AN INVESTIGATION INTO THE STRUCTURE AND FUNCTION OF 2-KETO-3-DEOXY- 6- PHOSPHOGLUCONATE ALDOLASE OF PSEUDOMONAS PUTIDA USING l, FLUORO, 2,4 - DINITROBENZENE Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY LESLIE ROHIT BARRAN 1 96 9 dd r: “Hui” IJBRARY I I Michigan St; 3?: [JUNVETThgi 1: saw ' This is to certify that the thesis entitled AN INVESTIGATION INTO THE STRUCTURE AND FUNCTION OF 2-KETO~5-DEOXY-6-PHOSPHOGLUCONATE ALDOLASE OF PSEUDOMONAS PUTIDA USINC 1, FLUORO, ?,h- DINITROBENZENE presented by Leslie Rohit Barran has been accepted towards fulfillment of the requirements for Ph.D. degree in Biochemistry Major professor Date October 27. 1969 0-169 BINDING or ? HOME 8 SONS' BOOK BINDERY "I” LIBRARY HINDI . again,“ ABSTRACT AN INVESTIGATION INTO THE STRUCTURE AND FUNCTION OF 2-KETO-3-DEOXY-6-PHOSPHOGLUCONATE ALDOLASE OF PSEUDOMONAS PUTIDA USING l, FLUORO, 2,4- DINITROBENZENE BY Leslie Rohit Barren The enzyme 2-keto-3-deoxy-6-phosphogluconate aldolase (KDPG aldolase) cleaves 2-keto-3-deoxy-6-phosphogluconic acid (KDPG) to pyruvate and glyceraldehyde-B-phosphate (G-B-P). Reaction of the aldolase with l, fluoro, 2,4-dinitrobenzene at pH 8.5 led to the uptake of four moles of FDNB per mole of enzyme resulting in 85- 901 inactivation of the enzyme. The uptake of as many as nine moles of FDNB per mole of enzyme only gives approximately 95% inactivation. The sites of dinitrophenylation were identified as the g-amino groups of lysine. However, the azomethine lysine was shown not to be a site for FDNB reaction. Protection against the uptake of three moles of FDNB was shown to be substrate specific while the uptake of the fourth mole of FDNB may represent non- specific dinitrophenylation. The DNP aldolase did not appear to be dissociated when examined by polyacrylamide gel electrophoresis or by sucrose gradient centri- fugation. An analysis of the kinetics of dinitrophenylation suggested that the uptake of a single DNP mole per mole of enzyme resulted in 50% inactivation and a concomitant change in enzyme conformation. Leslie R. Barran - 2 Subjection of the dinitrophenylated enzyme to isoelectric focuss- ing revealed that two separate protein peaks were present. The minor peak focussed at pH 8.0 (the same p1 as native enzyme) and contained 1.6 DNP moles per mole of enzyme. The major peak constituted 95% of the total protein placed on the gradient. This peak focussed at pH 5.1 and contained approximately four DNP moles per mole of enzyme. Since the reaction of amino groups of lysine residues with FDNB should not result in an increase in pI, the increase in p1 was ascribed to the occurrence of a conformational change induced by the uptake of FDNB. This suggestion finds support in the kinetic analysis of the dinitrophenylation reaction. The enzyme constituting the major peak showed a twofold increase in the Km for KDPG and a fifteenfold decrease in Vmax. The rate of tritium exchange for the fully dinitrophenylated enzyme from T20 to pyruvate was not found to be rate limiting as judged by the inability of aldehydes to stimulate the rate of cleavage of KDPG. Two experiments were carried out to verify the suggestion that the uptake of FDNB by KDPG aldolase results in a conformational change. The first experiment involved a comparison of the heat inactivation rates of native and DNP aldolase. The second experiment was a comparison of the extent of uptake of Ellman's reagent (5,5'- dithiobis-[Z-nitrobenzoic acid]) for native and DNP aldolases. The results of both of the above experiments are in accord with the con— clusion that KDPG aldolase undergoes a conformational change on dinitrophenylation. AN INVESTIGATION INTO THE STRUCTURE AND FUNCTION OF 2-KETO-3-DEOXY-6-PHOSPHOGLUCONATE ALDOLASE 0F PSEUDOMONAS PUTIDA USING 1, FLUORO, 2,4- DINITROBENZENE By Leslie Rohit Barran A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1969 This thesis is dedicated to my wife, Mary Ann, and my son, Craig, who showed great forbearance and understanding during the course of this study. 11 ACKNOWLEDGMENTS The author extends his sincere appreciation to Dr. W. A. Wood for his guidance, encouragement, and support throughout the course of this research. The author is further indebted to his laboratory colleagues whose stimulating discussions helped to contribute to this research. Thanks are also extended to his wife, Mary Ann, for her assistance in the preparation of this manuscript. iii VITA Leslie Rohit Barran was born on July 29, 1939, in Guyana, South America. He graduated from Queen's Collegiate, Guyana, in July 1958. He then worked for the following year as a laboratory technician in the Soil Chemistry division of the Department of Agriculture, Guyana. He emigrated to Canada in September, 1959, and received the degree of Bachelor of Science in Agriculture from McGill University, Montreal, in May 1963. He received the Master of Science degree from the same institution in May 1965. He accepted a graduate research assistant- ship in the Department of Biochemistry at Michigan State University from January, 1965, until the present. Mr. Barran is married and has a son. TABLE OF CONTENTS LIST OF TABLESOOOOO0.0.0.0.0....OOOOOOIOOOOOOO0.0.0.000...0.... LIST OF FIG‘IRESOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO00......0...... ABBREVIATIONSOOOIOOOOOOOOOOOOOOOOOOOOOOOOOOOOO00.0.00...0.0.... CHAPTER I. II. III. IV. INTRODUCTIONOOOOOOOOOOOOOOOOOOOOOOOOOOO00......0...... LITERATURE REVIEWOOOOOOCOOOCOOOCOOOO...0.0.0.0000...O. The Chemistry of Fluorodinitrobenzene................. The Reaction of FDNB with Proteins.................... The Reaction of FDNB with Aldolases................... Physical and Chemical Characteristics of KDPG Aldolase....................................... METHODS AND MATERIAISOOOOOOOOOOOOO00.000.000.000...0.. BHCtQtiOlogiCalooooooooooooooooooooooooooooooooooooooo Chemical.............................................. Enzymatic Synthesis of 2-keto-3-deoxy- 6-ph05phogalaCtOnateooooooooooooooooooooooosoo DCterminationS and Procedures......................... Measurement Of FDNB Uptake...................... Reductive Coupling of Pyruvate-3-140 and KDPGal-3'14C to KDPG Aldolase................. Measurement of the Rate of Tritium Exchange..... Radioactivity Measurements...................... ISOEIECtriC FOCUSSingooooooooooooooooooooooooooo Heat Inactivation...o........................... SUlthdIYI Determinations....o.............o.... Enzymatic....................................... RESULTSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOO Dinitrophenylation of KDPG Aldolase................... Establishment of Conditions for Dinitrophenylation............................ Specificity of Substrate Protection Against Dinitrophenylation.................... Identification of the Sites of Dinitrophenylation............................ Page vii viii ix U1U) 14 18 18 19 19 26 26 27 28 28 29 3O 30 3O 32 33 33 39 44 CHAPTER Page PhYSical State Of DNP AldOlaseooooooooooooooooooooooo 52 Polyacrylamide Gel Electrophoresis............. 52 Sucrose Gradient Centrifugation...oo.o......... 56 Kinetics Of Dinitrophenylation....................... 56 Isoelectric Focussing of DNP Aldolase................ 60 Effect of Dinitrophenylation on the Catalytic Characteristics 0f KDPG Aldolase................... 67 Determination of the Km for Native and DNP AldOlaSeooooooooooooooooooooooooooooooooo 67 The Rate of Tritium Exchange of Native and DNP Aldolase............................. 69 Effect of Dinitrophenylation on Enzyme Conformation.. 71 Heat Inactivation.............................. 72 Availability of SH Groups in Native and FDNB Aldolase................................ 73 v. DISCUSSIONOOOCOCOOOOOI.0...O....0...OOOOOOOIOOOOOOOOO 77 SpeCifiCIty Of FDNB Uptake....o...................... 77 Sites Of Dinitrophenylationo......................... 78 Homogeneity Of DNP Aldolase.......................... 79 Tritium EXChange Of KDPG AldOlaSeoooooooooooooooooooo 80 Kinetics Of DinitrophCDYIationooooooooooooooooooooooo 81 Effect of Dinitrophenylation on Enzyme Conformation.. 82 Relation of FDNB Reactive Sites to the Enzyme Anion Binding Siteoo00000000000000.00000000000ooooo 84 A Model for Dinitrophenylation....................... 86 VI. SWARYOOOOOOOOOOOOOOOOOOOO...0.0.0.000...0.0.0.0.... 90 APPENDIXOOOOOOOOOOOOOOOOCOOCOOOOOOOOOOOOOOOO0.0...00.0.0000... 92 Relative Rates of Release of B-formylpyruvate from KDG and KDCal...o.................................. 92 Calculation of the Binding of KDPGa1-3-IAC to KDPG Aldolase...................................... 92 The Binding of KDPGal-3-14C to KDPG Aldolase Prior to Dinitrophenylation.............................. 94 The Reaction of KDPG Aldolase with FDNB-U-lac........ 95 Analysis of the Kinetics of Dinitrophenylation....... 95 Details for the 0RD of Native and DNP A1dolase....... 96 LIST OF REFERENCESOOOOOCO00.00.0000...OOOOOOOOOOOOOOOOOOOOOOOO 102 vi LIST OF TABLES TABLE Page 1. DinitrophenYIation 0f KDPC AldOlaseooooooooooooooooooooooo 36 2. Stoichiometry of Binding of KDPGal-3-14C to KDPG Aldolase........................................... 43 3o DinitropheHYIation 0f KDPCal-Coupled AldOIaseooooooooooooo 45 4. Stoichiometry of Binding of FDNB—U-IAC by KDPG Aldolase... 49 5. Paper Chromatographic Identification of the Amino ACid DinitrophCDYIatedoooooooooooooooooooooooooooooooooo 51 14 6. Stoichiometry of Pyruvate-3- C-coupled A1dolase.......... 53 7. Properties of "Peak 1" and "Peak 11" Aldolases Obtained by Isoelectric Focussing....................... 66 80 Km for Native and DNP AldOlaseoooooooooooooooooooooooooooo 68 9. Tritium Exchange of Native and DNP A1dolase............... 7O 10. Titration of Sulfhydryl Groups in Native and DNP AldOlaSe USing DTNBOO...OOOOOOOOOOOOOOOOOOOOOOO0.0.0.... 76 APPENDIX TABLE 4 1. The Reddction of KDPGal-B-1 C to KDPG Aldolase in the Presence Of SOdimn BOIOhYdrideooooooooo0000000000000 92 4 2. The Borohydride Reduction of KDPGa1-3-1 C to KDPG AldOlaseOOOOOCOOOOI.0...OOOOOOOOOOOOOOOOOOOOOOOC.00 9i+ 3. Second Order Plot for the Kinetics of Dinitrophenylation.. 96 4. The Rate of Reaction of the Fast-reacting Lysine and the Rapid Initial Loss of Enzyme Activity........... 97 vii FIGURE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. LIST OF FIGURES Dinitrophenylation of KDPG Aldolase at pH 9.1............ Dinitrophenylation of KDPG Aldolase at pH 8.5............ Dinitrophenylation of KDPG Aldolase at pH 8.5 in the Presence Of KDPGalooooooooooooooooooooooooooooooooooooo Absorption Spectrum of DNP Aldolase...................... Polyacrylamide Gel Electrophoresis of Native Aldolase.... Polyacrylamide Gel Electrophoresis of DNP Aldolase....... Kinetics of Dinitrophenylation........................... Isoelectric Focussing of Native KDPG Aldolase............ Isoelectric Focussing of DNP Aldolase.................... Temperature Inactivation of Native and DNP Aldolase...... A MOdel for DinitrophCUYIationooooooooooooooooooooooooooo APPENDIX FIGURE 1. 2. 3. Plot of the Relative Rates of Release of B-formyl- pyruvate from KDC and KDGaloooooooooooooooooococoo.oono Optical Rotary Dispersion of Native and DNP A1dolase..... The Moffit-Yang Plot of the 0RD of Native Aldolase....... viii Page 35 38 42 48 55 55 58 62 64 75 88 93 99 101 Pi p1 NADH FDNB CDNB DNP KDPG KDPGal KDGal F-l-P FDP DHAP G-3-P 5' AMP GTP NaBH DTNB T O N Ki ABBREVIATIONS inorganic phosphate isoelectric point reduced nicotinamide adenine dinucleotide l, fluoro, 2,4-dinitrobenzene l, chloro, 2,4-dinitrobenzene dinitrophenyl radical 2-keto-3-deoxy-6-phosphogluconate 2-keto-3-deoxy-gluconate 2-keto-3-deoxy-6-phosphogalactonate 2-keto-3-deoxy-galactonate fructose-l-phosphate fructose 1, 6-diphosphate dihydroxyacetonephosphate glyceraldehyde-B-phosphate lactic acid dehydrogenase adenine nucleotide-5'-phosphate guanosine triphosphate sodium borohydride 5-5' dithio-bis-(2-nitrobenzoic acid) (Ellman's reagent) trichloracetic acid tritiated water Michaelis constant for substrate binding Michaelis inhibition constant ix sRNA poly U poly C N-(Morpholino)ethanesulfonic acid transfer RNA polyuridylic acid polycytidylic acid CHAPTER I INTRODUCTION The enzyme 2-keto-3-deoxy-6-phosphogluconate aldolase (KDPG aldolase) cleaves 2-keto-3-deoxy-6-phosphogluconate (KDPG) to pyruvate and glyceraldehyde-B-phosphate. KDPG aldolase from £3. utida was crystallized and later shown to consist of three subunits. Mechanisti- cally, the cleavage of KDPG is believed to proceed via a Schiff base mechanism. The enzyme forms an azomethine with pyruvate which could be reductively coupled to the enzyme with sodium borohydride. Degradation of the labelled enzyme revealed that pyruvate forms an azomethine with the €>amino group of a lysine residue. Studies are presently underway to introduce labelled pyruvate into the active site of KDPG aldolase, and by the use of standard analytical protein techniques to ascertain the sequence of amino acid residues at the active site. Dinitrophenylation studies, using 1, fluoro, 2,4-dinitrobenzene, suggested that the sites of uptake of the dinitrophenyl groups on the enzyme may constitute the binding sites for the phosphate group of KDPG. Assuming the above hypothesis to be true, it should be possible to label the sites on the enzyme responsible for binding the substrate phosphate group, and to degrade the labelled enzyme with trypsin. Subsequent purification, and sequencing of the labelled peptide would give valuable information concerning the residues that constitute the active site. The results of such a study, in conjunction with the 2 amino acid sequence around the azomethine site, would greatly enhance our knowledge of the amino acid residues that participate in catalysis. The intent of this study is to examine the effect of dinitro- phenylation on the physical, and catalytic properties on KDPG aldo- lase, and to test the hypothesis that dinitrophenylation arylates lysine residues which participate in binding the phosphate group of the substrate. CHAPTER II LITERATURE REVIEW The Chemistry of Fluorodinitrobenzene Fluorodinitrobenzene (FDNB) and related compounds readily undergo nucleophilic substitutions, since they contain the electron withdrawing nitro-groups, in both the ortho and para positions. These substitu- tions are believed to proceed through an intermediate complex, formed by the addition of the nucleophile to the carbon undergoing substitu- tion, and converting that carbon to one with its substituents arranged in a tetrahedral configuration (1, 2). The chemistry of nitroactivated halobenzenes has not been fully explored and most of the work has been limited to reactions with amines. Many reactions of amines with nitroactivated halobenzenes and related compounds have been reported to be base catalyzed (3). However, not all of such reactions are base catalyzed and the occurrence of such kinetic effects are not yet well understood. Moreover, there is some controversy over their interpretation. Many reactions are mildly accelerated by bases while others are strongly catalyzed. The reactions of nitroactivated arylchlorides, bromides, and iodides with primary and secondary amines are mildly base catalyzed (1, 4). Some reactions of the fluorides are strongly base catalyzed while others show mild accelerations (5). Reactions of 2,4-dinitrofluorobenzene with n-butylamine and aniline in alcoholic 4 or water-dioxane solvents are mildly catalyzed (3). However, reactions of the aryl fluorides with amines in aromatic hydrocarbon solvents ar second order in amines under most conditions (3, 6, 7). Also, the reaction of FDNB with n-methylaniline in ethanol or 60% dioxane-4OZ water is very sensitive to catalysis by acetate or hydroxide ion (4). Bunnett and Garst (5) interpret these observations according to the following scheme: H x R \\ NR 2 No R ’ N x k2 2 + RZNH »-—1—A 2 B 3.1 q k 3 N02 N02 N02 ArX I II Reactions of 2,4-dinitrochloro- and bromobenzene are not sensitive to base catalysis since k2)> k This condition is also true for many 1. reactions of FDNB, but in hydrocarbon solvents the k2/k1 ratio is greatly reduced. Also, in the reaction of FDNB with n-methylaniline, steric compressions in the intermediate (1) increase k-l so greatly that it exceeds k2 and the reaction becomes susceptible to base cataly- sis (k3). As stated previously, the nature of the mild accelerations is a matter of controversy. Ross (2) has regarded the mild accelerations as evidence for base catalysis of the second step of the reaction (See above scheme). In this scheme, the breakdown of the intermediate complex to products can occur either uncatalyzed (k2), or with catalysis 5 by base (k3). Bunnett and Garst (3) maintain that the mild accelera- tions observed with substituted nitrobenzenes and amines do not warrant the description of the acceleration as base catalysis. Their objections are based on the fact that strong bases like hydroxide ion are not much more effective, or are even less effective than weak bases like amines. In addition, similar effects are caused by substances like nitro com- pounds, sulfoxides, and sulfones which do not normally display basic properties in alcoholic solutions. Kirby and Jencks (8) believe these mild accelerations are instances of base catalysis. The authors also suggested that the disappearance of detectable general base catalysis with increasing base concentration, in the reactions of amines with monosubstituted nitrobenzenes, could occur through a change in the rate determining step of the reaction, with increasing concentration of base catalyst. The Reaction of FDNB with Proteins FDNB and related compounds react with a number of functional groups of proteins including the thiol, imidazole, e-amino groups of lysine, and phenolic groups (9). The side chains most reactive with FDNB are those of cysteine. The reaction may be followed spectro- photometrically by measuring the resulting yellow color of the DNP derivative. Other methods for following dinitrophenylation include the use of labelled FDNB, and amino acid analysis. The course of dini- trophenylation may be followed readily by Spectrophotometry when one particular side chain is modified. Reaction with amino groups may be measured by estimating the absorption at 360 mu (9); the molar absorp- tivity varies with the nature of the group that is substituted. 6 Functional groups other than amino groups exhibit maxima at wave- lengths lower than 360 mic The absorption maximum for S-DNP groups is at 330 mil, while substitution at imidazole and phenolic side chains in proteins, show maximal absorption at 330-360 mU-(9). Shaltiel (10) has described a method for the removal of DNP groups from imidazole, phenolic hydroxyl, and sulfhydryl groups, thus conver- ting dinitrophenylation into a reversible technique for blocking func- tional groups. The conditions for FDNB removal are relatively mild and the reaction is a cleavage by thiols e.g. 2-mercaptoethanol. Conse- quently, at pH 8.0 and 22°C (one hour incubation), reactions with a number of DNP-amino acids are complete with 90-100% yield. This tech- nique for reversible dinitrophenylation should extend the scope and importance of FDNB in protein chemistry. Sanger (11) was the first to employ FDNB as a tool for protein structure determination in his now classical work on the structure of insulin. Although the role of FDNB in protein structure determination has since that time decreased greatly, the reagent is still widely used in the determination of N-terminal groups of various proteins. FDNB and related compounds have also been used as tools to study enzyme mechanism. Under suitable conditions certain groups on the enzyme react more rapidly than others, allowing for the selective modification of the enzyme by addition of a stoichiometric quantity of reagent. Therefore, the enzyme supplies the specificity permitting a relatively non-specific reagent to be used to study enzyme mechanism. A disadvan- tage in the employment of FDNB and related compounds in protein modifi- cation studies in their low solubility in aqueous media. 7 Massey and Hartley (12) were the first to use dinitrophenylation in the study of enzyme mechanism. Specifically, they were interested in finding out what amino acids were essential in catalysis. They showed that the reaction of a single histidine in a-chymotrypsin led to the inactivation of the enzyme. Hirs 35‘31 (13-16) showed that bovine pancreatic ribonuclease A was rapidly inactivated by dinitro- phenylation at pH 8.0. Inactivation was due mainly to the reaction of FDNB with lysine 41. However, substitution at lysine 41 causes a conformational change that results in the exposure of lysine 7 and sub- sequent reaction of the latter with the reagent. Enzyme activity was preserved if the dinitrophenylation was carried out in the presence of substrate, or inorganic phosphate (Pi). Hirs eg‘gl (13, 14) suggested that FDNB was reacting at lysines which are at, or close to, the cata- lytic site and may be responsible for binding the substrate phosphate group to the enzyme. The X-ray data of Kartha SE 31 (17) later showed that lysine 41 is close to the catalytic site, however it is not invol- ved in substrate binding (18). Pontremoli eg‘gl (19, 20) showed that fructose diphosphatase of rabbit liver reacted with FDNB at pH 9.2 resulting in the dinitropheny- lation of two cysteine residues and a single lysine. The resulting DNP-enzyme showed a threefold stimulation of enzyme activity. In the presence of the substrate fructose 1,6-diphosphate (FDP), dinitro- phenylation at pH 9.2 did not result in enzyme activation. The two sites of dinitrophenylation were the sulfhydryl group of a cysteine residue and the e-amino acid residue of a lysine. Dinitrophenylation at pH 7.5 led to the arylation of a single cysteine residue and con- comitant enzyme activation. At this pH dinitrophenylation was completely 8 inhibited by FDP. The dinitrophenylated enzyme showed an increase in the Km for the substrates and for metal ion activators. The authors _ suggested that FDNB was binding to an allosteric site on the protein, thereby inducing a change in conformation which modifies catalytic activity. The above results were interpreted from the point of view that dinitrophenylation provides a model for a physiological mechanism which could conceivably activate the enzyme under conditions of gluco- neogenisis (19, 20). In a later report, Pontremoli eg‘el (21) dis- covered that certain disulfide reagents (e.g. cysteamine) activate fructose diphosphatase and suggested that such compounds could be important in physiological activation of the enzyme. Rosen and Rosen (22) have observed that when FDP-ase from Candida utilis was treated with FDNB in the presence of substrate, the enzyme was desensitized to the allosteric inhibitor 5'-AMP. Treatment of the enzyme with FDNB in the absence of substrate led to enzyme inactiva- tion. Identification of the residues reacting with FDNB showed that two tyrosyl and two lysyl groups were involved. It was concluded that these residues may be directly involved in the binding of FDP and AMP to the enzyme. Ronca £2.3l (23) reported that adenine deaminase from calf intes- tinal mucosa is rapidly inactivated by FDNB which reacted with two @- NHz-lysines. The authors observed a lack of protection by substrate analogs and suggested that the e-NH2 groups which react with FDNB do not play a role in binding adenosine to the catalytic site. In an attempt to locate the site of binding of messenger RNA (mRNA) to ribosomes, Moore (24) employed a number of site specific chemical reagents including FDNB. In this study FDNB reacted solely with 9 ribosomal protein and no evidence was obtained for reaction with the amino groups of the RNA moiety. The dinitrophenylated ribosome retained the capacity to bind poly U or poly C but specific sRNA binding was seriously impaired. Formaldehyde and Woodward's reagent attacked the amino groups on ribosomal RNA resulting in impairment of mRNA binding. On the basis of these results, Moore (24) postulated that messenger binding to the ribosome was via hydrogen bond formation between riboso- mal RNA amino groups and the phosphate groups of mRNA. Haines and.Zamecnik (25) employed dinitrophenylation in an effort to locate the binding sites of sRNA on amino acid ligases. On reaction of a crude mixture Of amino acid ligases with FDNB, the arginyl, his- tidyl, and leucyl enzymes were 88-99% inhibited while the lysine ligase retained 80% of its activity. The purified lysine ligase bound 4 moles of FDNB per mole of enzyme and showed a loss of pyrophosphate exchange of 30% while the acylation reaction was inhibited by 48%. Di Prisco (26) has observed that FDNB inactivated glutamate dehy- drogenase from beef liver. This study showed that FDNB desensitized the enzyme to the allosteric activator ADP or inhibitor GTP. Both ATP and GTP protect against FDNB inactivation. Dinitrophenylation in the presence of ADP and NAD did not abolish ADP activation, whereas, the enzyme was desensitized to GTP. These results indicated that the sites for ATP and GTP binding were not identical. Evidence was obtained by Cold (27) that FDNB and CDNB reacted rapidly with glycogen phosphorylase b from rabbit muscle (pretreated with mercaptoethanol in barbital buffer). Both reagents reacted rapidly with four sulfhydryl groups and much more slowly with other sulfhydryl, amino, and phenolic hydroxyl groups. Substrate afforded little protection 10 against phenylation. The dinitrophenylated enzyme showed a greatly decreased affinity for AMP and glucose-l-phoSphate. Philip and Graves (28) using the same enzyme as Gold (27) but with different pretreat- ment (dialysis against tris buffer), found that FDNB reacted with e- NH2 groups of lysine and sulfhydryl groups of cysteine. Enzyme ac- _ tivity was preserved in the presence of a-D-glucose-l-phosphate (G-l- P), or adenosine-S-monophosphate (AMP). When G-l-P or AMP were pre- sent during phenylation their binding sites were preserved but the Km for glycogen was increased. Keech and Farrant (29) found that FDNB reacted with a single ly- sine in sheep kidney pyruvic carboxylase resulting in enzyme inacti- vation. The protection afforded by the allosteric effector acetyl CoA against dinitrophenylation suggested that the e-NH of a lysine 2 residue may be involved in enzyme-acetyl CoA interaction. Treatment of myosin with FDNB by Bailin and Barany (30) resulted in significant changes in the properties of myosin when 1.5 moles of the reagent was incorporated per 5 x 105 gm of protein. The treated enzyme showed activation of the Ca++-ATPase activity and inhibition of the EDTA-ATPase activity. Also, the ability of modified myosin to superprecipitate with F-actin in the presence of Mg++ and ATP, or Mg++ and CTP, was inhibited. Virtually all of the FDNB incorporation was restricted to the biologically active part of the myosin molecule i.e. heavy meromysin and subfragment l. Dinitrophenylation resulted in the modification of cysteine, lysine, tyrosine, and histidine residues. In the presence of ATP, both enzyme activity was preserved and cysteine and tyrosine residues were protected. The authors suggested that it seemed likely that the reaction of these cysteine and tyrosine residues 11 with FDNB resulted in a conformational change in the enzyme thus affecting the binding of ATP to myosin. It was further pointed out that the substrate may have induced a conformational change in myo- sin which resulted in masking of these residues, thereby making them unavailable for reaction with FDNB. The Reaction of FDNB with Aldolases Cremona‘eg‘gl (31) reacted muscle aldolase with FDNB at pH 6.0 and found that the uptake of 2 moles of FDNB per mole of enzyme resulted in enzyme activation. Hydrolysis of the DNP-aldolase showed that S- DNP cysteine was the sole DNP-labelled amino acid. In the presence of fructose 1,6-diphosphate (FDP) the activation by FDNB was prevented. The DNP-enzyme showed a shift in pH optimum from pH 7.8 to pH 6.8, and the Vmax increased threefold, while the Km was unchanged. When 5-6 moles of FDNB were added to the enzyme, the activity remained equal to that of native enzyme. Treatment of the activated enzyme with mild alkali resulted in a B-elimination reaction to give two residues of dehydroalanine. The resulting desulfoaldolase had a level of activity equal to that of native enzyme suggesting that the two cysteine residues that reacted with FDNB were not essential for catal- ysis. The authors suggested that the metabolic modification of spe- cific sulfhydryl groups in this enzyme may be associated with regula- tory mechanisms for their activity 33 3123. Rowley, Tchola and Horecker (32) reported that at pH 9.2 trans- aldolase bound two moles of CDNB per mole of enzyme resulting in 90% inactivation of the enzyme. The reactive sites were identified as the e-amino groups of lysine. Inorganic phosphate and sulfate ions afforded 12 substantial protection against dinitrophenylation. Dinitrophenylation of rabbit muscle aldolase at pH 9.0 resulted in the uptake of 6-8 moles CDNB per mole of enzyme and 90% enzyme inactivation. In a later report Kowal, Cremona and Horecker (33) observed that dinitrophenylation of FDP aldolase involved the sulfhydryl groups of cysteine residues. The dinitrophenylation reaction appeared to consist of two distinct phases; an initial phase in which there is rapid uptake of 3-4 DNP residues with no detectable loss of enzyme activity, and a second slower phase in which an additional 8 DNP residues react. It is during the second phase that the enzyme is inactivated. In either case, the only DNP- amino acid detected on subsequent enzyme hydrolysis was S-DNP cysteine. Substantial protection of enzyme activity against inactivation was afforded by the substrates fructose-l-phosphate (F-l-P), FDP, and di- hydroxyacetone phosphate, at a level of approximately 1.0 x lO-3M. Fructose-6-phosphate which is not a substrate only gave a slight amount of protection when employed at a similar concentration. On the basis of the above results it was suggested that FDNB might be reacting at the sites responsible for binding the substrate phosphate group to these enzymes. Ingram and Wood (34) described conditions for dinitrophenylation of KDPG aldolase. At pH 8.0 (in imidazole buffer), or at pH 9.2 (in sodium bicarbonate buffer), KDPG aldolase bound 4 moles of FDNB per mole of enzyme giving rise to complete enzyme inactivation. Uptake of FDNB was measured both by 14C-FDNB as well as by following the increase in absorption at 360 mp4 At pH 8.0, only 4 moles of FDNB were bound. However, at pH 9.2 FDNB uptake slowly increased after 4 moles of the reagent were bound and the enzyme completely inactivated. KDPG and Pi, 13 at a level of 0.02M afforded virtually complete protection against either phenylation or loss of enzyme activity. G-3-P (0.02M) also gave substantial protection against phenylation, while pyruvate did not protect the enzyme. The sites of phenylation were identified as the €- amino groups of lysine residues. The coupling of 2 moles of pyruvate in the presence of sodium borohydride to the azomethine lysine did not inhibit dinitrophenylation of the enzyme. In the reverse order, the fully phenylated enzyme bound up to one mole of pyruvate per mole of enzyme. The above results were interpreted in the following manner. Four moles of FDNB reacted with 4 lysine residues (at the e-NH2 group) at the catalytic site. The normal binding of FDNB to pyruvate-coupled enzyme indicated that FDNB did not react with the lysine residue at the azomethine site. Since the enzyme was assumed to contain two catalytic sites (based on pyruvate-3-14C binding to the enzyme in the presence of sodium borohydride), each catalytic site was assumed to contain two FDNB reactive lysines. The protection of enzyme activity and inhibition of phenylation afforded by Pi, G-3—P, and KDPG, and the lack of protec- tion by pyruvate, were interpreted as evidence that the reactive lysine residues participated in binding the phosphate group of the substrate to the enzyme. This hypothesis by Ingram and Wood (34) concerning the nature of FDNB reactive sites is attractive since it raised the possi— bility of labelling these sites and subsequently sequencing the labelled peptides. Thus, valuable information could be obtained concerning the amino acid residues that comprise the catalytic site. The foregoing literature review shows that much valuable informa- tion concerning the identification of catalytically important residues 14 has been obtained by using FDNB as a chemical probe. It is important, however, to point out that the data also shows that FDNB can cause changes in the conformation of an enzyme. As Hirs (9) points out, in such chemical modification reactions in which the properties of the side chains are altered, significant effects on the tertiary structure are to be expected. Furthermore, the interpretation of these pertur- bations produced by modification studies on the enzyme properties such as catalytic activity, substrate binding, and cofactors, must there- fore be treated with caution. Physical and Chemical Characteristics of KDPG Aldolase KDPG aldolase from £3. putida was first crystallized by Meloche and Wood (35). The enzyme was found to be somewhat heat stable and to possess an unusually high degree of stability on exposure to low pH. Extensive studies on the physical properties of KDPG aldolase was carried out by Moehler, Hammerstedt, Decker, and Wood (36). Sedementation equil- ibrium studies showed that KDPG aldolase has a molecular weight of 72,000 (36). The enzyme was shown to have an 5 value of 4.35. Dissociation of the enzyme in guanidine hydrochloride and mercaptoethanol indicated a subunit weight of 24,000. In contrast to rabbit muscle aldolase which is dissociated below pH 3.9, KDPG aldolase did not dissociate over the pH range 1-9. Reaction of KDPG aldolase with maleic anhydride-IAC resulted in the uptake of 21 malyl residues and complete enzyme inacti- vation. The malylated enzyme showed no evidence of enzyme dissociation when it was examined in the ultracentrifuge. In addition, treatment of the enzyme with 0.05M NaOH did not dissociate the enzyme. The above data indiCate that the subunits of KDPC aldolase possess strong mutual 15 interaction. Hammerstedt and Wood (37) obtained evidence to Show that KDPG aldolase is a three subunit enzyme. This conclusion was based on the following Observations. Treatment of the enzyme with maleic anhydride- “C'resulted in the malylation of 21 lysine residues and the complete inactivation of the enzyme. On reversible dissociation of the labelled enzyme in the presence of native enzyme and subsequent subjection of the renatured enzyme to disc gel elctrophoresis, four distinct protein peaks were obServed. Enzyme activity was associated with the first three peaks (native, and enzyme with one, and two malylated subunits respectively) whereas the 14C label was associated with the last three peaks (enzyme with one, two and three malylated subunits respectively). These data indicate that the enzyme consists of three subunits. Addi- tional evidence for a three subunit enzyme was obtained by Robertson and Wood (38) who labelled KDPC aldolase with 14C iodacetic acid; sub- sequent digestion of the labelled enzyme with trypsin led to the forma- tion of four 140 labelled peptides. Peptide mapping studies, and the calculation of the minimal molecular weight from amino acid analysis of the enzyme, also support a three subunit model with identical subunits of molecular weight 24,000. KDPG aldolase thus becomes the only enzyme found to date that consists of three subunits. Grazi 33 El (39) established that two moles of pyruvate were reducibly coupled per mole of KDPC aldolase in the presence of sodium borohydride. A later report by Ingram and Hood (34) demonstrated that pyruvate was bound to the e-amino group of lysine. AnaIOg studies re- vealed that the specificity of azomethine formation was fairly wide, the only restrictions were the absence of a hydroxyl group on carbon 16 three and the presence of a polar group at carbon one. The rate of azomethine formation by pyruvate and a-keto-butyrate was established to be similar, but the rate of proton exchange was slower for Q-keto- butyrate. These results were interpreted to mean that azomethine for- mation and proton exchange were separate and distinct processes (40). These studies also revealed that 2-keto-3-deoxygluconate (KDG) was capable of forming an azomethine with KDPG aldolase, however this ana- log was not cleaved. This finding suggested that the phosphate group contributed more than enhancement of binding to the substrate, and that cleavage was not a direct result of azomethine formation. It was also found that 2-keto-3-deoxy-6-phosphogalactonate could be reductively coupled to the enzyme with sodium borohydride, however, this substrate analog was not cleaved (40). These findings indicate that specific orientation of the hydroxyl at carbon four is necessary for cleavage. Further evidence to substantiate the Schiff base mechanism of KDPG aldolase was obtained by Rose and O'Connell (41) who demonstrated that the rate of oxygen exchange of pyruvate-2-18O is five to six times greater than the rate of tritium exchange. This finding is consistent with the occurrence of a Schiff base intermediate which requires that ketimine formation precede the dissociation of protons from pyruvate. It was further estimated that the rate of hydrogen exchange from pyru- vate was twice the rate of the cleavage reaction. The N terminal amino acid of KDPG aldolase was found to be three- nine while the C terminal amino acid was identified as asparagine (38). To date, the only residues that have been implicated in the catalysis of KDPC aldolase have been lysine residues. As pointed out earlier, azomethine formation occurs with two to three lysine residues per mole 17 of enzyme (34, 42). The possible role of e-amino groups of lysine residues in binding substrate phosphate groups to the enzyme as suggested by Ingram and Wood (34) has already been discussed. Decker, Moehler and Wood (43) have observed that five sulfhydryl groups of KDPG aldolase were readily accessible to the reagent 5-5'dithiobis- (2-nitrobenzoic acid) (DTNB) in sodium bicarbonate buffer without affecting enzyme activity. An additional seven sulfhydryl groups could be further titrated after the enzyme was unfolded in urea. When the titration was carried out in potassium phosphate buffer only six buried sulfhydryl residues could be titrated. 0n reassociation of the DTNB-treated enzyme about 30% of the enzyme activity was re- covered. These results indicated that the sulfhydryl groups of KDPG aldolase do not play a crucial role in catalysis, or in maintaining the conformation of the enzyme. Meloche (44) has reported that two moles of radioactively labelled bromopyruvate were stably incorporated per mole of KDPG aldolase leading to complete enzyme inactivation. The author has suggested that the results are consistent with the occurrence of a basic amino acid residue at the active site adjacent to the methyl carbon of py- ruvate. Thus, the implicated amino acid might possibly be the base responsible for labilizing thecy-methyl proton of pyruvate to initiate condensation with C-3-P. CHAPTER III METHODS AND MATERIALS Bacteriological Pseudomonas putida, strain A 3.12 (Stanier), was grown on agar slants composed of 2% potassium gluconate, 0.6% (NH4)2HP04; 0.3% KH2P04, 2 306H20. The organism was grown at 28°C. Large quantities of cells were grown in 0.05% MgSOA.7H 0, 0.06% sodium citrate and 0.005% FeCl a 130 liter fermenter (New Brunswick Scientific) on the medium des- cribed above minus agar. The gluconate, magnesium, and phosphate salts were sterilized separately. A five percent inoculum was used and growth was continued at 28°C until the cells attained maximum growth. Growth was measured by following the optical density at 660 mil; a 1810 dilution of the culture medium at maximum growth generally gave an 0.0 reading of 1.0 - 1.5. 660 Pseudomonas saccharophila, strain 8 105, was obtained from Dr. C. W. Shuster and stored on agar slants containing 2.5% sucrose, 0.1% NH c1, 0.05% MgSO .7H 0, 0.005% FeCl .6H 0 and 0.33M KH po4 buffer, 4 4 2 3 2 pH 6.8. The cells were grown at 30°C. When grown in large amounts, 2 the cells were grown in a 50 liter batch in a 130 liter fermenter (New Brunswick Scientific) in the above-described medium, except that the sucrose was replaced by 2.5% galactose. 18 19 Chemical Potassium gluconate was obtained from C. Pfizer & Co. and galac- tose was obtained from General Biochemicals. D,L-glyceraldehyde-3- P04 and l, fluoro, 2,4-dinitrobenzene were obtained from Sigma Chemical Co., the FDNB was twice recrystallized from ethanol prior to use. S-DNP cysteine and 0-DNP tyrosine were obtained from Fisher Scientific Co. A sample of 2-keto-3-deoxy-6-phosphogalactonate was generously donated by Dr. M. Doudoroff, University of California (Berkeley). KDPG was prepared by condensing pyruvate and glyceraldehyde-3-PO4 with KDPG aldolase as described by Meloche and Wood (45). A sample of 2-deoxy- D-xylose was a gift from Dr. J. Preiss, University of California (Davis). Tritiated water was obtained from Volk Radiochemical Company. Pyruvate-3-1AC and FDNB-U-IAC were obtained from Nuclear Chicago. The following enzymes were used as reagents. Crystalline lactic acid dehydrogenase, alkaline phosphatase from calf intestine and crude DNAse were obtained from Worthington Biochemical Corp. cy-glycerophos- phate dehydrogenase-triose phosphate isomerase was a product of C. F. Boerhinger and Son. Enzymatic Synthesis of 2-keto-3-deoxy-6-phosphogalactonate The method of synthesis of 2-keto-3-deoxy-6-phosphogalactonate (KDPCal) involved a new approach to the synthesis of this compound. This method is analogous to that used in the synthesis of KDPG from pyruvate and G-3-P in the presence of KDPC aldolase (45). Thus, KDPCal was synthesized from pyruvate and G-3-P in the presence of KDPGal aldolase. One disadvantage of this method of KDPCal synthesis was that KDPGal aldolase had to be first separated from KDPC aldolase 20 which is also present in the cell extracts. Therefore, the noteworthy feature of the enzyme purification is the separation of KDPCal aldolase from KDPG aldolase. This separation is effected by a series of acid precipitations at pH 5.0. The KDPCal aldolase is precipitated at this pH while KDPC aldolase remains in solution. KDPGal aldolase was purified from Ei' saccharophila according to the method of Shuster (46). Cell extracts were prepared by sonic dis- ruption of the cells. The cells were suspended in ten volumes of 0.05M potassium phosphate buffer, pH 7.6, and disrupted by sonication for 10 min. The resulting suspension was centrifuged for 30 min at 30,000 r.pwm. The pooled supernatant solution was then subjected to acid treatment. All subsequent steps in'the purification were run at 2-4OC. The crude extract was diluted with 0.05M phosphate buffer to a final protein concentration of 5 mg per ml and 1M acetic acid was added until the pH was exactly 5.0. After standing for 20 min the acidified suspension was centrifuged for 30 min at 16,000 r.p.m. The precipitate was suspended in one-half the original volume of 0.05M phosphate buffer. Since much of the precipitate was insoluble after the first acid pre- cipitation, the suspension was first clarified by centrifugation. The solution was acidified to pH 5.0, and the precipitate was taken up in 0.05M phosphate buffer. The acid precipitation step was carried out a total of five times. The pH of the dissolved precipitate was adjusted to pH 7.6 with 1.0M NaOH. The extract was then heated with stirring to 70°C and maintained for 5 min at this temperature. After rapid cooling, the extract was centrifuged and the precipitate was discarded. In the final extract, 99.9% of the KDPC aldolase originally present was removed. 21 Condensation of pyruvate and G-3-P with KDPGal aldolase was carried out as described by Meloche and Wood (45) for the synthesis of KDPG. When non-radioactive KDPGal was prepared, 1.3 millimoles of D,L-G-3-P was condensed with 4 millimoles of sodium pyruvate. However, in the synthesis of KDPGal-3-IAC from pyruvate-3-IAC the molar ratios of G-3-P and pyruvate were reversed. After condensation was complete [as deter- mined by the thiobarbituric acid test for keto-deoxysugars (47)], the product was precipitated with bariwm acetate and the pH was adjusted to pH 3.5 with concentrated HCl. The precipitate was removed by centri- fugation and five volumes of ethanol was added to the supernatent. The precipitate was collected by centrifugation and dissolved in dilute HCl. The pH was adjusted to pH 3.5 and a second barium pre- cipitation was carried out. The resulting precipitate contained KDPCal in 65% yield relative to the initial D-G-3-P, and contained about 8% pyruvate, 5% KDPG and 35% KDPGal. The barium precipitate was dissolved, the pH was adjusted to 8, and to this solution KDPG aldolase plus LDH- NADH was added to destroy the KDPG present. The mixture was then sub- jected to two barium precipitations and the precipitate was finally dissolved in dilute HCl, then placed on a Dowex-l-Cl column. The column was eluted with a linear gradient of 0-0.2N HCl. The semi- carbazide positive tubes were tested for KDPCal activity with KDPCal aldolase. The tubes containing KDPGal were subjected to barium pre- cipitation and the precipitate was dried in a dessicator. The syn- thesized KDPGal was 70% pure (uncorrected for moisture) when assayed with KDPC aldolase and the LDH-NADH coupled assay system; no measur- able pyruvate or KDPC was present. 22 Since no authentic sample of KDPGal was available initially, the authenticity of the synthesized compound was checked by a number of criteria. The compound was cleaved by KDPGal aldolase but was not cleaved by KDPG aldolase in accordance with the observation of Shuster (46). The synthesized compound was dephosphorylated with alkaline phosphatase in the following manner: KDPGal (2.5 umoles) was added to a test tube containing 0.05M MgCl and the pH was adjusted to 9.2. 2 A small amount of phosphatase (about 0.5 mg) was added from the tip of a spatula and the pH decreased to 8.5 within a few minutes. Fur- ther addition of phosphatase did not result in the further lowering of the pH. Conversion of KDPGal to KDGal was measured by the thio- barbituric acid test (47). KDGal reacts with thiobarbituric acid whereas KDPGal is relatively insensitive to this reagent. Preiss and Ashwell (48) have demonstrated that the relative con- figuration of hydroxyl groups on the C4 and C5 position of 2-keto-3- deoxyonic acids determined the rate of liberation of B-formylpyruvate from the compound during periodate oxidation. Applying the above rationale to the synthesized KDCal, the rate of release of B-formyl- pyruvate from KDG should be much faster than KDGal. B-formylpryuvate was detected by the bright red color obtained on reaction with thio- barbituric acid. 23 COOH T=° pooa EH2 c=o H 10 | thiobarbituric HO-f-H 5 6‘:’ CH2 x>g red colored I I . adduct H-C-OH CHO ac1d CHZOH KDGal B -formylpyruvate A solution containing 0.95L1moles of KDGal was pipetted into a number of tubes and 0.25 ml of periodic acid (0.025M) was added to give a final volume of 0.5 ml. The same operation was carried out with KDG. The tubes were incubated at room temperature and at regular intervals 0.05 ml of 2% sodium arsenite in 0.5 NHCl was added to destory the excess periodate. The resulting samples were then subjected to the thiobarbituric acid test. e-formypyruvate release from KDG was complete within 10 min, whereas the synthesized KDGal required 40-50 min for complete release. (See diagram in Appendix, page 92). These results are consistent with the findings of Preiss and Ashwell (48) for KDG and KDGal, and is con- sistent with C4-C5 transhydroxyl arrangement. In a further attempt to identify the synthesized compound, KDGal was subjected to sodium borohydride reduction. The resulting mixture of metasaccharinic acids was degraded with ceric sulfate as des- cribed by Ghalambour, Levine and Heath (49). (DGal should yield 2- deoxyxylose which on reaction with thiobarbituric acid yields a colored adduct with an absorption maximum of 530 mu. 24 fOOH COOH C = O (HO) H-C-OH (H) CHO $2 [”2 Ruff $112 1 . HO-C-H haBll4 HO-C- Degradation HO-C-H I —-———-> | > I H-C-OH H-f—OH CeSO4 H-C-OH CHZOH CHZOH CHZOH KDGal a- and B-meta 2-deoxyxylose saccharinic acids Two hundred moles of sodium borohydride was added to 5 umoles of KDGal (pH 7.0) and the mixture was incubated for 15 min at room temperature. The pH of the solution was adjusted to 1.0 to decompose any excess sodium borohydride still present. To the mixture was added 3 ml of acidified ethanol and the solution was evaporated to dryness in a Buchler Rotary Evapo-Mix. The residue was taken up in 3.0 ml of acidified ethanol and once again evaporated to dryness. This procedure was then twice repeated and the residue finally taken up in 1.0 m1 of water. The solution was then treated with 1.0 ml 0.2N ceric sulfate and incubated at 550C for one hour. After the solution was cooled to room temperature barium hydroxide neutralization was carried out and the precipitate was removed by centrifugation. The supernatent was deionizedtw'addition of mixed bed ion exchange resin, evaporated to 0.3 ml and an aliquot was subjected to thiobarbituric acid reaction. The final product gave a positive thiobarbituric acid test, however the yield was quite low probably due to the fact that the mixed resin 25 used was quite strong and absorbed much of the compound. tion maximum of the final compound is shown below. The absorp- Compound max of Thiobarbiturate Adduct Synthesized KDGal Authentic KDGal Product of Ruff Degradation ('2-deoxyxylose') Authentic 2-deoxyxylose 549 mp. 549 mil 528 mp, 530 mu The synthesized KDGal had the same RF values as authentic KDGal, when thin layer chromatography (silica gel G) was carried out in two different solvents. Compound Solvent Synthesized KDGal 2-butanone-HoAc-H20 Authentic KDGal (8:8:1) Synthesized KDGal Phenol-satd.-water Authentic KDGal In summary, the synthesized product has been shown to be KDPGal based upon the following data: 1. The product was synthesized from G-3-P and pyruvate in the presence of KDPGal aldolase. 2. The product was cleaved by KDPGal aldolase but not cleaved by KDPG aldolase. 3. The dephosphorylated product was thiobarbituric acid positive. (imax for thiobarbituric acid adduct = 549 mu - This is diagnostic 26 for 2-keto-3-deoxyonic acids). 4. The slower rate of release of B-formylpyruvate from KDCal compared with KDG is consistent with a 04-05 transhydroxyl arrangement. 5. Borohydride reduction of the synthesized product followed by Ruff degradation showed the expected shift in.lmax of the thiobarbituric acid adduct from 549 mLLto 528 ml» 6. Synthesized KDCal cochromatographed with authentic KDGal. Determinations and Procedures Alpha-keto acid determinations were performed by the semicarba- zide method of Macgee and Doudoroff (50). The thiobarbituric acid test was employed as described by Srinivasan and Sprinson (47) and modified by Dahms (51). This modification consists of adding 5.0 ml of water to each tube immediately after color development leading to enhancement of the color stability. The thiobarbituric acid spray reagent for chromatograms was prepared as described by Srinivasan and Sprinson (47). Measurement of FDNB Uptake The course of FDNB reactions was followed spectrophotometrically at 360 mp at a temperature of 330C. The reactions were measured in a total volume of 0.25 ml in cuvettes of path length 1 cm on a Gilford modified DU Spectrophotometer fitted with an auxillary dwell timer. The optical density was measured for a seven second interval at 5 or 10 min intervals. A control was run at the same time to determine the optical density change in the absence of enzyme. To determine the extent of dinitrophenylation, the change in optical density of 27 the control cuvette was substracted from the value obtained in the presence of enzyme. The uptake of FDNB was calculated using a molar extinction coefficient of 17,000 for e-DNP-lysine (33). When FDNB-140 was employed, an aliquot of phenylated enzyme was removed and precipi- tated by the addition of 30% TCA to give a final volume of 5% TCA. The precipitate was washed six times with 5% TCA and finally taken up in 0.1N NaOH. A suitable aliquot was then used for counting in Bray's scintillation fluid containing 5% Cab-O-Sil. Reductive Couplingof Pyruvate-3-14C and KDPGal-B- “c to KDPG Aldolase Pyruvate-3-14C binding to KDPG aldolase in the presence of sodium borohydride was accomplished as described by Ingram and Wood (40). The reaction mixture usually contained 1.0 x 10"2 umoles of enzyme and 1.97 “moles of pyruvate 3‘146 in 0.04M phOSphate buffer, pH 6.0, to give a final volume of 1.0 ml. The solution was incubated in an ice bath for 5 min then subjected to borohydride reduction in the following manner: 0.005 ml of sodium borohydride (1.0M) and 0.0025 ml of 2M acetic acid were added alternately at 3 min intervals. After three series of additions of borohydride followed by acetic acid, the enzyme activity was assayed. The enzyme was at least 99% inactivated at this stage when pyruvate was used. When KDPGal binding to the enzyme was carried out the reaction was run exactly as described above except that 0.05M MES buffer pH 6.0 was employed, and a second addi- tion of substrate was necessary, followed by a single borohydride reduction to obtain complete inactivation. In the above binding experiments, two controls were run; the first contained no substrate or analog, however borohydride reduction was exactly as described 28 previously. The second control contained all the regular components except that no addition of borohydride was made. In all cases, ali- quots of the reaction mixture were removed, 0.5 ml of bovine serum albumin (10 mg/ml) was added, and the mixture was precipitated by the addition of 30% trichloroacetic acid (TCA) to give a final concentra- tion of 5% TCA. After a total of six TCA washings, the precipitate was finally taken up in 0.1N NaOH. A suitable aliquot was counted in Bray's scintillation fluid (52) containing Cab-O-Sil. Measurement of the Rate of Tritium Exchange Measurement of the rate of tritium exchange from T20 into pyruvate by KDPG aldolase was carried out at 28°C. The incubation mixture con- tained 20 moles of pyruvate, about 3 x 10-4 umoles of enzyme, 100 moles of imidazole buffer (pH 8.0), and 20 millicuries of T20, in a total volume of 1 ml. Aliquots were removed at regular intervals and the pyruvate was immediately converted to lactate by LDH and NADH. The mixture was then freeze-dried and the residue was redissolved in 2.0 ml of water. The procedure was repeated a total of six times to remove all remaining traces of T20. The lactate was determined by the proce- dure of Barker (53) while tritium was measured in Bray's scintillation fluid (52). Radioactivity Measurements All radioactivity measurements were performed in a Packard Tri- carb Liquid Scintillometer. Internal standards were employed to mea- sure the efficiency of the counting system. 29 Isoelectric Focussing Isoelectric focussing was carried out inIa 110 ml column (LKB) at 2°C for 40-50 hours. The following solutions were used for making the gradient containing a final ampholine concentration of 2%. Dense solution - 1.4 ml ampholine of suitable pH range was added to 42 ml water containing 28 gm of dissolved sucrose Less Dense - 0.6 ml ampholine dissolved in 60 ml water Solution The gradient was made from the dense and less dense solutions as des- cribed by the instruction bulletin for the LKB 8101 electrofocussing column (LKB, Bromma, Sweden). In addition, the anode solution con- tained 0.3 m1 conc. H2804 in 21 ml water containing 18 gm of sucrose. The cathode solution contained 0.2 m1 ethanolamine dissolved in 10 m1 of water. Running gel for polyacrylamide disc gel electrophoresis was pre- pared as described by Davis (54). No sample or spacer gel was used. Trisglycine buffer (pH 8.0) was used for electrophoresis and was pre- pared as described by Davis (54). All gels were first subjected to a 30 min pre-electrophoresis period to remove ultraviolet absorbing im- purities. After addition of the protein the gels were subjected to electrophoresis for 30-60 min with a current of 5 milliamps per tube. The gels were scanned directly on a Gilford modified DU Spectrophoto- meter outfitted with a Gilford linear gel transport attachment. 30 Heat Inactivation Heat inactivation of KDPG aldolase was effected by placing the enzyme (1 x 10-4’umoles) into 0.005M phosphate buffer, pH 6.0 (pre- heated to 70°C) to give a final volume of 1.0 ml. The tube was placed in a bath set at 70°C and.sammles were withdrawn at regular intervals and were placed in cold 0.005M phosphate buffer (pH 6.0) prior to assay for enzyme activity. Sulfhydryl Determinations Sulfhydryl determinations were performed with 5,5'-dithiobis-(2- nitrobenzoic acid) (DTNB) according to the procedure of Decker, Moehler and Wood (43). The reaction mixture contained 0.04 ml lmM DTNB in 5w phosphate buffer, pH 7.0, 0.14 ml of 0.5M phosphate buffer, pH 8.0, about 100 ugm.ofKDPC aldolase in a total volume of 0.3 m1. A control was run in parallel containing all the above additions minus enzyme. The molar extinction for DTNB was calculated to be 13,600 in the above system using cysteine as a standard. Enz tic A coupled assay was used for the estimation of KDPC aldolase activity based on the oxidation of NADH as developed by Kovachevich and Wood (55) and modified by Meloche and Wood (35). The assay is based on the following equation: KDPG KDPG +- ;> D C-3-P + pyruvate aldolase excess pyruvate +~ NADH.H :> lacfi?te lactate dehydrogenase NAD 31 The reaction was followed spectrophotometrically in cuvettes of path length 1 cm in a Gilford Modified Beckman DU Spectrophotometer at 28°C. One unit of activity is described as the absorbance change of 1.0 per min in a total reaction volume of 0.15 ml. Specific activity is defined as the number of units per mg protein. Protein was deter- mined by the 280:260 ratio as described by Warburg and Christian (56), or by the method of Lowry ££_21 (57). The 280 um.reading for KDPG aldolase shows an approximate increase of 10% on dinitrophenylation, while the Lowry protein value remains unaffected. Crystalline enzyme was prepared by the method of Ingram and Wood (40). CHAPTER IV RESULTS Six phases of experimentation were involved in this study and are treated as separate sections. The first section deals with the reac- tion of KDPG aldolase with FDNB. This section is divided into three parts; (a) the establishment of conditions for dinitrophenylation; (b) the demonstration of the specificity of substrate protection against dinitrophenylation; (c) experiments to identify the sites of dinitro- phenylation. The second section examines the effect of dinitropheny- lation on the physical structure of the enzyme by the use of (a) poly- acrylamide gel electrophoresis; (b) sucrose density gradient centri- fugation. The third section constitutes an analysis of the kinetics of dinitrophenylation. The fourth section deals with the determination of the nature of the residual enzyme activity of dinitrophenylated enzyme using the isoelectric focussing technique. The fifth section includes experiments to show the effects of dinitrophenylation on some of the catalytic parameters of the enzyme. This section is divided into two parts; an examination of the effect of dinitrophenylation on (a) substrate binding, and (b) the rate of tritium exchange from T20 into pyruvate. The sixth section contains two experiments to assess the effect of dinitrophenylation on the enzyme conformation. This section includes (a) a comparison of the rates of temperature in acti- vation of native and DNP aldolase, and (b) an experiment to estimate the available sulfhydryl groups in the two enzyme preparations. 32 33 l. Dinitrophenylation of KDPG Aldolase (a) Establishment of Conditions for Dinitrophenylation All attempts to repeat the dinitrophenylation of KDPG aldolase under conditions described by Ingram and Wood (34) at pH 8.0 (imida- zole buffer), or pH 9.1 (bicarbonate buffer), led to the uptake of FDNB in excess of 4 moles of FDNB per mole enzyme. Figure 1 shows an example of overphenylation (i.e. FDNB uptake exceeds 4 moles/mole of enzyme) at pH 9.1. It should be noted that at 90 min FDNB uptake still continued although enzyme inactivation had levelled off at approximately 95%. Four moles of DNP groups were bound per mole of enzyme within 30 min, resulting in approximately 80% enzyme inactiva- tion; the next five moles of DNP groups bound by the enzyme led to an additional 15% inactivation of the enzyme. All attempts to obtain complete inactivation by the introduction of a limited number of dinitrophenyl groups onto the enzyme were unsuccessful. Dinitro- phenylation at pH 8.0 (in imidazole buffer) under the conditions described by Ingram and Wood (34) also led to overphenylation. New conditions had to be established in order to introduce a limited number of dinitrophenyl groups onto the enzyme. Table 1 shows the conditions finally attained for dinitrophenylation. Approximately 4 dinitrophenyl groups were introduced onto the enzyme resulting in 85-90% inactivation. Figure 2 shows that the course of dinitropheny- lation is essentially complete within 100 min while only 85% of the enzyme activity is lost. In the presence of 0.02M KDPG, or Pi, both dinitrophenylation and loss of enzyme activity are inhibited, as has been reported previously (34). 34 FIGURE l.---Reaction of KDPC aldolase with FDNB at pH 9.1 and room temperature. The Open circles relate to change in optical density at 360 leand the closed circles to the loss in enzyme act vity. The cuvette contained 200 moles bicarbonate buffer, 1.6 x 10' llmoles of dialyzed enzyme and 0.5‘jmoles FDNB in a total volume of 1.0 ml. Samples were withdrawn at regular intervals for determination of KDPG aldolase activity. A control was run under identical conditions except that enzyme was omitted. 35 oz_2_<_>_mm >._._>_._.o< 92>sz o\o m a a m m. 0. IC dlm N E / _ .m m 0 01m A L / \ L O Y D) O -N LIJ 0 I16 E A9 /0 0 Wu. CH / .O Prox. I O O R D / \ 4 T K O 0 .W F /o 0\_ m o o oV/o_ I’ll... \_ _ _/pKO 5 o 5 O 5 2 2 l. J O. O O O O O O 18 own I>._._mzwo |_sz wn_OS_\mn_DOmo azo n_O mMJOE IN MINUTES TIME 36 TABLE l.---Dinitropheny1ation of KDPG aldolase at pH 8.5 (33°C). The value shown below is the average of ten determinations. Moles FDNB bound/mole enzyme % enzyme inactivation 3.8 + 0.1 35 - 9o The reaction mixture contained 250 umoles of pH 8.5 imidazole buffer, about 1.0 x 10"3 umoles of aldolase, 25 x 10‘3 umoles of FDNB in a total volume of 0.25 ml. A control was run containing all the additions except enzyme. The value given above is corrected for FDNB hydrolysis of the control. The total increase in 360 mu absorp- tion of the control is approximately one sixth of that obtained when enzyme is present. 37 FIGURE 2.---R8action of KDPG aldolase with FDNB in imidazole buffer, pH 8.5 and 33 C. For details see Table l. The open circles relate to the change in optical density, the open triangles and squares relate to the change in optical density in the presence of 0.02M KDPG and Pi, respectively. The corresponding closed figures repre- sent loss in enzyme activity. 38 OZ_Z_<_>_wm >.:>_._.o< m_>_>N2w o\o O O 0 8 O 6 O 4 O 2 O 0. 4 O. 3 _ 0. 2 O. O IOO 60 IN MINUTES 40 20 llq ‘ _ D N E :_ .— I O S 1" II n.A a0 114 u A L I D L m _ / _. . -VN. at __-o J I_A_ w. 1. a._ -mww Allo Ila R D / _ T K : \O/ O D— - M F I O O\r .A l O _. / \ . D A _ 01/. a. F II|\_.\_ 0_ A a m. m m w 0. O. 0. O. O. O 1800 I >._._mzmo |_sz maoz\maaomo azo to mmaoz TIME 39 (b) Specificity of Substrate Protection Against Dinitrophenylation The protection against dinitrophenylation by KDPG and Pi has been interpreted to indicate that dinitrophenylation was occurring almost exclusively at the active site, probably at the substrate-phosphate binding site on the enzyme. During an investigation of substrate pro- tection against dinitrophenylation, it was found that glucose-l- phosphate (G-l-P), which is a weak competitive inhibitor (Ki is )) l x 10-1M) of KDPG aldolase activity, afforded substantial protection against dinitrophenylation at a level of 0.02M. This finding raised the question as to whether dinitrophenylation was occurring outside of the active site and the protection observed was a relatively non- specific process, which may in fact reflect a more general ionic binding of the substrate unrelated to substrate specificity. Such a situation would imply that the relatively high concentration (0.02M) of Pi, or the phosphate moiety of a phosphorylated substrate, e.g. KDPG (Km = 0.8 x 10-4M), may bind ionically to the potentially reactive FDNB sites on the outside of the active site, thus protecting these sites against dinitrophenylation. It was, therefore, imperative to demonstrate that protection of the potentially reactive FDNB sites against dinitrophenylation was substrate specific. This required conditions which allowed occupancy of the active sites by the substrate, or analog, but did not furnish substrate or analog molecules in solution which could bind non- specifically to the FDNB potentially reactive groups outside of the active site. These conditions could be achieved by the use of the analog 2-keto-3~deoxy-6-phosphogalactonate (KDPGal) which binds to 40 the active site (Ki = 1 x lO-ZM), but is not cleaved by KDPG aldolase (40). The use of KDPGal is ideal because it forms a Schiff base, and can be stably coupled to the active site of the enzyme with sodium borohydride. KDPGal-lac was first coupled to KDPG aldolase with sodium borohydride. The excess KDPGal can then be removed by dialysis, leaving the active sites, including the anionic binding sites, blocked by a stoichiometric amount of KDPGal. Since 14C-labelled KDPGal was used, the number of moles bound by the enzyme could be determined. Dinitrophenylation of the KDPGal-reduced enzyme could then be carried out. If uptake of FDNB was inhibited, this would constitute strong - evidence that protection against dinitrophenylation was specific. On the other hand, if FDNB was still taken up by the enzyme, then FDNB must have reacted at sites removed from the catalytic site. KDPGal was used in the next series of experiments to demonstrate the specificity of substrate protection. Figure 3 shows the dinitro- phenylation of KDPG aldolase in the presence, and absence, of KDPCal. At a level of 0.02M KDPGal, protection against dinitrophenylation and inhibition of enzyme activity was comparable to that obtained with KDPG and Pi. The number of moles of KDPGal-3-14C that can be coupled with sodium borohydride to KDPC aldolase was next determined. KDPG aldo- lase was subjected to borohydride reduction in the presence of KDPGal- 3-140 and the completely inactivated enzyme was precipitated with 30% TCA and washed six times in 5% TCA. The results of KDPGal binding to KDPG aldolase are shown in Table 2. In two experiments KDPG aldolase bound 2.63 and 2.72 moles of KDPGal per mole of enzyme. In a separate experiment (not shown) it was found that unlabelled KDPCal-coupled 41 FIGURE 3.---Dinitrophenylation of KDPG aldolase at pH 8.5. The con- ditions employed for dinitrophenylation are the same as described in Table l. The open circles and triangles relate to the change in optical density in the absence and presence of KDPGal (0.02M) res- pectively. The closed triangles relate to the loss of enzyme activity in the presence of 0.02M KDPGal. 4 \7~AI s 42 oz_z_<_2wm >._._>_._.0< m_2>NZw o\o .m m m m m o I O II‘cI _ m F O O G _ ‘ NW ~ n _o .nNUIKIIAu “IA_8 IE 2 NSF _ .AOIII IO VIOE NBC 2/ _ WWI—m ‘ O O OGE O RPR / _ .T PI 0 ID A IO IAZ WKE ‘— /O _ D m . {a I? _ O 5 O 5 O 5 2 2 J l O O O O O O. O 92>sz mI_O_2\mn_Domo QZQ “.0 mij—2 IN MINUTES TIME 43 TABLE 2.---The binding of KDgGal-3-IQC to KDPG aldolase on reduction with sodium borohydride at 0 C. Experiment % Enzyme Moles KDPGal-3-140 bound No. Inactivation Per mole of enzyme 1 100 2.63 2 100 2.72 Non-reduced control 3 0.0037 The incubation mixture contained the following: about 1.7 x 10-2 ”moles of KDPG aldolase, 280 moles of MES buffer pH 6.0, 1.97 umoles of KDPCa1-3-IAC in a total volume of 0.5 ml. The mixture was incu- bated for 5 min. 5 ul of NaBH4 (1.0M) was added and following a three minute interval 2.5 ul of 2M acetic acid was added. After three cycles of NaBH4 addition followed by acetic acid addition, the enzyme activity was assayed. A further addition of 1.97 umoles of KDPGal-3- 14C was made followed by a single NaBH4 addition. The non-reduced control was prepared in an identical manner to the reduced enzyme except that NaBHA was not added. (See calculation on Appendix Page 92). 44 enzyme does not bind pyruvate-B-IAC in the presence of sodium boro- hydride. This finding constitutes strong evidence that KDPGal and pyruvate compete for the same sites on the enzyme. The results of KDPGal binding to KDPG aldolase are consistent with the data obtained on pyruvate reduction (2-3 moles bound/mole enzyme) (41), and are in agreement with the recent finding that KDPG aldolase is composed of three identical or nearly identical subunits, and presumably three catalytic sites. It has now been demonstrated that all three cataly- tic sites can be blocked with KDPGal. The KDPGa1-3-14C-coupled enzyme was then reacted with FDNB. After KDPGal reduction, the enzyme was dialyzed and subjected to di- nitrophenylation. In two separate experiments shown in Table 3, KDPG aldolase bound 3.24 and 2.78 moles of KDPGal per mole of enzyme, respectively. In both experiments the KDGal enzyme bound approximat- ely one mole of FDNB, while the non-reduced controls bound 3.6-3.7 moles of FDNB per mole of enzyme. Experiment 1 showed that 0.02M KDPG prevented the uptake of the one mole of FDNB by KDPGal-reduced enzyme. These results demonstrated that blocking the catalytic site abolishes the uptake of 3 moles of FDNB. The one mole of FDNB taken up by the KDPGal-coupled enzyme, may represent non-specific phenyla— tion, i.e. phenylation outside of the active site. These results show that substrate protection against dinitrophenylation is specific and that there is possibly a non-specific uptake of 1 mole of FDNB. (c) Identification of the Sites of Dinitrophenylation A spectrum of the DNP aldolase was taken on the Cary double beam Spectrophotometer with DNP aldolase in the sample compartment and the 45 magma cw wonwnomov mcofluwpcoo woman can Sosa mm3 Cowuommu newumfi>cocmonDMCMp one .ao ommm xwpcoam< mom .mawmuop wonuusm you .COMDmamcwza Iowuwcwv ou nowun Ao.m :av wowuan oaonmpHEM Emmo.o umamwm vmumamwp coca mam manuasm anacoEEo zo.m nuws moewu o>ww moans: cozu .oumwaam EnacoEEm nufiB powwowmwoonm max oE>NCm o£H sows poumonu no: mm3 uw umsu umouxo mGOwuwppm onu Adm pocflmuCoo Honucoo one Imop umau ou Hmowucovw mp3 unusanoaxu o>onm onu cw maanwn 0 ca om.o mH.o omax Emo.o uo monomoum :« mahnco oeanmaamommx oHoE won mason mzom mogoz mn.m mm.o oc.m mm.o o8>u£o cadun Ifimumox 0208 won mason mzom moaoz mn.m qm.m oexmwo mgoe \pcpon 0 Hum IHmquM mwaoz Howucou mE>NCo escapes qzm Howucoo makeup pooppow «mm coflumwmmonm mEhncm N .uaxm H .oaxu mahuconamomnx mo cofiumH>Co£mouuwcHouII.m mqmHom m¢.o no.0 eo.o no.0 n¢.o eo.o I om.o I ao.o Hm.o I ¢~.o Nw.o ~m.o mmmdopflm scum ocwmouNu unwoumNo ocmeg ochNA o>wum>uuopImzo mzan mzoIm mzaIHp mzme woumahcosaouuHch egos ocfiam who no :ofiumOAHHucovu caznmuwoumEonno nodmmIII.n mqm._._>_._.o< ”22>sz mo ZOFOu_ 92>sz mo mmI_0_2 4O 60 80 20 TIME M l N UTES IN 59 loss of activity. The lysine residues show two degrees of reactivity. If that part of the curve for the slower reacting lysines is extra- polated to time zero, the amount of lysine residues calculated at that point corresponds to three (See Appendix for additional data, page 96). ‘ Thus, there is one fast reacting lysine and three slower reacting lysines. Subtraction of the rate of reaction of the slow reacting lysines from the initial rate of reaction of FDNB with the enzyme (includes the rate of fast and slow reacting lysines) at each point in time gives the rate of reaction of the single fast reacting lysine residue (plot 3). The rate of reaction of the fast lysine (k = 0.078 1min-l) was calculated to be five times greater than the rate of l M- reaction of the slow lysines (k = 0.015 M- min-1). A similar operation was carried out for the rate of loss of enzyme activity. Extrapolation of the slow rate to zero time inter- sects the ordinate at a point calculated to correspond to 50% inacti- vation (See Appendix for additional data, page 97). Subtraction of this extrapolated rate from the initial fast rate of inactivation at each point in time gives the rate shown on plot 4. Thus the rate of reaction of the initial fast lysine (plot 3) parallels the initial rapid loss of activity (plot 4), indicating that the reaction of a single fast reacting lysine results in 50% inactivation of KDPG aldo- lase. KDPG aldolase is composed of three identical subunits containing three substrate binding sites, and presumably three catalytic sites. It would be expected, therefore, that if a single critical residue is destroyed on any subunit, the extent of inactivation should not exceed 33% (one of three active sites destroyed). It must be concluded, 60 therefore, that the introduction of a single DNP group at the damino group of a lysine residue into KDPG aldolase results in a conforma- tional change in the enzyme leading to decreased enzyme activity of the two remaining catalytic sites. Experiments will be described later to detect the existence of such conformational changes. Before these experiments were carried out, it was necessary to determine the nature of the residual activity of the fully dinitrophenylated enzyme. The question to be answered is whether the residual activity represents native enzyme, partially phenylated enzyme, fully phenylated enzyme, or any combination of these species. 4. Isoelectric Focussing of DNP Aldolase The relatively recent introduction of isoelectric focussing to resolve proteins with closely related isoelectric points is of great value in protein separation techniques. Thus, it is possible to separate proteins having a difference in isoelectric point of no more than 0.02 pH units. Isoelectric focussing of native aldolase (m 3 mg) was carried out using a 2% ampholine concentration over a pH range of 4-6. As seen in Figure 8, native enzyme was found to have a pI of 4.8; recovery of enzyme activity placed on the column was 98%. DNP aldo- lase (3.5 mg) was subjected to isoelectric focussing under conditions identical to those described for native enzyme. The DNP aldolase which is relatively insoluble, precipitated and fell to the bottom of the gradient. The experiment was, therefore, repeated using a 4% ampholine concentration. Figure 9 shows that at this ampholine con- centration the DNP aldolase (3 mg) was resolved into two separate 61 FIGURE 8.---Isoelectric focussing of native KDPG aldolase in a gradient of pH range 4-6 and containing 2% ampholine. Approximately 3.5 mg of enzyme was placed on the column. The open triangles re- late to the pH gradient. The closed circles relate to enzyme activity.and the open circles relate to protein absorption at 280 mu. 0.? 1 ¢..v I 0.0 r AllSNEICI 'IVOLLdO ”w 082 on m? N? mm cm on 0 fl N0 m mmEDZ Zorro/mam 000.N V0 0.0 000.v 000.0 000.0 0.0 0.. mmdjooqd. oaox w>_._.Zm_IQ0m._._Z_Q MIL. mod JMQOE < __ SSE 89 would also suggest that although Site 2 lysines are close to the catalytic sites, however these sites are not so close as to completely disrupt the catalytic sites by the binding of relatively large dinitro- phenyl groups. This conclusion is based on the fact that the fully dinitrophenylated enzyme still retains one fifteenth of its original activity and its ability to bind substrate is virtually unimpaired. It is believed that the loss in enzyme activity on dinitrophenylation is due to the occurence of conformational changes induced by the introduction of four dinitrophenyl groups onto the enzyme that results in disruption of the catalytic site. It is obvious that the residues most affected at the catalytic site are not those responsible for binding (although these are slightly affected), but rather those residues that participate in catalysis. In conclusion, this study demonstrates quite clearly that although valuable information has been obtained by the use of chemical modi- fication studies for numerous enzymes, the results in many cases should be treated with a certain degree of caution in absence of further supporting data from X-ray analysis. In addition, the words of Cohen (67) are especially pertinent "...it is necessary to point out that no acceptable technique has yet been developed to detect a limited conformational change in a protein. As a result it is impossible to determine in the majority of cases whether the effect of modification on enzyme activity is the indirect result of a con. formational change, or of the blocking of a group which participates directly in binding, or in catalytic function". CHAPTER VI SUMMARY The dinitrophenylation of KDPG aldolase was investigated. Approximately four DNP moles were bound by the enzyme resulting in 85-90% inactivation of the enzyme. The sites of dinitrophenylation were identified as the e-amino groups of lysine residues. However, the azomethine site was not a site of reaction with FDNB. Protection against three DNP moles were shown to be substrate specific while the uptake of a fourth DNP mole may represent non-specific uptake of FDNB. The DNP aldolase did not appear to be dissociated. The kinetics of dinitrophenylation suggested that the uptake of a single DNP mole per mole of enzyme results in 50% inactivation of the enzyme and a con- comitant change in enzyme conformation. It was further shown that the DNP aldolase could be resolved into two separate peaks by isoelectric focussing. The minor peak focussed at pH 4.8 (same as native enzyme) and contained 1.6 moles FDNB per mole of enzyme. The major peak (95% of the enzyme placed on the column) focussed at pH 5.1 and contained approximately four moles of FDNB per mole of enzyme. It was estimated that the reaction of €-amino groups of lysine residues with FDNB should not result in an increase in pI. The anomalous pI contained was, therefore, ascribed to a conformational change. The enzyme constituting the major peak showed a twofold 9O 91 increase in Km for KDPG and a fifteenfold decrease in Vmax. The rate of tritium exchange for this enzyme from T20 to pyruvate was found not to be ratelimitingas judged by the inability of aldehydes to stimulate the rate of cleavage of KDPG. The results of heat inactiva- tion studies of native and DNP aldolase and those obtained with sulf- hydryl titrations of the two enzyme preparations were interpreted to be consistent with the occurrence of a conformational change of the enzyme on dinitrophenylation. The above results are discussed in . relation to the dinitrophenylation studies on ribonuclease and muscle aldolase. A model is present to account for the dinitrophenylation of KDPG aldolase. I. III It?! 1' I. In | El IlllIliI‘lilll-Illl’I IIIIIIIIII IIII ll. lull {III} III II III II II APPENDIX Relative Rates of Release of B-formyl- Pyruvate from KDG and KDGal The relative rates of release of Bbfonmylpyruvate from KDG and KDGal are shown on page 93. Further details are given on page 23. 'Calculation of the Binding of KDPGal-3-IAC to KDPG Aldolase As shown on Table 2, page 43, the binding of KDPGal-3-IAC to KDPG aldolase is shown with the details of the experiment. In Experiment 1, 2.63 moles of KDPGal-3-IAC were reduced per mole of KDPG aldolase. Further details of this experiment are shown on Appendix Table 1. APPENDIX TABLE l.---The reduction of KDPGal-3-lac to KDPG aldolase in the presence of sodium borohydride. Units of enzyme % loss of Enz e activity remaining enzyme activity prior to NaBHa reduction 10625 - after lst series of NaBHa reduction 1609 84.8 after 2nd series of NaBHa reduction 188 98.2 92 ‘III II... III l [‘11- I l I," II. lilll!I I]!!! II IIIIIIIIIIeIl‘I-l l I'll-.11 III I." Illll .IIIIIII II YIELD OF fl-FORMYLPYRUVATE °/o 93 RATE OF RELEASE OF fl-FORMYLPYRUVATE FROM KDG AND SYNTHESISED KDGGI III | l00 C O O O l / go I. / O .0 /_ O 20 o I o 20 so 40 TIME IN MINUTES 50 II" Illql .Inlsl .l1 I l‘ Ila-l ' ..III ..\II|ll.“l-IIII 1" r l‘ I III III l I I} I 94 Calculation: Amount of aldolase treated = 1.06 mg - 1.569 x 10"2 umoles The treated enzyme was taken up in a final volume of 1.0 ml 0.04 ml (6.3 x 10'"3 “moles aldolase) aliquot of the final solution bound = 3089 cpm Efficiency of counting (internal standard) = 65.82% 6.3 x 10'3 umoles of enzyme bound 308 x 100 65.82 = 4695 dpm Specific activity of KDPGa1-3-140:= 2.81 x 105 dpm/umole thus the no. of moles of KDPGal-3-140 bound 4695 2.81 x 10D _3 = 16.71 x 10-3 moles since 6.3 x 10 umoles of enzyme bound 16.71 x 10'3 umoles of KDPGal- 3-140, the no. of moles KDPGa1-3-140 bound per mole enzyme = 16.71 x 10-3 6.3 x 10'5 = 2e63 The Binding of KDPGal-3-14C to KDPG Aldolase Prior to Dinitrophenylation Details of this experiment are given on Table 3, page 45. Fur- ther details of the KDPCal binding are shown in Appendix Table 2. APPENDIX TABLE 2.---The borohydride reduction of KDPGal-B-IAC to KDPG aldolase * Enzyme % enzyme recovery 280/260 of the Preparation After (NH4)2304 pptn. pptd. enzyme Expt. 1 Test 75 1.73 Control 77 1.70 Expt. 2 Test 80 1.68 Control 62 1.72 * after dialysis against 0.025M imidazole buffer, pH 8.5 95 The Reaction of KDPG Aldolase with FDNB-0-140 Table 4, page 49, gives the details of FDNB-U-lac binding to KDPG aldolase. The calculation for the estimation of the number of DNP moles bound per mole of enzyme is shown below. 3.43 x 10"2 umoles of aldolase was used in a final volume of 12.5 ml An aliquot (2.0 ml) was washed with TCAeand made up to a final volume of 0.5 ml with 0.1N NaOH. A 0.2 ml aliquot of final solution was2 counted. Therefore, 0.2 ml of the solution contained 0.222 x 10 umoles of enzyme. 0.222 x 10"2 x 10"2 umoles of enzyme binds 11,722 cpm Efficiency of counting = 64.6% 0.22 x 10--2 x 10--2 Ilmoles of enzyme binds 11722 x 100 am 64.6 14 6 = 18146 dpm Specific activity of FDNB-U- CI= 2.122 x 10 dpm/ mole amount of FDNB bound by the enzyme = 18146 ‘Iimoles 2.122 x 100 = 0.851 x 10-2 umoles x 10‘211moles Moles FDNB bound per mole enzyme 85 22 x lO'Zjimoles 83 0. 0 3 Analysis of the Kinetics of Dinitrophenylation The data on which Figure 7, page 57, is based is shown on Appen- dix Table 3» The reaction of 4 moles of lysine residues per mole of enzyme constitutes 100% reaction. The values for the rate of the fast-reacting lysines and the rapid initial loss of enzyme activity (plots 3 and 4 respectively, Figure 7, page 57) are shown on Appen- dix Table 4. 96 museumsmu mzom .N >0 emaaqwuasa mp3 wdwemeow hua>wuom dammed mo mosam> one .uuon c0 mocowco>coo you «I 2000.0 00H.0 00~0.0 m0.0~ nmm.0 mn0.m 0m 0000.0 002.0 0m~0.0 00.0N qu.0 n~0.m 00 00~0.0 0-.0 00N0.0 -.- H0n.0 ome.m 00 0220.0 n~.0 0000.0 0¢.- 050.0 0-.m 00 0m~0.0 00.0 ~200.0 ~0.~N 00H.H mom.~ 0m m¢H0.0 Nn.0 0200.0 0~.- mem.H mm0.~ 0N mn~0.0 00.0 0500.0 0~.- 000.2 ~00.~ 0N 00H0.0 0m.0 0550.0 m0.- emu.“ 00N.~ 0» 0520.0 00.0 0N00.0 ~0.N~ nNH.N nmwa.~ 0H ~e~0.0 00.0 002.0 0N.mN mme.m m00~.a 0 ~0~0.0 00.0 h-.0 00.e~ m00.m nmo.0 a n~m0.0 0.0 00H.0 m0.¢~ em.m mq.0 N 00.0 0.H 02.0 000.0N e 0 0 .w:wcwm6ou mzam wdwchemu wcficmeow muoa x wchHmEou weaned mace ACAEV mGACHmEew huw>wuom mzam moHoz mewchEwu mcfim>a Add vason page >uw>wuom mEkuCo mo wdwcmedu 020m moaoe mmaoz 020m mofioz oe>n:o cowuomum scauomuh oedema mogoz s cofiumaxconmonuwnfip 0o mowuocfix emu wow uon wouuo escoomIII.m mqm<9 xHQmem< 97 APPENDIX TABLE‘I.---The rate of reaction of the fast-reacting lysine and the rapid initial loss of enzyme activity. Moles lysine remaining Enzyme activity remaining Time (min) Moles FDNB remaining Moles FDNB remaining 0 0.040 0.041 2 0.028 0.0295 4 0.020 0.0205 6 0.010 0.008 Details for the 0RD of Native and DNP Aldolase Optical rotary dispersion measurements were made with a DURRUM- JASCO recording spectropolarimeter under constant nitrogen flush. A cell with a 1mm path length was used and the 0RD was measured at room temperature. Native enzyme solutions were employed at a concentration of 0.9 mg/ml while DNP aldolase was used at a concentration of 0.67 mg/ml. Prior to each run the 0RD of a control was run containing all additions except enzyme. A plot of specific rotation (0’)A- vs wave- length (1.) is shown on Appendix Figure 2. The 0RD data was plotted by the method of Moffit and Yang (68). 4 (m'): ao )‘o + bo X0 2 2 2 2 2 (>1 -X) ()1 -7.) o o In this equation, (m') is the reduced mean residue rotation, a0, b0, and A0 are constants; Ab of 212 was used for calculations. 30 and b0 98 APPENDIX FIGURE 2.---Optical rotary dispersion of native and DNP aldolases. The circles represent the 0RD f6r native enzyme while the triangles represent ORD for DNP aldolase. The conditions for this experiment are on page 97. SPECIFIC ROTATION ([ofilx) 99 +5000 0RD 0F NATIVE AND DNP ALDOLASE +3000 +IOOO -I000 -3000 -5000 -7000 -9000 I60 200 240 280 300 WAVELENGTH — mu (A) 340 100. were calculated from the slope and intercept, respectively, of a plot of (m') (A2 -7\ (2’) against (1/ A2 -A:). b = slope/K 4 where k = 100 (n2 + 2) o o M 3 M = mean residue weight n == refractive index a0 = zero intercept/KR: The apparent helical content (+) from the Moffit-Yang equation was computed as follows: 630 assuming bo = -630 for 100% helix and b0 = 0 for no helical content. The Moffitt-Yang plot is shown on Appendix Figure 3; the a helix content of the native enzyme was found to be 30.71%. The DNP aldo- lase shows approximately 507. reduction of ahelix as estimated from the trough at 233 mp. It was assumed that (m')233 = -l6500 for a helical structure and (m')233 = -2000 for a disordered structure (69). The 0RD results for the DNP aldolase are meaningless in view of the fact that the DNP radical absorbs very strongly in the 233 mu region. In fact, the absorption observed at 233 mu is approximately twice the absorption at 360 mu. 101 m (>303) x I0"8 _ ‘I pnvngx 2.9 20.. a II 41m zomjkpzo 2.3 now 25% .809 proorpmm II / _ Iboo OIO o /o /0 1. 2. 3. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. LIST OF REFERENCES Bunnett, J. F., QUART. REV. (London) 12, l (1958) Ross, 3. 0., in PROGRESS IN PHYSICAL ORGANIC CHEMISTRY l, s. c. A. Streitweiser, Jr., and R. W. Taft, Ed., Interscience Pub- lishers, Inc., N. Y., 31 (1963) Bunnett, J. F., and Garst, R. H., J. AM. CHEM. SOC. 81, 3875 (1965) Bunnett, J. F., and Randall, J. J., AM. CHEM. SOC. 89, 6020 (1958) Bunnett, J. F., and Garst, R. H., J. AM. CHEM. SOC. 81, 3879 (1965) Zollinger, H., and Bernesconi, C., TETRAHEDRON LETTERS, 19535 (1963) Pietra, F., and Fava, A., TETRAHEDRON LETTERS, 1083 (1965) Kirby, A. J., and Jencks, W. P., J. AM. CHEM. SOC. 81, 3217 (1965) Hits, 0. H. W., in.METHODS or ENZYMOLOGY 19, ed. 0. w. H. Hirs, Academic Press, N. Y., 548 (1967) Shaltiel, Se, BIOCHEM. BIOPHYS. RES. COMMUNo‘Zg, 178 (1967) Sanger, Fe, BImHEM. Je 22., 507 (1945) Massey, V., and Hartley, B. S., BIOCHEM. BIOPHYS. ACTA. 21, 361 (1956) Hirs, c. H. W., BROOKHAVEN SYMPOSIUM 13, 154 (1963) Hirs, C. H. W., Halmann, M., and Kycia, J. H., ARCH. BIOCHEM. BIOPHYS. ACTA. 111, 209 (1965) Murdock, A. L., Grist, K. L., and Hirs, C. H. W., ARCH. BIOCHEM. BIOPHYS. ACTA. 111, 223 (1965) MurdOCk, A. L., GriSt, Ke We, and Hits, Ce He We, ARCH. BICEHEM. BIOPHYS. ACTA. 114, 375 (1966) Kartha, G., Bello, J., and Harker, D., NATURE 213, 862 (1967) Perutz, Me Fe, EUR. Je BIOCHEM. g, 455 (1969) Pontremoli, 8., Luppis, 8., Wood, W. A., Traniello, S., and Horecker, B. L., J. BIOL. CHEM. 240, 3464 (1965) 102 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 103 Pontremoli, S., Luppis, B., Traniello, S., Wood, W. A., and Horecker, B. L., J. BIOL. CHEM. 240, 3469 (1965) Pontremoli, 8., Traniello, S., Enser, M., Shapiro, 8., and HOIECker, Be Le, PRCXI. NATLe ACADe SCIe (UeSe) 2g, 286 (1967) R053“, 00 Me, and R058“, Se Me, PROC. NATLe ACAD. SCI. (UeSe) 2, 1155 (1966) Ronca, C., Ipata, P. L., and Bauer, C., BIOCHEM. BIOPHYS. ACTA. 122, 379 (1966) Moore, P. B., J. M010. 3100 .2_'2_, 1.45 (1966) Haines, J. A., and Zamecnik, P. C., BIOCHEM. BIOPHYS. ACTA. 146, 227 (1967) De Prisco, 0., BIOCHEM. BIOPHYS. ass. COMMUN. 29, 148 (1967) Gold, A. M., BIOCHEM. l, 2106 (1968) Philip, C., and Graves, D. J., BIOCHEM. l, 2093 (1968) Keech, D. B., and Farrant, R. K., BIOCHEM. BIOPHYS. ACTA. 151, 493 (1968) Bailin, C., and Barany, M., BIOCHEM. BIOPHYS. ACTA. 168, 282 (1968) Cremona, T., Kowal, J., and Horecker, B. L., PROC. NATL. ACAD. SCI. (U.S.) 23, 1395 (1965) Rowley, P. T. Tehola, 0., and Horecker, B. L., ARCH. BIOCHEM. BIOPHYS. 107, 305 (1964) Kowal, J., Cremmna, T., and Horecker, B. L., J. BIOL. CHEM. 240, 2485 (1965) Ingram, J. M., and Wood, W. A., J. BIOL. CHEM. 240, 4146 (1966) Meloche, H. P., and Wood, W. A., J. BIOL. CHEM. 239, 3515 (1964) Moehler, H., Hammerstedt, R. H., Decker, K., and Wood, W. A., Manuscript in preparation Hammerstedt, R. H., and Wood, W. A., Manuscript in preparation Robertson, D., and Wood, W. A., Manuscript in preparation Grazi, B., Meloche, H. P., Martinez, C., Wood, W. A., and Horecker, B. L., BIOCHEM. BIOPHYS. RES. COMMUN. 19, 4 (1963) Ingram, J. M., and Wood, w. A., J. BIOL. CHEM. 241, 3256 (1966) 41. 42. 43 . 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 104 Rose, I. A., and O'Connell, E. L., ARCH. BIOCHEM. BIOPHYS. 118, 758 (1967) _Altekar, W., Unpublished results Decker, K., Moehler, H., and Wood, W. A., Manuscript in preparation Meloche, H. P., BIOCHEM. BIOPHYS. RES. COMMUN. 18, 277 (1965) Meloche, H. P., and Wood, W. A., 1n.METHODS OF ENZYMOLOGY IK, ed. W. A. Wood, Academic Press, N. Y., 51 (1966) Shuster, 0. W., in METHODS OF ENZYMOLOGY 33, ed. w. A. Wood, Academic Press, N. Y. 524 (1966) Srinivasan, P. R., and Sprinson, D. B., J. BIOL. CHEM. 234, 716 (1959) P138188, To, and AShwell, Ge, Jo BIOL. CHEM. 237, 317 (1962) Ghalambour, M. A., Levine, E. M., and Heath, E. C., J. BIOL. CHEM. 241, 3207 (1966) Macgee, J., and Doudoroff, M., J. BIOL. CHEM. 213, 757 (1955) Dahms, S. A., Private communication Bray, G. A., ANAL. BIOCHEM. 1, 279 (1960) Barker, S. B., in METHODS OF ENZYMOLOGY III, eds. S. P. Colowick and N. 0. Kaplan, Academic Press, N. Y., 241 (1957) DaViS, Be Jo, ANN. Ne Ye ACAD. SCIe 121, 404 (1964) Kovachevich, R., and Wood, W. A., J. BIOL. CHEM. 213, 757 (1955) Warburg, 0., and Christian, W., BIOCHEM. Z. 314, 149 (1943) Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. V., J. BIOL. CHEM. 193, 265 (1951) Mahowald, T. A., BIOCHEM. 4, 732 (1965) Levy, A. L., and Li, C. H., J. BIOL. CHEM. 22, 487 (1955) Ray, W. J. Jr., and Koshland, D. E. Jr., in BROOKHAVEN SYMP. IN BIOLOGY 13, 135 (1960) Rutter, W. J., Richards, V. C., and Woodfin, B. M., J. BIOL. CHEM. 236, 3193 (1961) Spolter, P. D., Adelman, R. C., and Weinhouse, 8., J. BIOL. CHEM. 240, 1327 (1965) 63. 64. 65. 66. 67. 68. 69. 105 Rose, I. A., O'Connell, E. L., and Mehler, A. H., J. BIOL. CHEM. 240, 1758 (1965) Biltonen, R. L., and Lumry, R., J. AM. CHEM. SOC. 87, 4208-9 (1965) Monod, J., Wyman, J., Changeux, J. P., J. MOL. BIOL. 22, 405 (1965) KOShland, D. E. Jr., COLD. SYMPe QUANTe 3101.0 21, 437 (1962) COhen, Le A0, ANN. REVS. OF 31mm. 21, 8d. P. De Boyer, Anne Revs. Inc., Palo Alto, California 695 (1958) MOffit, We, and Yang, J. To, PRU}. NATe ACAD. SCIe UeSeAe fig, 596 (1956) Jirgensons, B., J. BIOL. CHEM. 240, 1064 (1965)