REvzassaLE Dissocmnomr f '~ PEG LWER GLYCERALDEHYDE .. 3 ~PHOSPHA’E'E .DEHYDROGEHASE ‘ Thais forfhefiegreeofMS ' ‘ . , > Macrame sms msvaaism » ‘ — * mmm‘x waits-mam ; L; . LIBRAR Y :2 Michigan State ; University 3" O . .IIGfiv-H ,5 ll 3- °‘ 800K BINQERY NC. ' [riff-6:: ABSTRACT REVERSIBLE DISSOCIATION OF PIG LIVER GLYCERALDEHYDE 3-PHOSPHATE DEHYDROGENASE BY ADENOSINE TRIPHOSPHATE BY Kathryn DePree Mascaro Glyceraldehyde 3-P dehydrogenase is inactivated and dissociated by ATP at 0°. The effects of incubation time, EDTA, protein concentration, pH, ATP concentration, mercaptoethanol concentration and temperature on this inactivation and dissociation were measured by the oxida- tive phosphorylation assay and by sucrose density gradient centrifugation respectively. Optimal conditions for inactivation and dissociation were shown to be 12 hours incubation, 0.1 mg/ml protein, pH 7.5 to 9.0, 15 mM ATP and 0°. The tetrameric enzyme dissociated to a dimer with a sedimentation coefficient of 4.9 to 5.2 S. If produced in the presence of 0.1 M EDTA and 40 to 80 mM mercaptoethanol, the dimer was active (50 per cent as much activity per mg protein as the tetramer). At low glyceraldehyde 3-P dehydrogenase concentrations (0.025 mg/ml), the major dissociation product was inactive monomers . «:{6 ) Kathryn DePree Mascaro Tifb By warming to 25° for only five minutes, glycer- aldehyde 3-P dehydrogenase dimers could be reactivated as measured by the oxidative phosphorylation assay. Reactivation required 0.1 M EDTA and 80 mM mercaptoethanol. Adding NAD increased inactivation; adding sucrose decreased it. In the absence of ATP, the reactivated enzyme was a reassociated tetramer with a sedimentation coefficient of 7.8 S as measured by sucrose density gradient cen- trifugation. This sedimentation coefficient is the same as that of native glyceraldehyde 3-P dehydrogenase. The amino acid composition of the enzyme was obtained by analysis of the acid-hydrolysate in a com- mercial amino acid analyzer. The sulfhydryl reactivity of native glyceraldehyde 3-P dehydrogenase was studied by the titration of sulfhydryl groups with 5, 5'- dithiobis-(2-nitrobenzoic acid). Both the amino acid composition and the sulfhydryl reactivity Were very similar to those of other glyceraldehyde 3-P dehydro- genases that have been studied. The requirement for EDTA in the production of active dimers and in reactivation suggested that glycer- aldehyde 3-P dehydrogenase is sensitive to metal ions and that a metal ion(s) may be present on the crystalline enzyme. The possibility that this metal ion is zinc and the effects such a metal ion could have on glyceraldehyde 3-P dehydrogenase were discussed at length. Kathryn DePree Mascaro The purification procedure for pig liver glycer- aldehyde 3-P dehydrogenase developed by Dr. Dagher of this laboratory was confirmed and expanded to a larger scale. REVERSIBLE DISSOCIATION OF PIG LIVER GLYCERALDEHYDE 3-PHOSPHATE DEHYDROGENASE BY ADENOSINE TRIPHOSPHATE BY Kathryn DePree Mascaro A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1974 ACKNOWLEDGMENTS This author wishes to express her appreciation to Dr. Deal for his guidance throughout the course of this work. The assistance of the other members of her guidance committee, Dr. Kindel and Dr. Wilson, is also gratefully acknowledged. The author thanks Dr. Kindel for teaching her the fine art of column chromatography. The technical assistance of Maria Shimamoto in analyzing the amino acid composition is greatly appreciated. This author also wishes to express thanks to her husband, Leonard Mascaro, for his assistance, suggestions and patience during her graduate career." ii INTRODUCTION TABLE OF CONTENTS LITERATURE REVIEW . . . . . . . . . I. II. III. IV. Reactions Catalyzed by Glyceraldehyde 3-P Dehydrogenase . . . . . . . . A. B. Positive and Negative Cooperativity and and Reductive DephOSphorylation. Other Reactions Catalyzed by Glyceralde- hyde 3-P Dehydrogenase. . . . Half-of-the-Sites Reactivity . . . Molecular Structure of Glyceraldehyde 3-P Dehydrogenase . . . . . . . . A. B. C. D. E. Amino Acid Composition. . . . Sulfhydryl Groups . . . . . Binding Sites for NAD, NADH, Glyceralde- hyde 3-P and 1,3-Diphosphoglycerate Amino Acid Residues in the Active Site x-ray Crystallographic Studies . Mechanism for Oxidative Phosphorylation Isozymes of Glyceraldehyde 3-P Dehydrogenase. A. B. Isozymes from the.Same Tissue . Isozymes from Different Tissues. iii Page 10 11 13 14 15 15 16 Page V. Dissociation and Reassociation of Subunits . 17 VI. Effects of ATP . . . . . . . . . . 22 A. Immediate Inhibition . . . . . . . 23 B. Time-Dependent Inactivation and Dis- sociation. . . . . . . . . . . 24 C. Stimulation of Proteolytic Digestion. . 26 VII. Evidence for and Against the Binding of Zinc to Glyceraldehyde 3-P Dehydrogenase. . 27 VIII. Physical and Chemical Properties of Pig Muscle and Pig Liver Glyceraldehyde 3—P Dehydrogenases . . . . . . . . . . 29 MATERIALS AND METHODS. . . . . . . . . . . 33 I. Reagents . . . . . . . . . . . . 33 II. Enzyme Preparation . . . . . . . . . 33 III. Enzyme Assay . . . . . . . . . . . 36 IV. Protein Determination . . . . a. . . . 36 V. Inactivation, Dissociation, Reactivation and Reassociation Procedures. . . . . . 37 VI. Amino Acid Analysis. . . . . . . . . 39 VII. DTNB Reaction. . . . . . . . . . . 40 RESULTS . . . . . . . . . . . . . . . 41 I. Inactivation and Dissociation of Pig Liver Glyceraldehyde 3-P Dehydrogenase . . . . 41 A. Effect of Incubation Time . . . . . 43 B. Effect of EDTA . . . . . . . . . 48 iv C. Effect of Protein Concentration . . . D. Effect of pH . . . . . . . . . E. Effect of ATP Concentration . . . . F. Effect of Mercaptoethanol . . . . . G. Effect of Temperature . . . . . . II. Reactivation and Reassociation of Glycer— aldehyde 3-P Dehydrogenase . . . . . . III. Amino Acid Composition . . . . . . . IV. Sulfhydryl Reactivity of Native Glyceral- dehyde 3-P Dehydrogenase. . . . . . . DISCUSSION I. EDTA Effects and the Possible Presence and Role of Zinc. . . . . . . . . . . II. Dissociation into Dimers. . . . . . . III. Enzymatic Activity of Dissociated Glycer- aldehyde 3-P Dehydrogenase . . . . . . IV. Comparison of Inactivation and Dissociation of Pig Liver GAPD with Those of Other GAPDs . . . . . . . . . . . . . V. Comparison of Reactivation and Reassoci- ation of Pig Liver GAPD with Those of Other GAPDS . . . . . . . . . . . VI. Comparison of Amino Acid Composition with that of Other GAPDS . . . . . . . . Page 53 58 61 68 73 80 91 94 98 102 105 107 110 111 Page VII. Comparison of Sulfhydryl Reactivity of Pig Liver GAPD with that of Other GAPDS O O O 112 SUMMARY AND CONCLUSIONS. . . . . . . . . . 115 LIST OF REFERENCES . . . . . . . . . . . 119 vi Table 1. LIST OF TABLES Dissociation and Reassociation of Glyceralde- hyde 3-P Dehydrogenase . . . . . . . . Reactivation of Inactivated GAPD in the Presence of Different Compounds . . . . . The Effect of Incubation Time at 0° on Reactivation . . . . . . . . . . . Amino Acid Composition . . . . . . . . vii Page 19 82 83 92 Figure 1. 10. 11. 12. 13. LIST OF FIGURES General Mechanism for Oxidative Phosphorylation and Reductive Dephosphorylation . . . . . . Effect of Incubation Time at 0° on Inacti- vation of Pig Liver GAPD by ATP . . . . . . Effect of Incubation Time at 0° on Dissoci- ation of Pig Liver GAPD by ATP . . . . . . Typical Sedimentation Patterns for Native and ATP-dissociated Pig Liver GAPD . . . . . . Effect of EDTA on Inactivation of Pig Liver GAPD by ATP at 0°. . . . . . . . . . . Effect of Protein Concentration on Inactivation of Pig Liver GAPD by ATP at 0° . . . . . . Sedimentation Patterns for Native (Upper Frame) and ATP-dissociated (Lower Frame) Pig Liver GAPD at Various Protein Concentrations. . . . Effect of pH on Inactivation of Pig Liver GAPD by ATP at 0 o O O O O O O O O O O O 0 Effect of pH on Dissociation of Pig Liver GAPD by ATP a t O o O O O O O O O O O O O O Sedimentation Patterns for Native (Upper Frame) and ATP-dissociated (Lower Frame) Pig Liver GAPD at Different Values of pH . . . . . . Effect of ATP Concentration on Inactivation of Pig Liver GAPD by ATP at 0° . . . . . . . Effect of ATP Concentration on Dissociation of Pig Liver GAPD by ATP at 0° . . . . . . . Effect of Mercaptoethanol Concentration on Inactivation of Pig Liver GAPD by ATP at 0° . . viii Page 45 47 50 52 55 57 60 63 65 67 70 72 Figure Page 14. Effect of Temperature on Inactivation of Pig Liver GAPD by ATP . . . . . . . . . . 75 15. Effect of Temperature on Dissociation of Pig Liver GAPD by ATP . . . . . . . . . . 77 16. Sedimentation Patterns for Native and ATP- dissociated Pig Liver GAPD at Different Temperatures. . . . . . . . . . . . 79 17. Kinetics of Reactivation of Pig Liver GAPD Inactivated by ATP at 0°. . . . . . . . 86 18. Sedimentation Pattern of Reassociated Pig Liver GAPD Centrifuged Without ATP . . . . 88 19. Sedimentation Pattern of Reassociated Pig Liver GAPD Centrifuged with ATP . . . . . 90 20. Sulfhydryl Reactivity of Native Pig Liver GAPD O O O O O O O O O O O O O O 96 ix BSA CAMP Cys DPGA DTNB GAPD G3P His Lys ORD PCMB PGA SDS LIST OF ABBREVIATIONS bovine serum albumin adenosine 3',5'-cyclic monophosphate cysteine l,3-diphosphoglycerate 5,5'-dithiobis-(2-nitrobenzoic acid) enzyme glyceraldehyde 3-phosphate dehydrogenase glyceraldehyde 3-phosphate histidine lysine optical rotary dispersion p-chloromercuribenzoate phOSphoglyceric acid inorganic phOSphate sodium dodecyl sulfate INTRODUCTION Glyceraldehyde 3-ph05phate dehydrogenase (D- glyceraldehyde 3-phosphate:NAD oxidoreductase phosphory— lating, EC 1.2.1.12) is an enzyme in glycolysis. It catalyzes the only oxidative step in the pathway and pro- duces one of the two phosphorylated intermediates of gly- colysis that can be used directly to make ATP. The enzyme has been isolated from bacteria, fishes, birds, mammals (Allison and Kaplan, 1964) and plants (Duggleby and Dennis, 1974). From all sources tested, except plants (where three different GAPDS are found [Rosenberg and Arnon, 1955; Fuller and Gibbs, 1959; Hageman and Arnon, 1955]), GAPD is a tetramer of approximately 150,000 daltons which binds one NAD per monomer (Watson, Duee and Mercer, 1972). Because of its relative ease of preparation in large quan- tities, its importance as a model for oxidative phos- phorylation, its highly reactive sulfhydryl groups, the Cooperativity of its enzyme-coenzyme interactions, its multiple activities and its subunit-subunit interactions, GAPD has been a favorite enzyme for chemical, physical and kinetic investigations. Since few studies had been reported on GAPD from a gluconeogenic tissue, the aim of this research was to examine pig liver GAPD in hopes of gaining some under- standing of how liver (gluconeogenic) GAPD differs from muscle (glycolytic) GAPD. The examination consisted of studies on the dissociation-association properties and on the amino acid composition. Reversible dissociation and inactivation of GAPD by ATP at 0° had been demonstrated for the rabbit muscle (Constantinides and Deal, 1969) and yeast enzymes (Stancel and Deal, 1969). Dr. Dagher of this laboratory attempted to demonstrate similar behavior by pig liver GAPD, but was only partially successful; dissociation and inacti- vation were difficult to achieve reproducibly and little reactivation was observed. Since ZnCl2 precipitation was a step in the purification procedure for pig liver GAPD (Dagher and Deal, in preparation) and Keleti found zinc bound to several GAPDS (Keleti gt 21,, 1962), we reasoned that zinc might be bound to the enzyme, causing it to exhibit unexpected stability to inactivation, dissociation and reactivation. Consequently, we added 0.1 M EDTA to the reaction mixture in some preliminary studies on the dissociation of pig liver GAPD by dilution. The results of these experiments indicated that GAPD dissociated more readily in the presence of EDTA than in its absence. We then decided to re-attempt the ATP dissociation using EDTA. These experiments led to a characterization of the reversible dissociation and inactivation by ATP at 0° in the presence of EDTA. Amino acid compositions often reflect the dif- ferences between various enzymes because the primary sequence determines the enzyme's structure and conse- quently its function. Thus a difference in amino acid composition might be expected for muscle and liver GAPDS which perform different functions in_xizg; muscle GAPD mostly catalyzes oxidative phosphorylation because muscle is a glycolytic tissue, while liver GAPD catalyzes both oxidative phosphorylation and reductive dephosphorylation because liver is a gluconeogenic tissue. In hopes of finding such a difference, the amino acid composition of pig liver GAPD and the reactivity of the sulfhydryl groups in the native enzyme were studied, despite the fact that all mammalian GAPDS studied in detail thus far have nearly identical amino acid sequences (Harris, 1970) and that no major structural differences have yet been found between muscle and liver GAPDS (Bondi, Watkins and Kirtley, 1969; Smith and Velick, 1972; Lambert and Perham, 1974). LITERATURE REVIEW REACTIONS CATALYZED BY GLYCERALDEHYDE 3-P DEHYDROGENASE. Mechanism for Oxidative Phosphorylation and Reductive Dephosphorylation. The main physiological role of GAPD is to catalyze oxidative phosphorylation and reductive dephosphorylation. The general mechanism proposed for these two reactions by Duggleby and Dennis (1974) is shown in Figure l. acyl-E NADH I. NAD : ‘3 PGA (acyl-E-NADH) i acyl-E-NAD E . 4 T Pi NAD (G3P-E-NAD) acyl-E-NAD / Pi G3P } DPGA ELNAD Figure 1. General Mechanism for Oxidative Phosphorylation and Reductive Dephosphorylation This mechanism is based on the results of studies of pea seedling GAPD (Duggleby and Dennis, 1974), lobster and sturgeon muscle GAPDS (Trentham, 1971) and rabbit muscle and liver GAPDS (Smith and Velick, 1972). Oxidative phosphorylation involves the formation of an acyl enzyme thioester produced by NAD oxidation of a presumed thiohemiacetal adduct of G3P and GAPD (Racker, 1965); hydrogen is transferred directly from the l-position of G3P to NAD (Allison, Connors and Parker, 1969). The reaction is completed by transferring the acyl group to an orthOphOSphate ion. Bound NAD accelerates the formation and dissociation of the enzyme-G3P complex (Smith and Velick, 1972). Reductive dephosphorylation also proceeds via an acyl enzyme intermediate; it is formed by the displacement of phOSphate from DPGA. The acyl enzyme is then reduced by NADH (Smith and Velick, 1972). NAD is an activator for reductive dephosphorylation (Harrigan and Trentham, 1973; DeVijlders st 21., 1970). Since each monomer of GAPD binds one molecule of NAD or NADH competitively, this binding can affect the structural state of the enzyme and determine the rate of oxidative phosphorylation or reductive dephosphorylation. In the oxidative phosphorylation, the active form of the enzyme is the NAD-GAPD complex, and acylation promotes the release of NADH. According to Velick (Smith and Velick, 1972), in reductive dephOSphorylation the binding of NADH to a vacant active site inhibits the reaction (the inhibition occurs at low DPGA concentrations), but the binding of NADH to an acylated site simply results in reaction. In contrast, the binding of NAD to a vacant site prevents the binding of NADH to that site and pro- motes acylation at another site; thus NAD accelerates reductive dephOSphorylation at low DPGA concentrations. At high DPGA concentrations, bound NAD at the acylated site blocks the binding of NADH and inhibits the reaction. Other Reactions Catalyzed by Glyceraldehyde 3-P Dehydro- genase. In addition to the basic dehydrogenase reactions discussed above, GAPD catalyzes a number of other reactions-~transferase, phosphatase, esterase, diaphorase, transphosphorylase and NADH-NADHX activities. These reactions, their various requirements for the NAD-GAPD complex and for free sulfhydryl groups, and their sig- nificance have been summarized by Stancel (Stancel, 1970). Current research still continues on these reactions, their mechanisms and their possible importance i2_zizg (Francis, Meriwether and Park, 1971; Benitez and Allison, 1973; Foucault st 31., 1974). The arsenolytic reaction, a variation on oxidative phosphorylation, has become the basic assay for GAPD. Arsenate is substituted for phosphate to give an unstable acyl arsenate which is hydrolyzed immediately to the corresponding acid. This variation permits oxidative phosphorylation to be measured without competition from reductive dephOSphorylation. POSITIVE AND NEGATIVE COOPERATIVITY AND HALF-OF-THE-SITES REACTIVITY. Cooperativity in GAPD is related to pyridine nucleotide binding. Lobster muscle GAPD at 25° binds NAD with negative cooperativity (DeVijlders gt 31., 1970); rabbit muscle GAPD at 4° and 25° binds NAD (Conway and Koshland, 1968) and NADH (Boers, Oosthuizen and Slater, 1971; Boers and Slater, 1973) with negative cooperativity. Yeast GAPD at 4° and 25° shows positive c00perativity toward binding the second NAD and negative cooperativity toward binding the third and fourth (Cook and Koshland, 1970). Spectrosc0pic studies at 24° on the molecular basis for negative c00perativity in rabbit muscle GAPD indicate that the cooperativity arises from ligand-induced conformational changes on a previously symmetric molecule (Schlessinger and Levitzki, 1974). The significance of the above data has been brought into question as a result of investigations at more physiological temperatures. Upon increasing the temperature from 2° to 36°, Velick found the negative cooperativity exhibited by rabbit muscle and liver GAPDS disappeared (Velick, Baggott and Sturtevant, 1970). Also, Jaenicke and Kirschner have independently demon- strated that yeast GAPD at 40° binds NAD with positive cooperativity (Jaenicke, 1970; Kirschner gt 31., 1966). Von Ellenrieder has shown that yeast GAPD binds NADH linearly at 25° and 40° (Von Ellenrieder, Kirschner and Schuster, 1972). The question of whether GAPD exhibits negative or positive c00perativity is further complicated by Koshland's recent discovery that binding a single NAD to yeast GAPD can cause positive cooperativity within a subunit (accelerate the acylation and deacylation of its essential thiol) while causing negative cooperativity between subunits (decelerate NAD binding) (Stallcup and Koshland, 1973). In half-of-the-sites reactivity as exhibited by many multi-subunit enzymes, only half of the subunits can react with active site reagents; yet when these are reacted, all enzymatic activity is lost. Half-of-the- sites reactivity differs from negative cooperativity in that in the former, only half of the subunits can react, while in the latter, all the subunits can be made to react with high enough concentrations of reagents. In contrast, when each subunit reacts independently, it is called linear reactivity. Work by Stallcup and Koshland (1973) on yeast GAPD has shown that half-of-the-sites reactivity is induced by reacting the essential thiol with almost any sulfhydryl reagent (mercurials, alkylating or acylating agents, disulfides). Once two of the sulfhydryl groups have been reacted, the remaining two sulfhydryls cannot react and GAPD cannot catalyze any of its enzymatic activ- ities. But reaction of the active site lysines with alkylating and acylating agents permits all the lysines to be reacted and causes GAPD to exhibit linear reactivity toward addition of NAD to the tetramer. In contrast, with rabbit muscle GAPD (Levitzki, 1974) only certain alkylating agents can induce half-of-the-sites reactivity when reacted with the essential thiol. MOLECULAR STRUCTURE OF GLYCERALDEHYDE 3-P DEHYDROGENASE. Amino Acid Composition. Of the many GAPDS whose amino acid compositions have been examined, all have four peptide chains of approximately 330 residues (Harris, 1970). The amino acids in these chains are remarkably similar from one species to another: three-fourths of the residues have been conserved through five hundred million years of evolution (Elodi and Liber, 1970). Perhaps the most noteworthy conservations are those of the three tryptophan residues per monomer in all GAPDS (Perham, 1971; Allison and Kaplan, 1964), except rabbit muscle which has four (Harris and Perham, 1963) and Escherichia coli which has eight (Allison and Kaplan, 1964), and the four cysteine residues per monomer in all mammalian GAPDS, except monkey (Perham, 1969) and human muscle (Wolny, 1968) which have three. 10 In all GAPDS which have been sequenced (lobster muscle [Davidson 32 31., 1967], pig muscle [Harris and Perham, 1968] and yeast [Jones and Harris, 1972]), the amino acid sequences are remarkably similar; there is 60 per cent homology between the mammalian, yeast and lobster enzymes (Harris, 1970) and a much greater homology of 72 per cent between the mammalian and lobster enzymes (Perham, 1971). The differences in sequence that do exist are highly conservative substitutions that would not sig- nificantly disturb the three dimensional structure or mode of action of GAPD (Perham, 1971). Although isozymes of GAPD have been found, evidence for these isozymes having different sequences has not been discovered (Harris, 1970). Sulfhydryl Groups. Mammalian GAPDs (except monkey and human muscle) have four cysteines per monomer (Perham, 1969), lobster muscle has five (Davidson 33 21., 1967) and yeast has two (Harris, 1970). The two cysteines conserved are Cys-149, the essential thiol which forms a covalent bond with G3P and DPGA, and Cys-153, a close neighbor of Cys-149 (the numbers refer to the pig muscle sequence) (Harris, 1970). No disulfide bonds are found in native GAPD. Studies of rabbit, pig and lobster muscle GAPDS using PCMB (Boross, Cseke and Vas, 1969), o-iodobenzoate (Olson and Park, 1964; Ehring and Colowick, 1969), 11 iodoacetic acid (Perham and Harris, 1963; Park 2E’31., 1961; Harris, Meriwether and Park, 1963; Bernhard and MacQuerrie, 1971), fluorodinitrobenzoate (Shaltiel and Soria, 1969) or DTNB (Wassarman and Major, 1969) show Cys-149 residues react very rapidly. Upon further incubation of pig muscle GAPD with PCMB, Cys-153 residues react at a moderate rate and the remaining cysteines react very slowly (Boross, Cseke and Vas, 1969). But upon further reaction with DTNB, all four remaining cysteines per monomer in lobster muscle GAPD react at an equal rate (Wassarman and Major, 1969). Reaction with either PCMB (Szabolcsi, Biszku and Sajgo, 1960) or DTNB (Wassarman and Major, 1969; Boross, 1969) eventually results in denaturation of GAPD, possibly involving disulfide bond formation and precipitation. Reaction with PCMB also results in the loss of bound NAD (Velick, 1953). 7 Binding Sites for NAD, NADH, Glyceraldehyde 3-P and 1,3- Diphosphoglycerate. At substrate levels, NADH, G3P and DPGA each have only one binding site per monomer. But whether NAD had one or two binding sites per monomer is an open question; the problem is whether the NADs that are bound to the isolated mammalian GAPDS, that are part of the active NAD-GAPD complex and that show negatively cooperative interactions are also the catalytically active NADs. 12 The concept of two binding sites comes from studies of activity and negative cooperativity or activity and the NAD-GAPD complex. For example, changes in the amount of NAD bound in an NAD—GAPD complex and changes in the rate of NADH formation by oxidative phos- phorylation Operate independently (Chance and Park, 1967). This difference indicates that the NAD in the NAD-GAPD complex is not the same as the NAD being reduced to NADH; otherwise, the change in rates would be coordinated. Also, the dissociation constant of the NAD-GAPD complex is different than the dissociation constant for NAD during reductive dephosphorylation (Frieden, 1961); these two dissociation constants should be the same if only a single NAD per monomer is involved in both the enzymatic reaction and the NAD—GAPD complex formation. And lastly, NAD binds to GAPD with negative cooperativity, but oxidative phos- phorylation exhibits linear reactivity at every NAD con- centration (Peczon and Spivey, 1972). The concept of a single NAD binding site has arisen more recently. Schlessinger and Levitzki (1974) have strong evidence that the adenine moiety of NAD is solely responsible for the negatively COOperative binding in rabbit muscle GAPD; they argue that the rest of the NAD molecule is then available for catalytic activity and that the activity constants need not reflect the negative cooperativity constants. Duggleby and Dennis 13 (1974) have concluded that only analogues of NAD which can serve as substrates can form active NAD-GAPD com- plexes; they believe that identical binding requirements indicate use of the same binding sites. Boers has dis- covered that the Km for NAD and the K1 for NADH are com- patible with the negatively cooperative binding constants of the less tightly bound NADs; he feels that these NAD binding sites are the catalytically active sites (Boers and Slater, 1973). Amino Acid Residues in the Active Site. The amino acid residues in the active site include Cys-149, Lys-183, His-38, Cys-281 (Park, 1971), Cys-153 (Moore and Fenselau, 1972), Lys-191 and Lys-212 (Forcina 22.219! 1971; Zapponi st 31., 1973). Cys-149 binds GBP and DPGA (Park, 1971); the sulfhydryl group and the aldehyde carbon form a thioester bond (Racker, 1965). Cys-149 is also involved in the enzyme-coenzyme charge-transfer interaction (Boross and Cseke, 1967; Cseke and Boross, 1967). Lys-183 inter- acts with the pyrophosphate group on NAD and NADH (Buehner g£'§1., 1973); the adenine moiety is also involved in binding the pyridine nucleotides via its 6-amino group and the 2'-hydroxyl group on the adenine-linked ribose, but the nicotinamide moiety is not involved in binding (Yang and Deal, 1969a, b). His-38 is the nucleOphile responsible for removing or donating a hydrogen from or to Pi (Moore and Fenselau, 1972). Either Lys-191 or 14 Lys-212 is the anion binding site for Pi (Forcina 33 31., 1971; Zapponi 33 31., 1973), though Keleti contends that a histidyl residue is the anion binding site (Keleti, 1970). Cys-153 (Moore and Fenselau, 1972) and a dif- ferent lysyl residue (Foucault 33 31., 1974) are responsible for maintaining the structure of the active site. Although Cys-281 can accept the acetyl group from Cys-149 at high temperatures in rabbit muscle GAPD (Park, 1971), its importance in the active site is questionable because this residue is missing in the yeast (Perham, 1971) and monkey enzymes (Perham, 1969). X-ray Crystallographic Studies. Preliminary x-ray dif- fraction data have been reported for GAPD from lobster muscle (Watson and Banaszak, 1964), crayfish (Campbell 32 31., 1971), Bacillus stearothermophilus (Suzuki and Harris, 1971) and human muscle (Goryunov 33 31,, 1972). In addition, Watson has published a low resolution electron density map of human skeletal muscle GAPD (Watson, Duee and Mercer, 1972), and Rossman has pub- lished a high resolution electron density map of lobster muscle GAPD (Buehner 3E 31., 1974). The structural details described in all cases have been the same, despite the variety of GAPD sources. Molecules with bound NAD are nearly spherical o tetramers with a diameter of approximately 80 A (Watson, Duee and Mercer, 1972). Molecules without bound NAD may 15 be less symmetrical than molecules with NAD because the crystals formed without NAD are less symmetrical than the crystals formed with NAD (Goryunov 33 31., 1972). This change in symmetry correlates well with the conforma- tional changes caused by NAD binding which were observed in spectroscopic studies (Schlessinger and Levitzki, 1974). On the x-ray maps, a loop of electron density forms an integral part of two of the four subunit interfaces and relates the subunits in pairs (Watson, Duee and Mercer, 1972); thus, this provides a structural basis for half- of-the-sites reactivity (Stallcup and Koshland, 1973). The helices are right-handed, while the sheets are all left-handed twists (Chothia, 1973). The coenzyme binding site, an Open conformation associated with a parallel pleated sheet, is nearly identical to the coenzyme binding site of lactate dehydrogenase and very similar to those ofla number of other NAD-dependent dehydrogenases. But unlike those of lactate dehydrogenase, the coenzyme binding sites are located close to the subunit interfaces (Buehner 35 31., 1973), which may explain the observed negative coopera- tivity (DeVijlders 3E 31., 1970; Conway and Koshland, 1968; Cook and Koshland, 1970). ISOZYMES OF GLYCERALDEHYDE 3-P DEHYDROGENASE. Isozymes from the Same Tissue. Despite the fact that all chemical and physical studies indicate that the four polypeptide 16 chains in any tetrameric GAPD molecule are identical, researchers have found isozymes of GAPD in single tissues of turtle, perch, trout, spinach (Lebherz and Rutter, 1967), rabbit (liver and muscle) (Harris, 1970), pig (muscle) (Batke and Cennamo, 1972) and flounder (Marangos and Constantinides, 1974a) and from single strains of yeast (Lebherz and Rutter, 1967; Kirschner and Voigt, 1968; Holland and Westhead, 1973). These isozymes usually exist in sets of five, suggesting the possible combinations of two nonidentical subunits, and are observed by techniques which separate molecules according to charge (electrophoresis, isoelectric focusing, ion exchange chromatography), not according to size or shape. Only a single study has found other differences in a set of isozymes besides the difference in charge used to separate them; the flounder muscle GAPDS differ in sulfhydryl requirement, number of PCMB-reactive groups, heat stability and bound NAD content (Marangos and Con- stantinides, 1974a). Individual isozymes have not yet been identified as having separate activities or cellular locations from other isozymes. Isozymes from Different Tissues. Because most enzymes exhibit species-specific differences and because many glycolytic enzymes have tissue-specific isozymes, GAPDS from different species and from different tissues of 17 the same species have been examined for variation. Remarkably few variations have been found. Looking at GAPD from various species, the number of differences in amino acid composition and in antigenic determinants increase with the taxonomic distance between species. There are also differences in the amount of NAD bound and the reactivity of the enzymes toward NAD (Alli- son and Kaplan, 1964; Nagradova and Grozdova, 1974). Studies of GAPDS from different tissues of the same species have included comparisons of rabbit muscle, liver, brain, heart and kidney enzymes (Bondi, Watkins and Kirtley, 1969; Kochman and Rutter, 1968; Smith and Velick, 1972) and ox muscle and liver enzymes (Lambert and Perham, 1974). No major structural or kinetic dif- ferences have been found, although a few minor dif- ferences were noted (Specifically some chemical dif— ferences which Velick and Kirtley consider significant). Smith and Velick (1972) contend that the minor kinetic differences between rabbit muscle and liver GAPDS are sufficient to explain the respective glycolytic and glu- coneogenic roles of these enzymes. DISSOCIATION AND REASSOCIATION OF SUBUNITS. GAPD has been a favorite model for studying the forces binding enzyme subunits together. Such studies use various agents to dissociate and reassociate the enzyme; the process of dissociation is followed by hybridization or by separation 18 of the dissociated components in sucrose density gra- dients, by ultracentrifugation or on electrophoresis gels. A list of dissociating agents, their effect on GAPD, the reversal of this effect, the source of the GAPD and the major references is listed in Table 1. For GAPD, dissociation and inactivation are often linked; in many cases dissociation can cause partial or total inactivation and inactivation eventually brings about dissociation. Likewise, reassociation and reacti- vation are usually closely linked. The reassociated and dissociated forms of GAPD are tetramer, dimers, monomers and aggregates. The first three forms are interconvertible; the last is a product of irreversible denaturation. Aggregation is not to be confused with reassociation--reassociation is the for- mation of an active tetramer; aggregation is a random massing of unfolded monomers which have lost all ability to reassociate. Dissociation studies have given rise to speculation about the form of the active Species of GAPD. From active enzyme sedimentation velocity eXperiments using rabbit muscle GAPD, Hoagland and Teller have concluded that only the tetramer shows full dehydrogenase activity (Hoagland and Teller, 1969). Jaenicke has drawn similar conclusions from experiments showing that substrate and coenzyme binding shift the dissociation equilibrium for rabbit 19 mmma .cfimunmcmx.ao> cam Hermeo>mN .nofi>oxumz m mmma .muwcz paw Hans: m pmwosum no: mumfiwo mom mmma .fiooam m .m .m smficsum no: mumsflc zom onma .Hmumssom paw umcsomuflm M moms .aaxmeo>mu z mums .Hooam cam mxumpo>mN .moumxmq m gamma .umaama ocm pcmHmmom m Ammma mead unmade no Hexms mo.oc .>ocmmusm paw m>oumm¢ m mumcmmonm can adz paw mumaosoe cowusHHU onma .Hmumssom paw umcsomufim m .w Head unmmm mqfioswmu Lomuov .m>mmsw cam m>oomumwz m .mumnmmonm .oom mHmeU mnsumummamu 30H Huma mumfiflp .smEsomnom cam Sufism m HouwanDOHnuwp pom mHmEocoa mzom GOAUMHoommmmm Euom moocmummwm mmousom How mucmemuflsvmm UoDMHUOmmHo ucmmd msHDMfioommwo mmmsmmoupxsmo mum spasmoamumomHu mo sowumwoommmmm was GOHHMHUOmmwo .H manna 20 Guacamonm onma Iouwd umcflmmm mam .mmmammefls was mesa o usages 6cm aohusaho Laos no manna Ohms Ho umonm .mumuuho 2 H .Hmumsnom can Hmcnomnflm w omm ou msflfiums mumaflp ummoa umv numcmuum onma .mxoflcmmb w .m can uswmm msflosomu cam mumEocoE owcow sown cocoa cam» poms noun 30H .mhmsHMAc AHA mac ohmmn .cmEcoum cam mumnwzm w cocoa cwmuoum 30H moms .mocs .amm .\. Ho.o .mm can pflfinom .mxowcwmn m .M ed mmmmuocfi memn AN may owowom onma .oxoflcmmw w .m mumeocoe mm mEouuxm Huma ..mm mm acm>o m pmfiwsum pom mHmEHo mownpmnsm owmama 0.9Huma ..Hm no acm>o m mmma .Hmmo can mopflcwucmumcoo m mmma Ammouosm .ummmm Homv mumfiwo .mmma .Hme was Hmocmpm w usmmm mcwospmn .omm was mHmEOGOE Aoov mad amma .pcmanmom paw Hmmfima m gamma .mmoflcwu Icmumcou can woodcuts m moma .Hme m socoo camuonm 30H mmma .mosM .numcmnum UHGOH CHE Hum paw paenom .mx0ficmmn M .m .ucmmm mcaospmu .sz mHmEOGOE Imcfipwcmsm was won: mmocmummom mmonsom MMMHWMWMmMMMMWmm pmumWwwmmHo ucmmd mswumwoommwo nmaqflucoo .H magma 21 .Umflpsum nos cowumwoommmmmo .maco mHmEHU pmcamuno n .Umumum no: u 2 can maomsE smeHno n U .maomse Hmpssoam u m .maomsfi umu n d .maomsfi mcw>on u m .maomsE mam u m .pmmmm u M .maomse pagans u m "poms mGOMDMH>mHnnm ones Huma .m>mmsw can m>ocmnmwz m “mama Ho HE\mE ES .38 .mm on 3283 39:8 To .3 .2 3.8 can wmoflcflusmumcou m can “comm mcwospwu no mHmEosoa HUM can vommfivmzv cowumw00mmmmm Such mmocmummmm mwousom How mucwfiwuflswmm meMHOOmmHQ ucmmm osflumaoommwo posswucou .H manna 22 muscle GAPD from dimer to tetramer (Jaenicke, 1970). However, Ovadi has isolated an active dimer after incubating pig muscle GAPD with ATP and mercaptoethanol for two hours; the dimer is 50 per cent as active as the tetramer (Ovadi 33 31., 1971). Elodi (1958) and Kirshner and Schuster (1970) have obtained active dimers after dissociating with KCN and high ionic strength respectively. NAD protects the enzyme against inactivation and dissociation (Hoagland and Teller, 1969; Constan- tinides and Deal, 1969; Ovadi 33 31,, 1971). This pro- tection presumably comes from NAD maintaining an enzymic structure that is less susceptible to inactivation and dissociation or from NAD competing with the dissociating agent for the same binding site. NAD is also required for reactivation from urea, guanidine-HCl or dilute solution. Based on the requirements for NAD for 13 31333 reassembly, it has been speculated that NAD may be required for the 13 3133 folding of newly synthesized GAPD polypeptide chains and their association into tetramers (Deal, 1970; Teipel and Koshland, 1971). EFFECTS OF ATP. ATP, a well-known effector of various glycolytic enzymic activities and a partial structural analog of NAD, has three negative effects on GAPD activity (Stancel and Deal, 1968): (1) ATP produces an immediate inhibition by competing with NAD for the 23 NAD binding sites (Yang and Deal, 1969a), (2) ATP causes a time-dependent inactivation by dissociating the enzyme into subunits (Constantinides and Deal, 1969; Stancel and Deal, 1968, 1969), and (3) ATP stimulates a very rapid loss of activity in the presence of chymo- trypsin by causing structural changes which make the enzyme more susceptible to proteolytic degradation (Yang and Deal, 1969b). Immediate Inhibition. The immediate inhibition by ATP of oxidative phosphorylation was first studied with yeast GAPD at 25° (Yang and Deal, 1969a). Besides ATP, AMP, ADP and CAMP were found to be inhibitory; all the com- pounds showed competitive inhibition with respect to NAD. From this competition and a comparison of the structure of ATP and NAD, it was concluded that ATP and NAD compete for the same binding site (Yang and Deal, 1969a, b). Park's laboratory extended these studies to rabbit muscle GAPD using physiological concentrations of substrates and effectors. Their results were similar to those of Yang and Deal, except they found ATP to also be com- petitive with respect to Pi° From this additional competition, they concluded that the ATP-binding site covered the Pi-binding site as well as the NAD-binding site (Oguchi, Meriwether and Park, 1973). Adenine nucleotide inhibition has also been observed by Sapag- Hagar (1969) who used ATP, ADP and AMP to inhibit rabbit 24 muscle GAPD and by Nagradova, Vorona and Asriyants (1969) who used ADP and AMP to inhibit yeast and rabbit muscle and heart GAPDS. Park's laboratory has also studied the ATP- inhibition of the "nonphysiological" reactions of GAPD (reactions other than oxidative phosphorylation and reductive dephosphorylation). They have found three points where adenine nucleotides can inhibit the reactions: coenzyme binding, formation of enzyme- substrate complexes and transfer of S-acyl group from active site to acceptors (Francis, Meriwether and Park, 1971). Time-Dependent Inactivation and Dissociation. The time- dependent inactivation caused by the dissociation of GAPD in the presence of ATP was first discovered in Deal's laboratory (Stancel and Deal, 1968). Using yeast GAPD (Stancel and Deal, 1969), the optimal con- ditions for inactivation and dissociation into monomers were 1-2 mM ATP, 0°, 0.03-0.1 mg/ml GAPD, pH 9.0 and 0.1 M mercaptoethanol. AMP and cAMP did not dissociate the enzyme. The dissociation could be partially pre- vented by AMP, cAMP or any of the substrates and could be reversed by warming to 17°. The optimal conditions for reversal were 0.04 mg/ml GAPD, pH 7.0, 1-2 mM ATP, 17°, 10 per cent sucrose and 0.1 M mercaptoethanol. Extending the studies to rabbit muscle GAPD, Deal's 25 laboratory found a similar, yet different dissociation (Constantinides and Deal, 1969). The process still required 1-2 mM ATP, 0.1 M mercaptoethanol and 0°, but it occurred much more rapidly (in less than one hour instead Of taking 12-14 hr), was pH independent, happened at higher GAPD concentrations and resulted in dimers at high protein concentrations and monomers at low protein concentrations. This dissociation was prevented by 1 M (NH4)ZSO4 or 2 mM NAD. The reversal could still be accomplished by warming the enzyme, but Optimal con- ditions were 23°, no ATP and no sucrose. Feeling that 0° was not physiological, Park's laboratory attempted the inactivation experiments with rabbit muscle GAPD at 25° (Oguchi, Meriwether and Park, 1973). They were able to obtain results similar to those of Constantinides and Deal if they increased the ATP concentration to 6 mM. They did not study the dissociation or reacti- vation. In addition they demonstrated that the immediate inhibition by ATP and the time-dependent inactivation were two separate processes. There have been several other studies on the dissociation of GAPD by ATP. Keleti's laboratory dis- sociated pig muscle GAPD with ATP at 0° into a dimer which was active in the presence of mercaptoethanol. They demonstrated that four moles of ATP were bound per tetramer and that the dissociation caused the exposure 26 of a normally buried sulfhydryl group which, in the absence of mercaptoethanol, could form a disulfide bond with the same residue on the neighboring monomer (Ovadi 33 31., 1971; Ovadi, Nuridsany and Keleti, 1973). Lebherz, Savage and Abacherli (1973) demonstrated an ATP-mediated subunit exchange between isozymes of trout skeletal muscle GAPD at 25°. In addition to the subunit exchange, ATP caused a time-dependent inactivation which they did not characterize. Both the exchange and the inactivation were prevented by NAD. Darnall and Murray (1972) studied dissociation by ATP of rabbit muscle and human erythrocyte GAPDS using the physiological conditions for the human erythrocyte. Dissociation at 25° required a mixture of ATP and cacodylic acid and was reversed by 2,3-diphOSphoglycerate. Despite examination of glycolytic intermediates, tricarboxylic acid intermediates and several amino acids, they could not find a physiological compound that would replace cacodylic acid in their dis- sociation system. Stimulation of Proteolytic Digestion. The ATP-stimulated loss of enzymic activity in the presence of chymotrypsin has been studied only by Yang and Deal (1969b). They hoped to gain some insight into the mechanism of ATP binding and into the control of enzyme degradation. ATP, and to a lesser extent ADP and AMP, alter the structure of yeast GAPD in a manner which encourages 27 its digestion by chymotrypsin. This ATP effect is pre- vented by NAD and cAMP. The destabilization seems to be a function of the number and the charge of the terminal phosphate groups of the adenine nucleotide. It is interesting that the binding of 2,4,6- trinitrobenzenesulfonic acid (Foucault 33 31., 1974) and the removal of bound NADs (Cantau, Jaureguiberry and Pudles, 1970; Jaenicke, 1970; Fenselau, 1972) also increase the susceptibility of rabbit muscle GAPD to digestion by chymotrypsin. Since NAD binds competitively with ATP (Yang and Deal, 1969a, b) and 2,4,6-trinitro- benzenesulfonic acid binds competitively with NAD (Fou- cault 33 31., 1974), it seems likely that NAD, ATP and 2,4,6-trinitrobenzenesulfonic acid all cause the same structural alteration. EVIDENCE FOR AND AGAINST THE BINDING OF ZINC TO GLYCER- ALDEHYDE 3-P DEHYDROGENASE. Zinc is a functional com- ponent of several dehydrogenases, such as alcohol (Vallee and Hoch, 1955), malic (Harrison, 1963) and glutamic dehydrogenases (Adelstein and Vallee, 1958). Because zinc stabilizes the structure and participates in the Catalytic function of these enzymes by being involved in NAD binding, it has been prOposed that zinc might also be present in other pyridine nucleotide dependent dehydrogenases (Vallee, 1960). 28 By emission spectrochemistry and analysis for metal content, Vallee found metal ions, presumably zinc, in yeast and rabbit muscle GAPDS (Vallee 33 31., 1956). Subsequently, Keleti reported that pig muscle, cow muscle, crayfish muscle and yeast GAPDS contained two to three moles of zinc per mole of enzyme. Further studies indicated that removal of zinc from the enzyme led to protein denaturation and that chelation of zinc led to inhibition of enzymic activity; from these results Keleti proposed that zinc was essential for the catalytic activity and the maintenance of steric structure in GAPD (Keleti 33 31., 1962; Keleti and Telegdi, 1959a,b). Keleti has also studied the exchangability of zinc on various GAPDs (Keleti, 1964, 1966), and Polgar has studied the amino acid residues of GAPD and the areas of its substrates that bind zinc (Polgar, 1964). In opposition to this data, Ferdinand and Park, working independently, have not found significant amounts of zinc bound to rabbit muscle, pig muscle or yeast GAPDS (Ferdinand, 1964; Barkman, Sandstead and Park, 1970). Park has further shown that zinc is not required for coenzyme binding, substrate binding or catalytic activity (Barkman, Sandstead and Park, 1970). Other workers have demonstrated that GAPD is actually inhibited by heavy metals (Velick and Furfine, 1963) and that its activity is increased in the presence of chelating agents (Nagra- dova, 1965). 29 PHYSICAL AND CHEMICAL PROPERTIES OF PIG MUSCLE AND PIG LIVER GLYCERALDEHYDE 3-P DEHYDROGENASES. Pig muscle GAPD can be purified by two methods (Elodi and Szorenyi, 1956; Harrigan and Trentham, 1973). Pig liver GAPD has been purified by only one method (Dagher and Deal, in prepar- ation). Crystalline pig muscle GAPD has four molecules of NAD bound per tetramer (Ovadi 33 31., 1971; Harrigan and Trentham, 1973); crystalline pig liver GAPD has two or three molecules of NAD bound per tetramer (Dagher and Deal, in preparation). The isoelectric point for the pig liver enzyme is 8.8 (Dagher and Deal, in preparation); the isoelectric point for the pig muscle enzyme is 8.1 to 8.6 (Batke and Cennamo, 1972). The specific activity is 150 to 220 umoles/min/mg for both purified enzymes, and their pH optimums for oxi- dative phosphorylation are between 8 and 9. Kinetic constants for the pig liver enzyme include a Km for NAD of 12 uM and a Km for G3P of 240 uM in oxidative phos- phorylation and a Km for NADH of 23 uM in reductive dephosphorylation. Kinetic constants for pig muscle GAPD are similar to those for lobster muscle GAPD; approximate comparative values are estimated to be a Km for NAD of 13 uM, a Km for G3P of 90 uM and a Km for NADH of 3.3 uM. With both enzymes, NAD is a com- petitive inhibitor of NADH in reductive dephosphorylation and exhibits a Ki of 850 uM for pig liver GAPD and of 30 300 uM for pig muscle GAPD (Harrigan and Trentham, 1973; Velick and Furfine, 1963; Dagher and Deal, in prepar- ation). Pig muscle GAPD has a molecular weight of 145,000 1 6000 for its tetramer and 36,000 i 1500 for its monomer (Harrington and Karr, 1965). Pig liver GAPD has a molecular weight of 148,000 for its tetramer and 38,000 for its monomer (Dagher and Deal, in preparation). A partial specific volume of 0.737 ml/g at 20° has been measured for the pig muscle enzyme (Elodi, 1958). The amino acid composition and sequence of pig muscle GAPD have been determined (Harris and Perham, 1968), and the reactivity of its sulfhydryl groups has been studied (Boross, Cseke and Vas, 1969). The sulfhydryl residues can be divided into three classes based on their reactivity: Cys-149 is very reactive, Cys-153 is moderately reactive and Cys-244 and -281 do not react unless the subunits are unfolded. A number of measurements indicate that pig muscle GAPD has a compact hydrophobic core amounting to 40 per cent of its molecular weight and a 28 to 48 per cent a-helix content (Zavodszky, Abaturov and Varshavskii, 1966; Markovich, Zavodskii and Vol'kenshtein, 1966). These conclusions are based on the following evidence. ORD spectra for pig muscle GAPD with and without NAD and NADH and in the presence of SDS have been measured. 31 Using the parameters Ac, b0, a , Kc’ H and H 5, the 193 22 NAD-free enzyme is 34 per cent ordered structure; the O NAD-GAPD complex is 43 per cent ordered structure. The dispersion constants, Ac and b0, decrease 4 per cent on the addition of NADH to the NAD-free enzyme and decrease 5 per cent on the addition of SDS. Infrared spectro- photometric studies of hydrogen-deuterium exchange indicate that 32 per cent of the peptide hydrogens are inaccessible to water, 15 per cent are accessible within a measurable time span and 53 per cent exchange imme- diately. Addition of NAD or NADH increases the amount of inaccessible peptide hydrogens by 10 and 4 per cent respectively. The viscosity of native GAPD is 0.033 dl/g and of GAPD in SDS is 0.088 dl/g. Pig muscle GAPD exists as five isozymes which can be separated by isoelectric focusing (Batke and Cennamo, 1972). Smith and Velick (1972) Claim that such isozymes arise from differential binding of NAD, but the pig muscle isozymes all contain the same amount of NAD. The enzyme can be dissociated into dimers and/or monomers by KCN (Elodi, 1958), dilution (Lakatos, Zavodszky and Elodi, 1972) and ATP (Ovadi 33 31., 1971; Ovadi, Nuridsany and Keleti, 1973). Some preliminary studies have indicated that pig liver GAPD could be dissociated by ATP (Dagher and Deal, in preparation). One of the objectives of this thesis was to define this phenomenon in more detail. 32 According to Keleti, pig muscle GAPD contains two or three moles of zinc per mole of enzyme which are essential for catalytic activity and for maintenance of the steric structure (Keleti 33 31., 1962; Keleti and Telegdi, 1959a,b). However, Barkman, Sandstead and Park (1970) contend that there is no significant amount of zinc in pig muscle GAPD and that zinc is not required for catalytic activity; still, it is interesting to note that what they call an insignificant quantity of zinc in pig muscle GAPD (0.115 to 0.146 moles of zinc per mole of enzyme) is three to five fold higher than the amount they found in rabbit muscle or yeast GAPD. MATERIALS AND METHODS REAGENTS. Fresh pig livers were obtained from the Peet Packing Co., Chensaning, Michigan, and frozen within 15 min of slaughter. NAD and ATP (disodium salt) were purchased from P-L Biochemicals, Inc. Cysteine, tannic acid, gum arabic, ATP (disodium salt), BSA and DTNB were products of Sigma Chemical Co. Lactate dehydro- genase (dOg muscle) was purchased from Boehringer Mannheim Corp. G3P was obtained from Sigma Chemical Co. as the diethylacetal barium salt and converted into the free acid as previously described (Deal, 1969). Imidazole was purchased from Sigma Chemical Co. and was recrystal- lized once from chloroform:ether. DEAE-Sephadex A-50, SE-Sephadex C-25 and coarse Sephadex G-25 were products of Pharmacia Fine Chemicals. Norit A was obtained from Fischer Scientific Co. and was activated by acid-base washing (Heppel, 1963; Smith and Khorana, 1963). ENZYME PREPARATION. GAPD was crystallized from pig liver by adapting to a large scale the method of Dagher and Deal (in preparation). A whole pig liver (1100-1300 9) was thawed for 24 hr at 4°. The liver was cut into large pieces and 33 34 homogenized for one minute in a Waring blender with 1100-1300 ml cold 10 mM ZnCl2 and 25 mM mercaptoethanol. The homogenate was centrifuged at 13,800 x g for 30 min. Its supernatant was collected and made 50 mM in EDTA and 50 mM in mercaptoethanol. An ammonium sulfate fractionation at 0° was per- formed on the cold supernatant solution. The precipitate which formed between 65 and 95 per cent saturation (calcu- lated from Warburg's formula [Warburg and Christian, 1939]) was collected by centrifugation at 27,300 x g for 30 min and dissolved in 40 ml 10 mM imidazole pH 7.5, 25 mM mercaptoethanol. The resuspended pellet was divided into two batches and desalted on a column of coarse Sephadex G-25 (6 x 96 cm) equilibrated in distilled deionized water. The salt-free eluates were combined, made 50 mM in mercaptoethanol and concentrated to 40 ml in an Amicon TCF 10 ultrafiltration cell using a PM-30 membrane. The concentrated solution was made 25 mM in mer- captoethanol and adjusted slowly to pH 7.3 with 0.1 N NH4OH. The precipitate which formed was removed by centrifugation at 39,100 x g for 5 min. The supernatant was chromatographed on a column of DEAF-Sephadex A-50 (6 x 44 cm) equilibrated in 10 mM imidazole pH 7.3, 25 mM mercaptoethanol. The enzyme was in the first protein peak washed from the column by the buffer. This 35 peak was collected, made 25 mM in mercaptoethanol and concentrated to 20 ml in an Amicon TCF 10 ultrafiltration cell using a PM-30 membrane. The concentrated solution was made 25 mM in mer- captoethanol, 5 mM in EDTA and 0.2 mM in NAD and adjusted slowly to pH 6.6 with 0.1 N acetic acid. The precipitate which formed was removed by centrifugation at 39,100 x g for 5 min. The supernatant was applied to a column of SE-Sephadex C-25 (2.5 x 22 cm) equilibrated in 10 mM imidazole, 5 mM EDTA pH 6.6, 25 mM mercaptoethanol. The enzyme was eluted with a linear gradient of 0 to 0.3 M KCl, using 300 m1 of each solution. The majority of the enzyme appeared in the last protein peak off the column. This peak was collected, made 25 mM in mercaptoethanol and concentrated to 7 ml in an Amicon 52 ultrafiltration cell using a PM-30 membrane. The concentrated solution was subjected to the crystallization procedure of Jakoby (1971) using 2 ml solutions of various (NH4)ZSO4 concentrations containing 5 mM mercaptoethanol, 0.2 mM NAD and 50 mM Tris-HCl pH 8.4, instead of 1 ml solutions. The series of per cent saturation in (NH4)ZSO4 used was 91, 88, 85, 82, ,73' 75, 72, 69, 66, 63 and 60 per cent. The yield was 30 mg of crystalline GAPD with a specific activity of 160 to 220 umoles/min/mg. It could be stored in its crystallization liquor at a 36 concentration of greater than 1 mg/ml for at least four months at 4° without loss of activity. The crystalline enzyme gave a single band on SDS polyacrylamide gels (method of Fairbanks, Steck and Wallach [1971] as mod- ified by Welton [1974]). ENZYME ASSAY. Oxidative phosPhorylation was measured at 25° in a Beckman DU monochromator equipped with a Gilford multiple sample absorbance recorder. The assay was the ”low NAD-G3P assay" of Deal (1969), except the final con- centrations of NAD, G3P and GAPD were 2.5 mM, 0.7 mM and 0.25 ug/ml respectively and Tris-HCl pH 8.5 was sub— stituted for glycylglycine. The concentrations of the GBP and NAD stocks were measured spectrophotometrically by the method of Dagher and Deal (in preparation). An enzyme unit was defined as the amount of enzyme reducing one umole of NAD per minute at 25°. PROTEIN DETERMINATION. Protein was determined by the method of Mejbaum-Katzenellenbogen and Dobryszyka (1959). The tannin reagent was routinely warmed and filtered before each use. The 20 min incubation period was employed. A standard curve using crystalline BSA was constructed for each experiment. Turbidity was measured at 650 nm. A modification of this procedure was used to measure protein on sucrose gradients. Tannin reagent, 0.1 ml, was added to each 0.1 m1 sucrose fraction, and 37 the mixtures were incubated for 20 min at 25°. Gum arabic, 0.2 ml, was added, and the turbidity was measured at 650 nm. This method gave a relative protein concen- tration which could not be translated into mg/ml because sucrose and the various buffers used in the gradients affected the turbidity and the numerous standard protein solutions required for quantitation were not designed. INACTIVATION, DISSOCIATION, REACTIVATION AND REASSOCI- ATION PROCEDURES. For routine handling, a slurry of GAPD crystals was centrifuged at 34,800 x g for 10 min, and the pellet was resuspended in 50 mM imidazole, 0.1 M EDTA pH 7.5, 80 mM mercaptoethanol. The resus- pended enzyme was dialyzed against 50 mM imidazole, 0.1 M EDTA pH 7.5, 80 mM mercaptoethanol for 12 to 18 hr at 4°. To inactivate the enzyme, GAPD was diluted into the standard ATP inactivation buffer which consisted of 50 mM imidazole, 0.1 M EDTA pH 7.5 containing 80 mM mercaptoethanol and 10 mM ATP to a protein concentration of 0.1 mg/ml and incubated at 0°. These conditions were always kept constant except for the variable under exami- nation. The inactivation was followed by periodic assaying and usually required 10-12 hr. During inactivation, dissociation occurred. Dissociation was followed by sucrose density gradient centrifugation according to the procedure of Martin and 38 Ames (1961) as described by Constantinides and Deal (1969) with the following exceptions. The 5-20 per cent sucrose gradients were in 50 mM imidazole, 0.1 M EDTA pH 7.5, 80 mM mercaptoethanol, 10 mM ATP (the standard ATP inactivation buffer). These conditions could be slightly varied to reflect more accurately the inactivation con— ditions. An indirect marker of 0.1 mg/ml pig liver GAPD (7.8 S) or 0.125 mg/ml dog muscle lactate dehydro- genase (7.45 S) was used; it is called an indirect marker because the marker enzyme is not in the same gradient tube as the sample (and cannot be because the enzyme was located by protein determination as well as activity determination). The indirect marker was always run without ATP. A 0.25 ml sample was layered on each gradient; the large volume was chosen to minimize dilution. The gradients were centrifuged at 45,000 rpm for 13.5 hr at 0° in a Beckman Model L3-50 with a SW 50.1 rotor. Ten drop fractions were collected; they had a volume of approximately 0.1 ml. Each gradient tube yielded 34 to 37 fractions. The fractions were assayed before and after warming to 25°. The fractions were also analyzed for protein. To reactivate the enzyme, the inactivated GAPD (incubated at 0° for 10 to 12 hr) was warmed to 25° for 5 min. Any compounds which were added to aid the 39 reactivation were put in the inactivated sample before the warming started. The reactivation was measured by assaying. Reassociation of the reactivated enzyme was followed by sucrose density gradient centrifugation as described above. The samples were centrifuged at 45,000 rpm for 7.75 hr at 25°. AMINO ACID ANALYSIS. To prepare the enzyme for analysis, the bound NAD was removed by charcoal treatment. A mixture of 14 mg activated Norit A and 1.2 mg GAPD in 0.6 m1 crystallization liquor was centrifuged to remove the charcoal and NAD. The OD :OD of the supernatant 280 260 was 1.36, indicating that a little NAD still remained bound to the enzyme (Fox and Dandliker, 1956). It was later learned that more of the NAD could probably have been removed by incubating the activated Norit A-GAPD mixture for a longer period of time (Ovadi 33 31., 1971). The NAD-free enzyme was dialyzed against dis- tilled deionized water for 24 hr at 25°. The dialyzed enzyme was concentrated to 1.0 mg/ml on a rotary evapor- ator. Equal volumes of concentrated enzyme and 12 N HCl were combined, sealed into vials and heated at 110° for 18 and 56 hr to hydrolyze. After hydrolysis, the solutions were evaporated to dryness, and the residues resuspended in 0.2 M citrate buffer, pH 2.2. The analysis was performed on a modified Technicon amino acid analyzer 40 with a computerized integrator. The concentration of each amino acid was calculated by comparison to an internal standard, norleucine. The total number of cysteine residues was deter- mined by DTNB titration in SDS (Habeeb, 1972). The total number of trytOphan residues was determined by the spectrOphotometric method of Edelhoch (1967), using SDS instead of guanidine-HCl. The protein concentration of the samples had to be determined before the SDS was added. The partial specific volume was determined from the amino acid composition by the method of Cohn and Edsall (1943) as described by Kayne (1966). DTNB REACTION. The reactivity of the sulfhydryl groups in native GAPD was determined by DTNB titration (Habeeb, 1972). The reaction was measured in a Cary 15 recording spectrOphotometer. The reaction mixture contained 240 ug GAPD, 200 ug EDTA and 40 pg DTNB (65 fold excess compared to GAPD) in 0.08 M sodium phOSphate buffer, pH 8.0. The number of sulfhydryl residues titrated per tetramer was calculated with the molar extinction coef- ficient of 13,600 at 412 nm. RESULTS INACTIVATION AND DISSOCIATION OF PIG LIVER GLYCERALDEHYDE 3-P DEHYDROGENASE. The ability of ATP to inactivate and dissociate yeast (Stancel and Deal, 1968, 1969), rabbit muscle (Constantinides and Deal, 1969) and pig muscle GAPDS (Ovadi 33 31., 1971) at 0° has been established. Experiments by Dr. Dagher of this laboratory indicated that pig liver GAPD can also be inactivated and dissociated by ATP at 0° (Dagher and Deal, in preparation). Dr. Dagher followed the dissociation by sucrose density gradient centrifugation, analyzing the gradients for dissociated enzyme by first warming the fractions to reactivate the enzyme and then assaying each for activity (Constantinides and Deal, 1969; Stancel and Deal, 1969). Using this technique he had great difficulties obtaining repro- ducible results. This problem stemmed from the low activity of GAPD under these conditions and the inability of the dissociated enzyme to reactivate. The first part of this thesis work was designed to clarify and extend these preliminary results. Experi- ments indicated that GAPD is more readily inactivated and dissociated (by dilution) when 0.1 M EDTA is present. 41 42 Investigations of inactivation and dissociation by ATP at 0° showed that this process could also occur in the presence of 0.1 M EDTA. Furthermore, the native GAPD retained most of its activity in the presence of 0.1 M EDTA, and the dissociated enzyme could be reactivated under these conditions. Hence, the next logical step was to characterize the dissociation and inactivation of pig liver GAPD by ATP at 0° in the presence of 0.1 M EDTA. Unless otherwise indicated, the standard conditions chosen for inactivation and dissociation were (1) 0.1 M EDTA, (2) 0.1 mg/ml GAPD, (3) pH 7.5, (4) 10 mM ATP, (5) 80 mM mercaptoethanol, and (6) 0°. The enzyme was routinely incubated under these conditions for 10 hr. Inactivation and dissociation were followed as described in Materials and Methods. The gradients used to study dissociation contained 50 mM imidazole, 0.1 M EDTA pH 7.5, 80 mM mercaptoethanol and 10 mM ATP. The fractions from the gradients were analyzed for both protein and GAPD activity so that inactive, as well as active, enzyme could be located. Activity determination involved assaying the fractions before and after warming to 25°. The recovery of activity from the gradients was generally poor, especially in the presence of ATP. For control samples at 0° (no ATP), 40 to 60 per cent of the activity applied was recovered; for standard samples (see above paragraph), 43 11 to 27 per cent of the activity applied was recovered; and for other samples under nonoptimal conditions, 0 to 10 per cent of the activity applied was recovered. The poor recovery could have been caused by the presence of sucrose, since sucrose decreased reactivation (see section "Reactivation and Reassociation of Glyceralde- hyde 3-P Dehydrogenase"). Warming the fractions to 25° increased the recovery of activity by 1.2 to 3.2 fold in the gradients with ATP, but did not increase recovery in gradients without ATP. Effect of Incubation Time. When GAPD was incubated under the standard inactivation conditions for various periods of time, it lost activity as shown in Figure 2. Virtually all of the loss occurred during the first 12 hr, after which the activity tended to remain the same. GAPD incubated under similar conditions without ATP also showed some loss of activity, but it was small compared to the loss in the presence of ATP. Inactivation was accompanied by dissociation as shown by sucrose density sedimentation velocity experi- ments (Figure 3). Native pig liver GAPD had a sedimen- tation coefficient of 820,w = 7.8 S (lactate dehydroge- nase as a marker), while maximally inactivated enzyme had a sedimentation coefficient of approximately 5.0 S. In the absence of ATP, the sedimentation coefficient of GAPD did not change with incubation at 0° for at 44 .m9< mo wocmmnm may OH maouucoo ucmmwudmu maonswm ammo may oHflns .me¢ ou Oomomxw mmHQEmm ucmmmummu maonfimm OOHHHM mna .mmfl3umnuo pmumoaocfl mmoacs .mwflosum ocflsoaaow mnu mo umoE How coaumosocfi mo 06“» pumpcmum may mm OODQOOM was H: OH .mH0mmumsa .musmomxm u: maloa umumm musooo coflum>HDOMCH who mo umoz .Acoflumosocfl wu0mmov muw>fluom OHMHOOOm Hmswmfluo mean map ou pmuomuuoo mums mosam> Had .moamEMm umou map mm mmz 08mm on» manomxo CH owumwuu mm3 fined ocv Houucoo d .OODMOHOGH mmEHu who now Antonuwz cam mameumpmz mmmv Hommso cowum>fluomsfl med oumpcmum on» GM pmumnsosfl mums .moosuwz paw mamaumumz OH pmnfiuommo mm mamummuo xooum Eoum OOMMQOHQ .mmHmEmm .mad mo omdw um>fia mam mo cowum>auomcfi co 00 um mafia coaumndocfl mo pommwm .m musmflm 45 N ousmwm 283%: um mm mm em cm 3 2 w v _ _ 1 _ _ _ _ A T I W. ............. IOI ........... I I'll I I. I . U. I. ’9 1| ’0 ll ’9 u/ a: 2:. a z. ..| I o: .. ER 3 I. _ _ _ V _ r _ _ 1* 8H , o2 All/\IlOV OIJIOHdS 46 .Auxmu ommv 30H on was MC o no ucwwowmmmoo CowumucmEHOOm OCH .Aowuuon DOCV m m.> u 3.0mm mo quwOHmmwoo CoHumquEwOOm m on: ACBC OCV HOHuCOO one .COHummsmanquo «0 mafia 0C» OOCHOCA DOC Op OCm Coaummsmwuucmo ou HOHHQ mmeflu mCu mum OODMOHOCH mosey wCB .COHDMCHEHOCOO Campoud mCu Eoum Omumasoamo omOCu mum OOOMOHOCH mucowowmmooo CofiumquEflOOm OCH .mEmNCo m>HuOMCH may no CODE pmum>fluommn maamumcmm 0mm um COHuMQCOCH map “0mm um COHuthOCw umumm pCm OHOMOQ meHEumumO was >ua>fluo¢ .COHDMCHEHODOO Camuoum UCm wuw>fluom mo UODMOOH mumz mquwOmnm OCD CH mxmmm mahnCo mCu mo mCowuwmom mCB .AOOOOOC mm powwowon on Hafiz scans muCoE Ifluwmxo “mama Camuuoo CH Csu maouuCoo HmCuo Op cowuaoom CH Oman OCmV mad “COCuaz Houucoo mCu ou COfluwpom Cw mmz meanCw umxume one .ACOHuwCHEHoumO Camuoum an OODMOOH was memqu mCu mmsmomo mo DOCCMO OCmv mamsmm OCD mm onsu quHOmum 08mm mCu CH DOC ma Oshqu meume may omsmomo meHME uomeOCH Cm Omaamo ma DH “CBC msCHE Common COHDM>HDOMCA CBC OHMOCmum mCu OCm omouosm mCHCHmuCOO wasp DCOHOmum m CH Csu mm3 ammo um>wa mam mo meHME uomuflOCfi CC .mmouosm pmpom OCD How ammoxm .COHusHom Coauw>quMCw maOCommouuoo mCu mm mama 0CD mm3 moan ucmwpmum Como mo CowuflmomEOO mCB .Amoonumz OCm mamwuoumz mmmv 00 um MC m.ma now Emu ooo.mv um mquwomum mmouosm o\o omlm Cw pmmsmwuusmo wum3 mmamamm one .m wusmwm HOW UCOOOH may CH Umhwnommo mmHmEMm mahnCm mCu Co uso pmwuumo mums mmflosum mmmCB .mad as omflw um>fla mam mo CoauMHOOmmHO Co oo um mEflu CoflumoCOCw mo uommmm .m ousmflm 47 zofifizmzsomm oz_zz_omm mmobm amzmaoz. maze: m OCCmfim ms .v 1%: .I .1 mad [mud S 'lNEI IO |:|:|EIOO NO ‘llVlNElW l 03S m‘oz 48 least 36 hr, although slight inactivation did occur. In the presence of ATP, the enzyme inactivated and dis- sociated rapidly during the first 12 hr; after that time, it did not inactivate further, but continued to dissociate slowly. The sedimentation coefficient in ATP at 0° for 0 hr is probably low; presumably the enzyme was dis- sociating during the 10 hours of centrifugation in the gradient run under dissociation conditions (ATP, 0°). A typical sedimentation pattern for native and ATP-dissociated GAPD is shown in Figure 4. The nearly exact correspondence between the location of the enzyme peaks as determined by protein analysis and the location as determined by activity assays is clearly evident. The dissociated enzyme was still active after centrifugation, but had much less activity than the native enzyme. Effect of EDTA. EDTA partially protected GAPD against inactivation (Figure 5), but not dissociation, in ATP solutions at 0°. After 10 hr in solutions containing both ATP and EDTA, GAPD lost 50 per cent of its activity, while in solutions containing ATP and no EDTA, the enzyme lost 82 per cent of its activity. And from the gradient tube containing ATP and no EDTA, no activity was recovered. EDTA also protected GAPD against inactivation in the absence of ATP. From the gradient tube containing EDTA and no ATP, 71 per cent of the activity applied was 49 .Aomuuon HO>OC mum COHCB mmHmEmm umxume on» CuHs OomswCoo ma on DOCV mHmEMm HouuCoo m muCommHmmu Honemm Cmmo Cm OCm .Hmmmso COHum>HHOMCH mad OanCmum OCH CH OGDMHoommHO OCm omum>HHOMCH omdu HE\mE H.o mqummummH poo OHHom m .Aoa UCM n mmusmHm CH omdw m>HHMC mo mammum mCu umooxmv meanCm OOHMHOOm IwHO now mCumuumm COHumquEHOmm OCm mm>uso CoHum>HDOMCH mo mmHCmHm mCH I3OHHOM 0C» HHm CH mm .OHCmHM mHCu CH .ov um Omnsmmwa puny mH OOHHOHO muH>Huom 0C9 .m musmHm you OCOOOH on» CH OmnHuommO mm mucmHOmum muHmCmO mmouosm CH OmmsuHuquo CmCu mums mmHmEmm OCH .HC OH was OCH» COHumnCOCH 0C» “N oHCOHm How OCOUOH mCu CH OOQHHOmOO mm coummuu OCM venom loud oum3 HOHuCoo O>HuMC on» OCC OHQEMm COHuMHOOmmHO OCH .Omdw um>HH mHm OOHMHOommHOImad OCM m>HuwC “Om mCkuumm COHumquEHOmm HMOHQMB .v wusmHm 50 -')’ ACTIVITY (UNITS/ML) 592 :z 20 :95.”— v ousmHm 225m . 23.212 r. s m 2 s a- a . a + R s _ _ .. 15:... 1 _ L I q ..._. 13.9 .. H N T o I 85 .v - m l was: 5.283% l 25 v 1.. I. 0.70 m I m . m m: 1 as m ms T. P p b _ _ _ _ ' ) 'a 'o) NOIIVHlNZ-IONOO NII-llozld (. (059 51 ~.C9om HCOCHH3 Omouosm CH O>HHOC mm HOC mH omdw HO>HH mHm OCC OmCOO mmOH OHO3 deem HCOCHH3 mHCOHOCHm OCH OmsmOOC OmOCOmouprOO OHCHOCH OHOmCE moo Ho Hm OHCmHm OOmH HOCHOE HOOHHOCH Cm OOC «Bow HCOCHH3 mOHmEmm OCBV .HxOH OCH CH OOHCOm (Ohm Ohm Am OusmHm How OCOmOH OCH CH OOCHHOOOO mm COHHCOCHHHHCOO HCOHOCHO MHHmCOO OmOHOCm SC OOCHCHCOV mCOHHCHom OemnCO OmOCH How OOCCmmOE mHCOHOHmmOOO COHHCHCOEHOOm OCH .EOCH mo OEOm EOHH OOHHHEO mm3 Chou HmOOxO .N OquHm How OCOmOH OCH CH OOCHHOOOO mm OOHCOHH OCC OOHOQOHQ OHO3 mHOHHCOO OCO mOHmEmm COHHOHOOmmHo .oo Hm mad SC omCU HO>HH mHm mo COHHO>HHOCCH Co «Bow HO HOOHmm .m OusmHm 52 9.85 E: m OHCMHm 5.8 E To ./ . 92 o: 06.15 J :2 umuov DHIOEIdS 53 recovered at 0°, while from the gradient tube containing no EDTA and no ATP, no activity was recovered. Quite similar values, 5.0 and 4.6 S respectively, were obtained for the sedimentation coefficients of GAPD dissociated by ATP at 0° in the presence and absence of EDTA. Thus EDTA did not protect against dissociation as it did against inactivation. In the absence of ATP, sedimentation coefficients of 7.8 and 8.2 S respectively were observed in the presence and absence of 0.1 M EDTA at 0°. EDTA had no effect on the activity or structure of native GAPD at 25° in the absence of ATP. Effect of Protein Concentration. With decreasing protein concentration, GAPD was inactivated more readily by ATP at 0° (Figure 6). The increase in inactivation was not directly proportional to the decrease in protein con- centration. After 10 hr incubation under inactivating conditions, the 1.0 mg/ml sample lost 28 per cent of its activity; the 0.1 mg/ml sample (10 per cent as much pro- tein) lost 35 per cent of its activity; and the 0.025 mg/ml sample (4 per cent as much protein) lost 81 per cent of its activity. Not only inactivation, but also dissociation increased with decreasing protein concentration (Figure 7). At 1.0 mg/ml, the dissociation pattern showed a single peak with a sedimentation coefficient of 520 w = 6.7 S. I 54 .OOHCC> was COHHCHHCOOCOO OCCO OCH HmOOxO .m OusmHm HOH OCOmOH OCH CH OOCHHOOOO mm OOHOOHH OCO OOHOQOHQ OHOS mHOHHCOO OCC mOHmemm COHHCHOOOOHO .oo Hm CBC SC omdu HO>HH mHm mo COHHO>HHOOCH CO COHHMHHCOOCOO CHOHOCQ HO HOOwum .m OHCOHm 55 c . _ W , . m .4] .yy‘rw I afflnnl‘lmn-WWEEBCE I 585 L2: 2 w e s N _ H H _ __ 4H .................. c: ........... “W i ms.o\‘u ..... ..... CI 0 IIIIIIIIII o Ilfil, 1 . “Tr. I I «/ 9:3 :59: 3\V¢ l a: 8 o OHCMHC o2 All/\llOV OlleOEldS F391!” 56 Figure 7. Sedimentation patterns for native (upper frame) and ATP-dissociated (lower frame) pig liver GAPD at various protein concentrations. These studies were carried out on the enzyme samples described in the legend for Figure 6. The samples were centrifuged in sucrose density gradients as described in the legend for Figure 3. The native enzyme samples (upper frame) are the controls described in the legend for Figure 6. The curves for the native enzyme at 1.0 and 0.1 mg/ml were identical, so only one curve is drawn for both sets. The O.D. values for 650 0.1 mg/ml and 0.025 mg/ml have been multiplied by 10 and 40 respectively to adjust them to the same scale as the 1.0 mg/ml values. This multiplication results in an over-correction because with each lower protein concen- tration, less reagent was used in protein determination and the resulting higher sucrose concentrations affected the O.D. 0 readings. 65 D' 650) PROTEIN CONCENTRATION (o. 57 1.0 _ I l V I .50 S native __ 7-61 5 (no ATP) 3 5517.85 '- /I’ T)» 1.0 __ 0. 025 28 MENISCUS Figure 7 2.0 _ 3.50 S ,l 4. 95 S dissociated .. .'\. (I- ATP i f "q‘ ‘.6.85$ " 0.1 I 1.0 mglml I f h . R ‘(6 AI"... ...."- ”x 24 20 16 12 :: :' T'II“'——O' l (10mg/ml ) —( 8 BOTTOM FRACTION NUMBER # 58 At 0.1 mg/ml, a single 5.0 S peak was observed. Native GAPD at both concentrations had a sedimentation coef- ficient of $20,w = 7.8 8. But at 0.025 mg/ml, two peaks in a 1:6 ratio with sedimentation coefficients of 5.4 and 3.5 S respectively were observed. At this lowest concentration, the sedimentation pattern for native GAPD showed three peaks with a ratio of 2:1:1 and sedimen- tation coefficients of 7.6, 5.4 and 3.5 S respectively. Even after warming to 25°, no activity was observed in fractions from gradients containing ATP-dissociated enzyme whose original concentration was 0.025 mg/ml. But at 0°, 19 and 51 per cent respectively of the activity applied was recovered from gradients containing ATP and 0.1 or 1.0 mg/ml GAPD. Effect of pH. The rate and extent of inactivation of GAPD by ATP at 0° was greater at pH values above and below pH 7.5 than they were at pH 7.5 (Figure 8). They were greatest at acid pH values. GAPD inactivated for 8 hr at pH 9.0 lost 45 per cent of its activity, GAPD inactivated at pH 7.5 lost 37 per cent of its activity, and that inactivated at pH 6.0 lost 88 per cent of its activity. The inactivation at pH 5.0 was even more extreme; all activity was lost within one hour. Without ATP, incubation of GAPD at 0° caused no loss of activity at pH 9.0 or 7.5, but there was significant loss at the acid pH values. At pH 6.0, GAPD incubated without ATP 59 .musmsaummxm nonuo 0:» ca us ed on wound IEoo mm H: m mace Mom touchdosw mums “Honucoo o.m mm on» umouxmv maouucoo tam mmHmEmm mmmnu umnu kuoc on casonm 9H .N unamflm now bummed map cw confluommp mm pmummuu can cmumamum mum: maouucoo can mmHmEMm coquHoommwp .mwwzumnuo .o.m mm .wumumom Edwvom can .o.m no m.h mm .mHoNM©HEw .o.m mm .mcflomam mo mcowusaom SE om cw uso wmwuumo was soaum>wuomcfl one .omwum> mums am can Hummus man ummoxo .tmms mos Emummm sowum>wuomsfl mad vumcsmum esp .mmflvsum omen» :H .00 um mfid an amdw um>fla mam mo cowum>fluomsa so we no uommmm .m musmflm 60 @505 ms: I D a: 277 m w opsmwm All/\llOV ouldads 61 lost 18 per cent of its activity after 8 hr; at pH 5.0 it lost all its activity after 10 hr. Thus the increased inactivation at acid pH values was caused by the acidic media as well as the presence of ATP. The enzyme sample at pH 7.5 showed a 5.6 8 peak (Figure 9), while that at pH 9.0 yielded at 5.8 S peak. There was little difference in degree of inactivation or dissociation in the samples exposed to various pH values in the range of pH 7.5 to 9.0. In contrast, GAPD at pH 6.0 dissociated to a 4.9 8 peak, suggesting it was more dissociated than GAPD at pH 7.5. However, GAPD incubated without ATP at pH 6.0 and 0° was also par- tially dissociated; it had a sedimentation coefficient of 6.8 S. The sedimentation patterns for these dis- sociations are shown in Figure 10. Enzyme incubated at pH 5.0 and 0° with and without ATP continuously dis- sociated throughout the experiment; they were spread throughout the entire gradients. From the gradients with ATP, the recovery of the activity at 0° was quite poor. From gradients at pH 5.0 or 6.0, no activity was recovered. From those at pH 7.5 and 9.0, 27 and 10 per cent respectively of the activity applied was recovered. Effect of ATP Concentration. Increasing the ATP concen- tration from O to 15 mM in the incubation medium increased the inactivation of GAPD at 0° (Figure 11), but increasing “i va'urn IMmAMM m on. | i 62 .musmflm mnu Ga Umucmmmummu uos mum moan “mxmmm coaumu IcwEHUmm on was was possumcmp mumz Houusoo was mHmEMm o.m mm one .m musmflm mom vcmmma may :a UmQHuommc mm mucmflvmum mufimsmp wmouosm CH Ummsmwuusmo mum3 memEMm one .m musmflm mom bummed wnu CH confluommp mmamEMm mahusm map so uso wmflnumo meB mmflvsum mmmsa .00 um mefl mp amdw um>HH mam mo Gowumwoommflw so mm mo uomumm .m musmwm 63 ed 9m In out. o opsmfim od m S ‘1N3 IO IJflOO N0 llVlNEIWIGEIS M ‘02 64 Figure 10. Sedimentation patterns for native (upper frame) and ATP-dissociated (lower frame) pig liver GAPD at different values of pH. These peaks are the sedimentation patterns obtained with inactivated and dissociated enzyme samples described in the legends for Figures 8 and 9. The native enzyme samples (upper frame) are the controls described in the legend for Figure 8. The curves for the native enzyme at pH 9.0 and 7.5 were identical, so only one curve is drawn for both sets. The 00650 values for pH 6.0 have been multiplied by 4 in an attempt to adjust them to the same scale as the values for pH 7.5 and 9.0. The protein determination does not work well at acid pH values. PROTEIN CONCENTRATION (0.0. o) 65 65 I 1 I I 0.16- native 6.85 S 37.8 S _ (no ATP) I“ W '\ pH 9 o .. 6. : - . ' 0.12 0 f 1&15- \_~ '- I .' ,‘ 'I 0.08— 0"! i ‘4..." — Ii" I *5 0.04" J'Ji ;‘ ‘3 .- n' P "'- l I J L I I ‘r 016— dissociated ’ _ ' 4.85 S (+ ATP) [<— pH 6.0 5.55 s 0.12— ,: I, _ ,‘(f‘4 pH 7.5 ,- ' 5.76 s ‘ 0.08% pH 9.0_) 0.04” .4 . . 4.1.. I 1‘ 28 24 20 16 12 3 MENISCUS BOTTOM FRACTION NUMBER —> Figure 10 66 .N musmflm How bsmmoa may CH pwnfluomov mm nwuwmnu was woummmum mums maonusoo can mmHmEmm coapmaoommflp .sowumuucmosoo mam ca coaumfium> map Eoum mpflmd .00 um mad an omdw Hmbwa mam mo sowum>wuomgw so sowumnucmosoo mad mo wommmm .HH onsmflm 67 585 E: as magmas All/\llOV OHIOHdS 68 the ATP concentration beyond 15 mM did not produce more inactivation. After 10 hr, the per cent activities lost from enzyme in solutions containing various ATP concen- trations were as follows (activity loss given in paren- thesis): 0 mM (30%), 2 mM (34%), 5 mM (42%), 10 mM (46%), 15 or 20 mM (56%). The dissociation of GAPD also increased when the ATP concentration of the incubation medium was raised from 0 to 15 mM (Figure 12). Similarly, ATP concen- trations above 15 mM caused no further dissociation. The sedimentation coefficients of GAPD dissociated at various ATP concentrations were as follows (sedimentation coefficients given in parenthesis): 0 mM (7.8 S), 2 mM (6.0 S), 5 mM (5.6 S), 10 mM (5.0 S) and 15 or 20 mM (4.7 S). The rate of decrease in sedimentation coef- ficient with increase in ATP concentration was more pronounced than the rate of loss of activity. The recovery of activity from the gradients at 0° decreased steadily with increasing ATP concentration. The per cent recoveries at 0° of the activity applied to the gradients were as follows (per cent recovery given in parenthesis): 0 mM (42%), 2 mM (15%), 5 mM (12%), 10 mM (11%), 15 mM (10%) and 20 mM (5%). Effect of Mercaptoethanol. Mercaptoethanol partially protected against inactivation of GAPD in solutions con- taining ATP at 0° (Figure 13). Up to 40 mM, protection 69 .m musmflm Mom Usmmoa may ca cmnwuommc mm mucmwpmum huwmcop wmouosm cw wmmsmwuuamo oumz mmHmEMm one .HH ondmwh Mow Ucmmma may as Umnwuomow mmHmEMm osmmsm map so #50 Umflunmo mum3 mmwcsum omega .00 um mad an ammo um>wa mam mo coaumwoommflt co soaumuusmoaoo mad «0 poowmm .NH ousmflm 70 92:5 FE ho zo:<~=2mozoo NH onsmsm cm 2 .o _ _ _ S 3 J N W m" m o. m OI ..... .l I.‘I.I. 0 I6! I.c.l e m z. m z m... .hr x mm ZS L S M F _ 71 .mumm anon now csmuw ma m>nso 0:0 Mano om .Hmowusovw onB Hosmnumoummouofi SE cm can 2E ov um demand UmumwOOmmwp on» How mm>uso one .uxou on» CH poucommum mum Am madman MOM Ucomma may cw Umnwuommc mm coaummsmwuusmo ucoflpmum huflmsmw omOHUSm an Umsflmunov mcowusHOm oshnso mmonu “0m pmusmmoe mucowowmmmoo coaumu Ismaflvom one .Uowum> mm3 sowumuucoosoo Hogwauooummoume on» umouxm .m musmflm Mow Ucmmma map cw confluomoc mm topmouu was woummwum mums maouusoo was mmHmEmm sofluMHOOmmHo .oo um mam an omdw nm>HH mam mo soflum>fluomsfl so sodumnucoocoo Hocmnumoummouos mo poommm .ma ousmflm 72 A285 ms: 3 w o e ms ossmsa T \M 655838.55 .25 cm .8 T .:< o: om om“ o2 All/\llOV OHIOBdS 73 increased with increasing mercaptoethanol concentration, but concentrations greater than 40 mM gave no better protection than 40 mM. In 0 mM mercaptoethanol GAPD had a sedimentation coefficient of 4.6 S, while in 40 mM and 80 mM mercaptoethanol it had sedimentation coef- ficients of 5.3 and 5.2 S respectively. The recovery of activity from the gradients at 0° was 5 per cent of the activity applied in 0 mM mercaptoethanol, 17 per cent of the activity applied in 40 mM, and 23 per cent of the activity applied in 80 mM. Effect of Temperature. Increasing the temperature of incubation from the normal 0° to 25° caused a marked decrease in inactivation by ATP (Figure 14). After 10 hr, GAPD incubated with ATP at 25° had lost no activity, but GAPD incubated at 12° and 0° had lost 43 and 50 per cent of its activity respectively. The rate of loss of activity at 12° was more linear with time than the rate of loss at 0°. The sedimentation coefficient decreased linearly with the decrease in temperature (Figure 15). The GAPD at 0° had a sedimentation coefficient of 5.2 S, that at 12° had a sedimentation coefficient of 5.5 S and the enzyme at 25° had a sedimentation coefficient of 5.8 S (Figure 16). It is perhaps surprising that GAPD incubated with ATP at 25° showed no loss in activity, but showed a substantial decrease in sedimentation coefficient 74 .N madman new osomma onu :H wonfluommw mm omummuu was vmummoum mums maouucoo was moamEMm coaumwuommflp .mmwsuonuo .Umwum> mm3 mnsumumafimu may umooxm toms was Emumwm coflum>fluomsfi mad vumccmum onu .mmfltsum mmmnu cH .mad an ammo uo>fla mam mo coaum>wuomsfi so ousumummfiou mo poommm .wa ousmflm 75 as assess 58$ “2: S w c v N All/\UOV OlzllOEldS 76 .Auxmu momv 30H on was omm um unoeoflwmooo noeuwunoaevom one .ntmuuon pony m m.n u 3.omm mo munwwowmmmoo sowumusmeflpom won maouunou on» Hna .nn m.mn you any ooo.me um cmmsmnnuamo mums mucmncmum .o on» was «un m.oa MOM Emu ooo.mv um cmmSMfluunmo mnoB munmflpmnm oma mnu «an me.h MOM Emu ooo.mv um Ummsmwuunmo mums munmfivmum 0mm one .noeuo>wuomne How poms pmnu mm ousumummfimn mEMm on» no uso Umfluumo mos monEmm «0 now nomm mo cowummSMHuunmo umnu umooxo .m musmflm Mom Gammon mnu cw vmnwuomwv mm munmfipmum euemnmp mmouosm CH UmmDMHHusmu oum3 mmHmEMm one .va ousmflm MOM vcmmma mnu nH confluommt monEmm oEMNnm mnu so uso pmwuumo mums mmflvsum mmmne .me4 en omnu Hw>fla mam mo nowumwOOmmHU no musuwummfimu mo pummmm .ma ousmflm 77 I T ‘8 ‘. \O \O \. '— \O \0 \c \0 \‘ 9. \‘ I \0 \O \. - \O \0 \O \ cl? 3‘ * 0. \O In m M 0as ‘lNEIlOHflOO NOI1v1N3wmas 10 HBWPERATURE Figure 15 78 .no>uso oEom onu o>mm maouunoo onp Haov ea ousmwm now unomoa on» ma ponfluomov nonunoo onu we oameom o>euoc one .ma van «a mousmem MOM monomoa onu Ge wonflnomoo moamfiom ofienno touofloommflo paw wouo>flpoonfl onu nuH3 nonwouno nuouumm nofluounoeflcom onu ouo mxoom omone .mousuoquEou unouommep no Gena Ho>wa mam pouoHUOmmHUImed was o>wuon MOM mnuouuom noeuounofiwvom .ma ousmflm 79 59222 20 :05.”— on onsmna 225m 88 .52 he s w - s s a a a + mm s - . .1- o. _ _ _ . . r . 4 x a oo 11 ..... 0W . .AIONH ..I 3.0 o o ’ M. .N/OmN . a .. . a 3% m a m ,1 . I m a; ../ I 2 a 3:2). . 932835 I m S m I as I .. I as _ _ _ m w : _ _ _ _ 'CI '0) No |1V81NEION03 N131oaa (059 80 (compared to native GAPD). Although these samples were not assayed in the presence of ATP, there was no evidence for reactivation during the assay. Hence, reassociation and reactivation in the assay seems ruled out as an explanation for the results. A more likely explanation is that the GAPD dissociated during centrifugation. This explanation is supported by the facts that GAPD incubated with ATP at 25° and GAPD incubated without ATP at 25° had identical activities (within experimental error) before sedimentation, but enzyme with ATP had 80 per cent less activity than enzyme without ATP after sedimentation. Similar results were observed with fully reactivated GAPD centrifuged in the presence of ATP (see next section). REACTIVATION AND REASSOCIATION OF GLYCERALDEHYDE 3-P DEHYDROGENASE. The conditions required for the reacti- vation and reassociation of yeast (Stancel and Deal, 1969) and rabbit muscle GAPDS (Constantinides and Deal, 1969) after inactivation and dissociation by ATP at 0° are well defined. These conditions, however, are not adequate to reactivate and reassociate similarly treated pig liver GAPD (Dagher and Deal, in preparation). The idea that EDTA might be the missing required compound was suggested by the discovery that EDTA increased the activity of pig liver GAPD under inactivating and dissociating conditions (see section "Effect of EDTA"). 81 Consequently, a short study was undertaken to characterize the reactivation and reassociation of GAPD in the presence of EDTA. The importance of EDTA for the reactivation of GAPD can be seen in the first two lines of Table 2; the per cent activity regained was two fold greater in the presence of EDTA than in its absence. To affect the Ireactivation, the EDTA had to be present throughout both the inactivation and reactivation; it could not be added for just the reactivation by warming to 25° (first three lines, Table 2). Since regain of activity was not 100 per cent even in the presence of EDTA, a systematic study to see whether the addition of other compounds to the EDTA- containing reactivation solution could produce complete recovery of activity was undertaken (lines 4, 5 and 6, Table 2). Adding NAD improved the reactivation; adding sucrose decreased it. Together they cancelled each other's effects and gave no change in reactivation. The last line of Table 2 illustrates the importance of reducing agent to reactivation; reactivation in the absence of mercaptoethanol was only slightly better than in the absence of EDTA. A study of the effect of incubation time at 0° on reactivation is shown in Table 3., In most cases, little difference was seen between inactivating for 12 82 .ouo>wuoonw haouonEoo non UHU moamsom one .nwe m mom omN on onwauo3 an touo>wuooou can u: «H on on “on man an“: .o um manumnsoan an aoua>nuoaaw mum: monogamo .nowuo>euooou you omm ou msflsuos ouomon umsfl snow wouo>auomsfl oo on» on poops ouoz mussomeoun .med SE oH can Hononuooumoouofi SE om .m.n mm «eam s H.o .oHONMnHEw SE om nonwounoo Eswpos sowuo>fluoonw cuovnoumo m.mm mmouosm won .naz as m I + ago a.mm mmouosm won .aaz as m + + oco «.mn omouosm won + + ono can I m.om onz SE m + + osu m.mv neam z a.o + I ono m.mm I m.om nowufipwm on + + 03» e.mm I m.mm nowuwowo on + I oounu ovonwomom unfl>wu04 nnowuo>wuooom oaononuo moamaom Hocflmfluo ucou mom can on wound mossomsou Ioumoouoz oneom mo Honfisz mussomaou ucouommwo mo monomoum onu aw omdw vouo>wuoonH mo sowuo>wuooom .N oanoe 83 .oom>wuoonw eaouoameoo no: cap moamfiom one .Uouooflonfi mosey on» you oo um mnwuonsonfl an wouo>wuoonw mos ofihunmo .nowuo>wuooou Ugo noHuo>HuoonH nuon Mom #50 umoa ouo3 monsomfioo wouueEO .nowuo>fluooou mnwcnwmon oHOMon posh woosoouuna ouos monsomfioo novnd .men as oa can Hosonuooumoonoe SE om .m.e we deem z H.o .oHoNoUHEH SE om voawmunoo nowusaom sowuo>fluuoce Unovnoumn .GHE m mom omm on mGHEuoz en wouo>euooou mo3 ofihnnmo Hononuooummuuoe I m.mv e.Hm m.mm .anz 2E m ~omouosm woa + o.vm v.vn e.mm anz :8 m .omouosm woa + H.@@ m.vm ooa de SE m + m.~m v.ve «.me omouosm won + «.me o.om m.mm noeusHom wuownoum onowuo>fluoocH um onowuo>wuoone H: onoauo>wuoone Hm nnowusaom nofluo>wpomne mm moped wonwomom em Houmm ponwomom NH Houmn nonwomom tuopnoum on» Some woupwao >ue>wuo¢ unou mom >uw>wuom ucoo Mom muw>fiuod unoo mom no on covvfl monsomeou osowuo>wuooom :0 oo um oEHe sowuonsonH mo uoommm one .m oanoe 84 and 24 hr, but a significant loss in per cent activity regained occurred after inactivating for 36 hr. An exception to this trend was reactivation in the presence of NAD; it showed a steady decline in per cent activity regained upon additional incubation at 0° for several hours. The kinetics of reactivation in the presence of EDTA after inactivating for 12 hr are shown in Figure 17. The kinetics after inactivating for 24 and 36 hr were very similar. Maximal activity was reached in 3.5 to 5 min, a remarkably short time compared to the 12 hr required for inactivation (Figure 2). The sedimentation behavior of reactivated GAPD was examined in sucrose density gradients to see whether the reactivated enzyme had also reassociated. Fully reactivated enzyme centrifuged in gradients without ATP showed a pattern like that in Figure 18. The sedimen- tation coefficients of native and reassociated GAPD were identical, within experimental error, indicating that the enzyme was completely reassociated as well as reactivated. However, fully reactivated GAPD centrifuged in gradients containing ATP showed a pattern like that in Figure 19. Native GAPD had a sedimentation coefficient of 7.8 S, but reactivated GAPD had a sedimentation. coefficient of 5.4 S. 85 .HoHHEHm onoz Cofluo>euooCH an mm no em uoumo Coeuo> Ifluooou mo mofluoCflx one .omOHOSm no onz mCHUvo en pomsoo oECHo> CH ooCoquMHU onu How wouoouuoo ouos moCHo> eufi>euom one .ouo>wuooou on own um wouoam Con» wCo omouosm wmn UCo adz.z ~.o mCHmC mCoHuouuCoo ICoo omouosm UCo anz ouowumoummo onp on woumsnpo ouos moamfiom onu .oo no COHuonCOCH en Coauo>wuomCH Hound .Hn NH moa oEHu CoaumnCOCH.one .HOConuooumoouoE 0C won oamfiom oCo umooxo .N ousmflm Mom UComoH onu CH confluomot mo wouoouu cCo wouomoum ouo3 moamfiom Codoowoommfla .oo um med an oouo>wuooCH omdw Ho>HH mam mo Coeuo>auooou mo mowuoCHM .eH ousmwm 86 on s 352.5 us: an onsmnm 92 5:: m .3203 3.. S m s. m N _ _ _ 855838.55 oc/r \\ (M... ' "L‘Afl oooooooooo . $8.53 a: S .............. ............... Alr- LicoEuum o: 4 \. o . . d . F .1 o 1:2 2:. m_\v.q.\ All/\llOV OHIOJdS 87 .noouuon uOCv CoHuomaHuuCoo Monmo >UH>Huom HMUHpCoUH won monEom one .me< woCHouCoo uCoHpon HonuHoC nCo 0mm no Mn me.e “Om Emu ooo.mv no mos COHuoqCMHHuCoo on» umooxo .m oquHm How UComoH onu CH conHHomov mo muCoHpon euHmCou omouusm CH womCMHuuCoo ouo3 HouuCoo oCo onEom on» .CHE m How 0mm on mCHEHMB Houm< .de z N.o mCHmC.Q¢z SE m ou woumsnvo mo3 onEom COHuoHUOmmHU onu .COHuonsoCH Hound .un NH mos oEHu COHuonCoCH on» “N onsmHm How UComoH onu CH wonHuomoo mo oouoouu UCo wouomoum onos HonnCoo muH UCm onEom CoHuMHUOmmHU d .men “ConuHs womCMHuuCoo omnw uo>HH mHm vouoHUOmmoou mo Cuouuom COHuouCoEHvom .mH oHCmHm 88 an ossmHa $5222 20 :05: 55:8 mama“: s I m s s a I a s H mm om — ’c O — _ — O I l‘ql‘ I — C ox \ C. o .4 , . .g. II. o .o . I136 t H I . I85 . H. . 82682: I e x INS ... EEZIT. . I m a: I as _ _ _ n _ _ _LIL..IIL_ '0 '0) NOI1VII1N33NOO NI31ozIa 059 ( 89 .uxou onp CH wouComoum mH mH wco mH moHCmHm Coosuon mCuouuom COHuouCoEHUom CH oOCoHoMMHU onu mo COHmmsomHv a .omm um nonsmoos ponu mH couuon muH>Huoo one .omm on us me.» now sou ooo.mv on was aoHummsoHuucmo one umouxo .m onsmHm uom 6ComoH on» omouoCm CH vomSMHuuCoo ouo3 ”MHHsHm can swap “aHe m mom .mm COHuoHUOmmHo onu .CoHuonCoCH onu «m ousmHm How 6ComoH on» CH CH conHHomow mo mUCoHvon auHmCow mCOHuCHOm oemnCo one .moHuH>Huoo ou onuos ouos HouuCoo pCo onEom Hound .un NH mos oEHu COHuonCOCH ponHuomoU mo vouoouu UCo vouomoum ouos HouuCoo muH UCo onEom COHuoHoommHU C .med nuH3 nomCMHHuCoo omnw uo>HH mHm touoHoommoou mo Cuouuom COHumuCoEHUom .mH oquHm 90 .) ACTIVITY (UNITS/ML) $9222 20 :05: $8.52 an ansmnm cm l 3.: mod NH .o S .o '0'0) NOI1va1NzIoNOO NI31ozIa( ) (059 91 One explanation for the different sedimentation pattern in the presence of ATP is that ATP prevents reassociation, but not reactivation. This explanation seems unlikely because reactivated GAPD and native GAPD had identical activities (within experimental error) before sedimentation, but reactivated enzyme had 50 per cent less activity than native enzyme after sedimentation. If ATP prevents reassociation but not reactivation, this activity loss would not be expected. A more likely explanation of the different sedimentation pattern is that the reactivated enzyme was inactivated and dis- sociated in the gradient containing ATP. This expla- nation also agrees with similar results obtained for dissociation of GAPD at 25° with ATP (see section "Effect of Temperature"). No experiments with native enzyme in ATP containing gradients were performed, so it is not known whether native enzyme will also inactivate and dis- sociate under these conditions. AMINO ACID COMPOSITION. To provide further chemical characterization of pig liver GAPD, the amino acid com- position was determined (Table 4). The values listed represent a summary of data obtained from 18 hr and 56 hr hydrolyses. Threonine, serine, methionine and tyrosine were gradually degraded during hydrolysis; the values obtained at 18 hr and 56 hr were extrapolated back to 0 time to yield the ratios used in calculating the 92 Table 4. Amino Acid Composition Amino Acid Residues per Monomer Pig Liver GAPD Pig Muscle GAPDa aSpartic acidb threonine serine glutamic acidb glycine alanine valine methionine isoleucine leucine tyrosine phenylalanine ammonia lysine histidine arginine proline cysteine tryptOphan 37.3 20.1 18.9 20.9 37.0 32.1 32.3 12.0 19.4 20 10.5 13.5 57.2C 23.5 9.0 10.2 21.1 3.8 4.7 38 22 19 18 32 32 34 9 21 18 9 14 18 26 11 10 12 4 3 aData of Harris and Perham (1968). b No information is available concerning the number of amidated residues, although typically they are about 20-30% of the total. cThis value may be high because no attempt was made to remove ammonia from the HCl used in hydrolysis. 93 values given in the table. Valine, isoleucine and lysine were not completely released by the 18 hr hydrolysis; the concentrations at 56 hr were used directly to calcu- late the values listed. All other amino acid values were an average of the sample hydrolyzed for 18 hr and that hydrolyzed for 56 hr. Residues were calculated from nanomoles by assigning the arbitrary value of 20 to leucine. (Leucine was chosen because its value remained constant for both hydrolyses. Twenty was chosen because the number of the various amino acid residues, calculated from 20 leucines, agreed favorably with the number expected, assuming all GAPDS are similar.) The total number of residues calculated per monomer was 346. Using the molecular weights of the individual amino acids, a monomeric molecular weight of 38,070 daltons and a tetrameric molecular weight of 152,300 daltons are calculated. These results are in reasonable agreement with the values of 38,000 (monomer) and 148,000 (tetramer) obtained by Dr. Dagher using SDS polyacrylamide gel electrOphoresis and high-speed sedi- mentation equilibrium respectively (Dagher and Deal, in preparation). A partial specific volume of 0.7376 cc/g was calculated from the amino acid composition (method of Cohn and Edsall [1943] as described by Kayne [1966]). This value is very similar to the partial specific volume 94 of 0.737 cc/g at 20° measured for pig muscle GAPD (Elodi, 1958). The partial Specific volume of most proteins is between 0.70 and 0.75 cc/g (Schachman, 1957). The amino acid composition of pig muscle GAPD is also presented in Table 4 for comparison with that of pig liver GAPD. The two compositions are quite sim- ilar, except that pig liver GAPD has more proline. Pig muscle GAPD has only 332 residues per monomer (13 less than pig liver GAPD) and its monomer has a molecular weight of 36,000 (compared to 38,070 for pig liver). SULFHYDRYL REACTIVITY OF NATIVE GLYCERALDEHYDE 3-P DEHYDROGENASE. Because sulfhydryl groups frequently play important roles in maintaining the structure of proteins and catalyzing enzymic reactions, determination of the number of sulfhydryl groups, their reactivity and their location within a given protein molecule are of interest. The number, reactivity and location of these groups are often determined by analyzing their reactivity toward sulfhydryl reagents since this reactivity depends heavily on the number of groups, how reactive they are and how accessible they are to the solvent. The titration of available sulfhydryl groups with DTNB is shown in Figure 20. The first four groups reacted almost instantly (in less than 5 sec). The second four sulfhydryl groups reacted at a moderate 95 .oonooon museum HanowanCm no Conn mCHuonnoom nanH no onsmoofi o oHOE mH Comon CoHnonHmHoon nonnm o>uso COHnoHan one .EC omm no mnHoHnHCn mCHHCmooE an ooEHHMCoo moz oEmNCo oonooon onn no COHnonHmHooum one .oom m Conn mmoH CH wounsooo museum Hanomnanm Hson anHn onn no COHnoHan one .EC mHv no oom.mH no nCoHoHnnooo CoHnoCano HoHOE onn nnH3 oonoHCoHoo mos noEmunon Mom oonoHan mosonon HmuoennHCm no HonECC one .o.m mm .nonnsn ononmmonm EsHoom z mo.o CH namdo on oomomfioo mmooxo CHOH mmv mzeo on ov UCo Ceca on com .QCCG m: ovm ooCHonCoo onsanE COHnooon one .moonnoz oCo mHoHuonoz CH ooanomoo mo mzeo nnH3 oononan mm3 Dena .omdo no>HH mHQ o>HnoC no hnH>Hnooon HmuomnnHCm .om onCmHm 96 [50.D.412 m5 3 .. 352.3 ms: 8 0.0.9.0 ‘ m: 53 n n confiscate oN ansmnm NH 2.. HBWVHBL/SEITIOISEIH 'IMICMHIIOS 97 rate (in approximately 30 min). Two additional sulf- hydryl groups reacted very slowly, requiring 90 min from the beginning of the reaction to become fully titrated. At approximately 90 min, GAPD began to denature and precipitate. Light scattering off this precipitate gave the final rapid rate observed between 90 and 130 min. (The precipitation was confirmed by observing the turbidity at 650 nm.) Whether the last six sulfhydryl groups were titrated during this rapid structural change was not determined, but it seems likely that these groups would have become available for titration during the unfolding which usually proceeds denaturation. It also seems probable that the two sulfhydryl groups titrated between 30 and 90 min were made available for titration by the enzyme slowly beginning to unfold. DISCUSSION EDTA EFFECTS AND THE POSSIBLE PRESENCE AND ROLE OF ZINC. In previous work by Dagher and Deal (in preparation), pig liver GAPD had appeared to be partially resistant to dissociation and to show almost no reactivation. The original objective in this research was to determine why the characteristics of the dissociation of pig liver GAPD by ATP at 0° appeared to be so different from those of yeast and rabbit muscle GAPDS. The key discovery in solving this problem was finding that EDTA was required for producing an active dissociated enzyme and for reactivation. This discovery was especially important because in the previous work activity determination had been the only technique used to analyze for dissociated (by reactivating) and reactivated enzyme. EDTA stabilized the activity of native pig liver GAPD and stabilized against activity loss in the dis- sociation system. When the enzyme was incubated at 0° with or without ATP, it retained more activity in the presence than in the absence of EDTA. And a comparison of enzyme dissociated by ATP at 0° with and without EDTA showed that in both cases the dissociation products were 98 99 similar in structure (based on their sedimentation coef- ficients), but the "dimer" in the absence of EDTA had much less activity. The stabilization by EDTA suggests that GAPD is sensitive to metals and that the mechanism for this stabilization of enzymatic activity may be related to the ability of EDTA to chelate metals. One possible mechanism is the relief of heavy metal inhibition of catalytic activity. This mechanism is supported by Velick and Furfine's observation (1963) of the inhibition of GAPD by heavy metals and by Nagradova's studies (1965) on the increased activity of rabbit muscle GAPD in the presence of chelating agents. However, the mechanism appears unlikely because adding EDTA to native pig liver GAPD did not increase its activity. A second possible mechanism--the removal of metal ions whose presence on the enzyme disturb its structure enough to make it more susceptible to activity loss under inactivating con- ditions (low temperature or the presence of ATP at 0°) and to interfere with reactivation--seems quite plausible. It is interesting to speculate about how these metal ions, that can be removed by EDTA, affect the activity of GAPD. From the data discussed in this thesis, it appears that the metal ions only affect reactivation and the activity of the dissociated enzyme; they do not affect dissociation or the activity of the native enzyme. 100 These selective effects suggest that there is a dif- ference between the way that metal ions are bound to the dissociated and the native forms of the enzyme because the metal ions affect the behavior of the former, but not the latter. Polgar (1964) and Boross (1965) have discovered that metal ions can bind to GAPD through the sulfur atom of the cysteine residues. Ovadi, Nuridsany and Keleti (1973) have strong evidence that at least one cysteine residue per monomer which is normally buried in the tetrameric GAPD is exposed by dissociation in ATP at 0°. Consequently, it seems likely that the metal ions on GAPD could bind the sulfhydryl residues that are newly exposed by dissociation, causing the subunits to be bound together in an incorrect conformation and caus- ing a decrease in the activity of the dissociated enzyme and in reactivation. (Heavy metal ions have numerous binding sites that are not always filled; for example, zinc can form tetrahedral and octahedral complexes.) If indeed EDTA is chelating metal ions, the most likely source of these ions would be the ZnCl2 precipi- tation step at the beginning of the purification pro- cedure (Dagher and Deal, in preparation). The idea that pig liver GAPD could successfully carry bound zinc through the purification and crystallization is supported by Keleti's contention that zinc is firmly bound to a number of GAPDS (Keleti EE.E£°' 1962). And though Park 101 has convincingly argued that active GAPD does not require (or even have) bound zinc (Barkman, Sandstead and Park, 1970), the large amounts of data on zinc binding to GAPD gathered by Keleti (Keleti and Telegdi, l959a,b; Keleti EE.§£°' 1962; Keleti, 1964, 1966) and the study of the mechanism of zinc binding to GAPD by Polgar (1964) strongly imply that zinc can bind to GAPD, even if the binding is nonphysiological. Whether or not zinc is bound to the crystalline pig liver GAPD, our data support Park's theory that zinc is nonessential (Barkman, Sandstead and Park, 1970) rather than Keleti's idea that zinc is required for catalytic activity and the maintenance of steric structure (Keleti and Telegdi, l959a,b; Keleti gt_al,, 1962). For example, chelation by EDTA had no observable effect on the activity or structure of native GAPD. Dis- sociated GAPD had less activity in the absence than in the presence of EDTA. And GAPD without EDTA lost more activity at low temperature or in the presence of ATP at 0° than GAPD with EDTA. Although zinc appears to be nonessential for normal functioning, the question of control of enzymatic activity by zinc should still be considered. There is approximately 12 ug of zinc per gram of tissue (wet weight) in the normal liver (Leucke, personal communi- cation). Zinc could aid in decreasing the activity of 102 GAPD in the cell by binding to the dissociated enzyme and decreasing its activity. (GAPD in the absence of EDTA was more inactivated by low temperature or ATP at 0° than GAPD in the presence of EDTA.) Chelation of zinc and reversal of the action of the dissociating agents would reactivate the GAPD when full activity was required by the cell. (EDTA was required for reacti- vation.) But such a mechanism is negated by our data showing that chelation must occur before dissociation begins if reactivation is to be successful. (EDTA added just before warming did not effect reactivation.) How- ever, since inactivation at low temperature is non- physiological, the significance of the mechanism and its negation cannot be assessed. No experiments have been conducted to ascertain whether or not zinc actually is bound to crystalline pig liver GAPD purified by the method of Dagher and Deal (in preparation). Until such experiments are performed, it is impossible to be certain that the EDTA effects observed are related to the chelation of zinc. DISSOCIATION INTO DIMERS. Based on the sedimentation coefficients of 4.9 to 5.1 S, the ATP-dissociated GAPD had a molecular weight of 74,900 to 77,200 daltons, assuming a globular protein. The expected dimeric molecular weight is 76,100 daltons, using the monomeric molecular weight calculated from the amino acid 103 composition. Therefore, the product of ATP dissociation under standard conditions is assumed to be a dimer. Theoretically, the 4.9 to 5.1 S enzyme could be the result of a tetramer-monomer equilibrium, instead of a dimer. This possibility seems unlikely because although sedimentation coefficients intermediate between those of tetramer and dimer were seen, no sedimentation coefficients intermediate between those of dimer and monomer were seen (Figure 7). In a tetramer-monomer equilibrium, the entire spectrum of intermediate sedi- mentation coefficients would be expected. Constantinides and Deal (1969) also observed a tetramer-dimer equilibrium, but not a monomer-dimer equilibrium in sucrose density gradients with rabbit muscle GAPD dissociated by ATP at 0°. With sedimentation velocity experiments in the ultra- centrifuge, they were able to resolve the peak, whose sedimentation coefficient was intermediate between those of tetramer and dimer, into two peaks whose sedimentation coefficients were identical to those of tetramer and dimer. The peak with the sedimentation coefficient value of 6.8 S (Figure 7) is intermediate between the sedimen- tation coefficients expected for tetramer (7.8 S) and for a folded dimer (5 S). This observation of a single peak with a sedimentation coefficient intermediate between those of a tetramer and dimer requires 104 explanation. The simplest explanation is that a rapid (or at least a semi-rapid) equilibrium is established between the tetrameric and dimeric species. Furthermore, the sedimentation coefficient is a weight average quantity; for example, as concentration or other changes cause a redistribution of mass between dimer and tetramer, the weighted average sedimentation coefficient would reflect the change in distribution. In order for this rapid (or semi-rapid) equilibrium explanation to be the correct one, the rate of tetramer-dimer interconversion must be comparable to the rate of sedimentation in the gradient. The rate of enzymatic dissociation (10-12 hr) and the rate of tetramer-dimer equilibration do not have to be the same to observe a single peak; the mixture applied to the gradient is very near to equilibrium, so if partial resolution into dimeric and monomeric species were to occur, kinetic forces would cause material in both peaks to be converted into the other_species. Under these conditions it is not surprising that we do not get resolution. In conclusion, we do not postulate that the peak with an intermediate sedimentation coefficient represents a system in complete rapid equilibrium; what we do postulate is that the occurrence of association- dissociation reactions results in sufficient mass sedi- menting at a position intermediate between those of the tetramer and dimer to prevent resolution of the tetramer and dimer components. 105 Unlike ATP dissociation of yeast (Stancel and Deal, 1969) and rabbit muscle GAPDS (Constantinides and Deal, 1969), ATP dissociation of pig liver GAPD did not readily produce monomers. However, at the very low con- centration of 0.025 mg/ml GAPD, dissociation produced 85 per cent monomers and 15 per cent dimers (Figure 7). Because the lower limit for our protein determination technique was 0.025 mg/ml and because the monomer (unlike the dimer) was inactive, we were unable to investigate whether even lower concentrations of GAPD would dis- sociate completely into monomers. The inactive monomer could not be reactivated using the standard conditions. ENZYMATIC ACTIVITY OF DISSOCIATED GLYCERALDEHYDE 3-P DEHYDROGENASE. Our data suggest the existence of an active dimer of pig liver GAPD. The 4.9 to 5.1 S product from the standard inactivation-dissociation system was active at 0° after sucrose density gradient centrifugation. In sucrose gradients of dissociated GAPD, the sedimen- tation coefficient calculated from the activity determi- nation was always that of a dimer, even when the sedimen- tation coefficient calculated from the protein determination was lower. A comparison of dimer and tetramer specific activity indicates that the dimer may be 50 per cent as active as the tetramer. In the reassociation studies when reactivated GAPD was centrifuged in a gradient containing ATP, the dissociated species formed during centrifugation 106 was 50 per cent as active as the native enzyme centrifuged in a gradient without ATP. In the inactivation studies, the enzyme lost approximately 50 per cent of its specific activity under most conditions (exceptions: without EDTA, at low protein concentration, at pH 6.0 and at 25°), and analysis of the size of this partially inactivated enzyme in sucrose density gradients indicated that it was a dimer. There is no evidence for an active monomer. The only monomer observed was found after inactivating GAPD at 0.025 mg/ml, and it had no activity. The existence of an active dimer has been proposed by other workers. Constantinides saw indications of an active dimer in reassociation studies of ATP-dissociated rabbit muscle GAPD using sucrose gradients containing ATP and in inactivation studies at 7° (Constantinides and Deal, 1969). His data were similar to our data described above. Keleti's laboratory reported an active dimer of pig muscle GAPD produced by incubation with ATP at 0° in the presence of mercaptoethanol and isolated on a Sephadex G-100 column (Ovadi 23 21., 1971). The dimer had 50 per cent as much activity as the tetramer. Active dimers have also been found in other types of dissociation systems. Elodi (1958) found active dimers after dissociation in KCN, and Kirschner and 107 Schuster (1970) found the same after dissociation in high ionic strength. However, the active enzyme sedi- mentation velocity studies of Hoagland and Teller (1969) using rabbit muscle GAPD dimers produced by dissociation at low temperature failed to show any active dimers. The fact that not all dissociated enzyme is active suggests that different dissociation techniques produce different conformations of dissociated enzyme with dif- ferent activities. COMPARISON OF INACTIVATION AND DISSOCIATION OF PIG LIVER GAPD WITH THOSE OF OTHER GAPDS. Comparing the inactivation and dissociation of pig liver GAPD with those of yeast GAPD (Stancel and Deal, 1968, 1969), few similarities are found. The pig liver enzyme dissociates to active dimers, the yeast enzyme to inactive monomers; both dis- sociated enzymes are fairly compact. Inactivation requires 12 hr for pig liver GAPD and only 50 per cent of the activity is lost; complete inactivation requires only 5 hr for yeast GAPD. Optimal protein concentration for both enzymes is 0.1 mg/ml; but concentrations as low as 0.03 mg/ml of yeast GAPD will give the same dis- sociated monomer, while 0.025 mg/ml of pig liver GAPD gives a mixture of monomers and dimers. The yeast monomers produced at both concentrations can be reactivated; the pig liver dimers produced at the higher concentration can be reactivated, but the monomers 108 produced at the lower concentration have not yet been reactivated. The pH at which the maximal rate and extent of inactivation and dissociation occur for yeast GAPD is pH 9.0; for pig liver GAPD it is a pH range of 7.5 to 9.0. The Optimal ATP concentration for dissociation is 15 mM for the pig liver enzyme and 1-2 mM for the yeast enzyme. Both enzymes inactivate best at 0°; but pig liver GAPD still inactivates well at 12°, while yeast enzyme shows only 50 per cent inactivation. And both enzymes are normally inactivated in the presence of mercaptoethanol to permit reactivation. Comparing the inactivation and dissociation of pig liver GAPD to those of rabbit muscle GAPD (Constan- tinides and Deal, 1969), several similarities are found. Both enzyme dissociate to active dimers at one protein concentration and to inactive monomers at a lower protein concentration. However, for pig liver the concentrations are 0.1 mg/ml and 0.025 mg/ml respectively, and for rabbit muscle they are 1.0 mg/ml and 0.1 mg/ml respec- tively. Also, rabbit muscle monomers can be reactivated, while pig liver monomers have not yet been reactivated. Both enzyme inactivate reversibly in the pH range 7.5 to 9.0; but the rabbit muscle enzyme still inactivates reversibly down to pH 6.5, while pig liver enzyme inactivates irreversibly at pH 6.0. Inactivation of rabbit muscle GAPD takes two hours and results in a 109 90 per cent loss of activity; inactivation of pig liver GAPD takes 12 hr and results in a 50 per cent loss of activity. Both enzymes inactivate best at 0°; but pig liver enzyme still inactivates at 12°, while rabbit muscle enzyme does not. Optimal ATP concentrations for dissociation are 2 mM ATP for rabbit muscle GAPD and 15 mM ATP for pig liver GAPD. Both enzymes are normally inactivated in the presence of mercaptoethanol to permit reactivation. The conditions required for inactivation and dis- sociation of pig muscle GAPD have not been carefully characterized (Ovadi gt $1., 1971), making it difficult to compare its inactivation and dissociation with those of pig liver GAPD. However, the following points can be noted. In both cases the process can occur at 0° and pH 8.5. Pig muscle GAPD at 1.4 mg/ml will dissociate to dimers at ATP concentrations as low as 0.4 mM ATP; pig liver GAPD at 0.1 mg/ml requires 10 mM ATP to dis- sociate to dimers. Both enzymes give active dimers only if mercaptoethanol is present. Under maximal inactivation conditions, complete inactivation is obtained in 5 hr with pig muscle GAPD and in 12 hr with pig liver GAPD. In summary, the inactivation and dissociation of pig liver GAPD are more similar to those of pig muscle and rabbit muscle GAPDS than to those of yeast GAPD. 110 This trend is not surprising since the former three are mammalian sources while the latter is microbial. COMPARISON OF REACTIVATION AND REASSOCIATION OF PIG LIVER GAPD WITH THOSE OF OTHER GAPDS. There are almost no similarities in the reactivation and reassociation of pig liver GAPD and those of yeast GAPD (Stancel and Deal, 1969). Yeast GAPD reactivates at 17°, pig liver at 25°. Reactivation of yeast GAPD takes one hour; reactivation of pig liver GAPD takes five min. Yeast enzyme requires 1-2 mM ATP and 10 per cent sucrose for reassociation; pig liver enzyme requires 5 mM NAD and 0.1 M EDTA, and ATP and sucrose hinder the reassociation. But both enzymes require mercaptoethanol to reassociate. And both enzymes show decreasing reactivation with increasing periods of inactivation. Many similarities are observed in the reactivation and reassociation behavior of pig liver GAPD and that of rabbit muscle GAPD (Constantinides and Deal, 1969). Both enzymes reactivate at room temperature, and reacti- vation takes approximately 5 min. Both enzymes require mercaptoethanol for reassociation, and reassociation is hindered by ATP. With both enzymes, the centrifugation of reactivated enzyme in a sucrose gradient containing ATP causes the appearance of a dissociated enzyme. The only difference is the requirement of 0.1 M EDTA and 5 mM NAD by pig liver GAPD for reactivation; rabbit muscle GAPD has no such requirements. 111 In summary, the reactivation and reassociation characteristics of pig liver GAPD are very similar to those of rabbit muscle GAPD and very different from those of yeast GAPD. The similarities and differences are like those for inactivation and dissociation. COMPARISON OF AMINO ACID COMPOSITION WITH THAT OF OTHER gaggg. A comparison of the amino acid composition of pig liver GAPD with those of pig muscle (Harris and Per- ham, 1968), rabbit muscle (Harris and Perham, 1963; Velick and Furfine, 1963; Allison and Kaplan, 1964), yeast (Jones and Harris, 1972), lobster muscle (Davidson g; 21" 1967), human muscle (Allison and Kaplan, 1964; Wolny, 1968), chicken muscle, Escherichia sell and halibut muscle GAPDS (Allison and Kaplan, 1964) revealed that pig liver GAPD has approximately the same ratio of acidic, basic and hydr0phobic amino acids as all the other GAPDS. However, pig liver GAPD may have 9 to 17 more residues per monomer than the other enzymes. Comparing the amounts of individual amino acids found in all the different GAPDS, pig liver GAPD has 30 per cent more methionine and 60 per cent more proline than the other GAPDS; it also has slightly more glycine (as does human muscle GAPD). The amounts of cysteine, aspartic acid and serine in pig liver GAPD are like the amounts found in other mammalian GAPDS. The number of threonines, valines, phenylalanines and histidines in 112 pig liver enzyme are similar to the numbers in all the vertebrate enzymes and in lobster muscle GAPD. And all the enzymes including pig liver GAPD have approximately the same amount of glutamic acid, alanine, isoleucine, leucine, tyrosine, lysine and arginine. Pig liver GAPD also has a small number of tryptophans as is common for all GAPDS (except E. 321;), but, like rabbit muscle GAPD, it has four or five residues rather than the typical three. This strong conservation of the amino acid com- position of GAPD, even among phylogenetically distant species, suggests a selection:against variations pro- duced by mutations. Apparently, this enzyme, as a member of the basic pathway for converting carbohydrates to energy and producing storable complex sugars, is so important that the cell will not tolerate alterations of its specificity and structure. COMPARISON OF SULFHYDRYL REACTIVITY OF PIG LIVER GAPD WITH THAT OF OTHER GAPDS. In pig liver GAPD, one cysteine per monomer reacts rapidly, a second reacts at a moderate rate and the two remaining cysteines are buried (observed by titration with DTNB). This pattern of sulfhydryl reactivity is the same as that observed with pig muscle GAPD (titrated by PCMB) (Boross, Cseke and Vas, 1969; Vas and Boross, 1970) and with lobster muscle GAPD (titrated by organic mercurials) (Wassarman, Watson and Major, 1969). 113 The extremely reactive cysteine residue is probably Cys-149, a major functional group in the active site; other workers have established a connection between the titration of one very reactive cysteine and the inactivation of GAPD, and have identified the cysteine involved (Boross and Cseke, 1967; Szabolcsi, Biszuku and Sajgo, 1960; Bernhard and MacQuerrie, 1971). The moderately reactive cysteine is most likely Cys-153 as proposed by Vas and Boross (1970) from their work on PCMB titration of pig muscle GAPD. The buried cysteines are probably Cys-281, the sulfhydryl group exposed after ATP dissociation of pig muscle GAPD into dimers (Ovadi, Nuridsany and Keleti, 1973), and Cys-244, which is exposed only after the enzyme is unfolded. In rabbit muscle (Smith and Schachman, 1971) and pig muscle GAPDS (Szabolcsi, Biszuku and Sajgo, 1960; Friedrich and Szabolcsi, 1967), the titration of two sulfhydryls per monomer with PCMB caused the enzyme to unfold and expose the buried sulfhydryls. In pig liver GAPD, the titration of two sulfhydryls per monomer with DTNB caused the enzyme to precipitate. Since pre- cipitation can result when unfolded protein aggregates, it seems likely that titration of two cysteines did cause pig liver GAPD to unfold. But because the pre- cipitate caused so much light scattering, it is impossible to tell whether the buried sulfhydryls 114 were exposed and titrated. However, the titration of one cysteine on two of the subunits (reaction between 30 and 90 min) before the precipitate became too heavy suggests that at least some of the buried cysteines were being exposed and titrated. The nearly identical sulfhydryl reactivity of native enzyme seen with all GAPDs studied is even more strongly conserved than the overall amino acid compo- sition of GAPD. This strong conservation indicates that the position of some or all of the sulfhydryl residues is extremely important to the maintenance of the structure and activity of the enzyme. SUMMARY AND CONCLUS IONS The main accomplishments in this research were (1) the complete dissociation of liver GAPD by ATP and the characterization and optimization of this process, (2) the discovery of the right conditions for reacti- vation and reassociation, and (3) the discovery of the sensitivity to metals which is so critical for reacti- vation and for activity of the dissociated enzyme. Pig liver GAPD was inactivated and dissociated by ATP at 0°. The Optimal conditions were 0.1 mg/ml GAPD, pH 7.5 to 9.0, 15 mM ATP and 0°. Under these conditions the process required 10 to 12 hr. At low pH values or low protein concentrations, the native enzyme dissociated at 0° in the absence of ATP. The dissociated product was,a dimer with a sedimentation coefficient of 4.9 to 5.2 S; if it was produced in the presence of 0.1 M EDTA and 40 to 80 mM mercaptoethanol, the dimer was active. The inactivated and dissociated enzyme was reactivated by simply warming to 25°. Reactivation required 3.5 to 5 min. The optimal conditions for‘ 115 116 reactivation were 0.1 M EDTA, 80 mM mercaptoethanol, 25°, 5 mM NAD and no sucrose. In the absence of ATP, the reactivated GAPD reassociated to a tetramer with a sedimentation coefficient of 7.8 8 (same as that of native GAPD). The amino acid composition of pig liver GAPD was found to be similar to that of other GAPDS, especially the mammalian enzymes. The sulfhydryl reactivity of native pig liver GAPD is also very similar to the reactivity of other GAPDS. The requirement of EDTA for the production of active dimers and for reactivation of the dissociated enzyme suggests that EDTA chelates a metal ion(s) on the enzyme. The metal ion could easily be zinc picked up during purification. From our data it was determined that zinc, if present, is nonessential, but it was specu- lated that zinc could play a possible role in an enzymatic control mechanism. The physiological significance of reversible dissociation and inactivation at 0° and low protein con- centrations in the presence of high levels of chelating agent is questionable. But the.importance of this research lies not in its questionable physiological significance, but in its commentary on the forces required to maintain the native structure of GAPD. These forces have been discussed by Stancel (1970) in 117 his study on reversible inactivation and dissociation of yeast GAPD by ATP at 0°. The binding of ATP disturbs the steric structure and eXposes a part of the hydro- phobic core of the enzyme. These hydrophobic bonds are destabilized by low temperature, and the enzyme dis- sociates. Warming the enzyme to 25° permits the hydro- phobic bonds to reform. Mercaptoethanol is required to ' keep disulfide bonds from forming (Ovadi, Nuridsany and Keleti, 1973), and EDTA keeps zinc from binding to the cysteine residues exposed by dissociation (Polgar, 1964); otherwise, reactivation and reassociation could not occur. Further experiments suggested by this research are (1) an analysis of whether or not zinc is present on crystalline pig liver GAPD, (2) a discovery of the source of the zinc if it is present, (3) an examination of possible GAPD dissociation at 25° and 37° (low protein concentration), (4) a search for the dissociation con- ditions required at high protein concentrations, (5) a study to see whether or not the catalytic properties of the dissociated product, active dimer, differ from those of native tetramer, (6) a search for the conditions required for complete dissociation into monomers, (7) a search for the conditions required to reactivate monomers, (8) a study of the dissociation of NAD-free enzyme to see if it requires less ATP than enzyme with bound NAD, (9) an analysis of the effect of various 118 EDTA concentrations on reactivation and production of active dimers, and (10) a study of the possible reversible dissociation of pig liver GAPD by other dissociating agents such as KCl, (NH4)ZSO4 and urea. LI ST OF REFERENCES LIST OF REFERENCES Adelstein, S. J., and Vallee, B. L., g. Biol. Chem., 233 (1958) 589. Agatova, A. I., and Kurganov, B. I., Dokl. Akad. Nauk. SSSR, 169 (1966) 1452. Allison, W. S., Connors, M. J., and Parker, D. J., Biochem. Biophys. Res. Commun., 34 (1969) 503. Allison, W. S., and Kaplan, N. 0., g, Biol. Chem., 239 (1964) 2140. Anne, K. C., and Timasheff, S. N., Biochemistry, g (1970) 1481. Barkman, D. E., Sandstead, H. H., and Park, J. H., g, Biol. Chem., 245 (1970) 1036. Batke, J., and Cennamo, C., Int. g. Biochem.nl§ (1972) 537. Benitez, L. V., and Allison, W. 8., Archives Biochem. Biophys., 159 (1973) 89. Bernhard, S. A., and MacQuarrie, R. A., Biochemistry, lg (1971) 2456. Boers, W., Oosthuizen, C., and Slater, E. C., Biochim. Biophys. Acta, 250 (1971) 35. Boers, W., and Slater, E. C., Biochim. Bithys. Acta, 315 (1973) 272. Bondi, E., Watkins, J., and Kirtley, M. E., Biochim. Biophys. Acta, 185 (1969) 297. Boross, L., Biochim. Biophys. Acta, 96 (1965) 52. Boross, L., Acta Biochim. Bithys. Acad. Sci. Hung., 4 (1969) 57. 119 120 Boross, L., and Cseke, E., Acta Biochim. Bioghys. Acad. Sci. Hung., 2 (1967) 49. Boross, L., Cseke, E., and Vas, M., Acta Biochim. Biophys. Acad. Sci. Hung., 4 (1969) 301. Buehner, M., Ford, G. C., Moras, D., Olsen, K. W., and Rossman, M. G., Proc. Nat. Acad. Sci. US, 19 (1973) 3052. Buehner, M., Ford, G. C., Moras, D., Olsen, K. W., and Rossman, M. G., g. M01. Biol., 82 (1974) 563. Campbell, J. W., Duee, E., Hodgson, G., Mercer, W. D., Stammers, D. K., Wendell, P. L., Muirhead, H., and Watson, H. C., Cold S rin Harbor Symp. Quant. Biol., 26 (1971) 155. Cantau, B. N., Jaureguiberry, G. J., and Pudles, J., Eur. J. Biochem., I6 (1970) 208. Chance, 3., and Park, J. H., g. Biol. Chem., 242 (1967) 5093. Chothia, C., Q. M01. 13101., 15 (1973) 295. Cohn, E. J., and Edsall, J. T., Proteins, Amino Acids and Fe tides, Reinhold PEBIisEing, New YorE, I913, p. 373. Constantinides, S. M., and Deal, W. C., Jr., g. Biol. Chem., 244 (1969) 5695. Constantinides, S. M., and Deal, W. C., Jr., g. Biol. Chem., 245 (1970) 246. Conway, A., and Koshland, D. E., Jr., Biochemistry: Z (1968) 4011. Cook, R. A., and Koshland, D. E., Jr., Biochemistry, g (1970) 3337. Cseke, E., and Boross, L., Acta Biochim. Bioghys. Acad. Sci. Hung., 3 (1967) 39. Dagher, S., and Deal, W. C., Jr., in preparation. Darnall, D. W., and Murray, L. V., Biochem. Bithys. Res. Commun., 46 (1972) 1222. Davidson, B. E., Sajgo, M., Noller, H. G., and Harris, J. I., Nature, 216 (1967) 1181. 121 Deal, W. C., Jr., Biochemistry, J_(1969) 2795. DeVijlders, J. J. M., Boers, W., Hilvers, A. G., Harmsen, B. J. M., and Slater, E. C., in P ridine Nucleo- tide Dependent Deh dro enases (edited By H. Sund) Springer VerIag, Ber in, e1 elberg, New York, 1970, p. 233. ' Duggleby, R. G., and Dennis, D. J., J- Biol. Chem., 249 (1974) 162, 167, 175. Edolhoch, H., Biochemistry, Q (1967) 1948. Ehring, R., and Colowick, S. P., J. Biol. Chem., 244 (1969) 4589. Elodi, P., Acta Physiol. Acad. Sci. Hung., JJ_(1958) 199. Elodi, P., and Liber, S., in P ridine Nucleotide Dependent Deh dro enases (edited By H. Sund} Springer Verlag, Ber in, Hei e erg, New York, 1970, p. 175. Elodi, P., and Szorenyi, E., Acta Physiol. Acad. Sci. Hung., 9 (1956) 339. Fairbanks, G., Steck, T. L., and Wallach, D. F. H., Biochemistry, lg (1971) 2606. Fenselau, A., J. Biol. Chem., 247 (1972) 1074. Ferdinand, W., Biochem. J., 23 (1964) 578. Forcina, B. G., Ferri, G., Zapponi, M. C., and Ronchi, 8., Eur. J. Biochem., 22_(1971) 535. Foucault, G., Traore, F., Levilliers, J., and Pudles, J., Eur. J. Biochem., 12 (1974) 43. Fox, J. B., Jr., and Dandliker, W. B., J. Biol. Chem., 221 (1956) 1005. Francis, S. H., Meriwether, B. P., and Park, J. H., J. Biol. Chem., 246 (1971) 5427, 5433. Frieden, C., Biochim. Bithys. Acta, 41 (1961) 428. Friedrich, P., and Szabolcsi, G., Acta Biochim. Bithys. Acad. Sci. Hung., g (1967) 59. Fuller, R. C., and Gibbs, M., J. Plant Physiol., 31 (1959) 324. 122 Goryunov, A. I., Andreeva, N. S., Baranowski, T., and Wolny, M., J. Mol. Biol., JJ (1972) 421. Habeeb, A. F. S. A., in Meth. Enzymol., Vol. XXV (edited by C. H. W. Hirs and S. N. Timasheff) Academic Press, New York, London, 1972, p. 457. Hageman, R. H., and Arnon, D. I., Archives Biochem. Biophys., 21 (1955) 421. Harrigan, P. J., and Trentham, D. R., Biochem. J., 135 (1973) 695. Harrington, W. F., and Karr, G. M., J, Mol. Biol., JJ (1965) 885. Harris, J. I., in P ridine Nucleotide Dependent Dehydro- enases (edited 5y H. Sund) Springer Verlag, Ber in, Heidelberg, New York, 1970, p. 57. Harris, J. I., Meriwether, B. P., and Park, J. H., Nature, 198 (1963) 154. Harris, J. I., and Perham, R. N., Biochem. J., J9 (1963) GOP. Harris, J. I., and Perham, R. N., Nature, 219 (1968) 1025. Harrison, J. H., Fed. Proc., J; (1963) 493. Heppel, L. A., in Meth. Enz mol., Vol. II (edited by S. P. Colowick and N. 0. Kaplan) Academic Press, New York, London, 1963, p. 530. Hoagland, V. D., and Teller, D. C., Biochemistry, J (1969) 594. Holland, M. J., and Westhead, E. W., Biochemistry, lg_ (1973) 2264. Jaenicke, R., in Pyridine Nucleotide Dependent Dehydro- . enases (edited by H. Sund) Springer Ver ag, BerIin, Heidelberg, New York, 1970, pp. 71, 209. Jaenicke, R., Schmid, D., and Knof, 8., Biochemistry, J (1968) 919. Jakoby, W. B., in Meth. Enz mol., Vol. XXII (edited by W. B. Jakoby) Academic Press, New York, London, 1971, p. 248. 123 Jones, G. M., and Harris, J., FEBS Lett., JJ (1972) 185. Kayne, M. S. (1966) BE. 2. Thesis, Michigan State Uni- versity. Keleti, T., Biochim. Biophys. Acta, JJ (1964) 422. Keleti, T., Biochem. Biophys. Res. Commun., JJ (1966) 640. Keleti, T., in Pyridine Nucleotide Dependent Dehydroggé nases (edited by H. Sund) Springer Verlag, Berlin, Heidelberg, New York, 1970, p. 103. Keleti, T., Gyorgi, S., Telegdi, M., and Zaluska, H., Acta Physiol. Acad. Sci. Hung., JJ (1962) ll. Keleti, T., and Telegdi, M., Acta Physiol. Acad. Sci. Hung., JJ (1959a) 281. Keleti, T., and Telegdi, M., Acta Physiol. Acad. Sci. Hung., JJ (1959b) 235. Kirschner, K., Eigen, M., Bittman, R., and Voigt, B., Kirschner, K., and Schuster, I., in P ridine Nucleotide Jependent Deh dro enases (edited 5y H. Sun Springer Ver ag, Ber in, Heidelberg, New York, 1970, p. 217. Kirschner, K., and Voigt, B., J, Phy . Chem., 349 (1968) 632. Kochman, M., and Rutter, W. J., Biochemistry, 7 (1968) 1671. '- Lambert, J. M., and Perham, R. NJ, FEBS Lett., JJ (1974) 305. Lakatos, S., Zavodszky, P., and Elodi, P., FEBS Lett., JJ_(1972) 324. Lebherz, H. G., and Rutter, W. J., Science, 157 (1967) 1198. Lebherz, H. G., Savage, B., and Abacherli, E., Nature New Biol., 245 (1973) 269. Levitzki, A., Biochem. Biophys. Res. Commun., JJ (1974) 889. 124 Marangos, P. J., and Constantinides, S. M., J. Biol. Chem., 249 (1974a) 951. Marangos, P. J., and Constantinides, S. M., Biochemistry, l; (1974b) 904. Markovich, D. S., Zavodskii, P., and Vol'kenshtein, M. V., Biokhimiya, El (1966) 873. Marti, G. E., and White, B. J., Proc. Iowa Acad. Sci., JJ (1968) 109. Martin, R. G., and Ames, B. N., J. Biol. Chem., 2 6 (1961) 1372. Mejbaum-Katzenellenbogen, W., and Dobryszycka, W. M., Clinica Chemica Acta, J (1959) 515. Moore, J., Jr., and Fenselau, A., Biochemistry, J; (1972) 3753, 3762. Nagradova, N. K., Biokhimiya, JJ (1965) 50. Nagradova, N. K., and Grozdova, I. D., Biochim. Biophys. Acta, 342 (1974) 105. Nagradova, N. K., and Guseva, M. K., Biokhimiya, JJ (1971) 841. Nagradova, N. K., Vorona, M. K., and Asriyants, R. A., Biokhimiya, JJ (1969) 627. Oguchi, M., Meriwether, B. P., and Park, J. H., J. Biol. Chem., 248 (1973) 5562. Olson, E., and Park, J. H., J. Biol. Chem., 239 (1964) 2316. Ovadi, J., Nuridsany, M., and Keleti, T., Biochim. Biophys. Acta, 302 (1973) 191. Ovadi, J., Telegdi, M., Batke, J., and Keleti, T., Eur. J. Biochem., JJ_(1971) 430. Park, J. H., in Probes Struc. Funct. Macromol. Membranes, Proc. CoII . Jofinson Res. Found., 5th (Edited By B. C ance) Academic Press, New York, 1971, p. 83. Park, J. H., Meriwether, B. P., Clodfeller, P., and Cunningham, L. W., J. Biol. Chem., 236 (1961) 136. 125 Peczon, B. D., and Spivey, H. 0., Biochemistry, J; (1972) 2209. Perham, R. N., Biochem. J., 111 (1969) 17. Perham, R. N., in Probes Struc. Funct. Macromol. Mem- branes, Proc. Collo . Johnson Res. Found., 5th (Edited by B. C ance) Academic Press, New York, 1971, p. 95. Perham, R. N., and Harris, J. I., J. Mol. Biol., 1 (1963) 316. Polgar, L., Experientia, JJ (1964) 408. Racker, E., Mechanisms i2 Bioener etics, Academic Press, New York, London, 1965, CB. 1-5. Rosenberg, L. L., and Arnon, D. I., J. Biol. Chem., 217 (1955) 361. Sapag-Hagar, M., Rev. Espan. Fisiol., JJ (1969) 201. Schachman, H. K., in Meth. Enzymol., Vol. IV (edited by S. P. Colowicka nd N. 0. Kaplan ) Academic Press, New York, 1957, p. 32. Schlessinger, J., and Levitzki, A., J. Mol. Biol., 82 (1974) 547. Shaltiel, S., and Soria, M., Biochemistry, J (1969) 4411. Shibata, Y., and Kronman, M. J., Archives Biochem. Biophys., 118 (1967) 410. Smith, C. M., and Velick, S. F., J. Biol. Chem., 247 (1972) 273. Smith, G. D., and Schachman, H. K., Biochemistry, JJ (1971) 4576. Smith, M., and Khorana, H. G., in Meth. Enzymgl., Vol. VI (edited by S. P. Colowick and N. 0. Kaplan) Academic Press, New York, London, 1963, p. 645. Stallcup, W. B., and Koshland, D. E., Jr., J. Mol. Biol., 80 (1973) 41, 63, 77. Stancel, G. M. (1970) Ph. D. Thesis, Michigan State University. 126 Stancel, G. M., and Deal, W. C., Jr., Biochem. Biophys. Res. Commun., Jr (1968) 398. Stancel, G. M., and Deal, W. C., Jr., Biochemistry, J (1969) 4005. Suzuki, K., and Harris, J. I., FEBS Lett., JJ (1971) 217. Szabolcsi, G., Biszku, E., and Sajgo, M., Acta Physiol. Acad. Sci. Hung., ll (1960) 183. Teipel, J. W., and Koshland, D. E., Jr., Biochemistry, JJ (1971) 792, 798. Trentham, D. R., Biochem. J., 122 (1971) 59. Vallee, B. L., in The Enrymes, Vol. III (edited by P. Boyer, H. Lardy and K. Myrback) Academic Press, New York, London, 1960, p. 252. Vallee, B. L., and Hoch, F. L., Proc. Nat. Acad. Sci. JJ, ‘1; (1955) 327. Vallee, B. L., Hoch, F. L., Adelstein, S. J., and Wacker, W. E. C., J. Jg. Chem. Soc., ZJ (1956) 5879. Vas, M., and Boross, L., Acta Biochim. Bithys. Acad. Sci. Hung., J (1970) 215. Velick, S. F., J. Biol. Chem., 203 (1953) 563. Velick, S. P., Baggott, J. P., and Sturtevant, J. M., in Pyridine Nucleotide Dependent Dehydrogenases (edited by H. Sund) Springer Verlag, Berlin, Heidelberg, New York, 1970, p. 229. Velick, S. F., and Furfine, C., in The Enz mes, Vol. VII (edited by P. D. Boyer, H. Lardy and K. Myrback) Academic Press, New York, London, 1963, p. 243. Von Ellenrieder, G., Kirschner, K., and Schuster, I., Eur J. Biochem., JJ (1972) 220. Warburg, O., and Christian, W., Biochem. J., JJ (1939) 677. Wassarman, P. M., and Major, J. P., Biochemistry, J (1969) 1076. Wassarman, P. M., Watson, H. C., and Major, J. P., Biochim. Biophys. Acta, 191 (1969) 1. 127 Watson, H. C., and Banaszak, L. J., Nature, 204 (1964) 918. Watson, H. C., Duee, E., and Mercer, W. D., Nature New .Biol., 240 (1972) 130. Welton, A. (1974) JJ. 2. Thesis, Michigan State University. Wolny, M., Acta Biochim. Polon., JJ_(1968) 137. Yang, S. T., and Deal, W. C., Jr., Biochemistry, J (1969a) 2806. Yang, S. T., and Deal, W. C., Jr., Biochemistry, (1969b) 2814. lm Zapponi, M. C., Ferri, G., Forcina, B. G., and Ronchi, 8., FEBS Lett., JJ_(1973) 287. Zavodskii, P., Hung. Sci. Instrum., JJ_(1968) 15. Zavodszky, P., Abaturov, L. B., and Varshavskii, Y. M., Acta Biochim. Biophys. Acad. Sci. Hung., J (1966) 389. MICHIGAN STATE UNIVERSITY LIBRARIES IIII IIIII IIIIIIIII 3 1293 03 45 1937