, or both) are originally present in plaama, or that partial degradation occurs during purification* Very accurate analyses of the different forms would be required to detect such variations, and such data are not available at present. Ganrot and Laurell (53) have found that treat ment with neuraminidase has no effect on the electropho retic heterogeneity, which is therefore probably not due to variations in acetylneuraminic acid content. Another possibility is that part of the ^ plasma is combined with one oa™ more other proteins* Such complexes might sediment faster than free due to their larger size, and added trypsin, for example, would not be readily bound because the binding site would already be occupied. If ^ 2^ were bound to a molecule with esterase activity, the complex would be expected to retain partial activity. Esterase activity is found in the III-O Fraction, and apparently the protein in which this activity resides has properties quite similar to those of txypsin-binding because its relative concentration increases during the earlier fractionation steps, until it is removed at low ionic strength (Table 2). The low-ionic-strength precipi tate contains a relatively high proportion of the faster- sedimenting component (Fig. 5), which is consistent with the possibility that it is a combined form of How ever, such a complex would be expected, due to its larger size, to " b e somewhat retarded on starch gel electrophoresis at pH 8.6, unless the net negative charge of the complex were greater than that of the free 04 2* * • data show that fractions enriched in the faster-sedimenting component, such as the low-ionic“Strength precipitate, actually have a higher concentration of the faster-moving band on starch gel electrophoresis (Fig. 6). Schultze et al. have reported that saturated with plasmin is faster-moving (18). Farther, C^M mixed with tiypsin or chymotrypsin also moves slightly faster, even though these enzymes have a positive charge at pH 8.6. A third possibility is that the apparent hetero geneity of ot2^ arises from different conformations or states of aggregation of a single protein. These forms might be originally present in plasma, and relatively sta ble under the conditions used for purification, or they might be interconvertible so that a particular isolation procedure would stabilize one form preferentially. If this is the case, until conditions are determined tinder which each form is stable it would be difficult to obtain an entirely homogeneous preparation. For example, treatment of a purified fraction with NaOH appears to increase the relative proportion of faster-sedimenting material, without producing any immediate marked change in binding activity, however (Fig. 10). 81 At present the data do not point clearly to any single explanation of the apparent heterogeneity of so the question must remain open. The appearance of an 11 S peak in the ultracentri fuge pattern, with the 17 S peak still present but dimin ished in size, and no other material present which could be considered "breakdown products" (Pig. 7), indicates that o*2m dissociates into discrete subunits. The 17 S peak probably represents undissociated material, and since only one additional peak appears, the subunits must be identical, at least in sedimentation behavior. Scheraga (5*+) has pointed out that a marked change in sedimentation coeffi cient can occur on conversion of a rigid sphere to a random coil, but in such a case D would also decrease proportion ately. The 11 S component of acid-treated has fact a larger D, so it seems evident that dissociation occurs. An estimate of the relative molecular weights of the material in the two peaks, based on sedimentation and diffusion constants calculated from the same schlieren pattern, indicates that the subunits are about half the size of the undissociated material. Sedimentation coeffi cients for particles of various sizes can also be predicted by assuming that they are rigid impermeable spheres, for which s is proportional to the 2/3 power of the molecular weight. If a sphere of a given molecular weight has 82 s = 17 S, then a sphere of one-half this weight would have s * 10,7 S. This supports the idea that the 17 S protein dissociates into half molecules* It is not surprising to find a subunit structure in a protein of this size* If macroglobulin dissociates on treatment with mercaptoethanol, apparently by reductive cleavage of disulfide bonds (55)* on "^he other hand, appears unaffected by mercaptoethanol in its native state (56, 11), but it can be dissociated by H+, indicating that the 11 S subunits are probably not covalently bound* This dissociation is apparently augmented by acetate. The only literature pertinent to the acetate effect appears to be a recent report of a specific dissociating effect of acetate on hemoglobin (57), and earlier evidence for binding of undissociated carboxylic acids to proteins (58). The 11 S subunits probably contain more than one polypeptide chain* Treatment of 06 M with urea produces, in addition to the original material, a substantial concen tration of a subunit about half its size, and other poorly- defined peaks in the ultracentrifuge pattern as well (7). Reduction of in urea appears to cause further disso ciation (59)» suggesting that disulfide linkages between polypeptide chains are susceptible to reductive cleavage only after partial denaturation. Extension of the preliminary studies reported here 83 could be expected to yield further information about the structure of the trypsin-binding " tlie relation of structure to activity* Results obtained by both assay methods show that at a pH near 8 trypsin associates stoichiometrically with oi2M (Pig. 17), and does not appear to dissociate on dilution or in the presence of a high concentration of substrate. Recent evidence (M-0) that the interaction of trypsin with STI may involve limited proteolysis of the inhibitor raises the question of whether binding of 0* 2* * trypsin is also accompanied by changes in its primary structure. Definite information on this point is lacking, but the relatively slow rate of complex formation would be consistent with such an idea. The ultraviolet difference spectra of tiypsin com bined with various protein inhibitors have been presented by Edelhoch and Steiner (60). These are much smaller pro teins than 01 2M so higher molar concentrations (trypsin concentrations about 6.5 times as high as those used in the present study) can be used, and their spectra show much more detail than was possible with the trypsin- OtgM mixture (Pig. 9 )* Three of the inhibitors used by these workers contain no tryptophan, so differences in the region from 292-300 1191 are probably due to perturbations of the trypto- phyl residues in trypsin. Spande and Witkop (6l) have 8*f i shown that the sensitivity of tryptophan residues in tryp sin to oxidation by N-bromosuccinimide is altered by com bining trypsin with pancreatic trypsin inhibitor. Since cL2M undoubtedly contains tryptophan (Pig, 8), the differ ence spectrum presented here might include perturbations in both and trypsin, but it is interesting that it cor responds so closely with those reported by Edelhoch and Steiner for interactions of trypsin with protein inhibitors containing no tryptophan. This suggests that there may be similarities in the binding of trypsin to ot2*“ I " t * 1 ©8® other protein inhibitors. In contrast with trypsin bound to the other protein inhibitors mentioned, trypsin associated with 0^2^ 8^OWB appreciable catalytic activity in the standard BAPNA assay. In other words, the complex itself is an enzyme and its kinetic properties can be investigated. It was hoped that comparison of the behavior of free trypsin and the complex might reveal information about the changes which occur with binding of trypsin to and give some insight into the nature of this binding. The observation that product, as well as substrate, activates the trypsin-catalyzed hydrolysis of TAME (Pig. 12) confirms the proposal of Trowbridge et al. (30) that activation occurs when a second molecule is bound, and that the dissociation constant of this second molecule is 85 much higher than the Km so activation Is not observed at low substrate concentrations. Product inhibition of some of the synthetic trypsin substrates has been reported (36, 37) 9 so it is not surprising that BA and TA are inhibitory at low substrate concentrations. The fact that their affinity for the hydrolytic site is much lower than that of substrate explains why the net effect on the rate can be activation, at high substrate concentrations, and the high degree of activation of TAME hydrolysis by TA probably accounts for the failure to find any significant inhibition by TA even at very low TAME concentrations, Bechet and Yon (62) have also studied substrate activation of trypsin by TAME, and have proposed that the trypsin molecule contains a catalytic site and an auxiliary, or allosteric ( 6 3 ), site. Interaction between the sites is indirect; a molecule of substrate bound at the latter site acts as an allosteric effector to induce a discrete rever sible change in the structure of the enzyme, so that the catalytic site takes an optimal conformation. Such a picture is inadequate to explain the data summarized in Tables 3 and If, If TA is a good activator or allosteric effector for TAME hydrolysis, it should also activate BAEE hydrolysis; instead, it is inhibitory. In order to fit the results to an "allosteric" model, it is necessary to postulate that each effector alters the 86 catalytic site in a slightly different way, and that the alteration induced by TA, for example, is more favorable for TAME hydrolysis but less favorable for BAEE. Con* versely, the alteration induced by BA is less favorable for TAME hydrolysis than for BAEE. It seems equally possible that the activator bind ing site is not allosteric but is part of or adjacent to the catalytic site. Occupation of both sites would lead to activation, if substrate were present at the catalytic site, but some combinations of substrates and potential activators might not fit together well. For example, presence of TA at the activator site might make the cata lytic site less accessible to BAEE, so the potentially "activated" hydrolysis could not occur readily. The com plexity of a system in which reaction products can both activate and inhibit, and substrate can also activate, makes any attempted explanation rather speculative. Whatever the mechanism of activation, the fact that the rate of hydrolysis by trypsin of these ester substrates can be modified suggests a control mechanism which may also apply to tryptic hydrolysis of proteins. In certain pro teins, some lysyl or arginyl bonds seem to be unusually susceptible to trypsin, while others are quite resistant. (An example is the activation of trypsinogen and chymotryp- slnogen by trypsin). Undoubtedly, some peptide bonds are 87 much more available for hydrolysis than others, due to their location in the protein substrate, but an additional factor might be the presence of amino acid residues which could serve as modifiers at suitable positions in the sub strate molecule. Such Mbuilt-in activators" could function in the same way as excess TAME or TA added to the TAME assay system, but their effect would be confined to bonds having favorable geometric relation to the activator. It is likely that, for trypsin-catalyzed hydrolysis of both BAEE (37) and TAME (31)? the deacylation step is rate-limiting, so that kcat = k^, the rate constant for this step. Since deaciylation is probably catalyzed by an uncharged imidazole in trypsin, pH changes in the region of the imidazole pK would change k^, and the pH-rate profiles would be expected to resemble the titration curve of imidazole. The pE's are 6.95 for free imidazole, and 6.0 for the imidazole side chain of histidine ( 6* 0 ; additional shifts may occur on incorporation into a protein, and on binding of other molecules to the protein in the vicinity of the histidine residue. Such effects might explain the differences in the pH-rate profiles of BAEE and TAME (Fig. 16). Fhrther, as the reaction proceeds by several steps, pH changes would undoubtedly affect the constants describing these steps also. This is seen in the differ ence in of BAEE at pH 6.0 and 8.0 (Fig. l*t). The data 88 on pH effects in the present study axe too limited to per mit definite conclusions to be drawn, but they do point out possible differences in tryptic hydrolysis of TAME and BAEE which may or may not be related to the differences in their response to the modifiers, previously discussed. The kinetic study of the trypsin- complex shows that the reduction in catalytic efficiency of the complex, compared to free trypsin, is due to changes in both and kcat* partly explains why the "per cent inhibition" varies with substrate concentration (Fig. 17)• If the initial rate of hydrolysis by a given amount of free tryp sin is designated v(f), the rate for the same amo^it of trypsin combined with I® ▼(!)> their respective kcat'8 are ^cat^^ ^cat^*^* then according to the equa- kcattS] tion v = [g-J * ' tl ie ratio v(i)/v(f) would be equal to ^cat^^^cat^a constant, for any value of [S], if inhibition were due only to a change in kca^. However, binding to o^M changes also, so that v(i)/v(f) would not be expected to be constant except when [S] »Em. It seems reasonable to assume that binding to alters both the E^ and kcat of trypsin for other substrates, as well, for example the arginyl and lysyl peptide bonds of proteins. It is difficult to define ^ and [S] for protein substrates, but by the above reasoning it can be seen that, regardless of the numerical value of En, the ratio v(i)/v(f) can 89 become very small at sufficiently low [S], Since per cent inhibition * x 1 0 0 , this probably explains why v(f ) the reported inhibition by oigM of trypsin-catalyzed hydrolysis of casein varies from 86% (see "Results") to 95% (65) and 100# (66) for assays that were probably done under somewhat different conditions. Tryptic hydrolysis of protamine, which contains a higher concentration of trypsin- susceptible bonds due to its high arginine content, would be expected to show less inhibition by ©tgM than an equal weight of casein, and this seems to be the case (66). The above discussion demonstrated that degree of inhibition of trypsin by varies with substrate concen tration at low [S] because Km is altered. The observation that substrate activation by TAME and BAKE is relatively greater with the complex than with free trypsin (Pig. 18), indicates an additional reason why degree of inhibition would not be expected to be constant even at high concen trations of these substrates. It was mentioned earlier that o* 2M can interact with other proteolytic enzymes besides trypsin. Samples of ^ 2* * a" ^ successive stages of purification were also tested for thrombin-binding activity by Lanehantin et al. (52). This was found to increase parallel to' trypsin-binding activity, so apparently the same protein binds both enzymes. Further, the combining ratio of and thrombin was about the same as that with trypsin, suggesting that the same binding sites may be involved. The function of 0 < 2M in plasma is not known, but it seems likely that its ability to bind thrombin and plasmin has more physiological sig nificance than binding of trypsin. Nevertheless, the interaction with trypsin provides a convenient method for isolation and investigation of this protein, leading to a better understanding of its properties, which in turn should eventually lead to a clarification of its physiolog ical role. CHAPTER VI SUMMARY The subject of this study was the macroglobulin of human plasma which binds tiypsin and other proteolytic enzymes in such a way that partial enzymatic activity is retained; the residual activity is not inhibited by STI. This trypsin-binding protein has been purified by methods which do not appear to alter its ability to combine with proteolytic enzymes. The starting material was Cohn III-O Fraction, and purification steps included removal of lipo protein by centrifugation in a high-density solvent, frac tional precipitation with PEG, column chromatography on EEAE-cellulose, fractional precipitation at low ionic strength and pH, and adsorption of inactive protein on bentonite. Slight TAME esterase activity was found in the III-O Fraction, apparently associated with a protein whose behavior during successive fractionation steps was almost identical with that of trypsin-binding until it was separated by precipitation at low ionic strength. About four-fold purification of trypsin-binding activity over that of Cohn III-O Fraction was achieved (or about *fO-fold 91 over plasma); the yield of activity was about 50%• The product consisted of a major component with sgo^v * 17*3 S, thought to be the active binding protein, and a small amount of other material which appeared as a slightly faster-sedimenting shoulder on the schlieren pattern of the main peak. On starch gel electrophoresis at pH 8 . 6 , two bands were seen, both in the region; the major compo nent again moved more slowly. The molecular weight of the purified <*2M was estimated from sedimentation velocity data to be between 800,000 and 1,000,000. Its ultraviolet absorption spectrum has A maT at 278 191 and a stoulder at 283 n^i. At 280 191, a(l#,l cm) = 8.7. The trypsin- complex is about 65% as active as free trypsin in a standard assay using the substrate BAPNA. The combining ratio was approximately 3^6,000 grams of <*2^ per mole of trypsin, indicating that there are at least two trypsin-binding sites on the molecule. < 7 * - 2^ was found to dissociate incompletely into sub units of about half the original size at pH values below 5; this effect was augmented by acetate, and subsequent neu tralization induced partial reassociation. Acidification also resulted in partial or total loss of trypsin-binding activity, and this, too, was reversible under some condi tions, but there did not appear to be a direct correlation between loss of activity and extent of dissociation. 93 A pH-stat assay was used to compare the kinetic behavior of free trypsin and the 2**-trypsin complex. Substrates were TAME and BAEE. Rate measurements at pH 8.0 and 25° over a wide range of substrate concentrations showed that hydrolysis of both substrates by free trypsin followed Michaelis-Menten kinetics at concentrations below 0.2 mM (for TAME, = 1 x 10”^ M, kca^ = 62.5 sec"'*’; for BAEE, K* = 3 i 10 ^ M, ^cat * 12.6 sec”^), but substrate activation was evident at higher concentrations. The reac tion product TA was also found to activate TAME hydrolysis, but was inhibitory to BAEE hydrolysis, at high substrate concentrations. BA activated TAME hydrolysis slightly, and had no effect on BAEE, under these conditions. It seems likely that activation is due to formation of ternary ESS (or EPS) complexes that yield product at a faster rate than binary ES complexes (30). The apparently contradictory effect of the two products suggests either that EPS com plexes, and possibly even ES complexes, cannot form as readily with some combinations of substrates and potential activators, or that the net effect of formation of EPS com plexes is not "activation" for all substrate-modifier com binations • At low substrate concentrations the products were inhibitory, with apparent near 1 mM; inhibition appeared to be largely competitive. 9V When trypsin was preincuhated with oi2M» binding appeared to he irreversible under the assay conditions used. The complex was enzymatically active, with a higher Km and lower kca^ than those of free trypsin for both TAKE and BAEE. Substrate activation also occurred with the com plex, so if activation involves an additional binding site on trypsin this site is not blocked by The degree of apparent inhibition of trypsin by excess ot2M varied from over 8 0# with casein substrate ( 1# concentration) to less than 30# with high concentrations of TAME. This is con sistent with a "mixed" type of inhibition (alteration of both and kca- t ) for which the degree of inhibition would be expected to vary with substrate concentration, becoming very large at low concentrations. LIST OP REFERENCES LIST OP REFERENCES 1. Mutzenbecher, P. von, Biochem. Z. 266. 226, 250 (1933). 2. Cooper, G. R., in P. W. Putnam, Editor, The Plasma Proteins. Vol. I. Academic Press, New York, 1 9&0 , p. 5Tr^ 3. Oncley, J. L., Scatchard, G., and Brown, A., J. Phys, ' Coll. Chem. ^1, 18b (19^7). b. Cohn, E. J., Gurd, P. R. N., Surgenor, B. M., Barnes, B. A., Brown, R. K., Berouaux, G., Gillespie, J. M., Eahnt, P. W., Lever, W. P., Liu, C. H., Mittelman, B., Mouton, R. P., Schmid, K., and Uroma, E., J. Am. Chem. Soc. 22, b65 (1950). 5. Brown, R. K., Baker, W. H., Peterkofsky, A., and Kauffman, B. L., J. Am. Chem. Soc. 7 6 , ^+2Mf (195^). 6 . Schultze, H. E., Gollner, I., Heide, K., Schonenberger, M., and Schwick, G., Z. Naturforsch. 10b, 1+63 (1955). 7. Schonenberger, M., Schmldtberger, R«, and Schultze, H.E., Z. Naturforsch. 13b. 761 (1958). "-L~’ 8 . Brattsten, I., Arkiv Kemi 8 , 3^7 (1955). 9. Wallenius, G., Trautman, R., Franklin, E. C., and Kunkel, H. G., Fed. Proc. l£, 378 (1956). 10. Kunkel, H. G., in P. W. Putnam, Editor, The Plasma Proteins. Vol. I. Academic Press. New York. 1 9 6 6" , p. 2?$. 11. Schultze, H. E., Haupt, H., Heide, K., Moschlin, G., Schmldtberger, R., and Schwick, G., Z. Naturforsch. 17b. 313 (1962). 96 12. 13. l*f. 15. 16. 17. 18. 19. 20. 21. 22. 23. 2if. 25. 26. 97 Haverback, B. J., I^yce, B., Bundy, H. P., Wirtschafter, S. K., and Edmondson, H. A., J. Clin. Invest. |+1, 972 (1962). Mehl, J. W., 0*Connell, W., and BeGroot, J., Science 1^5. 821 ( 196^ ) . Oncley, J. L., Melin, M., Richert, B. A., Cameron, J. w., and Gross, P. M., Jr., J. Am. Chem. Soc. Zk> 5^1 (19^9). Jacobsson, X., Scand. J. Clin. Lab. Invest. 5* 97 (1953 5. Jacobsson, K., Scand. J. Clin. Lab. Invest. 7, Suppl. li, 5h (1955). Steines, W. J., and Mehl, J. W., J. Lab. Clin. Med., in press. Schultze, H. E., Heimburger, N., Heide, £., Haupt, H., Storiko, K., and Schwick, H. G., Proceedln^a Ninth Congress of European Society of Haematology. Lisbon. 1963. 5. KargerT Basel/ tfew York, 1963» p. 1315. Bergmann, M., and Pruton, J. S., Ad van. Enzymol. 1, 63 (19^1). Bergmann, M., Pruton, J. S., and Pollok, H., J. Biol. Chem. 122, 6>3 (1939). Schwert, G. W., Neurath, H., Kaufman, S., and Snoke, J. E., J. Biol. Chem. 122, 221 (19^8). Balls, A. K., and Jansen, E. P., Advan. Enzymol. 13. 321 (1952). Bender, M. L., and Kdzdy, P. J., Ann. Rev. Biochem. St, 1*9 (1965). Gutfreund, H., and Sturtevant, J. M., Biochem, J. 63. 656 (1956). Green, N. M., and Neurath, H., in H. Neurath and K. Bailey, Editors, The Proteins. Vol. IIB. Academic Press, Inc., New York, 195*+, p. 1057. Bernhard, S. A., Biochem. J. ^2, 506 (1955). 27. 28. 29. 30. 31. 32. 33. 3*+. 35. 36. 37. 38. 39. *t0. bl. b2. *♦3. bb. b5. 98 Zemer, B., and Bender, M. L., J. Am. Chem. Soc., 86, 3669 (196*0. Sturtevant, J. M., Brook haven Symp. Biol. 13, l5l (I960). Erl anger, B. P., Kokowsky, N., and Cohen, W., Arch. Biochem. Biophys. 25* 271 (1961). Trowbridge, C. G., Krehbiel, A*, and Laekowskl, M.,Jr., Biochemistry 2, 8*+3 (19o3). Baines, N. J., Baird, J. B., and Elmore, B. T., Biochem. J. 22, *t70 (196*0. Bechet, J.-J., J. Chim. Phys., 58*+ (196*0. Scheraga, H» A., Ehrenpreis, S., and Sullivan, E., Nature 182, *+6l (1958). Martin, C. J., Golubow, J., and Axelrod, A. E., J. Biol. Chem. 1718 (1959). Ronwin, E., Biochim. Biophys. Acta 33, j 326 (1959). Harmon, K. M., and Niemann, C., J. Biol. Chem. 178. 7*+3 (19^9). Schwert, G. W., and Eisenberg, M. A., J. Biol. Chem. 179. 665 (19*+9). Inagami, T., J. Biol. Chem. 239. 787 (196*0. Laskowski, M., and Laskowski. M., Jr., Advan. Protein Chemistry 2* 203 (195*0. Pinkenstadt, W. R., and Laskowski, M., Jr., J. Biol. Chem. 2|f0, PC9 6 2 (1965). Mehl, J. W., Park, M. Y., and O'Connell, W., Proc. Soc. Exp. Biol. Med., in press. Worthington Bescriptive Manual No. 11, 1961. Kunitz, M., J. Gen. Physiol. 30, 291 (19^7). Mehl, J. W., Anal. Biochem. 170 (1963). Poulik, M. B., Nature 180, l*+77 (1957). 99 *f6. Clarke, J, T., Ann. N. Y. Acad. Sci. 121. *+28 (1961 *). *+7. Bundy, H. P., and Mehl, J. W., J. Clin. Invest. 37, 9^7 (1958). *+8. Poison, A., Potgieter, G. M. Largier, J. P., Mears, G. E. P., and Joubert, P. J., Biochim. Biophys. Acta 82, *f63 (196*0. - ^9. Howard, S. M., and Mehl, J. W., Biochim. Biophys. Acta 105, 59** (1965). 50. Dixon, M., Biochem. J. 170 (1953). 51. Steinbuch, M., personal communication to Dr. J. W. Mehl, 1965. 52. Lanchantin, G. P., Plesset, M. L., Friedmann, J. A., and Hart, D. W., Proc. Soc. Exp. Biol. Med., in press. 53. Ganrot, P. 0., and Laurel1, C. 3 . , Clin. Chim. Acta, in press. 5*+. Scheraga, H. A., Protein Structure. Academic Press, New York, 1961, pp. 20-22. 55. Deutsch, H. P., and Morton, J. I., Science 125. 600 (1957). 56. Hitzig, W. H., and Isliker, H. C,, in H. Peeters, Editor, Protides of the Biological Fluids. Proceedings of the Seventh Colloauium. truces. Elegvier. Ameter^ip. 1%6.V ^8. 57. Chiancone. E., and Gilbert, G. A., J. Biol. Chem. 2*f0. 3866 (1965). 58. Cann, J. R., and Goad, W. B., J. Biol. Chem. 2*»0. l*+8 (19o5). 59. Poulik, M. D., Biochim. Biophys. Acta M+, 390 (i9 6 0 ). 60. Edelhoch, H., and Steiner, R. P., J. Biol. Chem. 2*t0, 2877 (1965). 61. Spande, T. P., and Witkop, B.. Biochem. Biophys• Res, Commun. 21, 131 (1965). 62. Bechet, J.-J., and Yon, J., Biochim. Biophys. Acta 89. 117 (196*0. 100 6 3 . Monod, J., Changeux, J.-P.. and Jacob, P., J. Mol. Biol. 6 , 306 (19635. 6*+. Edsall. J. T., and Wyman. J.. Bionhvsical Chemistry. Vol. I. Academic PressTTTSw York, 1958, p. 65. Ganrot, P. 0., Clin. Chim. Acta 1^, 5l8 (1966). 6 6 . Khodorova, E. L., Veremeyenko, K. N., and Antonian, A. A., Ukr. Biokhim. Zh. 3 6 , 6^3 This dissertation has been microfilmed exactly as received 66 —1 0 ,5 4 3 HOW ARD, SaH y M a y field , 1 9 2 8 - ST U D IE S O F T R Y PSIN -B IN D IN G OC^ M ACROGLOBULIN O F HUMAN PL A SM A . U n iv e r s ity o f S ou th ern C a lifo r n ia , P h .D ., 1966 B io c h e m is tr y University Microfilms, Inc., Ann Arbor, Michigan

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Purification and identification of trypsin inhibitors in human serum

PDF

Effects of sex hormones on cholesterol metabolism

PDF

Studies on acyl transfer reactions on n-acylimidazoles

PDF

Metabolism of 5-hydroxyuracil compounds

PDF

Studies on the transport, metabolism and chemistry of iron-sugar chelates

PDF

On the mechanism of ACTH stimulation of adrenal steroidogenesis

PDF

Galactose metabolism in human blood cells

PDF

Studies of steroid 11β-hydroxylation and cholesterol side chain cleavage in adrenal cortex mitochondria

PDF

Studies of human ɑ₂ macroglobin: physical and enzyme binding properties

PDF

The separation of complete and incomplete blood group antibodies

PDF

Steric effects in imidazole and ɑ-chymotrypsin catalyzed ester hydrolysis

PDF

Microheterogeneity of fetuin

PDF

Interaction between 11β-hydroxylation and respiration In bovine adrenocortical mitochondria

PDF

Kinetic investigation of the hydrolysis of aryl β-D-glucopyranosiduronic acids by β-glucuronidase

PDF

Chemical, physicochemical, and biological studies on the mucoproteins of plasma and serum

PDF

Pyrimidine Nucleoside Metabolism And Antimetabolites

PDF

Metabolism of mammalian deoxyribonucleic acid and nuclear ribonucleic acid

PDF

Studies On The Mechanism Of Action Of 5-Hydroxyuridine

PDF

Studies On The Metabolism Of Iron

PDF

Studies On The Enzymatic Formation Of Amino Acid - Ribonucleic Acid Compounds

Asset Metadata

Creator Howard, Sally Mayfield, 1928- (author)

Core Title Studies of trypsin-binding ɑ₂ macroglobulin of human plasma

School Graduate School

Degree Doctor of Philosophy

Degree Program Biochemistry

Degree Conferral Date 1966-06

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry, biochemistry,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Mehl, John W. (committee chair), Brown, Ronald F. (committee member), Visser, Donald W. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-213509

Unique identifier UC11360170

Identifier 6610543.pdf (filename),usctheses-c18-213509 (legacy record id)

Legacy Identifier 6610543.pdf

Dmrecord 213509

Document Type Dissertation

Rights Howard, Sally Mayfield

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA