A PHYSICAL ANALYSIS or THE EFFECTS or TEMPERATURE SUBSTRATES AND COFACTORS ON THE STRUCTURAL AND CATALYTIC PROPERTIES or RABBIT MUSCLE PYRUVATE KINASE ' Thesis for The Degree of Ph. D. MICHIGAN STATE UNIVERSITY V Fredrick .I. Kayne‘ 7 V ' 1966 , ll“llllllll‘lllllllllllfllllll ‘ “BRA n v 3 1 93 01107 4857 Michigan Sm: Rivers“,- This is to certifg that the thesis entitled A Physical Analysis of the Effects of Temperature Substrates and Cofactors on the Structural and Catalytic Properties of Rabbit Muscle Pyruvate Kinase presented by Fredrick J; Kayne has been accepted towards fulfillment of the requirements for Doctor of Philosophy daymghlBiochemistry WMLMfi Major professor [hm November 18, 1966 0-169 MA.” 0 9 7999 6T s “New? “all 0028 (2931 ABSTRACT A PHYSICAL ANALYSIS OF THE EFFECTS OF TEMPERATURE SUBSTRATES AND COFACTORS ON THE STRUCTURAL AND CATALYTIC PROPERTIES OF RABBIT MUSCLE PYRUVATE KINASE By Fredrick J, Kayne Rabbit muscle pyruvate kinase (ELC. 2.7.1.40) has been shown to undergo protein conformational changes under the following situations: (1) Upon the addition of the catalyt— ically required monovalent or divalent cations, (2) Upon the addition of the substrates phospho(enol)pyruvic acid or pyruvic acid, and (3) Upon changing the temperature of a solution of the enzymeo These conformational changes are observed by the production of ultraviolet difference spectra (characteristic of the solvent perturbation of tryptophan and tyrosine), changes in optical rotatory dispersion parameters and the sedimentation velocity of the enzyme, However, solvent perturbation studies and reaction of the protein tryptophans with N—bromosuccinimide or 2-OH—5— -N02-benzyl bromide have failed to show differences in the degree of exposure of this amino acid residue to the external environment. The temperature-dependent conformational change studied under varying solution conditions, follows the behavior expected for an equilibrium between two forms of the enzyme. Fredrick J. Kayne — 2 Decreasing pH, increasing ionic strength, 2H20 substitution for H20, and binding of activating cations favor the presence of the low temperature form of the enzyme. The high temperature form is favored by the addition of protein structure-disrupting agents. The results suggest that the low temperature form of the enzyme is more compact than that at the higher temperatures. The catalytic activity of the enzyme exhibits an anomalous Arrhenius plot with a curvature from a high energy of activation at low temperatures to a low energy of activation at high temperatures. The temperature about which this curvature occurs depends on the solution pH, 2H20 substit- ution and ionic strength. The changes found in these cases parallel those observed for the temperature-dependent conformational change. A relationship between both is suggested and this is incorporated into a model used for the explaination of the observed changes. The effects of solvent environment on protein structure and activity are discussed. A PHYSICAL ANALYSIS OF THE EFFECTS OF TEMPERATURE SUBSTRATES AND COFACTORS ON THE STRUCTURAL AND CATALYTIC PROPERTIES OF RABBIT MUSCLE PYRUVATE KINASE BY 3 Fredrick J9 Kayne A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1966 fl RES/<25 1// 7/9 7 s ‘3‘" a ,. \ AC KNOWLE DGEMENTS The author wishes to express his sincere appreciation for the guidance and assistance from Dr. C.H. Suelter throughout the course of his graduate studies. Teaching and research assistantships from the Department of Biochemistry during the course of this work are gratefully acknowledged. l w i ii .1 .1 .1 To my Parents TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . LITERATURE REVIEW . . , . . . . . . MATERIALS AND METHODS . . . . . . . . . . . . Enzymes . . . . . . . . . . . .. . . Chemicals . . . . . . . . . . . . . Instrumental . . . . . . . . . . Experimental . . . . . . . . . . . Enzyme Assays . . Difference Spectra . . . . . . Temperature-Perturbed Spectra Optical Rotatory Dispersion Measurements RESULTS . . . . . . . . . Difference Spectra. . . Solvent Perturba.tion of Model Compounds Tryptophan— Tyrosine Determination . . . Cation Interaction with Pyruvate Kinase Substrate Interaction with Pyruvate Kinase . Temperature— Dependent Changes Arrhenius Plots Pyruvate Kinase Fumarase . Sedimentation of Pyruvate Kinase . . Optical Rotatory Dispersion Measurements . Tryptophan Exposure . . . Solvent Perturbation of the Protein . Dena tura tion . . . . . . Reaction with N- Bromosuccinimide . . Reaction with 2— OH— 5— N02— Benzyl Bromide DISCUSSION . . . . . . . . . . . . . . . . . SUMMARY . . . . . . . . o . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . iv Page . 105 LIST OF FIGURES Figure 10. 11. i2. 13. 11+. 15. 16. 17. Spectrophotometer cell thermostatting system Solvent—perturbed ultraviolet difference spectra . . . . . . . . . . . Cation-perturbed protein protein difference spectra Substrate—perturbed protein difference spectra Temperature-perturbed protein difference spectrum . . . . . . . . . . . . . . . . Temperature—dependent difference absorbance of pyruvate kinase . . . Van't Hoff plot for pyruvate kinase . . . . . Temperature—dependent difference absorbance of pyruvate kinase in 2H20 solution . . Van't Hoff plot for pyruvate kinase in 2H20 Transition temperatures for non—activated pyruvate kinase under varying conditions Arrhenius plot for pyruvate kinase Arrhenius plot for pyruvate kinase in 2H20 Deuterium isotope effect on the pyruvate kinase reaction . . . . . . . . . Arrhenius plot for fumarase in 2H20 Sedimentation velocity of pyruvate kinase in the presence and absence of activating cations . . . . . , . . . . Moffitt plots for pyruvate kinase Hypothetical scheme illustrating observed conformational changes in pyruvate kinase Page 20 7 / 59 62 64 67 70 71+ 91+ Table II. III. IV. VI. LIST OF TABLES Dissociation constants for the interaction of substrates with pyruvate kinase in the presence of various cations . . Additional transition temperature data for pyruvate kinase . . . . . . Pyruvate kinase transition temperatures in the presence of various cations . . Arrhenius plot characteristics for pyruvate kinase . . . . . . Moffitt parameters for pyruvate kinase solutions . Relationship of Arrhenius plot "breaks” to conformational transition . . vi Page 55 56 65 75 ADP ATP EDTA NBS nmr ORD PEP ¢'—Br g: S20,w TMA+ TMACl Tris LIST OF ABBREVIATIONS Adenosine—5'—diphosphate Adenosine—5'-triphosphate Ethylenediaminetetraacetic acid N-Bromosuccinimide Nuclear magnetic resonance Optical rotatory dispersion Phospho(enol)pyruvic acid 2—hydroxy—5—nitro—benzy1 bromide 2-hydroxy-5—nitro—benzyl group Sedimentation coefficient adjusted to 200 and water Tetramethylammonium cation Tetramethylammonium chloride Tris(hydroxymethyl)amino methane Ultraviolet vii INTRODUCTION Rabbit muscle pyruvate kinase (E.C. 2.7.1.40) catalyzes the phosphoryl transfer from phospho(enol)pyruvic acid (PEP) to adenosine—5l—diphosphate (ADP) in the glycolytic pathway with the formation of adenosine-5'—triphosphate (ATP) and pyruvate. It was the first enzyme for which a definite cofactor role for potassium was established (Boyer §£_§1., 1945) and is among a number of enzymes which require both a monovalent and a divalent cation for catalytic activity. Research in the field of monovalent cation requirements has been covered in a recent review by Evans and Sorger (1966). The enzyme was first crystallized from rat muscle by Negflein.(see Bficher and Pfleiderer, 1955) and later from human muscle (Kubowitz and Ott, 1944), and rabbit muscle (BHcher and Pfleiderer, 1955). More recent preparations of the rabbit muscle enzyme have been given by Tietz and Ochoa (1958) and Kupiecki and Coon (1960). These were derived from their studies on "fluorokinase” and "hydroxyl— amine kinase” respectively, both of which were demonstrated to be activities possessed by pyruvate kinase. Other preparations of the enzyme and physical and catalytic properties of the rabbit muscle enzyme in particular are well covered in the review by Boyer (1962). Since that review, the enzyme has also been extensively purified from t a1., 1966; Hunsley, 1966), however, this yeast (Hess discussion will chiefly cover work done with the rabbit muscle enzyme. Siebert §£_al,(1965) have studied the enzyme from a number of mammalian tissues, especially with regards to its occurrance with, and relationship to the enzyme alkaline phosphatase. Studies of nucleotide binding kinetics have been presented by Melchoir (1965) and Plowman and Krall (1965), both working with the rabbit muscle enzyme. The subunit structure, dissociation intermediates and amino acid composition of this enzyme have just been reported (Steinmetz and Deal, 1966; Steinmetz, 1966). Physical and kinetic studies of the cation and substrate binding to the enzyme have been described in a number of recent papers (Cohn, 1963; Mildvan and Cohn, 1965; Suelter §£_al., 1966; Mildvan and Cohn, 1966). The latter authors have suggested a scheme for the conversion of the substrate complex (enzyme—Mn2+—ADP-PEP) to the product complex (enzyme-Mn2+-ATP—pyruvate) which envokes a bridge structure for the divalent cation between both substrates and the enzyme. Their studies have utilized nuclear magnetic resonance (nmr) techniques (proton relaxation rate and electron paramagnetic resonance) and these have indicated that a conformational change in the protein occurs upon interaction with the divalent cations. This confirmed earlier work (Suelter and Melander, 1963) which showed a prc¢ein ultraviolet difference spectrum was produced upon kfiJnding either the monovalent or divalent cations. A conformational change in the enzyme upon binding the mono— valent cations was also later suggested by the work of Sorger \N ._£.§l. (1965) using a completely different method, that of immunoelectrophoresis. The work reported in this thesis is a development of initial studies on the mechanism of the monovalent cation activation of the enzyme. The ultraviolet difference spectra first found by Singleton (1963) led to these studies. It is hoped that these will shed some light on the effects environ— mental changes can have on protein structure, and the possibility of subtle conformational changes effecting the catalytic activity of enzymes. Preliminary reports of this work have been presented (Kayne and Suelter, 1965; Suelter t 11., 1966; Kayne, 1966). LITERATURE REVIEW Conformational changes in proteins can be detected and followed in a number of different ways. This review will be an attempt to present examples of as many of these as possible which have been documented in the literature. The different physical methods will be emphasized. Those observations suggesting conformational changes entirely on the basis of the kinetic behavior of the enzyme or its activity changes as a function of various parameters will not be discussed. This of course, eliminates much of the published work on confor— mational changes in enzymes. This is especially true for the case of the allosteric enzymes where a considerable amount of work has been done based on the models and suggestions of Monod §£_al, (1965). To further limit the scope of this review, we will consider only those conformational changes which are not the result of association or dissociation phenomena of the protein molecules. Some changes in the relative spatial orientation of amino acid residues must be documented or inferred. It seems appropriate at first to mention some recent review articles covering the area of protein and polypeptide conformation in general. These are by Schellman and Schellman (1964) and Harrington §£_§l.(1966). Hydrogen exchange as a method of determination of protein conformation is discussed in another recent review by Hvidt and Nielsen (1966). The reversible denaturation of ribonuclease (Hermans and Scheraga, 1961) was one of the first systems studied and has been subsequently well documented. The initial observations were of ultraviolet (UV) spectral changes and changes in optical rotation. Green (1963) demonstrated tryptophan difference spectra upon binding biotin to avidin and showed a concurrent change in exposure of tryptophan to oxidation by N—bromosuccinimide. Binding of activating cations to pyruvate kinase caused a similar tryptophan difference (Suelter and Melander, 1963). Metal ion binding was shown to effect the conformation of carbonic anhydrase (Coleman, 1965). UV and optical rotatory dispersion (ORD) changes with pH were detected. Protein UV difference spectra caused by substrate binding have been detected by a number of workers. Yankeelov and Koshland (1965) have confirmed this with phosphogluco— mutase by showing corresponding changes in fluorescence, Sgoyw t al. (1964) and rate of chemical inactivation. Hayashi have shown the tryptophan differences in muramidase are caused by varying exposures of that residue to the solvent. Changes in the H—meromyosin system upon binding ATP have been discussed by Iyengar §£_§l. (1964) and Morita and Yagi (1966). Indirect evidence for substrate changes which can be affected by the allosteric modifiers of phosphorylase is given by Bresler g3 g1.(1966). UV differences were also used to observe the kinetics of unfolding of chymotrypsin and some derivatives in urea (Martin and Bhatnagar, 1966). Chymotrypsin and trypsin were also observed to show UV differences upon binding substrates (Benmouyal and Trowbridge, 1966). In earlier papers, Bigelow and Krenitsky (1964) observed UV differences during the acid titration of a number of proteins as did Inada _£ _1. (1964) for pepsin. Recently, Wellner (1966) has shown UV differences between the active and inactive forms of L—amino acid oxidase and correlated these with ORD differences. No differences were observed in reaction with antibody, $20,w or electro— phoretic mobilities. Conformational equilibria in yeast enolase have been studied by a number of workers (Westhead, 1964; Rosenberg and Lumry, 1964; Hanlon and Hesthead, 1965), finding ORD and sedimentation changes in addition to UV differences. Brewer and Weber (1966) have shown 520,w’ fluorescence and polarization of fluorescence changes in this enzyme upon binding Mg2+. As seen in the preceding papers, optical rotation is also a frequently used tool for following conformational changes. Havsteen and Hess (1962) first used this to show the changes produced upon the formation of acetyl chymotrypsin and the thermal stability of this modified protein. Fasella and Hammes (1964) used ORD in conjunction with pyridoxal phosphate spectral shifts for a study of the urea denaturation of aspartate aminotransferase. ORD changes verified hydrogen exchange studies on pH—dependent changes in ribonuclease by Leach and Hill (1963) and in bovine serum albumin by Benson _£_§1. (1964). Conformational changes seen in this manner upon binding heme to apomyoglobin were reported by Breslow (1964) and Harrison and Blout (1965). The first author confirmed these observations with sedimentation changes. The pH—dependent denaturation of muramidase was measured by changes in rotation at a single wavelength and viscosity changes (Sophianopoulos and Weiss, 1964). ORD changes induced by coenzymes were observed in glut— amate dehydrogenase by Magar (1965) and glyceraldehyde-3— —phosphate dehydrogenase by Listowsky £3 a1. (1965). Havsteen (1965) observed these in the latter enzyme with pH changes, solvent changes and substrate binding. Myers and Edsall (1965) have shown protein UV Cotton effects in carbonic anhydrase to disappear when the enzyme structure is disrupted. Torchinsky _3 a1. (1965) demonstrated that small Cotton effects in muscle phosphorylase caused by binding coenzyme can be detected by circular dichroism measurements. Hedrick (1966) has also shown ORD changes in this enzyme upon activation or interaction with AMP. ORD has also been used to show conformational changes in D—amino acid oxidase t g” 1966). following the binding of FAD or FADH2 (Aki Another important optical tool is that of emission spectroscopy. This includes fluorescence, polarization of fluorescence and phosphorescence. Steiner (1964) used differences in fluorescence and polarization of fluorescence to show structural transitions in the denaturation of muramidase. Corresponding difference spectra and optical rotation changes were shown. Churchich (1966) also demon— strated fluorescence changes in this enzyme upon diSulfide bond reduction and reoxidation and used phosphorescence decay times to verify these. Fluorescence changes during the denaturation of ribonuclease have been extensively studied by Cowgill (1964, 1965). In an earlier paper, Lovrien (1963) reported on the interaction of dodecylsulfate anions with bovine serum albumin as detected by changes in fluorescence, viscosity, UV absorption and potentiometric titration curves. One of the few studies using phosphorescence decay times is that on the denaturation of bovine serum albumin, fii-lacto— globulin and peroxidase by Stauff and Wolf (1964). Frattali ‘_£.al. (1965) have used fluorescence polarization, fluorescence, viscosity and solvent perturbation studies to show pH—depend- ent structural transitions in pepsinogen. Conformational changes associated with the regulation of glutamate dehydro— genase activity have been reported by Bayley and Radda (1966). They observed fluorescence and polarization of fluorescence changes in NADH while bound to the enzyme and additional changes upon binding GTP to the enzyme. These were confirmed by ORD studies. Massey and Curti (1966) have shown protein fluorescence changes upon binding FAD to D-amino acid oxidase are related to changes in absorption in the FAD spectrum. The availability of protein amino acid side chain residues to reaction with various reagents is another means of following conformational changes. The sulfhydryl-specific dithio(bis)nitrobenzoic acid has been used to show changes brought about by urea in citrate condensing enzyme (Srere, 1965) and by salts at high concen— trations in a number of enzymes (Warren and Cheatum, 1966). Amelunxen §£_a1, (1966) have shown changes in the suscepti- bility of protein sulfhydryls to air oxidation in glycer— aldehyde—3—phosphate dehydrogenase with substrate or cofactor interactions. According to Steiner (1966), when trypsin combines with its inhibitors (protein), its exposed trypto- phans are shielded from oxidation with N-bromosuccinimide and tyrosines from iodination with 1;. Chymotrypsin, its zymogen and various deratives have also been shown to exhibit differences in the availability of tryptophan residues to N-bromosuccinimide oxidation (Spande §t_al,, 1966). Changes in chymopapain during activation by CN_ were shown by differences in reactivity with diisopropylfluorophosphate by Ebata §£.il- (1966). These authors also showed UV differ- ence spectra and optical rotation changes, along with differ- ences in susceptibility to urea denaturation. Habeeb (1966) has used the accessibility of the disulfide group of proteins as an indication of changes in bovine serum albumin and Q—lactoglobulin. For the study, he employed the reagents 2-mercaptoethanol, sodium sulfite and peracetic acid. Finally, Lui and Cunningham (1966) have shown substrate protection of creatine phosphokinase from inactivation by iodoacetamide or pfnitrophenyl acetate. In this case, no changes were observed in viscosity or Sgo’w and little change 10 in the rate of peptide deuterium exchange as measured by infrared spectroscopy. Other optical methods have been used to detect conformat— ional changes in proteins. Among these, Mercouroff and Hess (1963) showed by solvent perturbation techniques, that difference spectra in a chymotrypsin—substrate complex were caused by "burying" the substrate chromophore. Solvent perturbation and tyrosine reactivity with I; were used in investigations on glyceraldehyde—3-phosphate dehydrogenase and lactic dehydrogenase (Libor e: al., 1965). An investi— gation of the tryptophan exposure in metmyoglobin as a function of pH was carried out by Williams (1966) using solvent perturbation and N—bromosuccinimide oxidation. A general discussion of the use of so—called "reporter groups" in protein conformation studies is given by Burr and Koshland (1964). ChromOphore binding studies have been used to a limited extent. Ullmann e: 11. (1964) demonstrated that AMP increased the affinity of muscle phosphorylase b for the dye bromthymol blue. Proflavin was shown to bind to trypsin and chymo— trypsin (Glazer, 1965) with spectral shifts in the dye, but did not bind to the respective zymogens. Bernhard §§_§1, (1966) showed this dye was not bound to chymotrypsin upon acylation of the enzyme. Swinehart and Hess (1965) have described a method to detect differences in the number of exposed tryptophans in native and denatured chymotrypsin. This method uses the visible absorption spectrum to determine 11 the amount of riboflavin—tryptophan complex formed. As examples of some of the less frequently used tech— niques for detecting conformational changes, Sorger §£_al. (1965) have shown different immunoelectrophoretic behavior between pyruvate kinase in the presence and absence of activat— ing monovalent cations. Differences between native and aged phosphorylase and enzyme with AMP and glycogen were shown by Michaelides and Helmreich (1966) with the use of antibody fragment inhibition. Changes in the nmr spectrum due to aromatic protons were seen in the ribonuclease transition by Mandel (1964). The use of another nmr technique, that of proton relaxation rate, was discussed by Cohn (1963). Conformational changes have been observed more directly in at least two cases. Slayter (1965) has observed changes in shape of shadowed electron micrographs of bovine serum albumin caused by lowering the pH. Changes produced by binding inhibitors to alcohol dehydrogenase were seen by symmetry changes in x—ray crystallography (Branden, 1965). Differential thermal analysis has been used to show thermal transitions in creatine phosphokinase (Watts, 1966) and ribonuclease (Beck §§_al,, 1965). Kresheck and Scheraga (1966), using an adiabatic solution calorimeter, have confirmed their original values for the 45H of the ribonuclease transition which were based on optical measurements. Naguchi __§__l. (1964) used dilatometers to show that obvious large changes in volume did not occur during the superprecipitation of myosin upon adding ATP. Differing half—lives for the exchange of radioactive zinc from carboxypeptidase were 12 observed when inhibitor complexed with the metalloenzyme (Coleman and Vallee, 1964). The reversible interactions between bovine serum albumin or B —lactoglobulin and hydro— carbons were detected by pH changes upon binding (Wetlaufer and Lovrien, 1964). These observations were confirmed by changes in viscosity, optical rotation and UV absorption. Finally, Massey §t_al, (1966) have shown that a break in the Arrhenius plot for D-amino acid oxidase is related to a conformational change by the demonstration of changes in Sao,w: fluorescence (of enzyme and coenzyme), UV and visible absorption around the temperature at which the break occurs. The references given here cover the most recent attempts to demonstrate protein conformational changes by physical methods. The list is not exhaustive and undoubtedly, other examples do exist. There are of course, many more examples of the use of chemical methods to detect changes, only a few of which were covered here. This is especially true for the case of studies showing protection against loss of enzymatic activity by various protein reagents. It seems opportune at this time to make some comments on these various methods. The observation of protein conformational changes should ideally be made in such a manner that it does not give rise to a perturbation of the system. The least desireable methods from this standpoint are those involving chemical reactivity of amino acid side chains. A critical analysis of these would require consider- able discussion, however, one ambiguity is readily apparent. That is the question, did the initial reaction make additional conformational changes in the protein molecule possible, which then resulted in a further reaction with the reagent? Careful study of the individual situation may reduce the possibility of this, but it probably can not be eliminated from consid- eration. This author believes that the use of the solvent perturbation technique of Herskovits and Laskowski (1962) results in a similar ambiguity. Other methods which may perturb the system include the preparation of the sample for electron micrographs or x—ray crystallography. But, in these cases, changes observed will have considerable validity since the preparation method will probably not vary for the samples under differing conditions. The various thermal analysis and pH titration methods are, of course, limited to observing effects on conformational changes brought about by changes in temperature or pH. These methods are probably best applied to the relatively large changes, gg.denaturation. The various hydrodynamic measure- ments do not obviously perturb the protein system but seem to lack the sensitivity needed to detect the smaller changes. The optical methods, absorption, emission and rotation, would seem to be the most desirable. They do not involve a perturbation of the system, and, at this stage of development, seem to be the most sensitive. However, these methods do present some difficulties. Since they measure average properties of the system, differences can thus be cancelled 14 out by opposite changes occurring simultaneously. Comparison of effects observed in proteins with those in model systems is difficult. And finally, since only a small portion of the system is usually undergoing a change, it is difficult, if not impossible, to identify the particular member contributing to the observation. This thesis will demonstrate the use of some of these methods in the study of a series of conformational changes. MATERIALS AND METHODS Enzymes The pyruvate kinase used in these studies was isolated from frozen rabbit muscle (Pel—Freeze Biologicals, Rogers, Ark.) by a modification of the method of Tietz and Ochoa (1958). In the first step, the cut—up muscle was homogenized in a large Waring blendor for 30 sec. This mixture was centrifuged in the large head of a Servall RC—2 refrigerated centrifuge prior to the second step of the Tietz and Ochoa "Methods". Protein concentrations were determined using an extinction coefficient of 0.54 (mg/ml)-1cm-1 (€LMI=1.28 x 105 cm-l) based on the data of Bficher and Pfleiderer (1955) and on the molecular weight of 237,000 (Warner, 1958). The value for the extinction coefficient was checked by drying a salt-free sample of the protein to constant weight. Dialysis tubing (Union Carbide Corp.) was boiled for 15 min in each of two changes of 0.001 M EDTA (ethylenediaminetetraacetic acid) before use. After the (NH4)ZSO4 precipitation in Step 6 of Tietz and Ochoa, an additional fractionation of the enzyme between 40 and 55% saturation (calculated at 200, run at 00) with (NH4)2SO4 was used instead of the crystallization. The enzyme prepared in this manner has a specific activity of 160 - 200 pmoles substrate/min/mg protein, is homogenous in the ultracentrifuge and shows only a very small percentage 15 l6 (<:2%) of a faster moving component on polyacrylamide disc electrophoresis. The enzyme was stored in the refrigerator in 0-02.M imidazole, pH 7.0 at a concentration of 50 - 90 mg/ml. Prior to use, a 0.5 - 2.0 ml sample of this solution was desalted by elution from a 1.7 x 17 cm column of Sephadex G-25 (Medium Grade, Pharmacia) at 00. The Sephadex was first equilibrated with 0.005 M Tris—HCl (tris(hydroxymethyl)amino— methane), pH 8.6 and this buffer was used for the elution. Fumarase (L—malate hydro—lyase, E.C. 4.2.1.2) was obtained as a crystalline suspension in 2.0 M (NH4)ZSO4 from Calbiochem. Chemicals Common inorganic and organic reagents were reagent or analytical reagent grades obtained from commercial sources. Distilled water which was passed through a bed of mixed ion exchange resins (Crystalab "Deeminite") was used in all procedures. "Special Enzyme Grade (NH4)2SO4 (Mann Research Laboratories) was used in the final steps of enzyme purifi— cation. Cacodylic acid (dimethylarsinic acid) and Tris ("Trizma Base") were obtained from Sigma. EDTA (disodium salt) and imidazole were obtained from both Eastman and Sigma. The imidazole was recrystallized when necessary from a mixture of chloroform and petroleum ether. Urea was recrystal- lized twice from 95% ethanol and fresh solutions were made prior to use. TMACl (tetramethyl ammonium chloride) was a 1'? product of Eastman. L-tryptophan, L—tyrosine, L—phenyl— alanine, N—acetyl-L-tryptophan, N—acetyl-L-tyrosine and fumaric acid were obtained from Calbiochem and used with- out further purification. N-acetyl-L-tryptophan amide and N—acetyl—L-tyrosine amide were products of Yeda Research and Development Co. and obtained through New England Nuclear Corp.. N—bromosuccinimide (NBS) was obtained from Matheson Coleman and Bell and recrystallized from benzene. Pyruvic acid was obtained from the same source and redistilled before use. Sodium dodecylsulfate (SDS) was also a product of Matheson Coleman and Bell and Triton X-100 was a product of Rohm and Haas. 2-hydroxy—5—nitro benzyl bromide (Ci-Br) was obtained from Sigma as were ordinarily the following biochemicals: PEP (tricyclohexylammonium salt), ADP (sodium salt) and ATP (sodium salt). These were the best grades obtainable and were used without further purification. Deuterium oxide was obtained from a number of commercial sources and in all cases was greater than 99.6 mole% 2H20. Other solvents included absolute ethanol from Commercial Solvents Corp., pfdioxane redistilled over CaH2 and dimethyl- sulfoxide which was shaken with NaHCOs and redistilled 1Q vacuo. Instrumental A Cary Model 15 spectrophotometer kindly made available by Dr. W.A. Wood was the instrument used for most of the l8 spectral studies. It was equipped with holders for convent— ional cells and tandem cells and thermostattable cell holders. The latter were used in experiments when the temperatures of the sample and reference cells differed from ambient. The sample cell holder was mounted on a plastic base to minimize heat transfer with the cell compartment and was equipped with a small thermistor. This was positioned to allow stirring by a plastic paddle attached to a slow (15 rpm) synchronous motor (Hurst Mfg. Corp.). This equipment is illustrated in Figure 1. All cells used were of fused silica and manufact- ured by Pyrocell Inc.. They included cells of 10 mm path lengths with 5 mm and 7 mm widths, tandem cells previously described (Kayne and Suelter, 1965), and 5 mm and 2 mm path length cells. Temperature control was achieved with the use of constantly stirred baths continuously cooled by varying amounts of isopropanol—water circulated through copper coils from a Sargent refrigerated water bath cooler. 250 - 1000 watts of heating capacity in immersion heaters was controlled by electronic relays actuated by variable contact thermometers. Temperatures of from —1 to 650 were achieved in this manner. Circulating bath temperatures were determined with ordinary thermometers whose readings were checked against a calibrated thermometer. Temperatures in spectrophotometer cells and reaction vessels-were measured with the use of glass probe thermistors obtained from Sargent and Fenwal Electronics Inc.. Calibration curves for these were prepared l9 .Hamo mnu CH coausaom Mo Hm>oa on“ BOHmQ HHOB Una AEmma pbmfla may m>onm haunmaam “Hmnuflum map mo Hm>ma was no Haoo may CH UUCOHpHmom ma MoanEHmnu Gnu Roms mcflusm .HmumEOUOEQOHpUmmm ma SHMU mo usoEnHmQEoo UHQEMm CH UUHHMmeH ma s30£m ucmfimfisvm .Empmhm manUMumoEHmnu Hamo HmumEOpogmouuommm .H ousmflm 20 .12: i 1.915 .,1,,./..__.:_.,._c .25 .l ”59.5: .Emo omemzuxzfi , .e ’-.¢ 539: 283225; moSEfiE. ‘- 11‘ 7A.: Homzzoo :94: 37.51.5301 1 U mmmmHHm moeoz ozHaaHem 21 from resistance measurements determined at various temperatures in a battery operated Wheatstone bridge circuit. The read- ability and accuracy achieved under optimal conditions in these experiments was g§,i 0.050, control was usually better than 1.0.10. A Radiometer TTT-l/SBR2/SBU1/TTA31 automatic recording titrator was used for routine pH measurements, pH-stat determinations of enzymatic activity, and for the deter— mination of titration curves. It was equipped with a manual temperature compensator, jacketed reaction vessels and the standard and semi-micro glass and calomel electrodes. Although the semi-micro electrode holder was shielded, it was usually necessary to connect all components to ground (including the water bath) during pH—stat operations. Routine absorption measurements and optical assays were carried out on a Beckman DU spectrophotometer equipped with a Gilford multiple sample absorbance recorder and dual thermospacers. Some preliminary spectral data were obtained on a Beckman DB spectrophotometer with a Sargent SRL recorder. Optical rotatory dispersion (ORD) measurements were carried out on a Durrum-Jasco ORD/UV/CD-5 spectropolarimeter and circular dichroism recorder. Variable and fixed path silica cells supplied with the instrument were used, along with a jacketed silica polarimeter cell obtained from Pyrocell. Sedimentation velocity measurements for pyruvate kinase under varying conditions of cation and temperature were made in a Beckman Spinco Model E analytical ultracentrifuge 22 equipped with the Schlieren optical system and the rotor temperature indicating and control unit. Experimental Enzyme Assays Pyruvate kinase activity was routinely determined by the use of a pH-stat assay made possible by the uptake of a proton by the methylene carbon of PEP according to the equation: pyruvate kinase K+, Mget H*+ ADP + PEP > PYruvate + ATP The proton uptake is directly proportional to the pyruvate formed at any pH and is almost stoichiometric at pH 7.5 (Melchoir, 1965). The standard assay reaction mixture contained: 0.1 M KCl, 0.008 M MgClg, 0.002 M ADP and O-OOl.M PEP at a pH somewhat above that at which the assay was to have been run. Approximately 1 ml of this was placed in the jacketed reaction vessel of the titrator and the pH was adjusted to that desired by the addition of HCl from the titrator. Commercial buffers were used to standardize the pH meter. Standardized HCl solutions were used at concent— rations ranging from 1.85 to 2.0 x 10—8 M, The reaction was started by the addition of sufficient enzyme to give an easily measured rate (usually 22» 1 pg in 5 to 25 p1). The rate of HCl addition required to maintain the set pH was then recorded. Duplicate or triplicate assays were always performed and specific activities were directly calculated in 23 nmoles H+ uptake/min/mg protein. The pH-stat assay allowed much flexibility in activity determinations since, unlike the commonly used coupled assay (pyruvate formation measured by coupling to lactic dehydro- genase with corresponding oxidation of NADH (Kubowitz and Ott, 1944)) controls did not have to be run to determine the effect that the varying conditions would have on the coupling enzyme. Furthermore, conditions could be used which would probably result in the loss of activity of the coupling enzyme during the assay. Activities determined by this method are comparable to those determined by the use of the coupled assay, and usually ranged from 22- 160 to 200 pmoles/min/mg protein at 250 and pH 7.5. This procedure allowed the assays to be conducted in 2H20 solutions with only slight modifications. First, the reaction mixtures were lyophilized, then dissolved in 2 ml of 2H20, allowed to stand at room temperature for 30 min, lyophilized again, and diluted to their final volume with 2H20. The titrant used in this case is 2HC1 at a concent— ration of 9a. 2 x 10_3 M, prepared by diluting concentrated HCl with 2H20 and then standardizing. The p2H of the solution was determined from the relationship p2H=:pH meter reading+-0.40 (at 100% 2H20) for the glass electrode (Glasoe and Long, 1960). This relationship is approximately linear with relative 2H20 concentration. Activation energy for the catalytic reaction was deter- 24 mined by using the pH—stat assay at various temperatures, according to the following procedure. The reaction vessels were maintained at the desired temperature and the pH meter was standardized with buffer using the temperature correct- ions supplied with the buffer. The temperature compensator of the titrator was set at this same temperature. This was repeated for each temperature during the series. At low temperatures it seemed best to keep the absolute rates low because of a slower electrode response and increased overshoot of the individual titration steps. Fumarase activity was determined by a spectrophotometric assay described by Massey (1953a). The decrease in absorption of fumarate at 296 mp (efM 9a. 60) due to conversion to malate was followed in solutions of 0.0167 M sodium fumarate in 0.067 M phosphate buffer. The reaction was started by the addition of microliter quantities of the enzyme to the solution in the cell at the appropriate temperature, and the initial velocity was recorded. Fumarase concentrations were determined by the method of Warburg and Christian (1942) on protein solutions in 0.01 M sodium phosphate, pH 7.3. Specific activities were calculated as pmoles fumarate converted/min/mg of protein, and were equal to ga.300 at pH's around 6.4 and a temperature of 250. For activity determinations in ZHZO solutions, the above reaction mixtures and buffers were lyophilized, dissolved in a small amount of 2H20, lyophilized again and diluted to the 25 appropriate volume with 2H20. The p2H's were determined as described previously. Difference Spectra Difference spectra were recorded on the Cary 15 spectro— photometer by the usual single cell technique or the tandem cell method of Herskovits and Laskowski (1962). Solvent perturbation studies of the model chromophores and protein were made with the tandem cell method in the following general manner: chromophore chromophore +perturbant +solvent +solvent solvent perturbant +solvent sample reference beam beam 1 The above method is applicable also the protein and the perturbant is‘a characteristics in the region being allows for titration of the protein the absorption due to the substrate reference beam. In the cases where substrates ADP and ATP were used as length cells were used in tandem in i when the chromOphore is substrate with absorption studied. This method with the substrate, since is blanked out in the the strongly—absorbing perturbants, 2 mm path place of the 10 mm cells. When the perturbant had no significant absorption in the 26 region under study, the single cell technique was used. An example of this is the titration of the protein with the various cations. In all cases titrations were performed with the use of microliter syringes (10 and 50 p1 capacities, Hamilton) for the additions of substrate or cofactor, followed by stirring with a small plastic rod. Equal amounts of the titrant were added to the cross-blanking cell when necessary, and the chromophore concentrations were kept equal by the addition of an equivalent volume of solvent to the reference solution. Spectra were then recorded over the absorption range of interest and, where necessary, were corrected for the dilution of the chromophore. Under certain circumstances precipitation of pyruvate kinase occurred during the addition of some of the perturbants. This seemed to depend on a number of factors including the solvent conditions, titrant and the length of time since the enzyme had been passed over Sephadex. This could usually be eliminated by preincubating the enzyme at temperatures above 300 for 15 min or longer and then centrifuging the small amount of resulting precipitate. The specific activity of the enzyme did not seem to change with this treatment and further precipitation did not occur during the titrations. Since the measurements being made involved very small absorption changes, and the light scattering intensity increases with the inverse 4th power of the wavelength (Doty and Edsall, 1951), even this amount of precipitate 27 could not be tolerated in the ultraviolet. The number of tryptophan residues in the pyruvate kinase molecule was determined by the spectrophotometric method of Bencze and Schmid (1957). Temperature—Perturbed Spectra Difference spectra produced upon varying the temperature of the enzyme or model chromophores were measured with the equipment described previously. The general procedure was to maintain the cell in the reference compartment at a constant temperature with one circulating bath while the temperature of the cell in the sample compartment was varied. The circulating bath for the sample cell was set to the approximate desired temperature and the spectrum was recorded after the temperature indicated by the thermistor in the sample cell reached a constant value, which was also recorded. Dry air was circulated through the cell compartments when high relative humidity caused condensation on the cell windows at low temperatures. During thermal equilibration, the solution in the sample cell was constantly stirred at 15 rpm with the overhead paddle. Stirring during the recording had no effect on the observed spectra. Precipitation of the enzyme usually occurred at elevated temperatures, so the preincubation procedure described previously was employed. The temperature of preincubation was of necessity, 2 — 40 above the highest temperature to be observed during a run. At temperatures above 9a. 400, the 28 precipitation could usually not be alleviated by preincuba— tion at a higher temperature. These same experiments were carried out with the enzyme in ZHZO solution. This was accomplished by lyophilization of the appropriate enzyme solution and subsequent dissolution in 2H20. Best results were usually obtained when the enzyme was lyophilized in the appropriate salt solution with the buffer solution in 2H20 being added later. In this case the enzyme was not completely solubilized in the 2H20 solution and centrifugation was necessary before the spectra were recorded. The specific activity of the enzyme was not appreciably altered by this treatment. In all cases where the temperature was varied, cacodylic acid buffer (TMA cacodylate) was used, since its acid dissociation constant is very insensitive to temperature changes (Datta and Grzybowski, 1961). Protein concentrations were 1.5 - 3.0 mg/ml, the same range used in the studies of the effects of the various perturbants. Optical Rotatory Dispersion Measurements Optical rotatory dispersion determinations were carried out in the usual manner on the Durrum—Sasco instrument. The instrument was purged with prepurified nitrogen during the experiments (ordinarily carried out at room temperature; Si- 230). Maximum absorbance of the protein solutions was kept below 1.0, and the scanning speed and pen gain controls were set for optimum response. A sensitivity of i 200 milli— 29 degrees full scale was used for most of the work, and wave— length scale expansion was used in detecting small differences in the rotatory dispersion parameters. The instrument was adjusted to zero rotation with the cell compartment empty and solvent baselines were usually recorded at the beginning of each series of runs. In the experiment where the temperature of the sample was varied, the jacketed cell was connected to a circulating bath and solvent and sample spectra were recorded as described above, at each temperature. Circular dichroism measurements were carried out in a similar manner with the instrument set to record in the CD mode with a full scale sensitivity of i C.005. ORD data were plotted according to the Moffitt—Yang equation (Moffitt and Yang, 1956): [a] : 5107\6 + boAd (Az-Aé) «(Ra-NEW? as: (Ag-A8) Kg, A3 , b0 was determined from (AZ-K8) the slope of this line, and a0 from the intercept. RESULTS Difference Spectra Solvent Perturbation of Model Compounds When the ultraviolet spectra of the aromatic amino acids are determined in one solvent XE: another, difference spectra are produced which are characteristic of the absorbing chromophore, its state of ionization, degree of hydrogen bonding and the properties of the respective solvents. Side chain substitution and incorporation of the chromophore into a large molecule, as well as temperature, will also affect the characteristics of the difference spectra. Figure 2 shows the results of pfdioxane perturbation of the spectra of some model protein chromophores. The basic forms of these spectra are characteristic of the solvent perturbation of the benzene, phenol and indole chromophores, and approx— imate the protein residues phenylalanine, tyrosine and tryptophan, respectively. In this study attention will be focused principally on the indole difference spectra characteristic of tryptophan. As seen in Figure 2, the solvent perturbed difference spectrum is relatively large. Differences of a similar magnitude are seen when the hydrogen on the imino nitrogen is involved in hydrogen bonding (Suelter and Weber, 1967). The differences caused by side chain protonation (Hermanséi.lo 1960) and temperature changes are much smaller. The latter 30 .osvflccuou HHUU Senses ocb mcflws ONE CH ouocmoEouco 2/ mo coaumnpcoocoo DEMm ecu Dmcflmmm cocuooon muuoomw map UCm coaufimom UHQEMm map EH ucm>aom mcflnnsuuom coumoaccfl one CH muoB monocmoEOHQU USE .msuoogm mucoumwwflo DUHOH>6HuH5 UDQMSDHUQIDQU>Hom .m ousmflm 52 as .xpozu4m><3 omN _ owN _ => wzzw1m 0 (D 9 I_Ol ‘ AllAlldUOSSV EDNEHBJJIG UVWOW .m wusmflm as .xpozu4m><; oon omN OQN o- OwN _ q — _ _ \\\| wz._.-4-._>._.uo<.z _ _ _ _ _ Z'OI 1‘ All/\lldHOSQV 33N383$$l0 BVWOW :5 .xsozu4u><; ONn con ODN OQN q . _ a _ . v T JozmhIJ-J>kuu<-z J o. N. V. w. 2-0l ‘ ALIAILdUOSBV BONBHBESIO HV'IOW \N \N of these may only amount to SE» 8% of the maximum dioxane perturbed difference with a 250 temperature differential. Tryptophan-Tyrosine Determination A tyrosine/tryptophan molar ratio of 2.4-2.5 was determined from the values given in Table I of the paper by Bencze and Schmid (1957), with a total tryptophan content of £3, 13 moles of tryptophan/mole of protein, based on the molecular weight of 237,000. This data gives a tyrosine content of 31-33 moles/mole of protein, which is in very good agreement with the value of 32 moles/mole obtained from conventional amino acid analysis (Steinmetz, 1966). Cation Interaction with Pyruvate Kinase When the activating cations KI, Mn2+, or both, are added to aqueous solutions of pyruvate kinase and the spectra recorded relative to the enzyme in TNmI, a non-activating cation (at the same ionic strength), the difference spectra shown in Figure 3 are produced. Perturbation of tryptophan residues in the protein is clearly indicated. The maximum molar extinction difference (AE max) for the cation activated enzyme is attained in 0.1 M KCl and 0.005_M MnC12(approximately the kinetic optimum (Kachmar and Boyer, 1953)), and this is 22.- 4400 cm"1 at 295 mp. The Aemax per active site of the enzyme, based on two sites as found by Reynard t al. (1961), is therefore 2200 cm‘l. This is approximately the maximum Figure 3. Cation-perturbed protein difference spectra. Solid line, pyruvate kinase in 0.1 M KCl‘T MnC12 XE, pyruvate kinase in 0.1 M TMACl. The spectrum shown is calculated from titration curves to be that which would be observed at saturation with Mn2+. Dashed line, pyruvate kinase in 0.1 M KCl yg, pyruvate kinase in 0.1 M_TMAC1. All solutions were in 0.05 M Tris, pH 7.8. Figure 3. NTQ\.< xkitltommfi MUEMQWKIiQ QGQQE I I I I KVOJM)+Mn?WSArD) T I I KflOJM)--- _ _ _ _ _ _ _ _ . _ _ no nv 00 .0 AW 0; AU ... 4 nlbx x \ftxikthmm—x WUEMkaka QQQQE -48- 250 270 290 3|O 330 WAVELENGTH,m# 230 f 5 36 value for the solvent perturbation of the tryptophan models. It was also found that Ca2+, a competitive inhibitor of the enzyme with respect to the divalent cation does not produce a difference spectrum at concentrations where it should be bound to the enzyme. When Mn2+was rapidly added to the enzyme solution in 0.1 M TMACl and the absorption measured at 295 mp, the maximal value for the change was reached within the time resolution limits (£1. 1 sec) of the conventional instruments (DU). A single experiment demonstrated that the comercially available stopped-flow instrument (Durrum) could not detect absorption changes of this magnitude without modificationl. Substrate Interaction with Pyruvate Kinase When the substrate PEP is bound to the activated or non—activated form of the enzyme, the difference spectra shown in Figure 4 are produced. Of interest here is the tyrosine difference seen upon binding PEP to the activated enzyme. Addition of TMA pyruvate to the non—activated enzyme shows no difference spectra, while addition to the activated enzyme shows tryptophan and/or tyrosine differences depending on the level of activation of the enzyme. Titration with the substrates gave data which, when plotted as convent— ional titration curves, allowed estimation of the apparent 1 This experiment was conducted with the help of Dr. V. Massey with an instrument in his laboratory. Figure 4. Substrate—perturbed protein difference spectra. Solid line, non—activated (0.1 M TMACl) pyruvate kinase in both sample and reference positions, with PEP in the sample position, calculated for saturation with PEP; dashed line, activated (0.1 M KCl + 0.001 M MnClg, corrected for saturation with Mn2+) pyruvate kinase in both sample and reference positions, saturated with PEP in the sample position. MOLAR DIFFERENCE ABSORPTIVITY x IO'2 Figure 4. 30- - 20_ —TMA _ ---- K'+ Mn" -22() _. .43()._ I 1 l r l r l 1 260 280 300 320 WAVELENGTH , mp 38 dissociation constants for the substrates. Curves were fitted to the relationship pKD(app):= p(S) + log i:—, where I‘d pS is the negative log of the substrate concentration and o( = AA = fraction of enzyme-substrate complex. AAmax Corrections for the free substrate concentration were made when necessary. These results are presented in Table I. Table I. Dissociation constants for the interaction of substrates with pyruvate kinase in the presence of various cations. Enzyme concentrations of 2.0—2.5 mg/ml in 0.05 M Tris-HCl, pH 7.8. Substrate Cations Present pKDa PEP 0.1 MTMA’r 3.6i0.2 PEP 0.1 MK+ 4.1101 PEP 0.2 MK+ 3.9-1.0.1 b PEP 0.1 M W+0.005 M Mn2+ 5.03:0.1, 4.70 PEP 0.1 M TMA*+ 0.005 M Mn2+ 5,010.1 Pyruvate 0.1 M TMAT .. g Pyruvate o.1 MK+ .. b Pyruvate 0.1 M.Ktt0»01.fl Mn2+ 3.6, 3.77 Pyruvate 0.1 M TMAT+0.005 MMn2+ 3.51:0.1 Pyruvate 0.1 M Kt+0.01 M Mga+,pH 8.5 3.10-3.62e aNegative log of the dissociation constant. bValue obtained by Mildvan and Cohn (1963) from proton relaxation rate measuremen s. No ultraviolet difference spectrum (250-300 mp) observed. Difference spectrum too small to determine pKD. Values obtained by Reynard §£_al, (1961) by the ultra— centrifuge method in the presence of MgSO4 (0.01 M), KCl (0.1 M), and glycylglycine buffer at pH 8.5 and 26. Substrate perturbation studies with ADP and ATP under varying activation conditions show no apparent protein difference spectra. In these cases nucleotide concentrations up to 6 x 10'4 M were used, which is approximately six times the reported dissociation constant (Mildvan and Cohn, 1963). Titrations of the enzyme with fluoride, hydroxylamine, and bicarbonate, which are substrates and cofactor for the 40 ”fluorokinase” (Tietz and Ochoa, 1958) and ”hydroxylamine kinase” (Kupiecki and Coon, 1960) reactions catalyzed by muscle pyruvate kinase, under conditions where their binding would be likely, did not result in the appearance of protein difference spectra. Temperature—Dependent Changes When the temperature of a non—activated sample of pyruvate kinase was lowered in the sample compartment of the spectrophotometer and the spectrum recorded against the same solution at a higher temperature, the spectrum shown in Figure 5 was observed. This is presented in comparison with that observed upon saturation with the monovalent and divalent cations at room temperature. The cacodylate buffer used in the temperature—perturbation studies does not seem to have any effect on the production of the cation perturbed spectra. When this tryptophan difference is determined as a function of temperature by recording the spectra at various temperature intervals, the points shown in Figure 6 are found. The maximum absorption changes at 295 mp are estimated from the observed data. This allowed determination of £L§_ for A Amax each point. If we assume that the temperature perturbation involves the interconversion of two forms of the enzyme, then the equilibrium constant at each temperature can be calculated according to the relationship: Keq :: 2§7€lL7fiZET' . max Figure 5. 41 Temperature—perturbed protein difference spectrum. Dashed line, non-activated (0.1 M TMACl) pyruvate kinase with sample position at low temperature relative to reference, corrected to theoretical maximum; solid line, pyruvate kinase in 0.1 M KCl4-Mn2+ Mg. pyruvate kinase in 0.1 M TMACl, calculated for saturation with Mn2+. Figure 5. K'+ Mn" ---- TEMPERATURE _ 1— 30* 25- 20-4 9— . l5- IOb 5 O 5 40.. -|5r _ O 5 ac . -3‘).. .455 — ~-o_ x >t>_Emomm< muzwmuuua «<42... -40 .. -45.. 290 3:0 330 WAVELENGTH, mp 270 250 42 Figure 6. Temperature—dependent difference absorbance of pyruvate kinase. Non—activated (0.1 M TMACl) pyruvate kinase in both sample and reference positions with 0.05 M TMA cacodylate, pH 7.8. The reference was maintained at 250 and the sample temperature varied as shown. The solid line is a theoretical curve calculated as described in the text. (4‘295’ (AA295 )max Figure 6. . I . I . I . 1 LG 1 0.9 - _. O 0.8 - _ 0.7 - -4 0.6 - - 0.5 - .. 0.4 - - 0.3 h- _. 0.2 - -. 0.l .. .. O . l . 1 . I . t o 4 e I2|620242832 TEMPERATURE, °c 44 45 A van’t Hoff plot (1n Keq.1§- l/T) for this data is presented in Figure 7 with a line fitted by the method of least squares. The linear van't Hoff plot indicates no apparent changes in heat capacity as a function of temperature and justifies the calculation of the theoretical curve shown by the solid line in Figure 6. The thermodynamic parameters calculated from the data. are: AH=33.1 kcal/mole; AFC: -1.1 kcal/mole and A.SO==107 cal/deg/mole, with an inflection point at 15.50. In all cases tested, the temperature-depend— ent absorbance change curves are completely reversible, and after thermal equilibrium is reached, no additional spectral changes are observed. Non—activated solutions of pyruvate kinase in 2H20 demonstrated a similar behavior. Figure 8 shows the temper— ature dependence of the production of the tryptophan differ— ence spectrum with a theoretical curve calculated from the van't Hoff plot of Figure 9. Thermodynamic parameters in this case are: AH =4l.4 kcal/mole; AFO=-0.504 kcal/mole; A.SO==137 cal/deg/mole, with an inflection point at Ea. 21.50. The temperature of the inflection point of the observed change (transition temperature) is apparently a characteristic of a prOcess which is sensitive to environmental effects. Experiments such as those described above were conducted with pyruvate kinase solutions under varying conditions of pH, p2H, ionic strength and urea, all of which are shown in Figure 10. On the left hand side of the figure, it is 46 Figure 7. Van't Hoff plot for pyruvate kinase. Data and conditions are those shown in Figure 6. Line drawn is calculated by the method of least squares. Ln 2.0 -|.O 3.3 3.4 —1|— )1? Figure 8. 48 Temperature—dependent difference absorbance of pyruvate kinase in 2H20 solution. Non-activated (0.1 M TMACl) pyruvate kinase in both sample and reference positions with 0.05 M TMA cacodylate, p2H 93. 8.15. The reference position was main— tained at 23. 250 and the sample position varied as shown. The solid line is a theoretical curve calculated as described in the text. Uo MKDF9 nonmasoamo ma QBEnc mafia .w 50 ousmflm ca csocm omocu mum mGOADHccoo Ucm mama .Owa as ommcflx opm>5H>Q How poam wmom u.cm> .0 musmflm 51 $0M Oren min ¢m.n ._. n0_x _ m_ .m I a _ a _ a _ MJOZ\J101 52 .mm comeHUcH asp um mpmsseoumo ass a mo.o can H0429 a H.o CH mecho QDHB coaumnpcoucoo wow: so mococcomoc Acosuumm ucmflm .mcoaumupcmocoo HUmSB mcH>Hm> ch M.N mm .opmahcoooo <29 fl mo.o CH UESNcm SDHB bumcwnpm OACOA co mucoccommc nCOHuowm prcmo .mm UUDEOHUQH ocp pm mpmahcoumo <26 fl mo.o cam Homzfi fl H.o EH UESNcm spas mwm Ucm mm co mocmwcmmmp ncospoom pwmq .mQOADHUcoo mcflhum> Hons: ommcflx mum>5nwm cmpm>fluomlcoc Mom monnwmuomEop QOHUHmQEHB .os musmsm .2 TEE; 0.0 0.0 2 Ihozmmkm 220. 0._ 0.0 0.0 00 ____._____ 0N .VN mm mm .oH magmas BHHIVH 3dW31 NOLLIS NVHI 3o 54 seen that decreasing pH markedly increases the transition temperature of the enzyme. The same effect is seen in '2H20;at equal pH's and p2H's, however, the transition temperature is raised in 2H20. An increase of around 50 in the transition temperature would still be noted if we make allowance for the degree of ionization of enzyme dissociable groups, which is of course, lowered in 2H20 solution. For example, titration of imidazole in 2H20 with 2 x 10"3 MFHC1 using the automatic titrator shows a value for pKapp(2H20)e£7.25 which is 9a. 0.3 higher than the value in H20 solution (Edsall and Wyman, 1958). The enzyme solutions were in 0.1 M TMACl with 0.05 M TMA cacodylate at the indicated pH's (measured in the presence of enzyme). In the center section of Figure 10, it is shown that the transition temperature increases with increasing ionic strength (varying the TMACl concentration) at a constant pH. The right hand side shows a significant decrease in the transition temperature as the urea concentration is increased at three different pH’s. In this case a constant ionic strength of 0.1 M was employed (TMACl). The data used for all transition temperatures indicate AsH values for the change in the range of 35 — 40 kcal/mole. Additional transition temperature changes are presented in Table II. Triton X-100, a non-ionic detergent, causes a lowering of the transition temperature as does urea. 55 Table II. Additional transition temperature data for pyruvate kinase. Enzyme Solution pH Transition Temp.OC 0.1 M TMACl 7.6 19.0 ” +0.02% Triton x—100 7.6 17.5 ” +0.001 M Gael2 7.6 25.5 0.1 M TMACl 7.8 15.5 " +0.001 M CaC12 7.8 17.1—17.4 ” " +0.002 M EDTA 7.8 14.9 Ca2+, which as previously stated does not cause a cation perturbed difference spectrum, nevertheless does produce an increase in the transition temperature at pH 7.6 and 7.8. Addition of EDTA to the same solution reverses this effect and also shows that small differences in transition temper— ature can be determined. In the presence of saturating levels of the activating cations K+ and Mn2+, the transition temperature was found to be increased from 150 to 320 (pH 7.8). Data on the effect of cations on the transition tempera- tures at pH 7.8 are presented in Table III. Illustrated here are the differences observed in the presence of monovalent and/or divalent activating cations, the weak activator Na+, and the non-activating and inhibiting cations Li+ and Ca2+ and the effect of urea on the fully activated enzyme. 56 Table III. Pyruvate kinase transition temperatures in the presence of various cations. All solutions at pH 7.8. Cation (molar concentration) Transition Temperature 0C + TMA (0.1) 15.5 " " +Mn2“’(5 x 10‘6) 17.4 " " " (4.5 x 10—5) 22.2 " “ " (1 x 10‘4) 25.9 " " " (0.005) 26.7 " " +Ca2+(0.001) 17.5 K+ (0.1) 22.4 Eai (0.1) 21.0 Li+ (0.1) 22.4 K+ (0.1) +Nm?+(0.005) 32.4 " " " ” -+l.0 M urea 29.4 11* (0.l)-+Mn2+(0.005) 52.4 K+ (0.1)+Ca2+(0.001) 26.6 Arrhenius Plots Pyruvate Kinase In an attempt to assess the effect of temperature on the catalytic behavior of the enzyme, sufficient data were obtained for an Arrhenius plot which allowed calculation of the activation energy. The initial velocity of the reaction was determined under optimal conditions (near saturation with substrates) at the various temperatures as described previously. In this case, the achievement of maximal velocity was checked at the lower temperatures 57 by determination of the pH optimum and by doubling the concentrations of the respective substrates. No apparent changes in the velocity were observed. The Arrhenius plot for the reaction in H20 at pH 7.55 under the conditions described in the ”Experimental” section is shown in Figure 11. It is apparent that there is an "abnormal” effect of temper— ature on the reaction velocity since a linear relationship is not followed, (for a discussion of this behavior see; Hulett, 1964; Kistiakowsky and Lumry, 1949). The data does suggest, however, that a linear relationship is followed at the extremes of high and low temperatures, indicating that a two-state relationship may be responsible for this behavior. A theoretical curve based on the assumption that an equilib- rium existed between two forms of the enzyme, as in the case of the temperature-perturbed difference spectra, is shown by the solid line in Figure 11. This was calculated from the following modified form of the Arrhenius equation: kcalC=CKkle(E5ct/R)(l/Tl‘l/T) +(l_.x)k2e(E§ct/R)(l/Ta-l/T) where ki is the rate constant at temperature Ti of the form of the enzyme with energy of activation Eéct; x is the fraction of the enzyme predominant at low temperature and (1-°<) that fraction predominant at higher temperatures. The data in this figure were normalized so that the rate at 00 is equal to 1. The Eact was estimated at 13.5 kcal/mole ( A.H*=12.9 kcal/mole) above 300 and 18.0 kcal/mole .onp ecu cfl Confluowmo mm Uwpmasoamo 58 o>u50 Havsuouoocu m we GBEHU mafia 0:9 .pxou ca Confluomop one mcoapficcoo coaucmmm .00 um pmcp ma on 0cm mm.s mm as SDHUOHU> coauomos Hmflpflcfl ecu ma H .ommcflx opm>50>m How UOHQ msflcocunfi .HH onsmflm .....h_ mag. x 00h owfi Own .HH musmam U‘I .J—J 60 '( A H5: 17 . 4 kcal/mole) below 00 . A similar determination was carried out with enzyme in 2H20 solution at a p2H of 7.85. This data is presented in Figure 12 as an Arrhenius plot, but with the solid line representing in this case the energies of activation in the high and low temperature ranges. The values are 6 and 16 kcal/mole respectively, with the discontinuity centered at 9;. 22.50. It is interesting to note that a considerable deuterium isotope effect was observed in these measurements kH20/k2H20 215 (Figure 13). This experiment was conducted with the sum of the H+ and 2H+ concentrations approximately constant (p(H+2H)::7.55). This pH value was near the optimum even in ca. 100% 2H20. Arrhenius plots were also determined for the enzyme under other pH and ionic strength conditions and these results are shown in Table IV. Fumarase Since Arrhenius plot discontinuities had been previously noted in enzyme catalyzed reactions (for reviews see: Stearn, 1949; Massey t a;,, 1966) it seemed of interest to determine if deuterium substitution could affect the Arrhenius plots of other enzymes. The fumarase system studied by Massey (1953b) was chosen under conditions where the discontinuity was about the same temperature as that in pyruvate kinase. At pH 6.35, Massey found a break in the Arrhenius plot at 180 61 .momcmu endpmuomEou 30H Eco cmfln obs cfl coapm>flnum mo moflmnoco ecu pcmmosmou mmcfla pcmflmuum ocB .uxou CA Conauomoe mum mGOHUHUcoo coauomom .mw.w mmg um xpflooao> coapommu HEHUHca may ms H> .ONmN ca ommcax oum>snwm How uoam wsflcocun< .ma musmam Figure 12. 62 «e O «o co on 0 no (0 e 1 Al‘. to e O O O 000 O are «e e) ee 1 1 l 1 l 1 l 0. O. N. V. N N - o !A U1 3.30 3.38 3.46 3.54 x 103 3.22 314 .on CH Dash we 606H>H6 em>u0mno sufloon> HMHDHCH 0CD mH huH>HDOm OHMHUQO m>HDmHmm .uxmp CH ’6 UUQHHOmmp one mCOHuHUCOO mammm UHEUCEDm .COHDOEUH ommCHx opm>snmm ecu Co uoowww omouomH EDHHUDSEQ .MH ondem Figure 13. 64 1 1 1 1 1 o o o — O - 1 1 1 1 1 o 00. 0. st N. O -' o o o o Ail/“10V 'OBdS BALLV‘IBH 60 80 100 VOLUME °/. 21420 40 20 65 Table IV. Arrhenius plot characteristics for pyruvate kinase. Concentrations of substrates and cations as described in text for the standard assay. Solution Eact kcal/mole "Break" 0 high temp. low temp. Temp. C pH 8.0 11 -— not obsvd. pH 7.8 11 20 14 pH 7.6 14 18 16 pH 7.4 9 14 20-22 PH 7.6 01=C455)a 6 13 21 p2H 7 8-7 9 (2H20) 6 16 22.5 aAdditional 0.45 M TMACl with usual 0 lO.M KCl. in the conversion of fumarate to malate in 0.067 M phosphate buffer (Massey, 1953b). The Eact changed from 10.6 kcal/mole to 6.7 kcal/mole as the temperature increased. Reexamination of this work showed a break temperature at 93, 150 with the energies of activation slightly higher (1 to 4 kcal/mole) at the higher and lower temperatures respectively. When this experiment was repeated in 2H20 solution at p2H 6.45, the results shown in Figure 14 were obtained. A considerably higher break temperature is observed (gg. 22.30) with energies of activation near the values previously found for fumarase in H20 solution. 66 .mmmCmH UHDHEHEQEUD 30H pCm CmHC 0CD CH COHpm>HDOm mo monuoCm 0CD qummHmon moCHH pCmHmHum oCB .m:.© mo mmm m CDHB uxmp CH UoQHHOon mm mum mCOHpHUCOO .Ommm CH omMHmEDM How DOHQ msHCwCHH< .HH musmHm 67 P m0. X 4 8M VD” Omfi min N¢.m mmfi .vm.m 09m H _ _ _ _ _ _ l w.._O$_\1_Hpom mo EUCUQO UCE EUCUmon 0CD CH ommCHx mum>5nhm mo thOOHo> COHpmuCoEHpom .mH musmHm Figure 15. 70 l N (D 1N3/0/JJHOD NOIIVINEIW/OEIS (Woes) l2 IO (PROTEIN) MG/ML 71 In an attempt to determine if differences in Sgo’w could be detected between the protein at low and high temperatures, measurements were made on identical samples in 0.10 M TMACl, 0.05 M TMA cacodylate, pH 7.6, run separately at 40 and 300. 520,w values were calculated using tables for the viscosity of H20 as a function of temperature in the Handbook g§_Chemistry and Physics (39th ed). Density measurements were made with solvent at the above temperatures. A partial specific volume of 0.740 cms/g at 100 (Warner, 1958), with an assumed change of 0.001 cms/g /1.5O (Taylor and Lowry, 1956) was used. The calculated results were: TOC 53055 4 8.49 50 8.53 However, the partial specific volume factor in these calculations plays a very important part, since if we use a value of 0.001 cmS/g/l.0O for its temperature coefficient, the calculated values become: TOC Sgo?w 72 Optical Rotatory Dispersion Measurements ORD studies on activated and non—activated pyruvate kinase were carried out as described in "Experimental". Preliminary experiments indicated very small differences between the ORD of the cation activated and non—activated forms of the enzyme. These data were found to fit a linear Moffitt plot using a A0 of 220 mp. The observed values for DX]233, (—54500), the specific rotation at the Cotton effect maximum and.%CK-helix (29%) compared favorably with those reported by Jirgensons (1965), (BX1233='-56000 and 32% helix). The %0(—helix was calculated as the ratio of the observed b0 to that of 100% helical polyglutamic acid using the same A0 (Jirgensons, 1965). Differences in the ORD between activated and non—activated samples were verified by recording the ORD with the expanded wavelength scale. Observed rotations at the various wavelengths (310-245 mp) were converted to specific rotations: Bx]kle§%.°NCm .ommCHx oum>su>m How muOHm uHHmwoz .QH ousmHm c . l 3 .2 s. mlo.oo +3: .2 .6 +0. 2 .6 com." .61 omen our .2: .5- no.“ a- — b — — _ — — — n p — ON 6. m. m c 0 cu m. N. m e 9? 2 cuo_x mo u .x .©H onsmHm ¢m 0m mm mm w. v. v—Zv)[v]— 0 2 ,_mx( 75 at a concentration of 9a. 0.6 mg/ml with a 10 mm path length cell and a temperature of 230. A A0 of 220 mp was used; the value b0 was determined from the slope of the line drawn by eye through the points, and a0 from the y—intercept (intercept==aOA8). Differences between b0 values were checked by calculating least squares lines for the data. The effect of temperature on the ORD was also studied with non-activated solutions of the enzyme in 0.05 M TMA cacodylate, pH 7.8. These results are presented in Table V along with those determined for the enzyme under other conditions. Table V. Moffitt parameters for pyruvate kinase solutions. Data obtained at 230 except where noted. All solutions in 0.05 M Tris, pH 7.8,except as noted. Enzyme Solution boo aoo 0.1 M TMACl —102 —182 0.1 M KC1+ 0.005 M MnC12 —112 —190 KC1+ MnC12+ 0.001 M PEP —115 —188 Kc1+ MnC12+ 0.0001 M ADP —102 —228 TMACl‘l' 0.0001 M ADP -106 —167 TMAc1+ 0.1% dodecylsulfate —110 -278 KC1+ MnC12+ 1.5 M urea —119 -l32 TMACl, 29.60 (0.05 M TMA cacodylate) -111 -170 TMACl, 9.80 (0.05 M TMA cacodylate) —117 —149 Considerable scatter in the data was usually observed with solutions of the enzyme with added ADP. It is not known whether this was due to an anomaly of the instrument at the higher absorbances used, or possibly an extrinsic Cotton 76 effect. The latter possibility was checked by recording the circular dichroism spectrum of the activated enzyme (0.05 M Tris-HCl, pH 7.8, 7.3 mg protein/ml, 2 mm path length cells) in the presence and absence of 4 x 10‘4 M ADP. No changes in the elipticity were observed near the region of nucleotide absorption at the most sensitive instrument setting,i.0.005. Tryptophan Exposure Solvent Perturbation of the Protein Independent attempts were made to determine the number of tryptophan residues in the protein exposed to the solvent environment and whether or not any change in this number occurred during the process of activation of the enzyme. The first method of choice was the solvent perturbation technique devised by Williams g; QM. (1965). Perturbants used were: ethylene glycol, ethylene glycol monoethyl ether, 1,2—propanediol, 2H20, dimethyl sulfoxide and sucrose. 20 — 40% (v/v) concentrations of the first three solvents caused precipitation of the protein at room temperature, so that measurements of the differences between the protein in the perturbant and the protein in H20 could not be made. 2H20, although a weak perturbant, could be used with solutions of the non—activated enzyme; however, the values of Afggs are so small in 50% 2H20, that exact numbers of exposed residues could not be calculated. Moreover, consid— erable precipitation occurred in cation activated solutions 77 of the enzyme in 50% (v/v) 2H20. Dimethyl sulfoxide and sucrose caused perturbation of the exterior chromophores showing difference spectra typical of tyrosine and tryptophan. In the case of dimethyl sulfoxide (21 %), the indication is that approximately half (14) of the tyrosine residues are exposed to the solvent. However, the critical observation is that no appreciable difference spectrum is observed between the cation activated and non—activated samples of the enzyme when both are in either 21% dimethyl sulfoxide or 25% sucrose. The data consequently show no difference in the number of exposed chromophores between the two forms of the enzyme; and from the data obtained it is impossible to determine whether the perturbant has stopped a conformational change from occurring in the activated enzyme or caused the corresponding conformational change in the non—activated enzyme. Denaturation Significant protein unfolding, observed by the formation of tryptophan and tyrosine difference spectra, are produced in 3.0 M urea or 0.1% sodium dodecylsulfate. Virtually no difference was observed in 1.5 M urea. 78 Reaction with N—Bromosuccinimide Recent work has indicated that NBS can be used for the selective oxidation of peptide tryptophan residues at pH's around neutrality (Spande gg gi., 1966). A titration of activated and non-activated pyruvate kinase was attempted with NBS in 0.1 M Tris—HCl, pH 7.8. No significant reduction in the absorption at 280 mp (after correction for dilution with the NBS solution) occurred with either form of the enzyme, however, reaction did take place,since after the addition of NBS to 9a. 3 x 10‘4 M, precipitation of the protein began to occur. In 6 M urea a considerable reaction of tryptophan with NBS was observed. Titration of either the activated or non—activated enzyme with NBS while observing the tryptophan difference spectrum gave the following result: the predicted negative tryptophan absorption (due to loss of tryptophan) was observed when the difference spectrum was calculated for activated or non—activated enzyme after reaction with NBS lg. that not reacted (at 1 x 10-4 M NBS). The data indicated that reaction occurred to a slightly greater extent with the non—activated enzyme. Reaction with 2-OH-5—N02-Benzyl Bromide Cation activated and non-activated samples of the enzyme in 0.05 M Tris—HCl, pH 7.8 were reacted with ¢'—Br at a concentration of 3 to 5 x 10‘3 M, similar to the method 79 employed by Horton and Koshland (1965). The ¢'—Br was a 0.1 M solution in dry acetone and this was added slowly with stirring to the protein solution at room temperature. The product was placed on a 2.5 X 10 cm column of Sephadex G—25 and eluted with 5 x 10_3 M Tris, pH 8.6. Columns of fresh Sephadex were prepared before each use because the ¢'—OH formed by hydrolysis is somewhat strongly adsorbed to the Sephadex. Thin layer chromatography with isobutanol:H20:NH3 (6:2:2) showed this procedure resulted in only protein bound 0' recovered in the first fractions off the column (void volume). The recovered material had considerable pyruvate kinase activity (gg. 120 pmoles/min/mg), and still possessed the cation requirements for activity. The concentration of bound 0' was determined from absorption in 0.1 M KOH at 408 mu (Horton and Koshland, 1965) using a molar extinction coefficient of l x 104 cm‘l, which was determined by hydro— lyzing weighed amounts of the ¢'—Br in H20. Protein concen- trations after reaction with ¢'—Br and Sephadex treatment were determined from its absorption at 290 mp in 0.1 M KOH (phenolate ion) using a molar extinction coefficient of 1.44 x 105 cm_l. This allowed calculation of the ratio; moles 0' bound/ mole protein. These values were not very reproducible, however, average ratios of gg. 3.5 for the activated and 4.5 for the non—activated enzyme, with the latter consistantly higher were observed. . 11.1 ..4.11_....-_-.1.__.._g__-_M 80 Non—activated enzyme usually showed a greater amount of denaturation upon reaction with ¢‘—Br, iég., a greater amount of precipitate formed, and this, when dissolved in 0.1 M KOH had ¢'/protein ratios of Si- 10. The ¢'—enzyme showed a normal tryptophan difference spectrum when cations were added to a solution in the sample cell and, although these spectra were recorded from 500 to 240 mp, they showed no perturbation of the C' chromophore. DISCUSSION It is evident that the ultraviolet spectral changes observed upon interaction of pyruvate kinase with its required activating cations or the substrates, PEP or pyruvate, reflect structural changes in the protein molecule. These have been shown here and elsewhere (Suelter §£_§;,,1966) to be related to the kinetic behavior of the enzyme (activator and Michealis constants) by the agreement between the dissoci- ation constants obtained kinetically and by the difference spectra titrations. The similarities between the protein difference spectra and those spectra produced by solvent perturbation of the aromatic amino acid residues leads to the premise that the differences observed in the protein spectra are brought about by a solvent (environmental) perturbation of certain residues. The environmental properties known to give rise to the spectral shifts are, chiefly, dielectric constant, refractive index, polarity and structural order, with probably no single property being entirely responsible in this case. Since a tryptophan difference is observed in most of the cases studied here, the discussion will emphasize this residue in particular. The blue (bathochromic) shift observed upon addition of cations or substrates or lowering of the temperature could be caused by either a shift of the chromophore from a region whose solvent properties are those of the non—polar, low dielectric type into a relatively more polar or aqueous 81 82 environment or to the breaking of intramolecular hydrogen bonds between the chromophore and a nearby acceptor, for example, the peptide backbone. In the past, the former idea has been accepted almost without question, however, there is no means of determining if this is the case i priori. A direct interaction of the chromophore with a charged protein group or the cation cofactor is considered unlikely. Charge perturbation (pH 1) of the indole chromo— phore does not produce an appreciable spectral shift. Also, similar spectral shifts are produced in the protein in the absence of cationic cofactors. Attempts were made to show if a difference in the number of tryptophans exposed could be detected between two conformations of the enzyme. Unfortunately, the methods used, although being the principal ones available, are not without ambiguity. The results have indicated that the solvent perturbation techniques (with sucrose or dimethyl sulfoxide) in themselves give rise to conformational changes in the protein which either mimic or mask those changes which were under investigation. Chemical modification of the protein by oxidation with NBS or reaction with 2-OH—5—N02-benzyl bromide resulted in the uncertainty of whether the conformation of the protein is the same after reaction with the first equivalent of reagent and whether this affected the reaction with subsequent amounts of reagent. If this ambiguity is present, then no analysis can be made from the chemical modification results. Nevertheless, the results with NBS 83 show that tryptophans are relatively unavailable for reaction (at pH 7.8) under either activating or non-activating conditions. There was no indication that those reacting are involved in the production of the difference spectrum. The results of the~¢'—Br reaction with the protein can be discussed in slightly greater detail. Apparently, a reaction takes place with this reagent since three or more moles of the chromophore are bound to the protein. Since the resulting protein still has considerable enzymatic activity, it is unlikely that gross structural changes have taken place during or after the reaction. This is also supported by the appearance of the tryptophan difference spectrum upon the addition of cations to the modified enzyme. However, although ¢'-Br is one of the most specific protein reagents known, it does react to a considerable extent (9;, 10%) with available cysteine residues (Horton and Koshland, 1965). Since pyruvate kinase is believed to have some 35 cysteine residues/mole of protein (Mildvan and Leigh, 1964; Steinmetz, 1966), it is therefore evident that if most of these were exposed to the reactant, they could account for a considerable amount of the bound chromophophore. This is considered unlikely, since other data suggests that only four cysteines are available for reaction before extensive unfolding of the protein occurs (Steinmetz, 1966). The greater amounts of ¢' chromophore bound to the non— —activated enzyme would be consistant with the observation that the activating cations protect the enzyme from inact— 84 ivation with sulfhydryl reagents (Mildvan and Leigh, 1964). Thus, it is concluded that no appreciable differences in the number of tryptophans exposed to the external medium exist between the activated and non—activated forms of the enzyme. The results have shown that the blue shift caused by lowering the temperature of a non—activated sample of the enzyme is very similar to that produced by cation binding. Although this can not be proven without an independent method of identifying the tryptophans involved, they are probably the same ones in both cases. This statement is based on the fact that the observed tryptophan differences are not additive, i,§,, a blue shift can be produced by addition of cations or a lowering of the temperature, but once formed by either method, it is not affected by the other, and this is true in a converse manner for the case of a red shift. The temperature—dependent changes are of the same magnitude as those produced by cations and substrates. That these changes are true indications of protein conformational changes is evident by their magnitude, which is considerably greater than that which would be observed from a similar temperature perturbation of the isolated chromophores. The non-linearity of the change in the temperature ranges studied, and its control by protein effectors that should not directly influence the chromophore, provide additional 85 evidence. The thermodynamic parameters, especially (is, calculated for these conformational changes are what one may expect for a somewhat limited change in a large protein molecule. Since entropy values for the hydrophobic bonding of amino acid side chains are still open to question (Schneider e: g;., 1965), it is not possible at present to assign limits to the number of residues which may be involved in the bonds maintaining the conformations. The linear van't Hoff plots are of interest, since they indicate that the system is somewhat uncomplicated by other effects (g.g., denaturation). Other work of this type has yielded results with non—linear van't Hoff plots (Hermans and Scheraga, 1961; Brandts, 1964), especially in the case of the more extensive conformational changes. The changes in the transition temperature with varying solution conditions show that the forces maintaining these conformations are sensitive to external influences. The results obtained upon varying the pH and p2H show that protonation of the enzyme favors the presence of the low temperature form, ;.e., higher temperatures are required to effect the transition into the high temperature form at lower pH's. The pH dependence approximates a titration curve over the pH range studied, indicating that an amino acid residue(s) with a pK around 7.3 is responsible for this effect of protonation. The imidazole group of histidine 86 residues would most likely be involved under these circum— stances. The temperature range over which observations can be made is limited, since a transition temperature represents the midpoint of a curve. To determine this midpoint, data must be obtained at temperatures at least 10 - 120 on either side of the inflection point. Therefore, the limitations in these studies are physical; those of the freezing point of the solution and the point at which heat denaturation causes protein precipitation. The effect of substituting 2H20 for H20 also results in a favoring of the low temperature form. The effect of deuterium substitution can be explained in a number of ways. First, the actual substitution of readily exchangeable protons in the protein could lead to an increased strength of internal hydrogen bonding (Nemethy and Scheraga, 1964). Secondly, the structure of 2H20 is somewhat more ordered than H20 at the same temperature and transitions in these structures occur at higher temperatures in 2H2O (Nemethy and Scheraga, 1964). Finally, hydrophobic bonding, which is also probably dependent on the solvating water structure, is stronger in 2H20. The ionic strength dependence of the transition tempera— ture shows a marked favoring of the low temperature form at ionic strengths up to 1.0 M. An interpretation of this data would be very difficult since it would depend on the relative importance of ionic bonds, both intramolecular and those 87 between the protein and solvent, in maintaining the various conformations. Urea tends to disrupt protein structure (Tanford, 1964) and in this case favors the presence of the high temperature form. Although the effect is not as great as with pH changes, it should be considered significant since the urea concen— trations utilized were low (to avoid any dissociation phenomena, Steinmetz and Deal, 1966). Similar results were obtained at three different pH's. Triton X—100, which is known to be another structure—disrupting agent shows the same effect as urea on the transition temperature. Similar results were not obtained with SDS, an ionic detergent, because at the concentrations employed (0.05%) sufficient unfolding of the protein structure occurred so that the transition was not observed. The binding of cations to the enzyme always results in an increase in the transition temperature relative to the enzyme in TMA+, ;.g., the low temperature form of the enzyme is stabilized. KI, Na+ and Li+ all of which are known to bind, but not necessarily activate, by themselves show approximately the same transition temperature, an increase of 91. 60 over the value in TMA+. When the enzyme is nearly saturated with Mn2+ alone, the increase is gg. 110, and is dependent on the Mn2+ concentration. However, when it is near saturation with Ca2+ only a slight increase in the transition temperature is noted. It is interesting that optimal levels of a monovalent and saturating levels of a 88 divalent cation show a further increase in the transition temperature, to some 170 over that with only TMA+. This occurs when a non—activating monovalent cation (Li+) is used but not with the non—activating divalent cation (Ca2+). The effect of urea on the fully activated enzyme is similar to that observed with non—activated enzyme. It is apparent that in certain cases there is a definite relationship between the catalytic ability of the enzyme— -cofactor complex and the transition temperature. These results lend support to the idea that the conformational changes have some relationship to, or effect on, the activity of the enzyme. The temperature dependence of the catalytic activity of pyruvate kinase was studied as a test of this idea. Before extensive studies of the effect of pH on the temperature— —dependent conformational transition were conducted, a curvature in the Arrhenius plot for the enzyme was found (Kayne and Suelter, 1965). At the time,this was related to the transition occurring at that temperature (160), and the value ofAAH observed for the temperature—dependent confor— mational change was used for the calculation of the theoretical Arrhenius plot. Actually, the equilibrium constant at each temperature was determined (Figure 7) and this was used to calculatetx and (l-N), the coefficients in the extended Arrhenius equation. The studies presented here have shown that the initial assumption used was wrong, that is, the tempera— tures of the conformational transition and Arrhenius plot break were not the same under identical conditions. However, 89 the relatively constant values forlsH observed show that under most conditions the degree of curvature should be the same for all the theoretical Arrhenius plots based on this data. The original coincidence was due to the slightly different pH's used in the first conformational change study and in the normal assay. The degree of curvature in the Arrhenius plot will of course be related to the enthalpy change between the two forms of the enzyme, if this type of model is used for the explaination of the observed behavior. Massey has shown this in his recent publication on a similar effect in D—amino acid oxidase (Massey t al., 1966). The theoretical Arrhenius plots in his paper show a sharp break if A H's (conformational) of the order of 100 kcal/mole are used. The curvature seen in the plot of Figure 11 does correspond with Massey's calculations for a break relating to a transition with a A.H of some 25 - 50 kcal/mole. It can be noted here that at the present time, pyruvate kinase seems to be the only enzyme for which a curvature of the Arrhenius plot is well documented rather than a relatively sharp break as found in most other systems with anomalous Arrhenius plots. The Arrhenius plots for pyruvate kinase were studied under varying conditions relating to those studied for the conformational change dependence. Unfortunately, reproducible assays were obtainable only over a relatively narrow pH range, namely 7.4 — 8.0. The only other parameters easily varied were the ionic strength and deuterium substitution. 90 Of course it would be almost impossible to vary cations or inhibitors since the Arrhenius plot should strictly use Vmax limited. rates. However, Thus, Table VI compares data this portion of the study was somewhat from the "Results" section showing the relationship of the Arrhenius plot breaks found to the corresponding conformational transitions. The important observation here is that of the consistent relationship between the two, being gg. 20 / same conditions. the conformational transition higher than the Arrhenius plot breaks under the Table VI. Relationship of Arrhenius plot "breaks" to conformational transition. Solutions as in Table IV. Solution "Break" Transition Temp. 0C Temp. 0C a (estimated) b pH 8.0 not obsvd. (13-14) pH 7.8 14 15.5 pH 7.6 16 19 pH 7.4 20-22 (23) pH 7.6 (11.10.55) 21 (25) sz 7.8-7.9 (2H20) 22.5 25.5-29.5 aFrom data in Figure 10. bActual break may occur but at too low a temperature for accurate observation. The validity of these results could possibly be questioned since there are factors which temperature may effect other than the rate constant for the reaction. However, attempts 91 were made to see that at the extremes of temperature, the measured velocities were maximal, since, strictly speaking, Vmax should be plotted as a function of temperature. The results obtained in 2H20 are probably valid to the same extent with similar limitations. The reaction was run near the p2H optimum and this was checked at the extremes of temperature. The isotope effect studies suggest that the deuterium substitution may cause a protein conformation effect in addition to a rate effect on the catalytic reaction. However, earlier work with the enzyme in 2H20 showed that the deuterium substitution does not affect the relative activating ability of various monovalent cations (Coelho and Pinset-Harstrom, 1963). The large shift in transition temperature with deuterium substitution seen previously is paralleled by a corresponding shift in the Arrhenius plot break. A similar effect has also been demonstrated here in the case of fumarase where deuterium substitution shifts the Arrhenius plot discontinuity (in this case a sharp break) some 70 higher. The (acidic) pH dependence of the Arrhenius plot break found by Massey is not very large, which further suggests that in these cases the ZHZO effect is not due to a change in the degree of protonation of the enzyme, but due to solvent effects on the stability of the various conformations. A paper has recently appeared (Patat and Kolb, 1966) which suggeas that the results observed by Massey in the fumarase system are artifacts caused by the 92 temperature dependence of the extinction coefficient of fumarate and an irreversible inactivation of the enzyme at the higher temperatures. A discussion of the validity of these suggestions would be too lengthy for presentation here, but it can be stated that this explaination will not account for the observations on the fumarase reaction presented in these "Results". The conformational changes found in the pyruvate kinase system which have just been discussed can be interpreted using a number of models. For the purposes of this discus- sion, only one will be presented for an explaination. In Figure 17, the enzyme is represented by the wavy lines with the tryptophan residue(s) shown which is thought to be observed in the difference spectra. The lower line shows an equilibrium between two forms of the non—activated enzyme, one predominant at low temperature and the other at high temperature. The latter has what we shall call the "buried" tryptophan(s). This residue(s) may be exposed to a non— -aqueous environment and/or involved in hydrogen bonding, or, this simply may be the result of a number of tryptophans in the molecule becoming more constrained. The former has the same tryptophan(s) in the opposite condition (sense). It must be restated here however, that no significant differences in the degree of exposure of tryptophans to the external aqueous environment can be detected. pH, ionic strength, 2H20, urea and Triton are shown under this as they .ommCHx oum>ouhm CH momCmCO HMCOHDMEHOMCOO Um>nomno mCHpmemsHHH oEoCom HEOHuoCbomhm .RA assess 00.1x 20.1.5... ¢ _ u . 5 _ >5 \ \ as. + 22 + x x 02 L O ...4mh >mh .NH ohsmHm 95 are observed to affect the supposed equilibrium. The charged and potentially chargeable groups are shown only to illustrate that protonation of side chain groups and ionic strength effects can favor one protein form over another. This would be especially true if we are considering a relatively small conformational change. The 2H20, urea and Triton data have indicated, somewhat suprisingly, that the conformation with the tryptophan ”buried" is probably the more unfolded. To reiterate here, 2H20 should slightly favor hydrophobic bond formation and the more ordered protein structure in general. The upper line illustrates the effect of binding the activating cations on the equilibria involved. The bridged structure shown is not necessary for the interpretation. Basically the same transition is seen in this case, however, at a considerably higher temperature. This was shown in Table III to depend on the amount and nature of the cations bound. Their addition clearly favors the low temperature form. In relating the observed conformational transitions to the break in the Arrhenius plot we must realize that the latter can be observed only under conditions where both monovalent and divalent activating cations are present. At this point it can be seen that, in the temperature ranges where breaks on the Arrhenius plots do occur, no confor— mational changes are seen in the activated enzyme by differ- ence spectroscopy. The inference is that they do indeed occur, 96 but to a limited extent. Apparently, they are only seen by changes in the E and, are probably related to the act) conformational transitions observed in the non—activated enzyme by difference spectroscopy. The evidence for this latter point is admittedly only circumstantial, that is, it is based on the similar behavior of the observed transitions and Arrhenius plot breaks under varying conditions. However, it is felt that the evidence is strong enough for the suggestion of this model as an explaination for the observations. One observation should be noted here which was not presented in the results because of difficulties in its measurement and interpretation. In determining reaction velocities for the high temperature ranges of the Arrhenius plots, the measured rates became non—linear after a certain time. This time interval decreased with increasing tempera- ture until the rates were entirely non—linear. Parallel behavior was seen at differing pH's, the temperature at which this began being lower at the higher pH values. This behavior did not seem to be related to the initial velocities and did not effect the Arrhenius plot breaks since it is observed until temperatures considerably above the break point are reached. The explaination involves an irreversible inactivation of the enzyme with an apparent first order loss of activity. This is observed only when the enzyme is diluted to assay 97 levels and can be eliminated by the inclusion of bovine serum albumin at 0.1 mg/ml in the reaction (unpublished experiments in this laboratory by R. Kuczenski). This behavior may be related to the high temperature conformational change of the activated enzyme. Independent evidence has been found to support some of the aspects of the model just suggested. The results of the ORD and sedimentation velocity experiments indicate that the low temperature form of the enzyme has a more compact structure than the high temperature forms. The consist- antly higher sedimentation velocities in the presence of the activating cations indicate that this form of the enzyme is either more compact (less assymetric) or has a higher molecular weight. The differences in 520,w do not seem large enough to be characteristic of a molecular weight change. It must be pointed out though, that for a rigorous exclusion of this possibility, the corresponding values of the diffusion coefficient must be determined and the molecular weight determined for both cases. Alternatively, the molecular weights would have to be determined from sedimen— tation equilibrium experiments. It is also of interest here that the Szo,w values increase at the lower pH value. Although these differences are just outside the range of experimental error, they are consistent and agree with the model presented in that decreasing pH favors the more compact form of the enzyme. 98 It is unfortunate that the 520,w differences in these cases are as small as they are, since this makes it impossible to see any change in the value as a function of temperature. The experiment performed should have shown a maximal diff— erence between the high and low temperature forms of the non—activated enZyme. It can be seen from the results, that the partial specific volume term in the correction of s to 520,w gives rise to too much uncertainty in the corrected values. This depends on the temperature coefficient assumed for the partial specific volume which is unknown for pyruvate kinase. The values used show how greatly this effects the result and it is concluded that the differences are not large enough to see unless the partial specific volume of the protein is determined under each of the conditions. Optical rotatory dispersion measurements have been used as a means of estimating the degree of helicity of various proteins (Jirgensons, 1965; Kronman §:_a;., 1965). The slope of the Moffitt—Yang plot, b0 seems to be the parameter which best describes the d.-helix content of proteins (Carver e: a;., 1966). The ao parameter can also be used for helix estimation but it is much more sensitive to side chain effects than b0. Increasing 0(—helix content is usually indicated by ac values becoming less negative and b0 values becoming more negative. In general, this change would be towards the more ordered protein structure. The values for b0 and a0 in Table V, indicate an increased 99 degree of N.—helicity in the presence of the activating cations. Addition of ADP to the enzyme results in Moffitt parameters indicating some changes in protein conformation may be taking place. No additional protein difference spectra were observed under these conditions. The addition of SDS to the enzyme shows what is probably little change in the degree of helicity. But the large changes in a0 observed may indicate extensive changes in conformation, possibly in the degree of exposure of the chromophores. Addition of 1.5 M urea shows some side chain effects but in the opposite direction from SDS and no change in b0 which is what would be expected from the model in Figure 17 under these conditions. Lowering the temperature of the non—act— ivated enzyme is seen clearly to favor a relatively more ordered form of the protein. It is of some interest that under conditions where conformational changes seem to occur and are just detectable by changes in the ORD parameters, no anomalous dispersion (intrinsic or extrinsic Cotton effects) is seen which might be caused by a perturbation of the aromatic residues(Myers and Edsall, 1965). This is further substantiated by the absence of circular dichroism differences in the 250-300 mp range of protein and substrate absorption, which shows not only the absence of protein effects but the absence of any extrinsic Cotton effects upon binding ADP. Although the ORD differences seen here are very small, their consistancy is additional proof for the occurrance of the conformational 100 changes outlined in Figure 17. Support for the model with the more compact forms of the protein being present at low temperatures or in the presence of activating cations is given by studies on the polarization of fluorescence of the enzyme (Suelter, 1967). A marked increase in polarization is noted under these conditions which indicates that the emitting chromophores (tryptophans) are more constrained at the low temperatures or upon addit— ion of the activating cations. This is in complete agreement with the model of Figure 17. The preceeding results and discussion lead to the con— clusion that pyruvate kinase undergoes conformational transitions with changes in temperature. These can be directly observed by a number of physical methods and a portion of the total change seems to affect a parameter of the catalytic reaction, the energy of activation. The individual observations are adequately explained on the basis of a temperature-dependent conformational equilibrium between two forms of the enzyme. The free energy for this process is strongly influenced by environmental effects on the protein. The greatest effect on the equilibrium is that produced by binding the monovalent and divalent cations required by the enzyme for catalytic activity. In the consideration of the activated enzyme, a three state model is really required to explain all the observations. At any one temperature, however, it is most likely that only two 101 states predominate. The argument for a two state system is not very import— ant and here, as in the cases for the D—amino acid oxidase conformational change (Massey g: al., 1966) and Changeux's model explaining the allosteric kinetics of certain enzymes (Monod QE a;., 1965), it is principally based on the fact that it is the simplest model which can explain the observat- ions. The term ”two state” referring to the conformationally differing forms should not be taken too rigidly, since an obvious extension of this is to consider the ”states" actually as populations of conformers whose average differ— ences are those observed. Although muscle pyruvate kinase does not seem to behave as an allosteric enzyme in the definition as it is now accepted, the demonstration of this conformational change shows the enzyme to be capable of the same type of behavior characteristic of allosteric enzymes. That is, the cations (and probably the substrates as well) are preferentially bound to one form of the enzyme, and this binding serves to shift the equilibrium towards that of the somewhat stabilized form. This work may serve as an indication that a more critical examination should be made of many similar systems studied previously or those which will be studied in the future. It is very likely that many investigators have overlooked small conformational changes as the possible cause for their observations. These observations may be 102 one of any number, like activator or inhibitor binding, substrate binding or enzyme stability under certain cond— itions. This author has found a number of examples in the recent literature where small differences in physical parameters may have been observed, but were not recognized as being significant indication of a conformational change. This attitude may have been brought about by previous observations of gross conformational changes (g.q., assoc- iation or dissociation of protein subunits) in some enzymes under these types of conditions. In searching for these, the smaller changes may have gone unnoticed. There is no reason to expect that either the small or large conformat- ional changes would be more important. Indeed, Massey t QM. (1966) suggest that a relatively small temperature—dependent change in D-amino acid oxidase gives rise to association as a larger secondary change. As they have suggested and this work suggests, it may be fruitful to study enzyme systems where Arrhenius plot breaks are known to occur, as a source of material for conformational change studies. It may well be some time before it is known how wide spread this type of behavior is in the various enzymes. The actual significance of the observations presented in this thesis is not known in as far as their occurrance and meaning l2 yiyg is concerned. But it is of interest to note that some of the conditions under which the conformational Changes were observed to occur are not radically different 105 from what one would expect for the muscle enzyme 12 3322. These would be ionic strengths of around 0.15 M, the presence of potassium and divalent activating cations, pH values around 7.0 (Bittar, 1964) and protein concentrations greater than 1 mg/ml. Under these conditions the transition may take place around 36-400, certainly a physiological temperature range for mammalian muscle. The only large deviation lg vitro would be the lack of opportunity to observe effects produced by association of this enzyme with other cellular components. SUMMARY Rabbit muscle pyruvate kinase has been shown to undergo conformational transitions upon the addition of substrates, activating cations or changing the temperature of the solution. These changes were observed by: tryptophan and tyrosine ultraviolet difference spectra, changes in the sedimentation velocity of the protein and ORD Moffitt plot changes. No apparent differences in tryptophan exposure were detected chemically, but some changes, possibly due to to sulfhydryl exposure differences were observed. The catalytic activity of the enzyme shows Arrhenius plots which are linear at extremes of temperature, but exhibit a curvature around certain temperatures. This behavior has been related to the temperature—dependent trans— itions observed under various conditions. It was suggested that the same, or part of the same conformational change in the enzyme is responsible for these effects. A model has been proposed to explain the observations and this is used to illustrate the effects various environmental parameters may have upon the protein conformation. 104 REFERENCES Aki, K., Takagi, T., Isemura, T., and Yamano, T. (1966), Biochim. Biophys. Acta 122, 193. Amelunxen, R.E., Carr, D.C., and Grisolia, S. (1966), Biochim. Biophys. Acta 113, 184. Bayley, P.M., and Radda, G.K. (1966), Biochem. J._g§, 105. Beck, K., Gill, S.J., and Downing, M. (1965), J. Am. Chem. Soc. 81, 901. Bencze, W.L., and Schmid, K. (1957), Anal. Chem. 29J 1193. Benmouyal, P., and Trowbridge, C.G. (1966), Arch. Biochem. Biophys. 115, 67. Benson, E.S., Hallaway, B.E., and Lumry, R. (1964), J. Biol. Chem. 259, 122. Bernhard, S.A., Lee, B.F., and Tashjian, Z.H. (1966), J. Mol. Biol. E, 405. Bigelow, C.C., and Krenitsky, T.A. (1964), Biochim. Biophys. Acta 88, 130. Bittar, E.E. (1964), Cell pH, Washington, D.C., Butterworths, p. 63. Boyer, P.D. (1962), The Enzymes 6, 95. Boyer, P.D., Lardy, H.A., and Phillips, P.H. (1943), J. Biol. Chem. 149,-529. Brénden, C. (1965), Arch. Biochem. Biophys. 112, 215. Brandts, J.F. (1964), J. Am. Chem. Soc..8§J-429l. Bresler, S., Firsov, L., and Glasunov, E. (1966), Nature 211, 1262. Breslow, E. (1964), J. Biol. Chem. 239, 486. Brewer, J.M., and Weber, G. (1966), J. Biol. Chem. 241, 2550. 105 106 Bficher, T., and Pfleiderer, G. (1955), Methods Enzymol. A, 42. / Burr, M., and Koshland, D.E., Jr. (1964), Proc. Natl. Acad. Sci. U.S. 52, 1017. Carver, J.P., Schechter, E., and Blout, E.R. (1966), J. Am. Chem. Soc. 8Q, 2562. Churchich, J.E. (1966), Biochim. Biophys. Acta 120, 406. Coelho, R(, and Pinset—Harstrom, I. (1963), J. Chim. Phys. 69, 524. Cohn, M. (1963), Biochemistry 2, 623. Coleman, J.E. (1965), Biochemistry 4, 2644. Coleman, J.E., and Vallee, B.L. (1964), Biochemistry :, 1874. Cowgill, R.W. (1964), Arch. Biochem. Biophys. 294, 84. Cowgill, R.W. (1966),Biochim. Biophys. Acta l_g, 196. Datta, S.P., and Grzybowski, A.K. (1961), in Biochemists' Handbook, C. Long, ed., Princeton, N.J., Van Nostrand Co., p. 44. Doty, P., and Edsall, J.T. (1951), Adv. Prot. Chem. 6, 55. Ebata, M., Tsunoda, J., and Yasunobu, K.T. (1966), Biochem. Biophys. Res. Commun. 22, 455. Edsall, J.T., and Wyman, J. (1958), Biophysical Chemistry, New York, Academic Press, p. 464. Evans, H.J., and Sorger, G.J. (1966), Ann. Rev. Plant Physiol. Ml, Fasella, P., and Hammes, G.G. (1964), Biochim. Biophys. Acta 92, 630. Frattali, V., Steiner, R.F., and Edelhoch, H. (1965), J. Biol. Chem. 240, 1127122. Glasoe, P.K., and Long, F.A. (1960), J. Phys. Chem. 53, 188. Glazer, A.N. (1965), Proc. Natl. Acad. Sci. U.S. 54, 171 Green, N.M. (1963), Biochem. J. @9, 599. 107 Habeeb, A.F.S.A. (1966), Biochim. Biophys. Acta 115, 440. Hanlon, D., and Westhead, E. (1965), Biochim. Biophys. Acta 6. 537. Harrington, W.F., Josephs, R., and Segal, D.M. (1966), Ann. Rev. Biochem. 55, 626. Harrison, S.C., and Blout, E.R. (1965), J. Biol. Chem. 240, 299. Havsteen, B.H. (1965), Acta Chem. Scand. 29, 1643. Havsteen, B.H., and Hess, G.P. (1962), J. Am. Chem. Soc. 55, 491. Hayashi, K., Imoto, T., and Funatsu, M. (1964), J. Biochem. 554 516. Hedricks, J.L. (1966), Arch. Biochem. Biophys. 114, 216. Hermans, J., Jr., Donovan, J.W., and Scheraga, H.A. (1960), J. Biol. Chem. 235, 91. Hermans, J., Jr., and Scheraga, H.A. (1961), J. Am. Chem. Soc. 8:. 3283. Herskovits, T.T., and Laskowski, M., Jr. (1962), J. Biol. Chem. 237, 3418. Hess, B., Haeckel, R., and Brand, K. (1966), Biochem. Biophys. Res. Commun. 25, 824. Horton, H.R., and Koshland, D.E., Jr. (1965), J. Am. Chem. Soc. 81, 1126. Hulett, J.R. (1964), Quart. Rev. 1&. 227. Hunsley, J.R. (1966), M.S. Thesis, Michigan State University, East Lansing, Mich. Hvidt, A., and Nielsen, 8.0. (1966), Adv. Prot. Chem. 2;, 287. Inada, Y., Matsushima, A., Kamata, M., and Shibata, K. (1964), Arch. Biochem. Biophys. 106, 326. Iyengar, M., Glauser, S.C., and Davies, R.E. (1964), Biochem. Biophys. Res. Commun. 25, 379. 108 Jirgensons, B. (1965), J. Biol. Chem. 240, 1064. Kachmar, J.F., and Boyer, P.D. (1955), J. Biol. Chem. 200, 669. Kayne, F.J. (1966), Federation Proc. 25, 346. Kayne, F.J., and Suelter, C.H. (1965), J. Am. Chem. Soc. 51, 897. Kistiakowsky, G., and Lumry, R. (1949), J. Am. Chem. Soc. 1;, 2006. Kronman, M., Blum, R., and Holmes, L. (1965), Biochem. Biophys. Res. Commun. 59, 227. Kubowitz, F., and Ott, P. (1944), Biochem. Z. 312, 193. Kupiecki, F.P., and Coon, M.J. (1960), J. Biol. Chem. 235, 1944. Leach, S.J., and Hill, J. (1963), Biochemistry 2, 807. Libor, S., E18di, P., and Nagy, Z. (1965), Biochim. Biophys. Acta 110, 484. Listowsky, I., Furfine, C., Betheil, J., and Englard, S. (1965), J. Biol. Chem. 240, 4253. Lovrien, R. (1963), J. Am. Chem. Soc. 55, 3677. Lui, N.S.T., and Cunningham, L. (1966), Biochemistry 5, 144. Magar, M.E. (1965), Biochim. Biophys. Acta 95, 345. Mandel, M. (1964), Proc. Natl. Acad. Sci. U.S. 52, 36. Martin, C., and Bhatnagar, G. (1966), Biochemistry 5, 1230. Massey, (1953a), Biochem. J. 55, 67. (1953b), Biochem. J. 55, 72. Massey, , Curti, B., and Ganther, H. (1966), J. Biol. 41, 23477 2358. Massey, V V Massey, V., and Curti, B. (1966), J. Biol. Chem. 25;, 3417. V Chem. 2 Melchoir, J.E. (1965), Biochemistry fl, 1518. 109 Mercouroff, J., and Hess, G.P. (1963), Biochem. Biophys. Res. Commun. 2;, 283. Michaelides, M.C., and Helmreich, R. (1966), Enzymol. Biol. Clin. 1, 130. Mildvan, A.S., and Cohn, M. (1963), Abstracts of the 145th National Meeting of the American Chemical Society, New York, N.Y., Sept. 9—13, p. 82C. Mildvan, A.S., and Cohn, M. (1965), J. Biol. Chem. 40, 238. Mildvan, A.S., and Cohn, M. (1966), J. Biol. Chem. 241, 1178. Mildvan, A.S., and Leigh, R.A. (1964), Biochim. Biophys. Acta 92, 393- Moffitt, W., and Yang, J.T. (1956), Proc. Natl. Acad. Sci. U.S. 42, 596. Monod, J., Wyman, J., and Changeux, J.P. (1965), J. Mol. Biol. 12: Morita, F., and Yagi, K. (1966), Biochem. Biophys. Res. Commun. 22, 297. Myers, D.V., and Edsall, J.T. (1965), Proc. Natl. Acad. Sci. U.S. 5:, 169. Nemethy, G., and Scheraga, H.A. (1964), J. Chem. Phys. 4;, 680. Noguchi, H., Kasarda, D., and Rainford, P. (1964), J. Am Chem. Soc. 85, 2077. Patat, F., and Kolb, H.J.(l966), Z. Naturforschung 21B, 32. Plowman, K., and Krall, A. (1965), Biochemistry 4, 2809. Reynard, A.M., Hass, L.F., Jacobsen, D.D., and Boyer, P.D. (1961), J. Biol. Chem. 236, 2277. Rosenberg, A., and Lumry, R. (1964), Biochemistry 5, 1055. Schellman, J.A., and Schellman, C. (1964), The Proteins 2 (2nd. ed.), 1. Schnieder, H., Kresheck, G.C., and Scheraga, H.A. (1965), J. Phys. Chem. 69, 1310. 110 Siebert, G., Kesselring, K., and Fisher, F. (1965), Z. Physiol. Chem. 341, Singleton, R., Jr. (1963), M.S. Thesis, Michigan State University, East Lansing, Mich. Slayter, E.M. (1965), J. Mol. Biol. 14, 443. Sophianopoulos, A.J., and Weiss, B.J. (1964), Biochemistry 1, 1920. Sorger, G.J., Ford, R.E., and Evans, H.J. (1965), Proc. Natl. Acad. Sci. U.S. 54, 1614. Spande, T.F., Green, N.M., and Witkop, B. (1966), Biochemistry 5, 1926. Srere, P. (1965), Arch. Biochem. Biophys. 110, 200. Stauff, J., and Wolf, H. (1964), Z. Naturforshung 292, 87. Stearn, A.E. (1949), Adv. Enzymol. 9, 25. Steiner, R.F. (1964), Biochim. Biophys. Acta Z9J 51. Steiner, R.F. (1966), Biochemistry 5, 1964. Steinmetz, M.A. (1966), Ph.D. Thesis, Michigan State Univer— sity, East Lansing, Mich. Steinmetz, M.A., and Deal, W.C., Jr. (1966), Biochemistry 5, 1399- Suelter, C.H. (1967), Biochemistry, in press. Suelter, C.H., and Melander, W. (1963), J. Biol. Chem. 238, PC4108. Suelter, C.H., Singleton, R., Jr., Kayne, F.J., Arrington, 8., Glass, J., and Mildvan, A.S. (1966), Biochemistry 5, 131. Suelter, C.H., and Weber, G. (1967), J. Am. Chem. Soc. in press. Swinehart, J.H., and Hess, G.P. (1965), Biochim. Biophys. Acta 104, 205. Tanford, C. (1964), J. Am. Chem. Soc. 86, 2050. Taylor, J.F., and Lowry, C. (1956), Biochim. Biophys. Acta 29, 109. 111 Tietz, A., and Ochoa, S. (1958), Arch. Biochem. Biophys. lg: 477- Torchinsky, Y.M., Livanova, N.B., and Pikhelgas, V.V. (1965), Biochim. Biophys. Acta 110, 619. Ullmann, A., Vagelos, P.R., and Monod, J. (1964)) Biochem. Biophys. Res. Commun. 21, Urnes, P., and Doty, P. (1961), Adv. Prot. Chem. 25, 401. Warburg, 0., and Christian, W. (1942), Biochem. Z. 529, 384. Warner, R. (1958), Arch. Biochem. Biophys. 8, 494. Warren, J.C., and Cheatum, S.G. (1966), Biochemistry 5, 1702. Watts, D.C. (1966), Biochem. J. 199, 13C. Wellner, D. (1966), Biochemistry 5, 1585. Westhead, E.W. (1964), Biochemistry 5, 1062. Wetlaufer, D.B., and Lovrien, R. (1964), J. Biol. Chem. 239, 596- Williams, E.J. (1966), Arch. Biochem. Biophys. 115, 21. Williams, B.J., Herskovits, T.T., and Laskowski, M., Jr. (1965), J. Biol. Chem. 240, 3574. Yankeelov, J.A., and Koshland, D.E., Jr. (1965), J. Biol. Chem. 240, 1593. IIIIIIIIIIIIIIIIIIIIIIIII 1qu11mu111mum111111111mum