ABSTRACT MECHANISM OF ALLOSTERIC CONTROL OF L-THREONINE DEHYDRASE 0F ESCHERICHIA COLI BY ADENOSINE-S'-MONOPHOSPHATE by Kenneth Warren Rabinowitz The biodegradative L-threonine dehydrase of Escherichia 222i catalyzes the irreversible dehydration of L-threonine and L-serine. AMP functions as an allosteric activator of the enzyme, causing both a 25-35 fold decrease in the Km for L-threonine, and a 3-4 fold increase in'Vmax. The Km values for L-threonine with highly purified dehydrase were 2 mM and 50-70 mM. and'Vmax values were 360 umoles/min/mg protein and 110 umoles/min/mg protein in the presence and absence of AMP, reapectively. The Ka for AMP was found to be 0.5 mm. Velocity versus substrate and velocity versus AMP plots were hyperbolic; Hill coefficients for L-threonine were 1.02 and 1.0“ in the presence and absence of AMP, reSpectively. The Hill coefficient for AMP was 0.99. As assessed from sucrose gradient centrifugations and molecular sieve chromatography. the dehydrase exists as a monomer of 3.1-3.6 S, and #0,000 approximate molecular weight at low enzyme concentration and in the absence of AMP. Both.AMP and increased enzyme concentration cause association of the dehydrase to oligomers as large as 8 S, or 160,000 molecular weight; this Species correSponds to a tetramer. m~ Ine oxygen bub: II I nor inoreai oxidized de- 6 ‘ '°' 2a» .‘03‘. uU. & “eacuivated .. '1 a ., 9:4-.th mur at 413 m , CECE addi‘: *1? L55 :1; MC ‘ I 1521:: *8 ‘h Vite de3:.:d \. Kenneth Warren Rabinowitz The dehydrase was inactivated upon oxidation by oxygen bubbling, oxidized glutathione, or aging. The oxidized dehydrase was monomeric, 3.2 S, and neither.AMP nor increasing the dehydrase concentration caused oligo- merization of this Species. Reduction with dithiothreitol reactivated the dehydrase and restored the monomer-oligomer equilibrium. Threonine dehydrase exhibits an absorption maximum at 413 mu, and positive circular dichroism at 415 mu. Upon addition of L-threonine a new Spectral maximum appears at 455 mu. while circular dichroism at 415 mu is lost. The 455 mu Spectral absorbance decreases, and the 415 mu circu- lar dichroism returns with Substrate depletion. An investi- gation with competitive inhibitors indicated that the azomethine linkage between pyridoxal phoSphate and a lysyl—residue of the dehydrase is reSponsible for the 415 mu circular dichro- ism. Aminocrotonate-pyridoxal phoSphate azomethine was suggested to be the intermediate absorbing at 455 mu. The step in the catalytic mechanism which is altered during allosteric activation was assessed by examination of the effect of AMP on (1) the loss of 415 mu circular dichro- ism, and (2) the accumulation of the 455 mu absorbance, observed during catalysis. and (3) the binding affinity of the dehydrase toward substrate analogues of L-threonine. AMP markedly enhanced both the extent of loss of the 415 mu circular dichroism, and the accumulation of the intermediate absorbing 3 influence : mantis: : :eps; non: etior. with ! Tutti. Kn? nu . COE’JQVNQ are inc a; Wriioxa; Kenneth Warren Rabinowitz absorbing at 455 mu. These findings indicate that the influence of the allosteric alteration on the catalytic mechanism must lie with one or more of the first four steps; noncovalent enzyme-complex formation, transaldimin- ation with substrate, or one of the two steps of dehydra- tion. AMP enhanced the binding affinity for a number of competitive inhibitors capable of undergoing only the first two steps of catalysis, i.e.. non-covalent complex formation, and transaldimination. Further, binding affin- ity was substantially increased for several substrate analogues which, by virtue of the lack of an d-amino group, are incapable of undergoing transaldimination with the pyridoxal phoSphate of the dehydrase. These observations suggest that the major effect of AMP is exerted on non- covalent enzyme-substrate complex formation. In order to evaluate whether oligomer formation is a sufficient condition to cause allosteric activation, the K1 for certain competitive inhibitors, as well as, the Km for L-threonine and Vmax were determined for the oligomer induced by high dehydrase concentration. The K1, Km. and ‘Vmax values obtained were Similar to those for the AMP-free monomer. On this basis it was concluded that oligomer for- mation, by itself, cannot account for the activation. The activation of threonine dehydrase was found to be a relatively slow process which could be followed directly 1 aiiition c Eminent: the :olec: mSpect t: but in :1“: Ir. analogues :r:eor.ir.e § '3 furor . y A ‘er’,fi‘\ap it‘d; "“31: t cor taOS Kenneth Warren Rabinowitz directly in the coupled-Spectrophotometric assay, after the addition of AMP to the L-threonine-preincubated dehydrase. Examination of the kinetics of the activation revealed that the molecularity of the process is second order with reSpect to enzyme. Hence, oligomerization is a necessary, but insufficient requisite for allosteric activation. In order to characterize the allosteric binding site analogues of AMP were tested for their ability to activate threonine dehydrase. The phoSphoribosyl moiety was found to be of greater importance than the adenine base. Alter- ation of either the 1', 2', 3' or 5' position of this moiety deleteriously affected the ability of the nucleotide to cause allosteric activation. Considerable alteration in the base moiety was tolerated indicating that the substitu- ents on the adenine base may contribute in a graded manner toward binding. Certain nucleotide analogues were shown to mediate oligomerization of the dehydrase. A large number of L-threonine analogues were found to function effectively as competitive inhibitors of the dehydrase. Analogues of the L-series diSplayed appreciable dehydrase affinity only in the presence of AMP, whereas, for those inhibitors of the D-series binding affinity was not deleteriously affected by the removal of AMP. Structural requirements for D-analogue binding were much more restrictive than for the L-analogues. Only D- analogues possessing both an a-amino group and a B or y ‘ A“ 1 \Q‘b SF". Oxy. Sun . O1 ‘ .‘ P; 38110:??? ov, ‘ 7 .. 315:3 53m “‘ A nod ~ ‘- 554136 an Kenneth Warren Babinowitz hydroxyl substituent (or a substituted B-hydroxyl) effec~ tively bound to the dehydrase, viz., D-serine, D-threonine, D-allothreonine, D-homoserine, and.DL-0-methylserine. L-analogues altered at the d—amino, and the B-hydroxyl posi- tions still diSplayed appreciable affinity for the dehydrase. A model to account for the alterations in quaternary structure and the allosteric activation is proposed. mm ! I‘flw‘iliflufl ill MECHANISM OF AIIDSTERIC CONTROL OF L-THREONINE DEHYDRASE OF ESCHERICHIA COLI BY ADENOSINE-S'-MONOPHOSPHATE By Kenneth Warren Rabinowitz A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1970 “W 1'- .I .- L: ' .h I, . of '13 I I I 1- I o-‘I an :‘m- "an ‘3 “.I" . uv ‘ '.-'.e a: in U? Q g... V..e :ror‘F TERD‘wed t. ”" .f P“ '5 n . on. ‘Vy \IC 3 & 1 3"“"tea a 'flflp ‘ . ~QVL e: 8:“ - O I 0; '33s“ , «A V\ ’ Jo :- u. ‘ A‘. «I AV“: :2 1 ~ . G f~.e3. .“ ’A cud‘ “re . O as; b. i Q ‘I ‘- ‘V‘Q VITA ir. Rabinowitz was born on July 27, 1943 in the Bronx, New York. He attended the Christopher Columbus High School in the Bronx from September 1958 to February 1961. He received the degree Bachelor of Science in chemistry from the City College of New York in February, 1965. He accepted a graduate assistantship in the Department of Biochemistry at Michigan State University in the Spring of 1965. He has accepted a fellowship from the National Institutes of Health for postdoctoral study under Dr. D. E. Atkinson at the University of California, Los Angeles. Mr. Rabinowitz is a member of the American Society for the Advancement of Science, Sigma Xi, and a former member of local 802, American Federation of Muscians. He is married. 11 The a tier: to ‘u'. A zonfidence t ! kstitutes c. The a fro: this 13 810:5, e339: Co‘rr. 3e:- ixc'e‘eted to so: the p.- 5°15 Synthe s ACKNOWLEDGMENTS The author wishes to extend his sincere apprecia- tion to W. A. Wood for his guidance, encouragement and confidence through the course of this research. Support from the predoctoral training grant of the National Institutes of Health is also acknowledged. The author also is grateful to the many colleagues from this laboratory for their numerous helpful discus- sions, eSpecially to Mr. Mark Roseman. Mr. Howard Brockman, Mr. John Gerlt, and Dr. Donald Schneider. He is also indebted to Dr. Roy Hammerstedt for his advice on amino acid syntheses. The author would like to thank Carolanne Rabinowitz, Mark Williams and Nancy Williams for their assistance in typing the rough draft of this thesis. iii To C.W.R. and To My Parents iv .‘fi'A‘VYm . R‘- - I £douvuvv¢ ‘b., o'flm.firw~= . " H‘ “oi-It‘d V‘." “ O! . ”F‘IQ' .56- Goose: 'p U! " 0 (H 'P Ins-31",(ll TABLE OF CONTENTS INTRODUCTION 0 . . . . . LITERATURE REVIEW . . . Allosteric Regulation of Enzyme Activity Early DevelOpment CoOperativity Parameters . K and‘V System Enzymes . Concerted Transition Model The Sequential Induced Fit Yeast Isocitrate Dehydrogenase . PhosPhofruotokinase of Yeast . ASpartic Transcarbamylase Glyceraldehyde-BAPhOSphate Dehydrogenas Rabbit Muscle . 8611.386 0 o o o Threonine Dehydrases Yeast Glyceraldehyde-B-PhOSphate Dehydro- Model Biodegradative Threonine Dehydrase Escherichia coli BioEEgradative'THEEonine Dehydrase Clostridium tetanomorphum . METHODS AND MATERIALS . Bacteriological . . . Chemical . . . . . . AMP Analogues . . Amino.Acids . . . Determinations and Procedures Enzyfiat 1° C C C O C C Assays for'Threonine Dehydrase. Purification of Threonine Dehydrase Removal of AMP . . of or o e 000000000. 23 32 38 38 38 38 44 50 51 53 56 RES UIJTS C O O O O O C O O O O O I O O I O O I O O O O 1) 2) 3) 4) 5) Characterization of Dehydrase Kinetics and Protomer-Oligomer Interconversions . . . . . . Effect of AMP on Km and V a . . . . . . . . .Absence of L-Threonine angthP Cooperativity . Effect of AMP on Quaternary Structure . . . . . $20 Value of the Dehydrase at Catalytic Assay Concentration . . . . . . . . . . . . Effect of Sulfhydryl.Groups . . . . . . . . . . Structural Requirements for Nucleotide Allosteric Activation . . . . . . . . . . . . . Alteration of the Base Moiety . . . . . . . . . PhoSphoribosyl Group . . . . . Effect of AMP Analogues on Sedimentation Behavior . . . . . . . . . . . . . . . . . . Mechanistic Site of Allosteric Activation . . . Kinetic Evaluation . . . . Spectral and Dichroic Properties of Threonine Dehydrase . . . Effect of Competitive Inhibitors on Circular Dichroism at 415 mu . . . Effect of Competitive Inhibitors on Absorption Spectrum . . . . . . Interpretation of the Absorption Peak at 455 mu Elicited by LJThreonine . . . . . . Effect of AMP on the Spectral Absorption at 455 mu and the Circular Dichroism at 415 mu . . . . Effect of AMP on K1 for IpThreonine Analogues . Structural Requisites for Threonine and Threonine Analogue Binding . . . . . . . . . . Relation of Activation and Oligomerization Kinetics of the Oligomer Induced by High Dehydrase Concentration . . . . . . . . . . . . Determination of the K1 for IpThreonine Analogues by Circular Dichroism . . . . Direct Determination of Km and Vmax for High Protein-Induced Oligomer . . Molecularity of the Allosteric Activation with RosPect to Dehydrase Concentration . . . . Activation by AMP at High LJThreonine Concen— tration . . . . Kinetic Plots for Reversed Order of Addition . vi 61 61 64 64 78 79 82 83 87 89 91 92 94 103 104 111 114 122 128 131 132 137 142 147 152 “'AR"C\ ~ Vfi“ '-UVUUUA».' $.57. 82‘“ :ffer Qua: :e latic A AC:: Cozpa“ Prm nose fl? - .g .her E 215: :ffec Cat: 1. 89"“ Page DISCUSSIOBI C O O O O O O O O O O O O O O 0 O O O O O 160 Quaternary Structure Threonine Dehydrase . . . . 160 Effect of AMP and Enzyme Concentration on Quaternary Structure . . . . . 161 Relation of Quaternary Structure to Catalytic Activity and Allosteric Activation . . . . . . 162 Comparison to Thnoretical Models . . . . . . 166 Preposed Model for Activation of Threonine Dehydrase . . . 169 Other Regulatory Enzymes Displaying Association- Dissociation Equilibria . . . . . . 173 Effect of the Allosteric Alteration on the Catalytic Mechanism . . . 178 Structural Requisites for Inhibitor Binding to the Catalytic Site of Threonine Dehydrase . . 181 SUI’IFIARY O O O O I O O O O O O O O O O O O O O O O Q 183 APPmDIX . O O O O O O I O O O O O O O O O O O O O O 186 Derivation of Km and'Vmax for the Threonine Dehydrase Reaction . . . . . 186 Derivation of Equations for Determination of K by Circular Dichroism . . . 189 Trea ment of Threonine Dehydrase with Sulfhydryl. Reagents O I O O O O O O O O O O O O I O O I O 192 LIST. “.mERENCES O O O O O O O O O O O O O O O O O 199 vii -. 5‘“ TABLE 10 11 LIST OF TABLES Standardization of SB-283 Rotor for Sucrose Gradient Experiments . . . . . . . . . . . . Ef feet or AMP on Km and. Vmax O I C C O C O O 0 s20 Values Obtained at Low Dehydrase Concen- tration in the Absence of AMP . . . . . . . Concentration Dependence of Sedimentation Velocity for Oxidized Dehydrase . . . . . . Specificity of Threonine Dehydrase for Purine Base 0 I O C O O O O O O O O O O O O O O O 0 Specificity of Threonine Dehydrase for the Ribosyl MOlety Of AbiP o o o o o o o c o o 0 Effect of AMP Analogues on Quaternary Structure of Threonine Dehydrase . . . . . . . . . . . Effect of AMP on Km of Threonine and K1 of Threonine Analogues . . . . . . . . . . . . Effect of AMP on Km of Threonine and K1 of Threonine Analogues from Kinetic and Circular Dichroism Determinations . . . . . Determination of Km and'Vmax for.AMP-Free Dehydrase at Varying Protein Concentration . Regulatory Enzymes Exhibiting Association- Dissociation Equilibria . . . . . . . . . . APPENDIX TABLE 1 Reversal of p-Hydroxymercuribenzoate Inhibi- tion by Glutathione . . . . . . . . . . . . viii Page 62 72 81 84 88 90 125 136 141 174 198 “'0‘...“ I ._ H II n .8 U‘Abu U) U\ (I) ‘0 10 '.O {r O 0 fl! 9. u-o ‘c 'l: f) hi It! FIGURE 10 11 12 13 14 LIST OF FIGURES The Mechanism of Enzymatic Threonine Dehydration I I O O O O O O O O O O O I 0 Method for Kinetic Determination of K1 . . . Removal of AMP from Dehydrase Preparation . Reciprocal'Velocity'Versus lpThreonine Plots for Highly Purified Dehydrase in the .Absence and Presence of AMP . . . . . . . Hill Plot for Threonine Dehydrase with IJ-T hre onlne O O O O O C O O O O O O O O 0 Hill Plot for Highly Purified Threonine Dehydrase with AMP . . . . . . . . . . . Variation of S Value with Concentration of the High y Purified Dehydrase . . . . Absorption Spectrum of IpThreonine Dehydrase Circular Dichroism of LJThreonine Dehydrase. Time Course for a-Ketobutyrate Formation, for Appearance and Disappearance of the Spectral Intermediate at 455 mu. and for the Loss and Reappearance of Circular Dichroism at 415 mu . . . . . . . . . . . Effect of Competitive Inhibitors on Visible Spectrum of Threonine Dehydrase in the Presence of AMP . . . . . . . . . . . . . Effect of Competitive Inhibitors on Visible Spectrum of AMP-Free Dehydrase . . . . . Difference Spectrum of LJThreonine Dehydrase in the Presence of D-Threonine. Effect of AMP on LJThreonine Induced Loss of Circular Dichroism . . . . . . . . . . ix Page 27 48 57 65 67 69 74 96 98 101 105 107 109 115 .o—‘H ”OF I I . 5 at: V““ ‘1“ 1 f A... w... «a» (J 10 45 #1 JVA AV 8 Q; Cu Fe m. r. .A s s m 1 .. 1.. "v 2. VA T1. \ w AU 1 2 A1) ’1». S \u‘ A‘ RAIL 2 2 2 at find ha v . ‘LA FIGURE 15 Effect of AMP on Difference Spectra of Threonine Dehydrase After Addition of LJThreonine . . . . . . . . . . . . . . . 16 Effect of AMP on LJThreonine Induced Accumulation of Spectral Intermediate at “55 m C O O O O O O I O O O O O O O O O O 17 Determination of K1 for the Inhibitor D-Threonine by Measuring Loss of Enzyme Circular Dichroism at 415 mu . . . . . . 18 A Lineweaver-Burke Plot of Velocity Versus Threonine Concentration Measured at High Enzyme Concentration in the Direct Spectrophotometric Assay . . . . . . . . 19 Time Course for the Activation of Threonine Dehydrase by AMP . . . . . . . . . . . . 20 Activation of Threonine Dehydrase as a Function of Time with'Varying Enzyme Concentration . . . . . . . . . . . . . . 21 Determination of Order of Threonine Dehyd- rase Activation with ReSpect to Protein Concentration by the Differential Method. 22 Time for the Activation of Threonine Dehydrase by AMP at High L-Threonine Concentration . . . . . . . . . . . . . . 23 'Velocity'Versus LJThreonine and Hill Plot for Enzyme Preincubated with.IPThreonine. 2h Velocity'Versus.AMP and Hill Plot for Enzyme Preincubation with LJThreonine . . 25 ‘Proposed Model for Monomer-Oligomer Inter- conversions and Allosteric Activation . . APPENDIX FIGURE 1 Inactivation of Threonine Dehydrase by p-Hydroxymercuribenzoate . . . . . . . . 2 Protection by AMP from p-Hydroxymercuribenzo- ate Inactivation of Threonine Dehydrase . 120 134 139 143 lfi8 150 153 155 158 170 19h 196 :w Chezi ‘. U C, C l a. i a. D W I C O vnv i C a. m. S a u 9 AU a: F- ?” ti 1: Previous ‘Preliminary reports of this work were presented at the annual meeting of the American Society for Microbiology, Los Angeles, 1966; the Second International Symposium on the Chemical and Biological.ASpects of Pyridoxal Catalysis, Moscow 1966; the Seventh International Congress of Biochem- istry, Tokyo. 1967; and at the annual meeting of the American Society of Biological Chemists, Atlantic City, 1969 and 1970. Parts of this work are contained in three previous publications (6#, 71, 7h). xi ‘i 911 ins“ revealed a 1: whose pre; 2) Vatior. re; e"« A‘ ..t.er_c: 3) 132' Wig}: in the pn Wily”? INTRODUCTION The biodegradative L-threonine dehydrase of Escherichia so}; catalyzes the following irreversible c-B elimination reactions: Ipthreonine -—-—-—>» d-ketobutyrate + ammonia L-serine -———%>"pyruvate + ammonia Early characterization of threonine dehydrase revealed a number of interesting properties of this enzyme: 1) Pyridoxal phosphate is a tightly bound cofactor whose presence is required for enzymatic activity; 2) AMP is an activator of the enzyme. with the acti- vation resulting from both a small increase in catalytic efficiency, and a large increase in substrate affinity; 3) The enzyme readily dissociates into lower molecu- lar weight Species in the absence of AMP; and reassociates in the presence of AMP. Thus, AMP apparently shifts an equilibrium between protomeric and oligomeric forms of the dehydrase. Since there is no evidence that AMP is directly involved in the catalytic reaction. AMP can be termed an "allosteric effector" and threonine dehydrase an "allosteric enzyme." Activation is then seen as resulting from a con- formationally induced alteration of the catalytic site caused by (or concomitant with) the binding of AMP to a 1 distinct 5 probably c by examini reaction :echamis: lates ind steps in 2 distinct second site. Such a concept of AMP activation is probably correct but vague indeed. This is best illustrated by examining the prOposed mechanism for the dehydration reaction in Figure 1 (page 27). The complexity of this mechanism leads to the question as to whether AMP stimu- lates indirectly, one. several or all of the intermediate steps in the reaction. The problem of identifying as precisely as possible the specific step(s) in the mechanism affected by AMP would appear at the outset to be unusually difficult and necessar- ily dependent on indirect eXperiments. However, by virtue of the presence of pyridoxal phOSphate there is a unique and invaluable Opportunity to observe conversion of inter- mediates on the enzyme surface--for several pyridoxal phos- phate intermediates possess measurable and characteristic spectral and dichroic prOperties. Accordingly, the effect of AMP on the accumulation of "spectral" and "dichroic" intermediates has been examined. Further, the effect of AMP on dehydrase affinity for competitive inhibitors has also been investigated. Since substrate analogues are incapable of undergoing all of the partial steps of the catalytic mechanism this study provided an additional means to locate specific points of influence of AMP. The results of these studies taken together indicate that the major influence of the allosteric transition must be exerted on an initial step in the dehydration mechanism. i.e.. on non- covalent e Fro frdergoes pcstulatio structure This could alteration quence of :erely as l Alternativ Elem’a'Cior t0 the pr: Water)- a». ‘J Ettivatio. ,Y-u ”‘5 and of L’l‘v’esti .. AS; ks. 0‘ Ought a‘ V e. D‘OVein C o“ ' N “50:9... ‘ ‘ Fa ‘ t! ”310de Ten . inlaite 3 covalent enzyme threonine complex formation. From the earliest recognition that threonine dehydrase undergoes AMP-related changes in quaternary structure the postulation had been made that the changes in quaternary structure may be casually related to the activation process. This could be mediated in several ways. First, allosteric alteration and activation may occur exclusively as a conse- quence of enzyme oligomerization. and AMP may then function merely as an agent which encourages such oligomerization. Alternatively. the allosteric transition reSponsible for activation may be exclusively dependent upon binding of AMP to the protomer, in which case the observed changes in quaternary structure are an indirect consequence of the activation. Lastly, activation may require both.AMP bind- ing and oligomerization. Studies undertaken in the current investigation indicated that oligomerization could be brought about in the absence of AMP by increasing the protein concentration. It was then established that such oligomerization did not cause enzyme activation. Finally. in order to more extensively characterize the dehydrase. nucleotide and substrate analogues were employed in an examination of the structural and chemical requisites of both the activator and catalytic binding sites. lb of allost e1gables a. 0 It, LITERATURE REVIEW Allosteric Regulation of Enzyme Activity A near plethoric volume of literature in the field of allosteric regulation has accumulated. consequently, numerous reviews on the subject have appeared (1-6), two within the past year (5, 6). As a result, this review shall be somewhat limited. It is primarily intended to present a brief description of the development of this field, the prOposed allosteric models and several specific examples which have been useful in testing predictions of these models. Early DeveloEment In 1956 Umbarger (7), and Yates and Pardee (8) first demonstrated that end products of biosynthetic metabolic pathways could Specifically inhibit the action of certain key enzymes early in their pathways. The independent investigations of Umbarger with the biosynthetic threonine dehydrase, and Yates and Pardee with aSpartic transcarb- amylase paved the way for the very rapid recognition of many other regulatory enzymes (9). As detailed characterization was undertaken it became clear that these enzymes possessed many distinct preperties absent from non-regulatory enzymes. Indeed, in z. charger' threonine not be ob‘ l/velocit; 1/veloc it; linear crtf 5 Umbarger's first report (7) on the feedback regulation of threonine dehydrase it was noted that straight lines could not be obtained in the normal Lineweaver-Burke plots of 1/velocity versus 1/substrate, but only in plots of 1/velocity versus 1/(substrate)2. Dixon plots were linear only when 1/velocity was plotted against (inhibitor)2. Plots of velocity versus substrate concentration, if they had been presented, would have appeared "S" shaped or sigmoid: later, such sigmoid plots were reported by Gerhart and Pardee (10, 11) for aSpartic transcarbamylase. The characteristic sigmoid-shaped binding curve was known from the early work on hemoglobin (12, 13) to reflect coOperative ligand binding; that is, binding of each mole- cule of a particular ligand enhances the binding affinity for subsequent molecules of the same ligand. Thus, it was recognized quite early that regulatory enzymes may not only exhibit interactions between non-identical ligands, such as between and substrate molecules ("heterotropic effects"), but also cooperative (or "homotropic") effects between identical ligands. Changeux (14) with biosynthetic threonine dehydrase, and Gerhart and Pardee (10, 11) studying aspartic transcarb- amylase reported that these enzymes could be rendered insensitive to inhibition by their feedback modifiers, isoleucine and CTP, respectively. upon treatment with heat and mercurials. Moreover, the resulting desensitized enzyme no lorger c' their subs‘ Jacob (in) regulatory kg sites (10) prOpc interactic by cantor: with 81th: CEIAnaewu. allosteri Yer alte 6 no longer displayed cooperative binding behavior towards their substrates. These findings led Monod. Changeux and Jacob (in) and Gerhart and Pardee (10) to suggest that regulatory enzymes should possess separate. but interact— ing sites for substrate and effector. Gerhard and Pardee (10) proposed that both the heterotrOpic and homotropic interactions of aSpartic transcarbamylase may be mediated by conformational alterations of the protein associated with either substrate or effector binding. Monod, Changeux, and Jacob (15) termed the later process an allosteric transition and argued in favor of this model over alternative models which encompass interaction of separate sites without conformational alterations. Supplementary studies continued to substantiate the concept of regulation through separate sites and conforma- tional alteration. Gerhart and Schachman (16) were able to separate catalytic and regulatory subunits of aspartic transcarbamylase. The catalytic subunit had a higher specific activity than native intact enzyme, but was insensitive to inhibition by CTP. Reconstitution of the two types of subunits restored effector sensitivity. Changeux (17) demonstrated that the biosynthetic threonine dehydrase of‘g.‘ggli could be rendered less sen- sitive to urea inactivation by incubation with isoleucine, and conversely, more sensitive through incubation with the isoleucine antagonists Lavaline and L-norleucine. These find* .55 1 :crleucint threonine Fr: activatio: Hathaway 1 3P}.'+-isoc my er.2y Effector-s I'9-Elllattor '13!) in‘ OOV 7 findings suggested that L-isoleucine, L-valine and L- norleucine could all induce conformational alterations of threonine dehydrase. From.Mansour and Mansour's report on cyclic AMP activation of liver fluke phOSphofructokinase (18), and Hathaway and Atkinson's study of the AMP activation of DPN+-isocitrate dehydrogenase (19) it became apparent that many enzyme activators could be considered as allosteric effectors. In such cases it is hard to rationalize any regulatory model which does not necessitate separate sites with interactions through conformational alteration. CoOperativityjParameters It may be beneficial to mention several parameters that have been introduced to quantitatively express the degree of cooperativity for a ligand with a regulatory enzyme. One of these is the coefficient "n" of the Hill equation: log (V/V-V) = n.log (X) - log K where v is the velocity observed at any ligand concentra- tion X. V is the enzyme's maximal velocity, and K, the Michaelis constant for a substrate ligand, and a binding constant for an effector. The Hill equation was originally derived to account for the binding of oxygen to hemoglobin (12, 13). Values of "n" of greater than one are considered indicative of cOOperative binding behavior, and values below or. lizard b: AI be a 11m curve is value of function "15.6? CO v/ol A no». System 8 below one, indicative of negative cooperativity (where each ligand bound decreases the affinity for subsequent ligands). Although the Hill equation is normally considered to be a linear function, Hyman (20) has pointed out that this curve is in fact more complex, having a slope of unity at both high and low substrate concentration and exhibiting a maximum slope at its midpoint. As Atkinson noted (2), the value of the Hill coefficient "n" at the midpoint is a function of both the number of binding sites for the ligand under consideration, as well as, the degree of interaction between these sites. ‘Values of "n" are not necessarily expected to be integers and they do not in themselves indi- cate the number of such sites: however, they do yield a minimum value for the number of sites. For a fractional value of "n", the minimum number of interacting sites would be the next highest integer. In the absence of any coop- erativity the Hill coefficient would be unity. Koshland, Nemethy and Filmer (21) introduced the parameter RE which is the ratio of enzyme activity observed at 90% ligand saturation to the activity at 10% saturation. A non-coOperative system would have an RE value of 81, cooperative systems less than 81, and negatively cooperative systems greater than 81. K and V System Enzymes Monod. Hyman and Changeux (22, 23) noted that kinetic alteration of many allosteric enzymes is accomplished by exclusive tially ur. falls exo Herod SE enzymes 8 Concerted Ch 9 exclusive modification of the Km. leaving the Vmax essen- tially unaltered. In other regulatory enzymes modification falls exclusively on Vmaxo with Km remaining unaltered. Monod‘gt'gl.. termed the former class of enzymes K system enzymes and the latter group V system enzymes. Concerted Transition Model It is apparent that any model prOposed to explain allosteric enzyme behavior must account for the uniQue properties of many regulatory enzymes, such as, cooperative binding behavior and desensitization. The first model to gain wide recognition was introduced by Monod, Hyman and Changeux (23). The essentials of the model may best be described by first considering an.exp1anation for homotropic effects, where the binding of a particular ligand affects the action of the enzyme on another molecule of the same ligand. The Monod, Hyman and Changeux model requires that all allosteric enzymes be oligomers, made up of identical subunits which have been termed protomers. For the sake of simplicity an enzyme composed of two identical subunits, 1.9.. a dimer, shall be considered here. Monod.gt‘gl.. postulate that the free enzyme pre-exists in a number of equilibrating conformations at a minimum two. This equi- librium is illustrated in simplified form: 0., as 1 A 903:3 rrfi as 0“ 10 a T i Lo r- ..13 The ligand binding sites on the two conformations R and T have different prOperties, either in ligand affinity, or, as is possible in the case of substrate, in catalytic efficiency. Normally, R represents the more active con- former. V The Monod, Hyman and Changeux model assumes that the symmetry of the protein oligomer is retained at all times. The subunits of any one oligomer must then be entirely in either the R or the T conformation. The properties of all the binding sites change together and in the same way when going from the R to the T form. For this reason the model is often referred to as the concerted transition model. COOperativity is then affected as follows: the ligand in question binds more effectively to the R con- former and as the concentration of ligand molecules increases the equilibrium ratio of the conformers should shift toward the R form. This provides a greater fraction of oligomers present in the form with higher affinity. Binding of the subsequent ligand molecules would be enhanced compared to the binding of the first ligand molecules. which 8313 "‘5' 11 This simplified example can be expanded to cases in which each protomer contains sites for different ligands as shown below: Catalytic Site 2 K W / Effector Site Where the uppermost sites bind some ligand other than sub- strate (effector), and where the lower sites are the cata- lytic sites. Each protomer contains one and only one site for each type of ligand. If an effector molecule has greater affinity for, say the R conformer, the equilibrium ratio of R to T will shift as before to favor the R form resulting in a cooperative homotrOpic effect with reSpect to modifier. However, as shown in the figure, this also results in shifting the fraction of catalytic sites to the R conformer. If the R conformer is more active than the T conformer the net result of the effector binding will be to bring about a heterotrOpic activation. In a similar manner, an effector molecule which exhibits higher affinity for the T form will cause a heterotropic inhibition. This description of the concerted transition model immediately leads to a number of predictions concerning 12 allosteric enzyme behavior: (1) All heterotropic effector molecules are expected to exhibit homotropic cooperative binding effects. (2) In K systems, that is, where the R and T conformers each possess different affinities for the substrate molecule, the substrate is eXpected to exhibit homotropic effects. However, in cases where the R and T conformers each possess identical intrinsic affinities for the substrate, i.e., in V systems, no substrate homotrOpic effects should be observed. (3) Since the effect of posi- tive modifiers is to increase the fraction of enzyme in the more active conformation, such modifiers must reduce the degree of coOperativity observed in substrate binding. This should be discerned as a decrease in sigmoidicity, a decrease in the coefficient "n" of the Hill equation, and an increase in RB. On a similar basis, negative modifiers should increase the degree of cooperativity observed in substrate binding of a K system enzyme. (h) For a K sys- tem enzyme, as the substrate concentration is increased, the fraction of enzyme present in the R conformation (the "state function") should increase more rapidly than the fraction of catalytic sites bound with substrate (the "binding function"). Thus, if both the "state function" and the "binding function" were plotted versus substrate concentration on the same set of coordinates, the "state function" would be expected to proceed the "binding func- tion". (5) Negative coOperativity may not be accounted these 152‘ or t 115 t. n (‘1‘ One Dev-1.- ~ D ccount ’ .it rode Presente 13 for via a concerted transition mechanism. The demonstra— tion that not all allosteric enzymes behave according to these predictions has argued against unconditional support of this model. The Sequential Induced Fit Model The most prominent alternative model proposed to account for allosteric enzyme behavior has been the induced fit model. Detailed description of this type of model was presented by Koshland, Nemethy and Filmer (21). The sequential induced fit model differs fundamen- tally from the concerted transition model in two reSpects. First, no postulation is made for a pro-existing equilibrium of several enzyme conformational forms, each with differing binding and catalytic properties. Rather, the binding of a ligand is assumed to gauge a conformational alteration in the subunit to which it is bound. Second, once this principle is stated no further restrictions need be placed on the conformation of the entire protein. Indeed, any number of alternative events may follow. In one extreme case the ligand bound subunit may have no effect at all on the conformation of the other subunits. This would result in non-cOOperative ligand binding. In another extreme the altered subunit could cause conformational alterations in all the other subunits, resulting in maximum cooperativity. This would, in essence, amount to a concerted transition. In a i induce exhibi azount 14 In a third alternative case the ligand bound subunit could induce the remaining subunits into a conformation which exhibits decreased affinity for ligand binding. This would amount to negative coOperativity. The lack of conformational and symmetry restrictions should again be emphasized: any number of cases between these extremes are permitted. The ligand bound subunit may affect only some of the other subunits and need not affect them to the same degree. It is clear that the validity of any model can only be assessed by a comparison of the predictions of that model to the actual observations obtained in purified systems. The remaining pages of this section are devoted to discus- sion of some of the more carefully characterized allosteric enzymes. The vast numbers of allosteric enzymes studied make any attempt at an exhaustive review unpractical. It is important to emphasize that examination of the litera- ture reveals that the enzymes discussed here typify many other regulatory systems. A description of isocitrate dehydrogenase and phos— phofructokinase from yeast, E. 22;; aSpartic transcarb- amylase. glyceraldehyde-3-phosphate dehydrogenase from rabbit muscle and yeast, and two biodegradative threonine dehydrases follows. and Szit cofactor A}? to t be high: to enzy: cules th molecule this def 15 Yeast Isocitrate Dehydrogenase The kinetic properties of the AMP-activated NAD+- isocitrate dehydrogenase of yeast were investigated in detail by Hathaway and Atkinson (20) and Atkinson, Hathaway and Smith (2h). Binding of the substrate isocitrate, the cofactors NAD+ and Mg¢*, as well as, the positive effector AMP to this dehydrogenase was noted by Atkinson g£_§;.. to be highly cOOperative. They defined "order", with regard to enzyme behavior as the "minimal number of ligand mole- cules that must be assumed to interact with each enzyme molecule to explain observed kinetics." On the basis of this definition DPN+-isocitrate dehydrogenase was estab- lished as exhibiting an essential fourth order dependency with respect to isocitrate, and second order dependencies with respect to DPN‘, mg++. and AMP. It is particularly important to note that the binding of the activator, AMP, does not cause a decrease in the Hill coefficient "n" dis- played for substrate binding: that is, AMP does not evoke any change in isocitrate cooperativity. Instead, the effect of AMP is to shift the Hill curve into a range cor- responding to lower substrate concentrations on the abscissa. It would appear that the effector molecules in the case of yeast DPN+-isocitrate dehydrogenase act uniquely by altering the Km for substrate and do not effect the strength or nature of the homotrOphic interactions. This finding is clearly inconsistent with the predictions of the cones Ar Hathaway the yeast assrres t site bird postulate Catal. t1!- \- 16 the concerted transition model. An induced fit type of model was proposed by Atkinson, Hathaway and Smith to account for the observed behavior of the yeast DPN*-isocitrate dehydrogenase. Their model assumes two catalytic sites for the dehydrogenase, each site binding one DPN+ and one Mg++ molecule. The enzyme is postulated to possess two AMP effector sites and two non- catalytic isocitrate effector sites. A The binding of isocitrate at any site (catalytic or effector) is assumed to increase the association constant for isocitrate at the other sites by a factor of at least 20. Binding of DPN+ increases the association constant for isocitrate at that site only by a factor of 20 and binding of AMP at an effector site increases the association con- stant for all four isocitrate sites by a factor of 5. This model is in sufficient agreement with the observed experi- mental data. Phosphofructokinase of Yeast In examining the yeast phosphofructokinase, Atkinson, Hathaway and Smith (25) noted that this enzyme, like iso- citrate dehydrogenase, shows no decrease in substrate cooperativity upon activation by its heterotropic effector, AMP. Through reevaluation of earlier published studies the above authors revealed that effector molecule binding does not cause alteration in substrate binding cOOperativity for 17 a number of other phosphofructokinases, notably those from the liver fluke and from rabbit muscle. Aspartic Transcarbamylase Aspartic transcarbamylase 0f.§- ggli'has been, over the years, one of the most extensively characterized allo- steric enzymes. Early studies on this enzyme proved use- ful in formulating the concepts of allosteric control. The native aspartic transcarbamylase oligomer is a large protein sedimenting at 11.88, with a molecular weight of 310,000 (16). Treatment of the enzyme with the mercurial pHMB causes a dissociation of the native enzyme into cata- lytic and regulatory subunits with molecular weights of 100,000 and 30,000 respectively. The purified catalytic subunit has a higher specific activity than the native enzyme, shows non-cooperative kinetics, and is insensitive to either inhibition.by CTP or activation by ATP. The regulatory subunit lacks enzymatic activity. Recombination of the regulatory subunit with the catalytic subunit restores sensitivity to the effectors CTP and ATP. From equilibrium dialysis studies with the substrate analogue succinate, Gerhart, Schachman and Changeux (26, 27) suggested that native aspartic transcarbamylase possesses four catalytic sites, two on each catalytic subunit prepared through pHMB treatment. On this basis, Gerhart, Schachman and Changeux (26, 27) argued that the isolated catalytic 18 subunit should be composed to two more identical subunits; the native aspartic transcarbamylase should then be composed of four catalytic and four regulatory subunits (i.e., four protomers). Weber (28) and Wiley and Lipscomb (29) have challenged the validity of this molecular description. Molecular weight determinations from sodium dodecyl sulfate polyacrylamide gel electrophoresis and amino acid sequence analysis has led Weber (28) to assign a molecular weight of 17,000 for the regulatory subunit, and a molecular weight of 33,000 for the catalytic subunit. These findings indi- cate that the 30,000 molecular weight regulatory subunit isolated by pHMB treatment is actually composed of two identical polypeptide chains. The isolated 100,000 molecu- lar catalytic subunit should be composed of three, rather than two subunits, and the native oligomer 6 regulatory and 6 catalytic subunits (i.e., 6 protomers). In support of this model, x-ray crystallographic analysis by Wiley and Lipscomb (29) has shown that separtic transcarbamylase has a 3 fold axis of symmetry and contains at a minimum 6 of each polypeptide chain. Alternate conformations of a particular enzyme might be expected to have different chemical and physical proper- ties. Thus, it should be possible to determine through physical chemical procedures the fraction of enzyme present in a particular conformation. In the case of aspartic transcarbamylase, conformational alteration is accompanied by chart; to p33 the enz; the enzy changes reactior aentatic (30) we: carbazy] succina‘. with th l9 . by change in the reactivity of the enzyme sulfhydryl groups to pHMB (27-31), changes in the sedimentation velocity of the enzyme (27, 30), and change in the suseptibility of the enzyme to tryptic digestion (31. 32). Following the changes in the pseudo-first order rate constants for the reaction of pHMB with native enzyme and change in the sedi- mentation velocity of the enzyme. Gerhart and Schachman (30) were able to determine the fraction of aspartic trans- carbamylase present in the "R" conformation at varying succinate concentration. By comparing such "state" data with the binding data obtained from equilibrium dialysis studies with succinate, Changeux and Rubin (33) noted that at any particular succinate concentration the fraction of total enzyme in the R conformation was greater than the fraction of total substrate sites occupied. As previously stated, this would be required if a concerted transition model is valid for a particular allosteric model. McClintock and Markus (32), utilizing tryptic digestion and pHMB reaction rates as a means of assessing the fraction of enzyme in the "R" state, confirmed Changeux and Bubin's findings with the substrate analogue succinate. From pHMB and tryptic digestion experiments in the absence of the co-substrate carbamyl phosphate McClintock and Markus were able to obtain "state" data with the natural substrate aspartate. The state function had the identical shape as the binding function obtained from activity transca and Kos by Ge rh transca the enz CTP. x carted cm 1 A v r C; lith S. 20 measurements. Therefore, for the natural substrate, the fraction of catalytic sites filled, being identical with the fraction of enzyme in the "R" conformation, is clearly inconsistent with the predictions of the concerted transi- tion model, but not with the induced fit model. Support for an induced fit interpretation of aSpartic transcarbamylase has recently been presented by Levitzki and Koshland (3“). In re-evaluating kinetic data obtained by Gerhart and Pardee (10) on CTP inhibition of aspartic transcarbamylase, Levitzki and Koshland demonstrated that the enzyme displays negative cooperative behavior towards CTP. Negative cOOperativity is inconsistent with the con- certed transition model. Negative cooperativity may explain some of the unusual behavior of this enzyme, i.e., partial competitive inhibition kinetics (10, 11), and the detection of only four CTP binding sites by equilibrium dialysis in an enzyme with six identical regulatory subunits (26, 27). In light of these latter findings it appears that a concerted transition interpretation is no longer adequate to account for the behavior of aspartic transcarbamylase and an induced fit interpretation seems more appropriate. Glyceraldehyde-34Phosphate Dehydrogenase of Rabbit Muscle Koshland, Conway and Kirtley (35. 36) have investi- gated the allosteric behavior of NAD+ with rabbit muscle glycere ase bir decree: (negati that pi then t1 21 glyceraldehyde-3-phosphate dehydrogenase. This dehydrogen- ase binds 4 molecules of NAD+, the affinity of each NAD+ decreasing as the number of NAD+ molecules bound increases (negative cOOperativity). If a symmetry model similar to that prOposed by Monod, Wyman and Changeux (23) is to hold, then the relative values for three of the four'NAD+ bind- ing constants must for statistical reasons be related to each other by the following ratios: K3/K2 = 4/9, Ku/K3 = 3/8. The experimentally observed ratios of K3/K2 and K4/K3 were much larger than the values predicted by the concerted transition model. Koshland, Cbnway and Kirtley suggest that rabbit muscle glyceraldehyde-3-phos— phate dehydrogenase is best fit by a sequential induced fit model. Yeast Glyceraldehyde-3-Phosphate Dehydrogenase The allosteric behavior of NAD+ binding with the tetrameric glyceraldehyde-3—phosphate dehydrogenase of yeast was investigated by Kirschner, Eigen. Bittman and Voight (3?, 38). This enzyme which exhibits hyperbolic binding kinetics for NAD+ at 20°, diSplays sigmoidal coOperative behavior at 40°. Employing temperature Jump and stop-flow techniques the binding of NAD+ to glycer- aldehyde-B-phOSphate dehydrogenase was examined under con- ditions in which cooperative behavior is normally apparent. Three distinct relaxation processes were detected, the yeast 22 slowest of these apparently correSponding to a unimolecular isomerization reaction, i.e., enzyme isomerization without interaction with DPN+, the two faster reactions being bimolecular, presumably DPN+-enzyme interactions. The detection of two such bimolecular relaxation processes implies that yeast glyceraldehyde-3-ph08phate dehydrogenase should possess two different types of NAD+ binding sites. Since induced fit models do not explicity account for independent unimolecular steps, Kirschner (38) argued that the observations made with the yeast dehydrogenase were inconsistent with such models. Therefore, a concerted transition interpretation was favored for the dehydrogenase. Kirschner proposed that the two different types of NAD+ binding sites are located on two different conformers of the enzyme (R and T). The relative slowness of the isomerization reaction permitted investigation of the properties of the two forms of the enzyme. Among the observations, the difference Spectra of the T form was found to be distinct from that of the R form. Rapid spectrophotometric titration indicated that the T form, like the R form. possesses four identical NAD+ binding sites. The r form was found to possess vir- tually no catalytic activity. While concerted transition interpretations may still have merit for a number of regulatory enzymes, such as for yeast glyceraldehyde-3-phosphate dehydrogenase, this model CETJIO‘ isoci' trans: ase ac 23 cannot account for the varied behavior of such enzymes as isocitrate dehydrogenase, phosphofructokinase, aspartic transcarbamylase, rabbit muscle glyceraldehyde-3-phosphate dehydrogenase (and the biosynthetic and biodegradative threonine dehydrases). The much less restricted induced fit model may satisfactorily account for the behavior of these later enzymes, and may be a closer representation of the actual control mechanism for these systems. Threonine Dehydrases The large number of allosteric threonine dehydrases of microorganisms have been recently reviewed by Wood (39). In addition, Umbarger (40) has discussed recent findings on the biosynthetic threonine dehydrases of E. 321$ and S. typhimurium. For these reasons, no attempt shall be made to give an exhaustive review on the many threonine dehydrases. The discussion here shall be limited instead to a description of studies on the biodegradative threonine dehydrases of‘g.‘ggli and Q. tetanomorphum which are con- sidered pertinent to the current investigations. Biodegradative Threonine Dehydrase of Escherichia coli Anaerobic serine deaminase activity in resting cell suspensions of Escherichia coli was first observed by Gale and Stephenson (41). These authors found that this deamin- ase activity could be protected against inactivation by reducir as, by stories the dea early c Cepurif from or or blot TGQuire 15% act unusual in SEVe earlier L‘sel‘ir fmmd. One of by lac b1°83h. in 18g ingens 13m 18 24 reducing agents, such as glutathione and formats, as well as, by adenylic acid. Later studies in a number of labor- atories (42-44) implicated a variety of other cofactors in the deaminase reaction, notably zinc and biotin. These early observations were substantially clarified by Wood and Gunsalus (45), who carried out the first partial pur- ification of the serine and threonine deaminating activity from anaerobicly grown E.‘ggli. Both deaminating activities copurified which suggested that the two activities result from one enzyme. There was no requirement for either zinc or biotin; however, both AMP and reduced glutathione were required for activity. Wood and Gunsalus (45) observed that serine catalysis caused inactivation of the deaminat- ing activities toward both serine and threonine. This unusual type of substrate inactivation has since been noted in several other threonine dehydrases (46, 47). Umbarger and Brown (47) in attempting to clarify earlier conflicting reports on the presence and levels of L-serine and L-threonine dehydrases in‘E. 2211 (48-50), found that this organism elicits two L-threonine dehydrases. One of these enzymes is specifically inhibited and repressed by isoleucine. This constitutive enzyme has been termed the biosynthetic threonine dehydrase because of its involvement in isoleucine biosynthesis. The second dehydrase which is insensitive to isoleucine, is detected only when the organ- ism is grown in complex media lacking glucose and under anaerc enzy: Gunsal dehyd: 25 anaerobic conditions. This inducible dehydrase is the enzyme which had been demonstrated earlier by Wood and Gunsalus. Umbarger and Brown (47) suggested that this dehydrase probably had a catabolic function in the cell, and as a result it has since been referred to as the bio- degradative threonine dehydrase. The biosynthetic and biodegradative threonine dehydrases both catalyze serine deamination and are inactivated by such catalysis (47). Dehydration Mechanism - In their original studies on anaerobic serine deaminase activity Gale and Stephenson (41) had assayed serine deamination from the ammonia released upon catalysis. Chargaff and Sprinson (51, 52) later demonstrated that the products of both serine and threonine deamination were ammonia and either pyruvate or d-ketobutyrate, respectively. These authors also noted that substitution of the hydroxylic hydrogen by various alkyl groups, or by phoSphate or phoSphatidic acid pre- vented deamination. From this evidence, Chargaff and Sprinson suggested that the mechanism for both threonine and serine deamination involved desaturation to an d—B unsaturated amino acid, with subsequent rearrangement of this Species to an imino acid followed by spontaneous hydrolysis to yield ammonia and the keto acid. A more detailed prOposal suggested by Metzler, Ikawa and Snell (53) and independently by Braunstein and Shemyak1n_(54), incorporated pyridoxal phoSphate into the catalytic mechanism. This mechanism involved pyridoxal 26 phOSphate assisted d-proton elimination and subsequent B-hydroxyl group diSplacement. Phillips and Wood (55) obtained isotopic exchange data which was consistent with the latter scheme. The mechanism, presented here in Figure 1, is described by the following steps: First, (a) the substrate is presumed to occupy the catalytic binding site forming a non-covalent complex with the enzyme. Transaldimination (b) with the pyridoxal phOSphate of the enzyme then forms the threonine- pyridoxal phoSphate azomethine. Dehydration (c) of the substrate next ensues. Figure 1 shows this occurring in two steps, pyridoxal phOSphate assisted d-proton elimination and subsequent B-hydroxyl group diSplacement. The intermedi- ate at this point, the azomethine of aminocrotonate with pyridoxal phosphate undergoes a second transaldimination (d) to yield the non-covalent complex of aminocrotonate with the enzyme. .Aminocrotonate tautomerizes (e) to imino- butyrate, which is subsequently hydrolysed to d-ketobutyrate and ammonia. While the exact position of product dissocia- tion in the mechanism is uncertain, some evidence has appeared to suggest that the tautomerization of aminocroton- ate occurs on the enzyme surface, and hence may be enzyme catalyzed. First, Krongelb, Smith and Abeles (56) examined the stereochemistry of the tautomerization reaction from aminocrotonate to iminobutyrate with the enzyme homoserine dehydrase of liver. This enzyme catalyzes a somewhat 27 0:9 co: amass: no musm apes B 0 on osaaoonn no»: soaps aoahv 0: loo .Zom H shaman 28 A) (359:3 ( J+I 228:1 _ 29480»: @— e... -ooo-o-~zo..xo flighgxscxu‘ 3.525 F. 50.525.335.12. 3. greatest «:2 : .80-?o-£o pox 29 analogous reaction to that of threonine dehydrase, c-y rather than d-B elimination: ® 0 o HoCHZ-caz-ca-cf' --+> CHB'CH=f'C: i>~+> I O' 0' NBA? NH2 0 H20 0 I / CH -CH -c-c’ fi CH -CHp_-c-c’ 3 2 H ‘o- 3 H ‘0‘ NH NH3 0 When the dehydration of homoserine was carried out in D20, the d-ketobutyrate had incorporated D into the B-carbon asymmetrically. If dissociation of the amino- crotonate-enzyme complex had proceeded the tautomerization reaction, the a-ketobutyrate should not have been asymmet- rically labelled. Therefore, the results indicated that tautomerization did indeed proceed on the enZyme surface. Evidence reported by Flavin and Slaughter (57) indicates that a similar situation may exist in the case of the biodegradative threonine dehydrase. They found that N-ethylmaleimide reacts with a catalytic intermediary of threonine dehydrase, considered to be aminocrotonate, form- ing an asymmetrically labelled adduct. The asymmetric labelling of the adduct suggests that aminocrotonate does not freely dissociate from the enzyme. Thus, tautomeriza- tion in the case of threonine dehydrase may occur on the enzyme surface. ‘Phillips and wood (55) have demonstrated that 180 Tequii dehyd» 30 from H2180, and D from D20 are both readily back incorpor- ated into L—threonine during catalytic dehydration. The large extent of this incorporation (25% incorporation of D and 180 into L-threonine after 50% conversion of substrate to d-ketobutyrate) suggests that the intermediary steps of catalysis through B-elimination must be freely reversible, and that the rate limiting step of the dehydration mechan- ism should lie subsequent to B-hydroxyl group elimination, i.e., at the second transaldimination or some step associ- ated with product dissociation. Pyridoxal.Pho8phate - Umbarger and Brown (47) obtained the first confirmatory evidence that the biodegradative threo- nine dehydrase contained the cofactor pyridoxal phosphate. Through hydroxylamine treatment they ressolved the dehydrase of pyridoxal phoSphate and obtained a partial reactivation by re-addition of the cofactor. A pyridoxal phOSphate requirement had been earlier discerned for the L-threonine dehydrase of Neurospora by Reissig (58). Phillips and Wood (55) noted that partially purified threonine dehydrase exhibits an absorption maximum in the range considered characteristic for protonated Schiff bases of pyridoxal phosphate with amino acids. This maximum, at 402 mm, was substantially diminished upon reduction of the dehydrase with borohydride. Subsequent hydrolysis of the reduced enzyme yielded the fragment, Né-pyridoxyllysine, indicating that the e-amino group of a lysine residue formed 31 the azomethine linkage with the aldehydic group of pyridoxal phOSphate. Nakazawa 333 21. . (59) found that the biodegradative threonine dehydrase exhibits positive circular dichroism and a positive Cotton effect in the OED of the enzyme, both at 415 mu. The 415 mu circular dichroism was lost upon reduction with borohydride. The authors assigned the 415 mu dichroism to the same species absorbing at 402 mu, i.e., the azomethine of pyridoxal phOSphate and the a-amino group of a lysine residue. Since pyridoxal phosphate itself has no asymmetric center the authors suggested that the asymmetry necessary for dichroism arises from the interaction of the aldimine with the rest of the enzyme. Nakazawa gtmgl.. (59) noted that the 415 mu circular dichroism is totally lost upon addition of substrate to the enzyme, and returns as substrate is depleted. They suggested that this loss in dichroism results from the formation of a new optically inactive Schiff base between pyridoxal phos- phate and threonine. Tokushige 32 gl.. (60) observed that substrate addition induces a red shift of the spectral maximum to near 450 mu, with the maximum gradually returning to the low 400 mu range as substrate becomes exhausted. The authors suggested that this new absorption maximum may be ascribed to an enzyme-substrate complex. Involvement of AMP - Phillips and Wood (55, 61) examined the involvement of AMP in the activation of the dehydrase. enzyme struct veloci press: or suc accoun the pr 520 vs Versio 932210;; S‘JCI‘QS 32 Unable to assign a function to this nucleotide in the cata- lytic dehydration mechanism they concluded that AMP serves as a true allosteric activator. In sucrose gradient studies with partially purified enzyme, Phillips and Wood (61) found that AMP causes gross structural changes in the dehydrase. The sedimentation velocities obtained for the enzyme in the absence and presence of AMP were 4.88 and 7.68, respectively. Changes of such magnitude were considered to be too great to be accounted for solely by conformational alteration within the protein molecule. As a result they prOposed that the 820 value changes represented protomer-oligomer intercon- versions. When intermediary concentrations of AMP were employed single dehydrase peaks were detected in the sucrose gradients with 820 values between the limits of 4.88 and 7.68. As Gilbert has pointed out (62), such behavior would be expected for a system in rapid equilib- rium between two forms, with the position of the enzyme in the gradient dependent on the statistical distribution of the enzyme species among the two forms. Recognition that AMP causes quaternary structural changes as well as activation, led Phillips and Wood to speculate that these two processes may be causally related. Biodegradative Threonine Dehydrase of Clostridium tetanomorphum The biodegradative threonine dehydrase of Clostridium 33 tetanomorphum is similar in many respects to the catabolic enzyme from Escherichia coli. It is activated by an adenine nucleotide, ADP rather than.AMP, exhibits a molec- ular weight of approximately 160,000 (63), similar to molecular weight values reported for the E, 32;; enzyme (64, 65), and is comprised of dissociable subunits which exhibit less catalytic activity than the parent enzyme (63). The demonstration that ADP was not altered during the dehydrase reaction and that inorganic 32P was not incorporated into ADP (66) suggested an allosteric role for the nucleotide activator. ADP Activation - Three independent investigations on the kinetics of the dehydrase from g. tetanomorphum yielded results which are widely varient from each other. It appears that the employment of different assay conditions account for the disparity of the results of these studies. Nakazawa and Hayaishi (67) used Tris-RC1 buffer at pH 8.4 in their studies while Whiteley and Tahara (68) utilized Tris-H01 buffer at pH 9.6, and Vanquickenborne and Phillips (69), phosphate buffer at pH 8.0. Only the latter investigators employed crystalline enzyme in their studies. On the basis of the more detailed investigation conducted by Vanquickenborne and Phillips the kinetic analysis pre- sented by these authors would appear to be the most reli- able. Vanguickenborne and Phillips obtained Km values of 54 v. in the well w and 3. obtair observ centre of ADP press: hYper‘: Prese: 8.6 b1 l/L-tr ever, 34 54 mM for L—threonine in the absence of ADP and 3.5 mM in the presence of ADP. These values compared reasonably well with those reported by Nakazawa and Hayaishi of 37 mM and 3.6 mM, respectively. Much different values were obtained by Whiteley and Tahara. Nakazawa and Hayaishi, and Whiteley and Tahara observed sigmoidal kinetics in plots of L-threonine con- centration versus enzyme velocity obtained in the absence of ADP: normal hyperbolic plots were obtained in the presence of ADP. While Vanquickenborne and Phillips found hyperbolic plots of substrate versus activity in both the presence and absence of ADP, as the pH was increased above 8.6 bimodal Lineweaver-Burke plots of 1/velocity versus 1/L-threonine were detected, in the absence of ADP. How- ever, if the reaction cuvettes were allowed to preincubate and the assay rate not measured until 30 minutes after sub- strate had been added, the double reciprocal plots were again nearly linear. These observations suggested that a time-dependent phenomenon involving substrate in the absence of ADP was taking place. Vanquickenborne and Phillips examined the effect of pH and various buffers, as well as, ADP and substrate on the enzyme in sucrose gradients to determine if any gross conformational alterations of the dehydrase could be detected. In the absence of ADP, the native enzyme with an s20 value of 7.68 dissociated to a form of around 4.58 under the conditions of either high pH An 1 88;; bloc Ste :5 te e d: a k1): 35 (above pH 8.6) in phosphate buffer, or at any pH in Tirs- HCl or carbonate buffer. ADP or substrate could reverse the dissociation. The above results suggest that the sigmoidal kinetic behavior observed by Nakazawa and Hayaishi (67) and by Whiteley and Tahara (68) in the absence of ADP was due to enzyme dissociation and slow substrate induced reassociation, rather than from a true cOOperativity mechanism. The biodegradative threonine dehydrase of g. 3322- nomorphum apparently exhibits neither substrate nor acti- vator homotropic effects. In regard to heterotropic behavior, Vanquickenborne and Phillips (69) noted that the interactions between substrate and effector in the dehydrase are mutual. While ADP elicits a 15 fold stimu- lation in Km for L-threonine, L-threonine similarly elicits a change in Ka from 160 uM at 0.2 mM L-threonine, to 3.3 uM at 10 mM L-threonine. Vanquickenborne, Vidra and Phillips (70) undertook an investigation probing the chemical nature of the separate catalytic and regulatory binding sites of the biodegradative threonine of g. tetanomorphum. Employing such amino acid reactive reagents as N-ethylmaleimide, tetranitromethane, iodine, and fluorodinitrobenzene, and diazobenzene sulfonic acid chemical modification of certain kinetic parameter were attained. N-Ethylmaleimide at pH 6.5 36 and tetranitromethane at pH 6.3 brought about an altera- tion in the catalytic ability of the enzyme, as evidenced by a large (10 fold) decrease in the enzyme's Vmax' Treat- ment under such conditions did not affect the Km of the enzyme toward substrate. Inclusion of the substrate analogue, allothreonine, in the reaction mixture afforded protection against these changes in Vmax- Iodination at pH 6.3 resulted in an alteration of the dehydrase's ability to bind the effector ADP, and thereby to cause activation. This was attested by an extremely large increase in the K3 for ADP, and a concomitant four fold increase in Km. Iodination did not cause significant alteration in VmaX' ADP afforded protection from iodination. A number of other reagents, tetranitromethane at pH 8.3, fluorodinitro- benzene at pH 7.6, and diazobenzene sulfonic acid at pH 7.6, brought about modification in both the Km and‘Vmax of the enzyme. In most of these cases, allothreonine protected against Vmax modification and ADP against Km alteration. Vanquickenborne, Vidra and Phillips discussed these observations in a manner so as to implicate reactive groups on the enzyme which may be associated with either catalytic activity or activator binding. Tetranitromethane, when employed at pH 6.3. and N-ethylmaleimide are considered to react somewhat Specifically with the sulfhydryl group of cysteine. The observation that these two reagents modify vmax but not Km suggests that a sulfhydryl residue may be 37 present in the vicinity of the catalytic site. The relative non-Specificity of the other reagents employed in the inves- tigation makes the assignment of other reactive residues difficult. Further investigation will be required before other definitive assignments of catalytic and activator site residues can be made. (G835 METHODS AND MATERIALS Bacteriological IpThreonine dehydrase was prepared from an iso- leucine-requiring mutant isolated from Escherichia coli (ATCC 8739) and generously donated by Dr. A. T. Phillips, Pennsylvania State University Park, Pennsylvania. For the isolation of threonine dehydrase, the organism was grown at37° for 24 hours without aeration or agitation in a 120 liter New Brunswick Scientific Company fermentor, model CF-130. The growth medium of Wood and Gunsalus (45) as modified by Niederman gtwgl. (71) con- sisted of 2% Sheffield N-Zamine MAX, 1% Yeast Extract (General Biochemicals), and 0.5% dibasic potassium phos- phate. A 10-liter culture grown in this medium for 20 hours was used as inoculum. The organism was maintained in slab cultures of this same medium containing 1%% agar and trans- ferred periodically to assure viability. Chemical AMP Analogues A number of.AMP analogues were prepared from the correSponding nucleosides by enzymatic tranSphoSphorylation. The procedure employed was that of Rottman, Ibershof and 38 39 Guarino (72), and involved incubation of the nucleoside with UMP and an enzyme preparation from wheat seedlings. The analogues prepared were N6-ethyl-AMP, N6-diethyl-AMP, and a mixture of 1'-hydroxymethyl~AMP and 1'-phOSphohydroxy- methyladenosine (psicofuranine 1'- and 6'—phOSphateS, reSpectively). Nl-methyl-AMP, and N6-methyl-AMP were synthesized from AMP according to the procedure of Griffin and Reese (73). Details of the above mentioned nucleotide syntheses have already been reported (74). 7-DeazanAMP (tubercidin-5'-ph03phate), 1'-hydroxy- methyladenosine (psicofuranine), and cytidine arabinoside- 5'-phOSphate were donated by Dr. William J. Wechter of the Upjohn Company, Kalamazoo, Michigan; 7-deaza-7-cyano-AMP (toyocamycin-5'-phOSphate), 7-deaza-7-carboxamido-AMP (sangivamycin-5'-phOSphate), and N6-ethyl and N6-diethyl- adenosines were kind gifts of Dr. R. J. Suhadolnik. Albert Einstein Medical Center, Philadelphia, Pennsylvania; 3- iso-AMP was donated by Dr. Nelson Leonard, University of Illinois, Urbana, Illinois;.ApA, CpA, 2'-O-methylaAMP and 3-deoxquMP (cordycepin-5'-phOSphate) were gifts of Dr. Fritz Rottman, Michigan State University, East Lansing, Michigan. The 8-aza-9-deazanMP (formycin-5'-phoSphate) was a gift of Dr. S. Nishimura of the National Cancer Center Research Institute, Tokyo, Japan. azino prooe aoety of L- Corps allot (Hutr 10 ml 51 of ”lth PIVe And Afte- 40 Amino.Acids The L- and D-enantiomers of DL-allothreonine were ressolved through acetylation of DL-allothreonine and Specific enzymatic deacylation of the L-enantiomer. The amino nitrogen of DL-allothreonine was acetylated by a procedure Similar to that described by Nirata (75) for acetylation of DL-serine. The procedure for deacylation of L-allothreonine with hog renal acylase I (Sigma Chemical Company) is similar to that described for chloroacetyl-L- allothreonine (76, 77). Two grams of DL-allothreonine (Nutritional Biochemical Corporation) were dissolved in 10 m1 of 2 N NaOH, and 5.2 ml of acetic anhydride and 52 ml of 2 N NaOH were added in alternative aliquots of 0.2 ml and 2 ml each, while the reaction mixture was maintained at 0-5° with shaking. The mixture was allowed to incubate for two hours and then was passed over a column of Dowex 50W-X8 (ca. 1 liter of the acid form) and the fractions acid to congo red pooled. Under these conditions sodium ions and unreacted amino acid were retained on the resin. The effluent was concentrated to 60 ml, in 33222, at a temperature not exceeding 30°. The concentrated effluent was adjusted to pH 7.0 with 25% NHupfl, the volume adjusted to 75 ml with water. Five mg of hog renal acylase powder was added and dissolved, and the mixture incubated at 37° with occasional agitation. After 16 hours the precipitated protein was removed by WET! a dc a t? were l-oc r Av hot 41 centrifugation and an additional 2 mg of acylase powder was added. This procedure was repeated at 16 hour intervals, until the assay for L-allothreonine revealed that no additional acetyl-L-allothreonine had been deacylated. Details of the L-allothreonine assay are presented below. After completion of the enzymatic deacylation, 1 m1 of glacial acetic acid and 100 mg of acid-washed Norit were added, and the mixture filtered with suction through a double layer of No. 4 Whatman filter paper, covered with a thin coating of wet Norit. A few milliliters of water were used to wash the residue on the filter. One drop of 1-octanol was added to reduce foaming, and the combined filtrate and washing were then evaporated, AEHZEEESv at a temperature not exceeding 30°, to a volume of 6 ml. Upon addition of ethanol to 80% concentration L-allothreonine crystallized. After chilling for 6 hours at 4° the L-allo- threonine was recovered by vacuum filtration and washed successively with ethanol and ether. This preparation of L-allothreonine still contained some contaminating protein which was removed by the following procedure. The preparation was dissolved in a minimum volume of hot water and a small amount of Norit added. This mixture was brought to a boil for 1 minute, and then filtered. For recrystallization the clear filtrate was again brought to a boil. and boiling absolute ethanol added to 80%. The crystal suSpension was filtered with suction and washed 42 successively with ethanol and ether; 350 mg of L—allo- threonine were obtained. The earlier filtrate and washings, containing acetyl- D-allothreonine, was evaporated at room temperature by subjecting the solution to a stream of dry, purified nitro- gen directed at its surface until all the solvent was removed. The residue was dissolved in 10 ml of water and passed over a column of Dowex 50W-X8 (ca. 150 ml of the ‘ acid form) and the fractions acid to congo red paper pooled. A negative ninhydrin test assured the absence of free amino acid. The above eluate was evaporated, in 32322, at a tem- perature not exceeding 30° to a volume of 4 ml, then treated with 1 ml of concentrated HCl and refluxed for 2 hours. The solution was then evaporated, 22:22222v at 40° to dryness. Several more evaporations with water served to remove the excess acid. The residue was dissolved in 4 ml of water and redistilled analine was added to the disappearance of the Congo blue reaction. On addition of absolute ethanol to 80% concentration D-allothreonine crystallized. D-Allothreonine was recrystallized by the same pro- cedure employed above for L-allothreonine; 225 mg of D- allothreonine were obtained. A mixture of L- and L-allo-d-methylthreonine and D- and D-allo-a-methylthreonine were synthesized1 using a 1The author is indebted to Dr. R. H. Hammerstedt for undertaking the syntheses of the d-methylthreonines. 43 modified procedure of that described by Mix and Wilcke (78) involving condensation of acetaldehyde with the cOpper complexes of D- and L-alanine. The capper complex of D-alanine was prepared as follows. 10 grams (110 millimoles) of D-alanine (California Biochemical Corporation) was dissolved in 200 ml of water and 30 grams of basic CuCO3 added. This mixture was heated on a steam bath for several minutes, then cooled, filtered and concentrated to dryness. The alanine capper complex was dissolved in 100 ml of 11% NaOH and 62 ml of redistilled acetaldehyde was added drapwise over a period of 2.5 hours, in a dry ice bath with stirring. The solution was allowed to stir overnight at 4°. It was then acidified with acetic acid to pH 3.0-3.5 and concentrated to a syrup. The syrup was washed twice with water and dissolved in 200 ml of 1 N HCl; H28 was then bubbled through the solution for 45 minutes. At this point several milligrams of acid-washed charcoal were added and the precipitate removed by filtration on No. 50 Whatman filter paper and Celite. The filtrate was concentrated to a syrup, dissolved in 150 ml of water and again filtered. This solution was then passed over a column of Dowex-50W-X8 (ca. 250 ml of the acid form), the column washed with 2 liters of water and develOped with 1 N NHuOH. The tubes which were positive to ninhydrin were pooled, treated four times with a small amount of Norit and filtered. an The solvent was removed, iggzgggg, and the remaining syrup dissolved in 20 ml of water. The mixture of L- and L-allo- a-methylthreonine was crystallized by the addition of abso- lute ethanol to 80%. A mixture of D- and D-allo-d-methylthreonine was synthesized by the same procedure, except that L—alanine was employed in place of D-alanine. A sample of L-allothreonine was also a gift of Dr. A. Meister, Cornell University Medical College, New York, New York. I All other chemicals were obtained commercially. Determinations and.Procedures LmAllothreonine was determined by a procedure which involved dehydration of the amino acid to d-ketobutyrate with threonine dehydrase followed by Spectrophotometric measurement of the d-ketobutyrate produced. A 0.1 ml sample was added to a silica microcuvette containing 0.09 ml of 0.1 M potassium phOSphate buffer, and 0.01 ml of 0.1 M AMP. An initial absorbance reading was taken at 315 mu, and then 100 ug of biodegradative threonine dehydrase was added to the cuvette and the release of d-ketobutyrate followed until d-ketcbutyrate accumulation ceased. A final reading at 315 mu was taken and the millimoles of L-allothreonine dehydrated was calculated from the extinction coefficient ofCL-keto- butyrate (6315 = 20.8 M'l). 1+5 Inorganic phosphate determinations were made by the Fiske-SubbaRow method (79). Protein was routinely deter- mined by the method of Lowry.ggwgl.. (80). Density gradient centrifugations, performed essen- tially as described by Martin and.Ames (81), were carried out in either the SH-39 rotor (Spinco Division, Beckman Instrument Company), or the SB-283 rotor (International Equipment Company). Runs in the SW-39 rotor were made at #0 for 17 hours, while these in the SB-283 rotor were con- ducted at the same temperature but for a period of 31 hours. Linear sucrose gradients of 5 to 20% contained 0.1 M potas- sium phosphate buffer, pH 8.0, 1 mM dithiothreitol, and where apprOpriate, 5 mm AMP. Fructose diphosphate aldolase of rabbit muscle (s20 . 7.98 (82)), horseradish peroxidase (320 a 3.58 (83)), and beef liver catalase (320 a 11.4 (84)) were employed as markers in the gradients. Noll (85) has pointed out that sedimentation velocity determinations in 5-20$ sucrose gradients may not be valid for any rotor other than the sw-39. Therefore, it was necessary to standardize the gradients for runs in the 83-283 rotor. Three enzymes of known s20 value were centri- fuged together, and each one was used in calculations of s20 values for the other two. Extremely good agreement between known and calculated s20 value was obtained when the absolute difference between marker and unknown protein s20 value was less than u Svedberg units (Table 1). As TABLE 1 Standardization of 83-283 Rotor_£gr Sucrose Gradient Experiments The linear sucrose gradients (12 ml volume) contained 5 to 20% sucrose in 0.1 M potassium phOSphate buffer, pH 8.0. The sample applied to the gradient containing catalase, 5 ug, peroxidase, 5 ug, rabbit muscle aldolase, 10 mg, and cytochrome c, 1.0 mg was centrifuged at 39,000 rpm for 31 hours in the 8B-283 rotor. . ...___—' Marker Catalase Aldolase Peroxidase Enzyme 11.#B 7.98 3.53 Evaluated Catalase --- 11.38 10.98 Aldolase 7.98 --- 7.68 Peroxidase 3.758 3.728 --- Cytochrome c 3.268 3.248 3.138 I+7 long as this limitation is met sucrose gradient centrifuga- tions in 5 to 20% sucrose are valid in the 83—283 rotor. After centrifugation, the gradient tubes were punched and multiple drOp fractions collected. The frac- tions were stored in ice until the assays were completed. SpectrOphotometric assays were performed in silica microcuvettes of 1.0 cm light path, except where otherwise specified. Absorbanee changes were measured in a Gilford Multiple Sample Absorbance Recorder attached to a Beckman DU monochromator. Spectral studies were performed in a Cary model 15 spectrophotometer. Circular dichroism studies were conducted in a Jasco OED/UV-5 spectropolarimeter. Step flow studies were performed in a Gibson-Durrum Step-Flow spectrOphotometer, and transmittancies recorded on a Tektronic Storage Oscilloscope, Type 564. Oscillo- scope traces were photographed with a Polaroid camera; transmittancies were converted to absorbance by a computor program which also calculated reaction velocity by a least squares analysis of the absorbancy versus time data. K1 values for substrate analogues were determined kinetically by the method of Dixon (86). A typical plot is presented in Figure 2. The slope and intercept of the graph were obtained by computor least squares analysis. K1 values for several analogues were verified by plotting 48 Figure 2. Method for Kinetic Determination of K1 Microcuvettes contained 1 umole of dithiothreitol, 0.08 umcles of NADH, 0.06 mg of lactic dehydrogenase, 15 umoles of potassium phosphate buffer, pH 8.0, an apprOp- riate dilution of threonine dehydrase, plus L-threonine and D-threonine as indicated; final volume was 0.2 ml. The assay was followed at 3&0 mu in a Gilford spectrOphotometer, at 28°. K. [D-Threonine] (mM) I50 I K; = 8 mM |25 --+ 25 MM L-Threonine IOO ‘3 > | .".".: MM 0 L-Tiweonine % 75 \—‘— > _\_ 5. w . J] I l A 25 —— ' \ IOO um I w. ‘ L-Threonine _ 1 l 1 J -25 O 25 50 75 IOO 50 data according to the method of Lineweaver-Burke (87), with lepes and intercepts again obtained by computor least squares analysis. All radioactivity measurements were performed in a Packard Tricarb Liquid Scintillometer. Internal standards were employed to measure the efficiency of the counting system. Enzymatic Horseradish peroxidase, rabbit muscle aldolase, and beef liver catalase, which were employed as standard in the sucrose gradient studies, were obtained from Worthington Biochemical Corporation. Aldolase was assayed by the triose phosphate isomerase, c-glycerophosphate dehydrogenase method of Baranowski and Neiderland (88). Peroxidase and catalase were assayed as described by Maehly and Chance (89). Lactic dehydrogenase of rabbit muscle was obtained from Worthington. Hog renal acylase I was obtained from Sigma Chemical Company. Crystalline AMP deaminase from frozen rabbit muscle, employed in several studies to assure the rigorous removal of AMP from certain enzyme preparations, was a kind gift of Drs. K. Smiley, Jr. and C. Suelter, Michigan State Univer- sity, East Lansing, Michigan. 51 Assays for Threonine Dehydrase Coupled Spectrophotometric Assay - For routine measurements of enzyme activity and for examination of the kinetics of dilute enzyme solutions a lactic dehydrogenase-coupled spectrophotometric assay, essentially as described by Phillips and Wood (55), was employed. The coupled system is described by the following equations: Threonine j) c-ketobutyrate + ammonia a-ketobutyrate + NADH --€>’ a-hydroxybutyrate + NAD The loss of NADH was followed Spectrophotometrically at 3h0 mu. The assay mixture contained the following: 1.0 uncle of AMP, 1.0 umole of dithiothreitol, 0.08 umoles of NADH, 0.06 mg (0.25 units) of lactic dehydrogenase, “.0 umoles of L-threonine, 15 umoles of potassium phOSphate buffer, pH 8.0, and a suitable dilution of the enzyme. The reaction is initiated by the addition of either L-threonine or enzyme. The enzyme diluent contained 1 mM AMP, and 1 mM dithiothreitol in 0.1 M potassium phOSphate buffer, pH 8.0. All assays were run in a total volume of 0.20 ml. .Absorbance changes were measured at 340 mu; the reaction temperature maintained at 28°. End.Point_Assay -.An alternative assay was employed for the kinetic examination of threonine dehydrase at enzyme concen- trations higher than applicable with the lactic dehydrogenase- 52 couple spectrOphotometric assay. The procedure involved incubation of the dehydrase for a fixed period of time with substrate, and subsequent determination of the amount of a-ketobutyrate released during this time period. The initial assay mixture contained 1.0 umoles of AMP, 1.0 umoles of dithiothreitol, 15 umoles of potassium phosphate buffer, pH 8.0, L-threonine and threonine dehyd- rase. After a time interval, 20 ul of the assay solution was removed and the enzymatic reaction terminated by the addition of this aliquot to 2 ml of boiling 0.1 M potassium phosphate buffer, pH 8.0. An apprOpriate aliquot of this solution was added to a microcuvette containing 0.08 umoles of NADH, and 15 umoles of potassium phOSphate buffer, pH 8.0, in a final reaction volume of 0.2 ml. An initial absorbance reading was taken at 340 mu, and 0.06 mg of lactic dehydrogenase added. The reaction was allowed to run until no more absorbency loss at 3h0 mu was detected. From the initial and final absorbance reading, and the known extinction coefficient of NADH (6.22 x 103) the rate of dehydration was calculated. Spectrgphotometric Assaz_at High Enzyme Concentration - For assays at extremely high protein concentrations the dehydra- tion was followed directly in a Gibson-Durrum StOp Flow apparatus. d-Ketobutyrate production was measured by recording the decreases in transmitance at 320 mu. This wavelength was chosen rather than the absorption maximum 53 (310 mu) so as to minimize interferring absorption from dithiothreitol and protein. The reaction mixture contained 5 mM dithiothreitol, 5 mM AMP when apprOpriate, 0.1 M potassium phosphate buf- fer, pH 8.0, L-threonine and threonine dehydrase at con- centrations as high as 1 mg/ml. A unit of threonine dehydrase is defined as that amount of enzyme which will produce an absorbance change of 1.0 per minute under the conditions of the coupled spectrophotometric assay (55). Based on a molar extinction coefficient of 6.22 x 103 for NADH, this unit equals 0.032 umoles of threonine dehydrated per minute. The assay was shown to be highly reliable and linear with enzyme content for rates up to 0.2 absorbance change per minute. Purification of Threonine Dehydrase The procedures employed for the purification of threonine dehydrase were developed in this laboratory and involved at various times Dr. P. D. Whanger, Dr. J. H. Piperno, Dr. H. A. Neiderman and the author. The ultimate procedure developed as described below was largely the final contribution of Dr. Piperno and Neiderman. In routine purifications approximately 600 grams (wet weight) of.§.Iggli cells were suspended in 600 ml of 0.1 M potas- sium phosphate buffer, pH 8.0, containing 1 mM AMP and 1 mM 54 DTT. The cells were disrupted after two passes in a Menton-Gaulin Laboratory Homogenizer; cell debris was removed by centrifugation. The following purification steps were then undertaken. All procedures were carried out in the cold room, or at ice bath temperatures, and unless otherwise stated all buffers contained 1 mM AMP and 1 mM dithiothreitol. Protaminefgulfate Treatment - The protein content of the crude extract was diluted to 12 mg/ml with 0.1 M potassium phosphate, pH 8.0, and ammonium sulfate was added to a concentration of 0.1 M. One-fifth volume of 2% protamine sulfate, pH 5, was then added with stirring. The precipi- tate was removed by centrifugation at 12,500 x g for 20 minutes and discarded. Ammonium Sulfate Precipitatign - To the above supernatent enough solid ammonium sulfate was added to make the solu- tion 2.0 M in ammonium sulfate. The solution was centri- fuged for 30 minutes at 12,500 x g, and the precipitate dissolved in a minimal volume of 0.1 M potassium phoSphate, pH 8.0. Chromatography on.DEAE-Sephadex - The enzyme preparation was chromatographed on DEAE-Sephadex (A-50, Pharmacia Fine Chemicals, Inc.) following a procedure similar to that reported by Tokushige (90). The enzyme solution was dialysed overnight against 55 0.05 M potassium phosphate buffer, pH 7.3, containing 0.2 M KCl, in order to obtain the enzyme in the proper buffer for chromatography. After dialysis, the conductivity was measured and the ionic strength of the solution adjusted to match that of the column buffer, by addition of an appropriate amount of distilled water, containing 1 mM dithiothreitol and 1 mM AMP. The enzyme preparation was applied to the column (15 cm x 5.8 cm) with the effluent flow rate regulated at 90 ml/hr. The column was washed with 1 liter of 0.05 M potassium phOSphate buffer, pH 7.3 -0.2 M KCl at the same flow rate. Elution was performed with a linear gradient of 0.2 to 0.8 M KCl in 0.05 M potassium ph08phate, pH 7.3. One liter of each elution buffer was used; the flow rate was maintained at 120 ml/hr, by use of a bicycle chain pump (Technicon Company). 15 ml fractions were collected, assayed and those containing activity pooled. Lower Specific activity fractions were rechromato- graphed on DEAE-Sephadex with subsequent enzyme prepara- tions. The enzyme was precipitated by the addition of 3.75 M ammonium sulfate solution, to a final concentration of 2.6 M. The solution was centrifuged at 20,000 x g for 20 minutes, and the precipitate dissolved in a minimal volume of 0.1 M potassium phosphate, pH 8.0. Chromatography on Hydroxylapatite - Hydroxylapaptite was prepared as described by Levin (91). The column (16 cm x 3 cm) 56 was packed under 1 lb. of air pressure; all elutions were performed under 1.5 lbs. of air pressure. Prior to hydroxylapatite chromatography, the enzyme solution was dialysed overnight against 0.03 M phOSphate buffer, pH 8.0. After dialysis, the ionic strength of the enzyme solution was adjusted to match that of 0.03 M potas- sium phosphate. The enzyme was applied to the column and the column washed with 100 ml of 0.05 M potassium phOSphate, pH 8.0. 50 ml each of 0.07 and 0.09 M potassium phosphate, pH 8.0 was passed over the column, and the enzyme subse- quently eluted with 0.11 M potassium phosphate, pH 8.0. One to 2 ml aliquots were collected and the high Specific activity fractions were pooled and concentrated by ammonium sulfate precipitation as described above. The precipitate was dissolved in a minimal volume of 0.1 M potassium phos- phate, pH 8.0 for storage. Removal of AMP Traces of AMP were rigorously removed from the dehydrase by chromatography on Sephadex:G-25 (Pharmacia Fine Chemicals, Inc.). The column (0.6 x 4.6 cm) was equilibrated with 0.1 M potassium phosphate buffer, pH 8.0 containing 1 mM dithiothreitol and the sample eluted with 14C the same buffer. A typical profile with dehydrase and AMP, is shown in Figure 3. The final concentration of AMP in the dehydrase was assessed to be not greater than 10'6M; 57 .Amav cospoaeso one onsem mo posses on» an meaaadm mucosad Ha H.o mo wsapssoo soapmaaapsdom oasaaa an meanness as: mudbdpoeoaUdm .sasaoo came on» Hobo cwommea camsam d an peamdasa mflmsoaboaa mm: soapmasaoaa usanmzd one .HopaoanponpHp as o.a vengeance sods: .o.m ma .aomasn cumsamona adammMpon z a.o and: p90 defiance was Soapsam .Aao m.: H 0.0V assaoo mmlu Noumeaem s on oeaaaad was Aaao ooo.wmav oafiumz< no peace: o~.o and adopoaa ho we mN.o confidence scans “HE mN.ov menace no canadm 4 sodpmaeaoam omeapmsen scam mz< no Hebosem .m chamdm Onv 00. mm. mums—DZ memo an on AV d_e . a o. .I. ov.ud2< meowmm 253 02d: 253m ON 0d 00 .OM $1.an cOI 59 this corresponds to 0.07 mole of AMP per 160,000 grams of enzyme, for a protein solution of 4 mg/ml. RESULTS In the detailed investigation conducted by Phillips and Wood (55, 61), a Specific role for AMP in the catalytic dehydration mechanism was not revealed. It was concluded that AMP functions as a true allosteric activator of threonine dehydrase; that is, conformational alterations associated with.AMP binding are reSponsible for the observed changes in activity. Two primary research objectives were formulated at the onset of the current investigation. These were: (1) to determine the catalytic step(s) in the dehydration mechanism which is affected as a result of the allosteric transition thereby accounting for the observed changes in Km and‘Vmax, and (2) to determine the role of oligomeriza- tion in allosteric activation. These studies are considered in sections 3 and 5 of this chapter. The nature of the above investigations necessitated undertaking detailed characterization of the kinetics and quaternary structure of the dehydrase. Results of these studies have already been reported (64) and will be presented in summary form only in section 1. Since many of the earlier findings were obtained with partially purified enzyme preparations, these experiments have been repeated in certain cases with highly purified dehydrase. The results with pure enzyme will be 60 61 presented here. An investigation was also undertaken in which struc- tural analogues of AMP were used to assess the chemical and structural requisites of the activator site. The results of this study were reported at an earlier date (74) and are presented in abbreviated form in section 2 for the sake of completeness only. 1);Characterization of Dehydrase_Kinetics and Protomer-Oliggmer Interconversions Effect of AMP on Km and‘Vmax Km and‘Vmax values of highly purified threonine dehydrase were determined from the three common transfor- mations of the Michaelis-Menten equation (93): 1/v = 1/Vmax + Km/Vmax (l/S): S/v = Km/Vmax + (l/Vmax)S: v = Vmax - Km(v/S). The data were treated by the method of least squares with the aid of a computor program to give the necessary parameters. The Km and Vmax obtained are presented in Table 2; in series I of this Table AMP was removed by dilution, and in series II by chromotography on Sephadex G-25 (see below). For all determinations coefficients of correlation of 0.99 or greater were obtained with each of the three functions. As may be noted, AMP causes a 25-35 fold decrease in the Km for L-threonine and concomitantly, an increase of some 3-4 fold in'Vmax. These results indicate that the 62 TABLE 2 Effect of AMP on Km and'Vmax AMP was removed from the dehydrase sample by dilution of the dehydrase to the concentration range of assay (Series I), or on a Sephadex G-25 column as described in Methods (Series II). The specific activities of the untreated dehydrase preparation and the G-ZS-treated preparation were 11,500 and 10,900 units per mg protein. respectively. All assays were performed at 28°. The values of Km and‘Vmax were calculated by a linear least squares program from the three common transformations of the Michaelis-Menten equa- tion (93): l/v = Km/Vmax(1/S) + 1/Vmax: S/v = (l/Vmax)S + Km/Vmax3 v =‘Vmax - Km(v/8). L Km Km Ratio Vmax Vmax Ratio a -AMP +AMP Series qAMP +AMP IIMF’ -AMP +AMP3' 31M? mM umoles x mg ‘— protein' I 50 2.0 25 90 345 3.9 II 67 2.0 35 111 361 3.3 eAdded at a concentration of 5.0 mM. 63 activation of threonine dehydrase is primarily the result of a marked decrease in Km for L-threonine. However, AMP activation is accompanied by a moderate, but significant increase in'Vmax. The KIn values determined here of 50-70 mM without AMP, and 2 mM with AMP compare reasonably well with the earlier reported values of Phillips and Wood (55) and Whanger‘gt‘gl. (64), obtained with much less purified dehydrase. Phillips and Wood (55) reported values of 5 mM and 30 mM in the presence and absence of AMP, respectively; while Whanger gtflgl. (64), obtained values of 4'mM and 100 mM. However, the latter workers (64) concluded that‘Vmax did not change upon the addition of AMP. Considerably different values were obtained by Hirata.gg'gl. (94), who reported Km values of 4.2 mM and 20 mM. respectively, in the presence and absence of AMP. They further reported a 10 fold stimulation in'Vmax caused by AMP. Conditions of the assay employed by Hirata 23.2}: (94), were substantially different from those used in this laboratory in regard to pH and thiol content: no thiol was included in their assay system. In view of the sulfhydryl sensitivity of the dehydrase (see below) it would appear that the absence of an effective thiol reagent may have accounted for the large difference in‘Vmax detected in their assay. 64 Absence_of L-Threonine and AMP Cooperativity Rate-concentration studies with the highly purified dehydrase failed to reveal any cooperativity in confirma- tion of the results obtained with the partially purified enzyme and reported by this laboratory earlier (64). Substrate versus velocity plots were hyperbolic and recip- rocal Lineweaver-Burke plots of 1/velocity versus 1/sub- strate were linear (Figure 4). Hill coefficients of 1.0 were obtained from the Hill plots both in the presence and absence of AMP (Figure 5). The Ka for AMP was determined to be 0.5 mM. Plots of AMP versus velocity similarly failed to reveal any sigmoidal behavior, and a Hill plot for AMP had a slope of 0.99 (Figure 6). Therefore, it would appear that the biodegradative threonine dehydrase does not exhibit any homotrOpic effects. Hirata gt a}. (94), have reported that the K8 for AMP varies slightly as a function of L-threonine concentra- tion. This type of heterotrOpic interaction was not examined in the current investigation. Effect_ef AMP on Quaternary Structure Phillips and Wood (61) first noted that AMP causes appreciable changes in the sedimentation velocity of the dehydrase in sucrose gradient experiments. Under the con- ditions then used, an 320 value of 4.88 was obtained either 65 .oom.HH mes sages» omsapmncu on» go apabapos odmdooam .omN as poaaomaea one: masses “apogee: 2H condensed no one: memes oaanspso on» no escapapsoo mad as o.m no oosomoam use eosemp< on» ad omsaumaoa peamdasm hanwam how mpoam onusooanelg mamaob updooao>.asooaaaoom .3 shaman 66 .-(”"“’°"“‘W) Moons/v: N 0. Vi 0. ‘0 N v '0 [0 IO N' mi 0. 2 <1 5 3 0. E <_ O \ 3 8 8 8 8 ‘3 ,-(“’“”°"9\7V) MooneA/I V 0.3 0.2 0.l l/(L-Threonine) (mMT'I 0.025 0% 0.075 0 6? .: oaswam ad condensed me one msoapduaoo 22329.8 5:: onesoaeoo c5825 a8 3on :3 .m chum; 40.0 20.0 l0.0 v/V-v 68 W/O AMP Slope = I.O4 *— l o L 2 '46 L-Threonine (mM) IO 20 40 ICC 200 69 Figure 6. Hill Plot for Highly Purified Threonine Dehydrase with AMP Conditions of the catalytic assay were as described in Methods; assays were performed at 28°. Specific activity of the dehydrase sample was 10,000. 70 .l .2 .4 .6 .8 AMP (mM) LO 69 Figure 6. Hill Plot for Highly Purified Threonine Dehydrase with AMP Conditions of the catalytic assay were as described in Methods; assays were performed at 28°. Specific activity of the dehydrase sample was 10,000. v/ V-v 70 .I .2 .4 .6 .8 I AMP (mM) .0 71 with IMP, or when no nucleotide was present in the gradient. When AMP was included an 820 value of 7.68 was observed. These changes in 820 value appear to be too large to result exclusively from a conformational alteration of a protomeric molecule and it was concluded that the changes in 820 value probably resulted from a protomer-oligomer transition. Assuming an average partial specific volume for all dehydrase species, the 820 values stated above correspond roughly to molecular weights of 78,000 and 155,000, respectively (61). Further investigation has revealed that at low dehydrase concentration, in the absence of AMP, dehydrase species can be obtained which exhibit 820 values consider- ably lower than 4.88. Sedimentation velocities observed with low levels of highly purified dehydrase, in the absence of AMP, are presented in Table 3. The above observation suggested that both protein concentration and.AMP may influence the quaternary struc- ture of the dehydrase. Therefore, the effect of protein concentration on sea values in sucrose gradient experiments was examined. In the earlier investigations with low dehydrase levels, AMP was routinely removed from enzyme samples by dilution prior to sample application to the gradient. However, this method is ineffective when more concentrated dehydrase samples are employed in gradient experiments. 72 TABLE 3 Table of 820 Values Obtained at Low'Dehydrase Concentrations Without AMP Sucrose density gradient centrifugations were per- formed essentially as described by Martin and.Ames (81). Centrifugations were conducted for a period of 31 hours at 39.000 rpm in the SB-283 rotor (International Equipment Company) at 4°. AMP was removed from the dehydrase samples on Sephadex G-25, as described in Methods. 12 ml gradients, prepared in 0.1 M potassium.phosphate buffer, pH 8.0, con- tained 5.0 mM dithiothreitol. W Preggigziona ggégzegg 820 II 9.2 3.6 I 18 2.9 I 36 3.4 I 72 3.0 II 92 3.7 9Preparation I, 8,300 specific activity; Preparation II, 14,300 specific activity. 73 An alternative procedure had to be developed to remove rigorously all traces of AMP from the dehydrase prepara- tions. A number of procedures which were tried failed to satisfactorily remove AMP. Long term equilibrium dialysis, as well as repeated ammonium sulfate precipitations, could not lower the AMP concentration to less than 5 x 10"6 M. Further, appreciable enzyme activity was fre- quently lost during these procedures. Chromatography on Sephadex 6-25 was found to be highly effective for removing AMP from the dehydrase prepa- rations. This method invariably reduced the nucleotide concentration to 10"6 M (corresponding to 0.07 molecules of AMP/160,000 molecular weight, for a protein solution of 4 mg/ml). This procedure was routinely employed to remove AMP. Figure 7 indicates that the 320 value for the dehydrase in sucrose gradients increases as a function of protein concentration. The 820 values presented in Figure 7, obtained with highly purified dehydrase are plotted with both linear and logarithmic abscissa with reSpect to the number of dehydrase units applied to the gradient. The curves on the right, plotted with linear absciaa, approach the shapes of hyperbolas; however, with logarithmic abscissa straight lines were obtained. .At very low dehydrase concentrations in the absence of AMP a limiting 320 value 74 .msoapsApseosoo memes odphaspso pm omsaphsou I consume pomoao .oo>osoa soon hobo: we: mad nods: Soak omsapmaou I seamssaaa .soapstMaausoo op nodaa poosaaon was poboaoa soon ms: was sodas scam means» I moaoaao pomoao .pobosoa mamsoaowda soon we: mz<_soans aoau undamaged m24 I neHoaao some .popsodpsd when: mz< as o.m use HopHeHSpoanpdp :6 o.m ponds» Isoo no«ns .o.m ma .aomasn opsnamosa asasmspoa z H.o ad venomoaa one: museapsau .mponue: :a confluence as .mqu Nopdsaom so madness omsaehnev on» some peboaea mm: mz< .A.oo uneasnpmsH adaxoem .QOHmdbdQ oozaamv aopoa mmIsm can :« mason 5H no Amadaaoo pacemasdm assoapdzaopsHv Hosea nmNImm on» :« any ooo.mn one o: as mason an no weapon a non deposusoo one: msoapswsmaapsoo .Aamv head one causes no scenes 0:» an soapswsu Iaapsoo useapsam mpamseu seasons an eesaspno one: sesame own one masseuseo codudasm maswam one no soapsapseonou.£pas eaas> own no scandaas> .m eapwdm 75 000.0 Hzmadmo 20 mtz: 08... 83 o 80.0.80. 8. o. _ awn L a... e. a. _ ...-cIII . \. r. s. _ IO 0.0 225 9 as? mscfizm mmEdsa I. It) 0.0 .34 oz I o.» . I . D... Q54 2E J hula- IIIII _ I|anlld|mld d 0.0 _ Es 9 as? _ 92>sz oz_z_._._oo._u> zo_._.<._.zm2_omm 20 zo.._.OSea mamsoaowaa mad spas oHaSSm smeavmaep I mafia penmsa .eamuso paooxo mpsosoaaoo HHS oocampaoo moppobso cocoaomom .Ha m.o mes oedaop Honda one “concedes“ when: did as o.m one .o.m ma .aoumSp opssamona azammopoa z H.o .HopaoASSoanpap as o.H .Ha\ma m as saunas decadence moppobso oaaaom .amuoaoposaoapooae ma Hepoz memo s so pepaoooa one: caveman one emsaumnom enasoohsaIA ho sappooam soapaaomnd .m madman 9? 00¢ ONV 3:: Soc 2925 00¢ 03» 00¢ 000 ’— 00m own 00. 08. eouoqmsqv 98 .popsaosa me: mad as o.m use» paooxe obons confluence ms was: msoapapcoo I mafia paaom .Ha 0.0 ms: sadaob Honda cap snowman epssamona sadnespoa z H.o use .Hopaoanpofinpap as o.H .sadaoo mNIo Hoomsaem s so peboaoa mamsoaowaa Soon on: mz< sodas Scam .Ha\wa N.“ as meanno vocampsoo ounobfio causes I mafia eczema .aopea Iaasaoaoauooam mI>D\Qmo comma s so popaoooa me: Seaoanoap amazoaao one museumsom endsoeanan ho Smdoasoaa anaconda .m oaswdm 99 3,5 £95683 0N¢ 00¢ 08m 00m (ll 0 1,_OI X WW (0 8 100 The dehydrase exhibited a maximum in circular dichroism near 415 mg, and a shoulder at 438 mu. These dichorism observations agree with those reported by Nakazawa 93 a}. (59). The Spectral and dichroic maxima have been attributed to the formation of an aldimine between pyridoxal phOSphate and d-amino group of a lysine residue on the enzyme (55, 60, 71). Since this pyridoxal phoSphate aldimine lacks an asymmetric center, the cptical activity necessary for dichroism must result from the association of the pyridoxal phoSphate azomethine with the rest of the enzyme structure. Effect of L-Threonine on Absorption Spectrumgand Circular Dichroism of Threonine Dehydrase - Tokushige,‘23'gl. (60) first reported that addition of L-threonine to the dehydrase caused a shift of the absorption maximum to around 450 ms and loss of the 415 mu circular dichroism. The original absorption and circular dichroism returned as the substrate became depleted. Studies in our laboratory have confirmed these observations. In Figure 10, the time course for product (d-keto- butyrate) accumulation is compared to the time course for appearance and disappearance of the Spectral intermediate (455 mg) and for the disappearance and reappearance of circular dichroism at 415 mu. As the rate of c-ketobutyrate formation begins to fall rapidly, so does the absorbance at 455 mu, simultaneous with rapid increase in circular dichroism 101 .Hs 0.0 Smdoasoap seasoaao sou .Hs m.0 no: manoSoASmSes Hesperus pom ossaob Honda or» “soapoeoa on» opeapasa op moppebso ex» on 00008 was cadsooHSpIq as on .Aas\ws mo.ov onesoasoo mongoose» one .as< ss o.m .Hoeaosspoaspae as o.m .0.m ma .aommsp openamoaa SSammdpoa : «.0 decadence canvass caduceus 0:9 as we: as smaoanoan aeHSoaao no oossaeoaaeem use smog on» you use .38 mm: as openeosaeusH Heauoeam on» yo oossaeoaamman use oossaeoaa<.aom .sodpdaaoh opmHmpSDovoMId you omaaoo made .0H earmam 102 (96) "w 917 wsuosuom HV'anHIO 00. 0h 00 am 03.32:). 0 n e m m _ a. . l _ _ w\. ., 3:. a... .03 02.5020? _ asauuz>~zu . fl 4 4 3... one. 0255.20: . and I 34230502.: A _ a x _ as 0.9 20.528“. 0.255093% 0 d ~— azzaoq 080200 05:08:... 00 hcoeosnoos‘ 60:00 ._ «.0 ‘9. V. 0 0 "U1 0l€ AONVBHOSBV 0°. 0 0. 0.0.0 0N0 .0 0n0.0 . 0¢0.0 000.0 ”W 9917 AONVBHOSBV 103 at 415 mu. Tokushige gtflgl. (60) suggested that the formation of a new optically inactive azomethine of substrate and pyridoxal phOSphate may account for the loss of dichroism upon L-threonine addition; and, that the 450 mu absorbance might be ascribed to the formation of an enzyme-substrate complex. Competitive inhibitors of threonine dehydrase were employed in Spectral and dichroism studies with the dehyd- rase which lead to the interpretation that threonine- pyridoxal phOSphate azomethine formation accounts for the loss of 415 mu circular dichroism. Further, an attempt has been made to identify the catalytic intermediate reSpon- sible for the 455 mu Spectral absorbance. Effectwof Competitive Inhibitors on Circular Dichroism at 415 mu A number of competitive inhibitors incapable of under- going dehydration were studied to more clearly define the significance of the dichroism observation. All of the inhibitors tested possessing d-amino sub- stituents were found to effectively induce loss of the cir- cular dichroism. Included among these inhibitors were D-threonine, DL-allothreonine, L-a-aminobutyrate, L-alanine, and D-serine. Inhibitors Lacking c-amino substituents, i.e., L-lactate, Disc-hydroxybutyrate, DLps-hydroxybutyrate, 104 d-hydroxybutyrate, could not induce loss of the circular dichroism. Indeed, the intensity of the 415 mu circular dichroism increased slightly upon addition of inhibitors of this latter category. These findings indicate that competitive inhibitors with d-amino groups are capable of entering into some asso- ciation with the enzyme which inhibitors lacking d-amino groups are not; this is probably formation of an external azomethine. These results serve to substantiate the inter- pretation that transaldimination is reSponsible for the loss of circular dichroism. Effect of Competitive Inhibitors on;Absorption Spectrum The effect of various inhibitors on the absorption spectrum of threonine dehydrase is Shown in Figures 11 and 12. In the presence of AMP the inhibitors D-threonine and DL-allothreonine were found to induce a shift of the Spec— tral maximum from 413 mu to 418 mu. Further, the intensity of this new maximum was 40% greater than that seen with native enzyme; however, no new absorption was detected at 455 mm. The inhibitors, L-a-aminobutyrate, and DL-B- hydroxybutyrate effectively increased the intensity of the 413 mu absorbance 15-25% without inducing any shift in the position of this maximum. In the difference Spectrum with D-threonine a maximum was revealed at 423 mm (Figure 13). 105 .00000 speazpsn IasosoasIaIss as ooa seas I odds ooppoo looses oeesapssosaseIeIa as cos sea: I osaa sesame “ass as m use Hotnesspoaspao as a spa: .o.m so .aemmsn epssamona S H.0 ca Ha\ms m pd omsasmnov madaooasu I mafia dadom I enemas 0o code psmam .osasoosspoHHqus as on sea: use atone no I osaa cosmos uofisocsanm ss ooa and: use atone me I odds oeppoe ”as< as m .Hoeaosseoaseao as a .o.m so .soaesn opusoeosa ssaeeepoo s “.0 sh HS\ws m as museumaov enasooasp I mafia dadom I seawam mo evdm puma maw mo monomoam was ad omsa0m£0m enamooaae do ssseooan cabana> so necessassH opasaeeasoo do scones .Ha enemas 106 000 00¢ 0N¢ 0.. V m m a. 0.. O u 0 a 0m. 0N On. 38. £05.08; 00m 000 00¢ 0N¢ 000 0¢m a... a. a . a p .. ... .. f ... . 22:3 4...! Iaaeoar-‘-._o \«... a; e. w. ... .... ... .. ... ... 4.. 4“ .... f. 4.. .... . 2.52.: ”4... ! .... .. I0..D\Qm0 00000 0 ca 00x0» 0H0: mwSa000a Smaoa£0a0 a0HSoaH0 .Hs 0.0 003 085H0> Hecam 0:» .086» ones 00 0000000 0H0S00 one on 00000 003 0sdsooa£pIq a N.0 .00»00«0Sa 0a0£s mz< as 0.0 0:0 .Hop«0a£poa:pd0 as 0.0 .0.0 we .aohusn epmsamona adammepoa z a.0 .0NI0 H000£aom so 00>oaea mamaoaomah S009 00s ma4 sodas Scam “Hs\saopoaa we 0.0. 000a0m£00 00amaasa 00Saepsoo 00000050 Smaoazoam a0asoaH0 00 0000 00050SH emanceaseIA so 024 00 000000 .3« waswam 116 X _F h— o o N o O (D o — o m o O. . 2 . 4 o — o 2 v ° E o o m . . ./ m 0 Q g . 3 ..__ . N o g . 0 / o v”’|’ ’,-——--—"“’—-’- .L__‘=;'”"".L-' J 0 O. 0. O. O. 0. N c_>_ a) co <1- cu Time (minutes) 117 of circular dichroism is only 34% of that observed in the presence of AMP. Even when the initial L-threonine con- centration is increased to 0.5 M the loss of circular dichroism is still less than 50% of that obtained in the presence of AMP (71). Thus. it is apparent that AMP markedly enhances the extent of dichroism loss attainable at any L-threonine concentration. This finding indicates that one or more of the catalytic intermediates which lack the assymetric pyridoxal phosphate—lysine azomethine link- age necessary for circular dichroism (viz.. L-threonine- pyridoxal phOSphate azomethine. E82, the conjugate base of threonine-pyridoxal phoSphate azomethine, E83, or amino- crotonate-pyridoxal phOSphate azomethine, E34) is accumu- lated to a much greater extent in the presence of AMP. As the following mathematical relationship reveals: (E32 + E33 + E54) = EOS(1/K1K2 + 1/K1K2K3 + l/KleKBKL‘) (5) a decrease in the dissociation constant for any of the first four steps of catalysis (K1 through K4) would cause increased accumulation of these three intermediates. The effect of AMP on the intermediate absorbing at #55 mu is shown in.Figures 15 and 16. Figure 15 is a differ- ence spectrum obtained in the Cary Spectrophotometer, and Figure 16 was obtained by following the #55 mp absorbancy directly in a Gilford Spectrophotometer. The results indi- cate that AMP appreciably enhances the accumulation of the 118 .0aswam 0:: :a 00p00a0nfl 0d :6 0m: 00000: asapo0a0 2000 :000 :0a:z p0 0Edp 0:8 .Ha m.0 m0: m0pp0>30 00:0A0m0h 0:0 0Haa00 :00: mo 0&5Hob H0Qam 0:0 “:0a0000h 0:0 000apasd Op 00up0>00 0Haa00 0:» ca 00000 003 A28 omv 0zdso0h:pla 00 00Hoaa mm.0 .00o:p0z :« 00pa90000 00 mmlo N000:Q0m :o 00>oa0a mamsoaomah :00: 00: mz< :0H:3 Scam 000H0z:00 0:“:o0a:puq mo me n 0:0 .00000000: 000:: :25 0.00 020 00 00:05: 0.0 .o.0 ma .:2 0.00 000000 000000000 asdmm0poa 00 003a1 0.5m .::s 0.00 Hopd0usuoanpae ho 00Hoan m.m 00na0psoo 00pp0>50 00:0a0m0a 0:0 0Has00 0:9 00000000910 00 00000004 00000 000005000 0saso0a:a mo 0npo0am 00209000“: no mzq mo 900mmm 0:9 .ma onswam 119 000 000 00¢ 00... 00» 000 00¢ 8m 0%. 3:: 50204023 .0: _ 000 owe (r: 0. o id cl (5 BONVBHOSBV aid: "2 o i 120 Figure 16. Effect of AMP on L-Threonine Induced Accumula- tion of Spectral Intermediate at #55 mu Microcuvettes contained purified dehydrase (1.44 mg protein/ml) from which AMP had been rigorously removed on Sephadex G-25, 5.0 mM dithiothreitol, 0.1 M potassium phos- phate buffer. pH 8.0. and 5.0 mM AMP where indicated. 0.2 M L-threonine was added to the sample cuvette at zero time; the final Volume was 0.25 ml. Absorbancies were recorded on a Gilford Spectrophotometer at 28°. Absorbance (455 mp) 121 l I 0.2 M L-Threonine 5 mM AMP L6 32 4.8 Ti me (minutes) 122 455 mu intermediate, considered to be aminocrotonate- pyridoxal phOSphate azomethine (Esu). As revealed in equation (6): a decrease in the dissociation constant for any of the first four steps in the dehydration mechanism can account for the increased accumulation of aminocrotonate-pyridoxal phosphate azomethine, E34. The practical consequence of both the dichroic and Spectral observations is, then, that AMP must activate bio- degradative threonine dehydrase by stimulating some cata- lytic step which lies prior to c-aminocrotonate-pyridoxal phOSphate azomethine, i.e., non-covalent binding, transaldi- mination, or one of the two steps of dehydration. It might be argued that AMP causes accumulation of both the "Spectral" and "dichroic" intermediates by inhibiting a reaction step which falls subsequent to B-hydroxyl group elimination. In such a situation, however, the effect on Km and Vmax would be reverse of the actual observations; that is, there would be an increase in Km and decrease in‘V Therefore, max ' this alternative possibility may be eliminated. Effect of AMP on K1 for LJThreonine Analogues From the investigation of the Spectral and dichroic prOperties of threonine dehydrase it was possible to limit the steps at which AMP may cause allosteric activation to 123 the first four. In order to determine more Specifically which step or steps are involved in the AMP-associated acti- vation, the effect of AMP on dehydrase affinities for vari- ous competitive inhibitors was studied. It was decided for the following reasons that such an investigation may lead to an accurate assessment of the effect of AMP on the early steps of the dehydration mechanism. First, it should be clear that many competitive inhibitors may undergo several of the partial steps of catalysis and K1 values determined for such inhibitors should be some function of all of the steps in which the inhibitor is involved. Indeed, the results of the circular dichroism investigation suggest that amino acid competitive inhibitors can effectively undergo a transaldimination reaction with the dehydrase. It is even conceivable that these inhibitor-pyridoxal phos- phate complexes might enter into c-proton and S-substituent exchange on the enzyme where consistent with the structure of the analogue. However, the failure to detect "#90" mm and "#50" mm absorbing Species indicates that little or no c-elimination or c-B elimination intermediate is formed with these inhibitors. These partial reactions if they occur at all must have equilibrium constants that make accumulation of absorbing intermediates highly unfavorable. The K1 values obtained for an amino acid competitive inhibitor is, therefore, probably a function of only the equilibrium con- stants for non-covalent association and transaldimination. 124 In such a case the K1 is defined by the following equation: K1 = K1K2/(1 + K2) (7) In addition, it should be clear that the K1 for a competi- tive inhibitor lacking an c-amino group must exclusively reflect non-covalent association with the enzyme. Such threonine analogues as Lplactate, DL-a-hydroxybutyrate, DL-B-hydroxybutyrate, b-hydroxybutyrate, pr0pionate and butyrate can then be eSpecially useful for testing the effect of AMP on initial non-covalent complex formation. A competitive inhibitor investigation serves the second function of providing information about the chemical and structural requisites for substrate binding to the catalytic site. The results here shall therefore be dis- cussed in connection with both of these facets. To facilitate the discussion, the various L-threo- nine analogues tested have been divided into three groups in Table 8, i.e. (I) those inhibitors which show marked enhancement in affinity with AMP addition, (II) those inhibitors whose affinity for the dehydrase is independent of AMP, and (III) analogues which show little or no dehydrase affinity. The inhibitors listed in group (I) were found to exhibit appreciable affinity for the dehydrase in the presence of AMP, but poor binding affinity in the absence of AMP; that is, AMP markedly enhanced the affinity dis- played by the dehydrase toward these analogues of L—threonine. 125 TABLE 8 Effect of AMP on Km of Threonine and K1 of Threonine Analogues K1 values were determined by the method of Dixon (86) as described in Methods. Substrate or Inhibitor -AMP AMP 8‘ LpThreonine 50-70 2.0 Group I LmAllothreonine 200 0.8 IpHomoserine >1000 5.0 IpAlanine >1000 15 lpchminobutyrate >1000 15 L-Lactate 350 40 DL-d-Hydroxybutyrate >1000 37 DL—B-Hydroxybutyrate 350 36 c-Hydroxybutyrate 360 25 PrOprionate 180 80 Butyrate 500 70 DIpAlaninol >1000 75 LAValine 117 70 Group II Dqulothreonine 0.7 0.6 D-Threonine 10 20 D-Serine 14 7.5 D-Homoserine 50 32 DLPO-Methylserine 45 14 Group III Dquanine >1000 206 D-quminobutyrate >1000 >1000 D-Lactate >1000 >1000 Bquanine >1000 >1000 Isoserine >1000 >1000 L-d-Methylthreonine, L-allo-c-methylthreonine 200 120 D-c-Methylthreonine, D-allo-a-methylthreonine 132 310 Dips-Methylserine 200 400 IpAlanine methyl ester >1000 200 LmAlanine ethyl ester 300 200 a'AMP concentration was 5 mM. 126 These inhibitors were apparently all of the L-series. While the separated enantiomers were not available to test in the case of DLPa—hydroxybutyrate, the substantial affinity for only the L-isomer of lactate (the three carbon analogue of the above inhibitor) suggests that only the L-enantiomorph of d-hydroxybutyrate can function as a competitive inhibi- tor. Separated enantiomers were also not available for DL- B-hydroxybutyrate. The inhibitors 6-hydroxybutyrate, propi- onate, and butyrate in group (I) are not optically active. Surprisingly, it was observed that the dehydrase exhibited high affinity for a number of D-analogues (group II); this affinity was not significantly changed when AMP was removed.Lp For example, while both D- and L-allothreonine have K1 values of approximately 1 mM in the presence of AMP, the K1 for L-allothreonine dr0ps from 1 mM to 200 mM in the “It may be important to note that when a D-analogue is added to a reaction cuvette under conditions where AMP is absent from the reaction mixture, inhibition of the cata- lytic assay does not ensue immediately; rather, the rate gradually decreases to its final inhibited reaction velocity. The time period for this lag was of the order of one to five minutes and appeared to be a function of L-threonine concen- tration. In the presence of AMP no lag was apparent. In determining K1 values the rates from the initial lags were ignored. The nature of the inhibition after the lag period was established to be simple competitive. Although this observation indicates that the D-ana- logues must cause some conformational alteration before they can competitively inhibit the dehydrase reaction in the absence of AMP, the nature of the lag was not investigated in any greater detail. 127 absence of AMP, and that for the D-analogue remains near 1 mM. This behavior contrasts sharply with that exhibited by both the natural substrates L-threonine and L—serine, as well as those inhibitors of the L—configuration in group (I). Further discussion of the D-inhibitors is, therefore, reserved to later. The results of TableEB may now be used to evaluate the effect of AMP on the catalytic mechanism. As was noted, AMP increases the K1 for many of the L-analogues greater than 70 fold, 200 fold for L-allothreonine and L- homoserine. Since the K1 values for these analogues is considered to reflect only the first two steps of catalysis, i.e., non-covalent complex formation and transaldimination, these findings indicate that the effect of AMP in the cata- lytic dehydration mechanism must lie at one of these early steps of catalysis. Further, it can be noted that.AMP causes a 9 to 25 fold increase in the affinity of the dehydrase for analogues which possess no a-amino substituent, i.e., DL-a-hydroxybuty- rate,.DL-B-hydroxybutyrate, y-hydroxybutyrate, L-lactate, propionate, and butyrate. Examination of equations (1) and (3) reveals that an alteration in the equilibrium constant for non-covalent complex formation of this magnitude is sufficient to account for a 25 fold change in Km. It is even possible that the change in K1 might be greater for the better binding natural substrate than for these analogues. 128 Although these eXperiments have not established whether AMP might have some slight effect on other steps in the dehydration mechanism, these results do indicate that the major effect of AMP in the allosteric activation of threonine dehydrase is eXpressed on the initial step of catalysis, i.e., non-covalent complex formation. 4) Structural Requisites for Threonine and Threonine Analogue Binding The chemical and structural requisites for substrate binding to the dehydrase may be evaluated from the various substrate analogue affinity constants. The L-analogues which diSplay high dehydrase affinity only in the presence of AMP, and the D-analogues which exhibit appreciable affin- ity in both the absence and presence of AMP will be considered in turn below. No Single substituent on the carbon chain can totally account for the L-inhibitor affinity; that is, each substitu- ent on the substrate apparently contributes, to some degree, toward the binding. The K1 values for the analogues L- alanine and L-c-aminobutyrate possessing a-amino group sub— stituents and no hydroxyl substituents, or for DL-B-hydroxy- butyrate, possessing a fi-hydroxyl substituent and no a-amino substituent, are quite low. Remarkably effective binding affinity is also observed with the analogues L-lactate, Dina-hydroxybutyrate, y-hydroxybutyrate, and L-homoserine 129 in which the hydroxyl substituent is positioned on either the B or y carbon. Even pr0pionate and butyrate, lacking both amino and hydroxyl groups, as well as, DL—alaninol, the primary alcohol analogue of DL-alanine, bind fairly well to the dehydrase. However, analogues with an amino substituent at a position other than on the c-carbon have poor binding affinity; S—alanine and isoserine (2-hydroxy- 3-aminOpr0pionate) were completely ineffective as competi- tive inhibitors. Esterification of the carboxyl group of L-alanine, or substitution of a methyl group at the c-proton position of the substrate (d-methylserine, d-methylthreonine) results in considerable reduction in enzyme affinity towards these analogues. The Specificity requirements for the D-analogues (exhibiting appreciable dehydrase affinity in both the presence and absence of AMP) was much more restrictive than for the L-inhibitors. Only those D-analogues possess- ing both an c-amino substituent and a B-hydroxyl substituent, or substituted S-hydroxyl substituent, effectively bound to the enzyme. Thus, D-alanine, D-a-aminobutyrate, and D— lactate were poor inhibitors (group III). While the separated D and L enantiomers of O-methyl- serine were not available for test, the results obtained with the racemic mixture suggests that D-O-methylserine exhibits fairly high affinity (K1 estimated at 20 mM) which is not affected by AMP. D-O-methylserine is, therefore, 130 included in group II. Finally, it Should be noted that the importance of the behavior of D-analogues becomes apparent only when com- pared with the L-analogues. Hirata 33 El. (94) reported that the K1 for D-threonine and D-serine were essentially the same in the absence and in the presence of AMP. How- ever, the latter authors did not examine any competitive inhibitors of the L-configuration. 5) Relation of Activation and Oligomerization The effect of AMP binding and oligomerization on allosteric activation will now be considered. Phillips and Wood (61) first suggested that the oligomerization process might be influencing activation of threonine dehydrase. There are clearly several alternative mechan- isms by which this could occur: (1) allosteric activation may result exclusively from the oligomerization process with AMP functioning solely as an agent which encourages oligomerization. Activation is then only indirectly dependent upon AMP binding. (2) The allosteric transition reSponsible for activation may be caused by binding of AMP to the monomer with the observed changes in quaternary structure being an indirect consequence of the activation. (3) Activation may necessitate both.AMP binding and oligo- merization. A number of investigations were undertaken in order to distinquish among these alternative mechanisms. 131 Kinetics of the Oligomer Induced by High Dehydrase Concentration Sedimentation velocity (section 1) and molecular weight experiments (65, 108) have indicated that oligomeri- zation of threonine dehydrase may be achieved in the absence of AMP by increasing the dehydrase concentration. The data presented in Figure 7 suggest that the quaternary structure of AMP-free enzyme at concentrations of 1 mg/ml or higher is similar to that of enzyme at more dilute protein concen- trations and in 5 mM AMP. The opportunity, therefore, existed to test whether oligomerization, by itself, is a sufficient event to account for the 25-35 fold decrease in Km for L-threonine. The difficulties in measuring the kinetic properties of concentrated enzyme are obvious; catalytic rates of dehydration are so rapid that velocity measurements are virtually impossible to evaluate in the coupled-Spectro- photometric assay system. Therefore, alternative means had to be develOped to determine the activation state (Km for L-threonine) of the oligomer formed at high protein concen- tration. Two methods were used which involved (a) deter- mination of the dissociation constants of known competitive inhibitors of the dehydrase by methods not depending upon catalysis, i.e.. in the absence of L-threonine, and (b) determination of the Km for L-threonine in validated modi- fied assay systems in which high levels of dehydrase could be used. 132 Determination of K1 for LJThreonine Analogues by Circular'Dichroism Table 8 has indicated that the inhibition constants for a number of inhibitors are markedly different for acti- vated and non-activated dehydrase. Since amino acid competi- tive inhibitors of threonine dehydrase are known to evoke loss of the circular dichroism exhibited by native enzyme at 415 mu, we sought to determine K1 values for these ana- logues by following loss of circular dichroism with varying inhibitor concentration. All such studies were performed at sufficiently high protein concentrations (5-10 mg/ml) so that the molecular form of the dehydrase is oligomeric even in the complete absence of AMP. If aggregation had caused activation then the K1 for an inhibitor such as L-c-amino- butyrate, determined by the circular dichroism method Should be similar to that obtained in the catalytic assay in the presence of AMP and L-threonine. If activation has not been caused by protein concentration induced aggrega- tion the K1 for L-d-aminobutyrate obtained by circular dichroism should be quite high, as it is with AMP-free dehydrase in the catalytic determination of K1. In addition to determining K1 for L-c-aminobutyrate, inhibition constants for the inhibitors, D—threonine and DL-allothreonine, were also determined to check the valid— ity of the method. AMP was rigorously removed by chroma- tography of the dehydrase sample over a Sephadex G-25 133 column, as described in Methods. The K1 values were calculated from either of two plots (see Appendix II for derivation): 1 = C(K c (8) when - (2310.0) 01%, I + 15; AA. = K K K K (9) MW. - 0.0 ‘th'kf where QAAL,RO) is the circular dichroism value at 415 mu before the addition of any inhibitor,z§AL.R is the observed circular dichroism at any inhibitor concentration, K1 and K2 are the dissociation constants for the first two steps of the catalytic dehydration mechanism, E0 the initial con- centration of dehydrase and c, a proportionality constant. A typical plot using equation (9) is presented in Figure 17. The results of these determinations are presented in Table 9. As Table 9 indicates the K1 values obtained by the circular dichroism and kinetic methods are in close agree- ment, attesting to the validity of the circular dichroism assay. The values determined for L-d-aminobutyrate by the circular dichroism method are identical with those obtained in the catalytic assay. While K1 values for D-threonine, DL-allothreonine in both the presence and absence of AMP are somewhat higher (3—4 fold) than those values obtained kinetically, these discrepancies may result from the differing 134 .n0p0ada0aoaoapo0a0 0|>D\Qmo 00000 0 :: :0x0p 0:03 00::0009 Bmaon:0a0 HOH azonao .00dep:050 :0pdaoaod: 000daaoaaa0 :: 00000 003 0:::o0::9:0 .Ha 0.0 003 0asao> 0:0 “Ha\wa N.: 00 :o:p0:p:00:o0 0 p0 0shu:0 0:0 .0.0 we .nommsn 000:90o:a 0300000: 2 «.0 .02: z: 0.0 .Hopa0a:poa:pa0 z: 0.0 00::00:00 0ppobso :8 0H: 00 amaoa:0ua h0a50aao 0amu:m no 0000 w:«::000: a: 0:a:o0a:B:Q nopapa::H 0:» no: a: mo :oap0:aaa0p0m .mH 0:50am 135 véo Nfio 02.200050: ofio mod cod vod Nod d<<<+ mZ_ZOm¢I.To op 136 TABLE 9 Effect of AMP on XE of Threonine and K1 of Threonine Analogues From Kinetic_and Circular Dichroism Determinations Conditions for determination of K1 were essentially as described in Figures 2 and 17. _ ._¥ _- ——— Km or K1 (mM)a K1 (mM)b Substrate or Inhibitor -AMP +AMP -AMP +AMP LJThreonine 50-100 2,5 --- --- DLmAllothreonine 0.8 1.5 4.5 3.5 D-Threonlne 10.0 10.0 50.0 50.0 L-anmlnobutyrate >:1000 15.0 >»1000 15.0 aKinetic determination bCircular dichroism determination 137 conditions used for the catalytic and dichroism measure- ments; i.e., L-threonine is present for the catalytic determinations but not in the circular dichroism procedure. Again, the K1 for L-d-aminobutyrate obtained with AMP-free dehydrase at high protein concentration by the circular dichroism procedure, agrees with the value obtained at low protein concentration by kinetic means. This suggests that ologomerization does not, in itself, cause activation. Direct Determination of Km and V for max _§igh Protein:lnducgd Oligomer The most definitive means to determine whether oligo- merization causes activation is to compare the kinetic parameters of Km and Vmax of the oligomer induced by high protein concentration to that induced by AMP. Toward this end two assay systems were developed which could be employed to examine the kinetics of the dehydrase at high protein concentrations. The first of these was an end point assay, in which the dehydrase was incubated for a fixed period of time in a reaction mixture with L-threonine, 0.1 M potassium phosphate. pH 8.0. and 5 mM dithiothreitol, at 28°. The time interval for incubation varied between 30 seconds and 2 minutes. The a-ketobutyrate was subsequently determined by a procedure involving its conversion to a-hydroxybutyrate with lactic dehydrogenase and NADH oxidation (see Methods). 138 This assay was demonstrated to be quite reliable since it measured rates produced by 2000 dehydrase units in 0.2 ml. However, the above assay was limited in that the reaction rate was not followed directly with time and linearity of each assay therefore not guaranteed. Second, a stOp flow assay was employed in which the direct release of d-ketobutyrate was followed spectrophoto- metrically at 320 mu in a Gibson—Durrum Stop Flow Spectro- photometer. This assay was used at concentrations as high as 15,000 units per ml, which is equivalent to 1 mg of dehydrase per ml. Even at this extremely high dehydrase concentration valid catalytic velocities were obtained. as attested to by comparison of the specific activities deter- mined in the stOp flow spectrophotometric assay to those determined in the dilute coupled Spectrophotometric assay. All assays were performed at 28° with AMP rigorously removed from dehydrase samples by chromatography on Sephadex 6-25, as described in Methods. Km andvmax were evaluated from two of the transfor- mations of the Michaelis-Menten equation (93): i/v = i/Vmax + Km/Vmax'(1/S)8 S/v = Km/Vmax + (1/Vmax)s' A Lineweaver-Burke plot for the data obtained at 1 mg/ml (15,000 units/m1) in the stop flow assay is shown in Figure 18. The coefficient of correlation of this plot was 0.99. The Km and Vmax values for threonine dehydrase in the absence of AMP at varying protein concentration are summar- ized in Table 10. The Km values obtained at concentrations 139 Figure 18. A Lineweaver-Burke Plot of Velocity Versus Threonine Concentration Measured at High Enzyme Concentration in the Direct Spectrophotometric Assay Mixing chamber of the Gibson-Durrum st0p flow spec- trOphotometer contained 5.0 mM dithiothreitol, 0.1 M potas- sium phOSphate buffer, pH 8.0. threonine dehydrase at a concentration of 15,000 units per ml (1 mg of dehydrase/ml) and L-threonine at the concentration indicated; temperature was maintained at 28°. AMP was removed from the dehydrase sample on Sephadex G-25 as described in Methods. Reaction velocities were recorded as decreases in transmittance at 320 mu (c-ketobutyrate production) on a Tektronix Type 564 Storage Oscilloscope. Transmittancies were converted to absorbancies with the aid of a computor program which also calculated reaction velocities by linear least squares analysis. Coefficient of correlation of the plot shown was 0.99. 140 ISO THREONINE DEHYDRASE CONCENTRATION = l mg/ml l50r-Km = 225 mM —-—- IZO O) O l/Velocity (AA340,min)" 8 T 0 IS 32 48 64 80 l/(L-Threonine) (M4) 141 oma osnom oaapoaoposaonpomam lumaasoo Ha\w3 mmo.o oma mmm Azoaa aopmv oanpoaoponaoapomnm soapmapaoonoo manuao swam Ha\wa a son mma psaoaueam Ha\wa 0.0 «as and paaoaueam Haxwa mo.o Asaoponm.wa\aaa\moaoanv Azav mammw soapmnpsoocoo [I am onmnvhson. wma> F maampoo you mooSpoz op Houom .mmmmmm obopm on» no mGOapmnpnoocoo samponm wnuwnd> pa .omm um doahomnma one: madman HH< owmnumnon oonmumz< you use» use am no soapmsaahouon 0a mqmfia 102 of 0.05 mg/ml, 0.6 mg/ml, and 1 mg/ml are all 3-4 fold higher than the Km values obtained in the dilute coupled spectrophotometric assay at a protein concentration of 0.025 ug/ml. As may be recalled, a similar 3-h fold increment was noted with the K1 values which were deter- mined by circular dichroism at high protein concentration. It is apparent, nonetheless. that the dehydrase at concentrations as high as 1 mg/ml does not diSplay the high affinity observed with AMP-induced oligomer. This finding agrees with that obtained by the circular dichroism determinations. Oligomerization by itself is. thus, insuf- ficient to cause the allosteric activation. Molecularity of the.Allosteric Activation with Respect to Dehydrase ggncentration The activation of threonine dehydrase. under certain Specific conditions, is a relatively slow process which can be followed with respect to time directly in the coupled spectrophotometric assay. This is revealed upon examina- tion of Figure 19. Curve A in Figure 19 demonstrates the course of the reaction in the catalytic assay when AMP is added to the enzyme first, and the reaction is then initiated by the addition of L-threonine. There is an immediate onset of a-ketobutyrate production at the maximal linear rate. This same behavior is also noted when AMP and L-threonine are added simultaneously to cuvettes containing the dehydrase. 143 Figure 19. Time Course for the Activation of Threonine Dehydrase by AMP Reaction cuvettes contained 5.0 umoles of dithio- threitol, 0.04 umoles of NADH, 0.30 mg of lactic dehydro- genase. 75 umoles of potassium phosphate buffer. pH 8.0, and 0.7 units (0.0”? ug) of L-threonine dehydrase. First and second additions were then made to both reaction cuvettes as follows: Curve A, 5.0 umoles AMP first, followed by 20 umoles of L-threonine; Curve B, 20 umoles of L-threonine, followed by 5.0 umoles of AMP. The final volume was 1.0 ml. Loss of NADH was followed at 340 mu in a Gilford SpectrOphotometer, with temperature maintained at 28°. 114-1+ 3 l4 E T o 5!, 1.2 I— Curve A <1 AMP + L-Threo LU IO U) 55 a: 0.8 .. Q Is? Addmon UJ o 0.6 ———— — m 0 Curve B E: 1’ L-Threo 4' AMP m 04-— — - J a: ' , l 8 ’ 2nd Addition ‘13 L l 0.2 4 l 3 5 7 9 ll TIME (minutes) 145 Curve B of Figure 19, shows the rate when 20 mm L-threonine is added 3-5 minutes prior to AMP. The low activity observed initially is that characteristic of non- activated (high Km) dehydrase. Addition of AMP then causes a slow increase in dehydration rate which attains 65% the maximal velocity rate of Curve A. The instantaneous rate of threonine dehydration at any point along Curve B during activation should reflect the concentration of activated dehydrase. Hence, the increase in rate from near zero velocity to constant high velocity measures conversion of non-activated to activated dehydrase. By studying the rate of activation as a function of enzyme concentration, it should be possible to establish the molecularity of the activation with reSpect to dehydrase concentration, provided that the association step is rate limiting. Thus, another of the possible roles of oligomer- ization in activation can be tested; i.e.. whether oligo- merization is a necessary requisite of kinetic activation. The differential method to determine order is pre- sented below. It is recognized that this method is some- what limited since initial rates must be employed. A more detailed investigation was subsequently undertaken by Mr. John Gerlt in his laboratory in which order was examined with integrated, as well as, differential rate equations.5 5This work has been the subject of a presentation at the 1970 meetings of the Federation of American Societies for Experimental Biology (109) in Atlantic City, New Jersey. 146 The activation process can be eXpressed as: nB F.) An (Scheme II) where B is the concentration of the low activity form dehydrase (high Km), A is the concentration of the high activity form (low Km), and n is the stoichiometric coef- ficient: n will be greater than 1 if activation is depen- dent on oligomerization, and equal to 1 if activation is dependent only on isomerization. The kinetic studies were carried out according to the following procedure. The absorbance of the reaction mixture containing threonine, AMP-free dehydrase, lactic dehydrogenase, dithiothreiotol and NADH was measured for a short time (3—5 minutes) at 340 mu, with temperature main- tained at 28°. AMP was then added and the activation followed as an increase in absorbance until a constant rate of NADH oxidation was attained. Relatively low levels of enzyme were employed in order to obtain activation rates slow enough to measure with an ordinary Spectrophotometer. At any time during the activation process the con- centration of (A) is pr0portional to the rate of NADH dis- appearance: (A) = k' (’déiADH) (10) Values of 'déiépH) were determined from the chart recorder tracings of NADH loss versus time, where tangent lines were drawn to the curve traces at various times, t, with 147 the aid of a flat surface mirror. The slope of the tangent line at a particular time, t, correSponds to the lepe of -d(NADH) dt is plotted versus time for several enzyme concentrations the NADH loss versus time curve at that time. in Figure 20. From Scheme II the initial rate of activation should be: 3.1% = RB“ (11) or log.%%.= n log (B) + log k (12) 35 values were obtained from the initial slopes of plots of 'défiADHl versus time (Figure 20); n was then determined from a plot of log.%% versus log (B) (equation 12), Such a plot, with a slope of 2.03 is shown in Figure 21. This result indicates that at a minimum, a dimerization must be involved in the activation process. Findings of the kinetic investigation conducted by Mr. John Gerlt are in concurrence with this interpretation (109). Since earlier experiments had indicated that oligo- merization alone was insufficient to account for activation, activation must require both AMP binding and oligomerization. Activation bygAMP at High L-Threonine Concentration When dehydrase is preincubated with L-threonine at concentrations greater than 20 mM. the nature of the activa- tion is even further complicated than is the case at lower 148 .mmmnumsoo caduceus» mo coapmpapom no open on» paomoaaoa on com: ma Npo\AmQ42mmsosHm> one .uoHSmmoa ones mpsmwamp omen» mo mmaon on» cam aoanaa ooduhfim peak a mo pad on» Spas sedan one; mad» mamaob mmmoaoou mawz ho mobhso on» on nomad enemas» .pe\Ammdzvun mo mosamb sampno oe .Avu\fima Spas mafia no soapondm m we omeaomnon enanoonna mo scapmbapod .om oaswam $93525 met. w v o m... m a. o co _ _ 0.0 20.0 [slants Bo \ wnoo v _ .o 30.0 I 8.0 A A a 2.5 8.0 V8.0 In use: ONNG . . 8.0 80.0 \ moo ww/O‘Vevv 150 Figure 21. Determination of Order of Threonine Dehydrase Activation with Respect to Protein Concentra- tion by the Differential Method. Each point represents the initial slopes (AZA/minz) of the curves presented in Figure 20 plotted against initial protein concentration. 151 —,——.. Slope = 2.03 :. 0. m 0 8 Itl oo. 0. wee... 2 4mmm. 6. 4 2 , fo. x «2838343 8:384 so 23. 3:2. 4.0 8.0 (Units of Threonine Dehydrase) x 10' 2.0 I.O 152 threonine concentrations. As noted in Figure 22, after the addition of AMP there occurs a decrease in the rate of a-ketobutyrate production. This is followed by a slow increase in the rate of dehydration, with the final velocity attained, again, being equal to 65% the velocity observed with AMP preincubated dehydrase. This finding suggests that AMP may cause conforma- tional alteration of the dehydrase even before AMP-mediated activation (Km alteration) is achieved. The behavior of threonine dehydrase at high L-threonine concentrations was, however, not investigated in any greater detail in the current investigation. Kinetic Plots for Reversed Order of Addition The observation that AMP activation was markedly different where L-threonine is either present or not present prior to the addition of AMP. led to the Specula- tion that the kinetic behavior of the dehydrase might also be different for these two orders of addition. As a result, the reaction kinetics were examined further under conditions where L-threonine was added prior to AMP. As was observed with dehydrase preincubated in.AMP (section 1 of Results) a hyperbolic velocity versus L-threo- nine plot was obtained (Figure 23); a Hill plot of this data had a slope of 1.0 (Figure 23). The Km for L-threonine was 1.3 mM which agrees extremely well with the value of 2 mM reported in section 1. 153 Figure 22. Time Course for the Activation of Threonine Dehydrase by AMP at High L-Threonine Concen- tration Conditions were as described in Figure 19, except that the reaction cuvette contained 100 umoles of L-threonine. 15a l2 II/l .0. :9 e ...-I w e m m .m —I.I ( I. 6 E MW M mA T m. , 4 2 0 8 We 22 .. .. o w o. o 18 ovm ._.< mmdmmomo wozqmmommq 155 .msoapmHSono on» :« pom: one: mz< mo sodpaoem nouns vocamppm moves Hdaawma one .owm no: manpmnoaaou «m24 on aoaaa mopssaa w some ohms osdmoohspuq no msodpmapsoosoo opmanaoaaam on» no God» named on» page paooxo apogee: ad oonanomou we who: momma one no msoapdosoo osasooasalq spas oomeSoadoam oamwsm you poam Haam use osazoonneuq mamao> mpaooao> .mN onswdm 156 (JL- A /./\.) 60'] . On #0 N- 35:85-3 o3 m5 om. oh- m- 38v m. 36085.74 0. m ON- ro '0. T 8 CP 8, E ‘2 ‘2 to 0. 8 157 A plot of velocity versus AMP concentration for dehydrase preincubated in L-threonine was Sigmoidal, with an inflection in the curve noted at 0.5 mM AMP, the K8 for AMP (Figure 24). This behavior is in marked contrast to the hyperbolic dependence observed when.AMP was added prior to L-threonine. A Hill plot of the velocity versus AMP data had a maximum lepe of 1.84 (Figure 24); however, the shape of the Hill plot was atypical. Rather than exhibiting a maximum lepe at the mid-point of the plot (at the K8 concentration AMP) the slope increased continu- ally with AMP concentration. Further, when the Hill coeffi- cient was used to plot the Lineweaver-Burke relationship, i/velocity versus 1/(AMP)n a linear curve was not obtained, as would have been expected had the sigmoidal velocity dependence resulted from cOOperative binding interactions (97). These observations may indicate that the observed Hill slope of 1.84 does not reflect true cooperativity as seen with many other allosteric enzymes (1-6) but rather may reflect the protein dependent second order oligomeriza- tion process. .mQOapmHSono on» how names on» ooahom m2< mo coauappm on» songs dosamppm mopmh Hmaaxma one .owm was seapomoa on» no whopmaoaaop “mzw no :oaumapcoosoo opmaaaoaaam on» no 158 soapaoom on» on house mopszaa m moppobdo soapomma map on poops was ocaaomHSpnA 25 cm page paooxo moosuoz ca pmnahomop we who; momma 0;» mo mnoapapcoo ondzoohnala spas oopmnzosamam osmucm How poam Haam use mz< m5mho> mpaooao> .dN mazmdm 159 (A-A M) 50‘: $.23 8.. Eel n34 .O._ mud- Om..- DEN: v... 0.. _ 00 On. _ m5. 0. 0.0 05.. II: VQ. u 82m E:E_xo_2 O.” W P h - (”’w/OV‘EVV) Moons/x DISCUSSION The primary goal of this investigation was (a) to determine the relationship between activation and oligo- merization of the dehydrase, and (b) to determine the mechanistic step which is altered in the allosteric transi- tion to account for the 25-35 fold decrease in the Km for threonine. These latter studies, however, necessitated detailed characterization of both the kinetics of activa- tion and the oligomerization process. Thus, we began with an examination of the kinetics of the dehydrase and with sucrose gradient and molecular weight experiments in order to determine the nature and form of the oligomerization changes. Quaternary Structure of Threonine Dehydrase The changes in sedimentation velocity and molecular weight observed in sucrose gradients and by gel filtration (64) are considered to result from an equilibrium between monomer and oligomer forms of the dehydrase. Based upon s20 values the molecular weight of the monomer is around 40,000 while that of the oligomer approaches 160,000. These findings suggest that the oligomer is a tetramer composed of four monomers. Estimations of pyridoxal phosphate content (65, 108, 110) 160 161 indicated that the dehydrase contains four pyridoxal phos- phate molecules per tetramer. Shizuta gt_gl. (65) have suggested from equilibrium dialysis experiments that the dehydrase also possesses four AMP regulatory sites per mole of tetramer. The monomer would then be the smallest dehydrase Species which could possess both a catalytic and regulatory site. Effect of AMP and Enzyme Concentration on Quaterna;ygStructure The equilibrium between the monomeric and tetrameric forms of the dehydrase is Shifted toward the tetramer by either AMP or high dehydrase concentration. When the apparent 820 values are plotted against dehydrase concentra- tion a smooth transition is observed with the 320 values approaching a value around 8 S at high protein concentra— tions. This correSpondS to a molecular weight of approxi- mately 160,000 for a protein with average partial Specific volume. The intermediate 820 values observed probably result from a rapid equilibrium between monomeric and tetra- meric forms of the dehydrase, with the statistical distribu- tion among the forms being reflected in the observed 820 value. Gilbert (62) has considered this kind of equilibrium and Shown that rapid equilibrium between two Species results in a single peak. ~ A smooth transition in $20 value was observed by 162 Phillips and Wood (61) when.AMP concentration was varied in gradients over the range of 10‘7 to 10"3 M. It was of obvious importance to determine the molecu- lar form of the enzyme at the dehydrase concentrations present in the coupled-Spectrophotometric catalytic assay. With 0.2 units of dehydrase per gradient (an amount of enzyme which gives a valid assay rate in the coupled-Spec- trophotometric assay) the 820 value with.AMP was 7.4 S and without AMP 3.5 8. Experiments with 0.5 unit per gradient and with a composition similar to that of the catalytic assay gave s20 values of 7.2 S and 2.9 S with and without AMP. Therefore, under the conditions employed in the nor- mal catalytic assay, the dehydrase assumes the monomeric state in the absence of AMP and the oligomeric state in the presence of AMP. However, because of the long time period necessary for sucrose gradient runs it is not possible to establish from these experiments whether the oligomerization is achieved before or after the AMP-asso- ciated kinetic activation. Relation of Quaternary Structure to Catalytic Activigy and Allosteric Activation The results of the sucrose gradient investigations, as well as, ultracentrifuge analysis (65, 108) indicate that at concentrations of 1 mg/ml the AMP-free dehydrase may have the same quaternary structure as dehydrase in the presence 163 of 5 mM AMP. However, the AMP-promoted oligomerization is accompanied by a 25-35 fold decrease in the Km for L-threo- nine. If oligomerization were a sufficient requisite for allosteric activation, a large decrease in Km should also occur when the dehydrase associates as a result of increased dehydrase concentration. Therefore, the activation state of the oligomer induced by high protein concentration was determined by two means (a) evaluation of K1 for certain competitive inhibitors by a non-kinetic method, that is, from circular dichroism observations, and (b) kinetic determination of Km at concentrations as high as 15,000 units/ml (1 mg/ml) with an end point and stop flow assay. The K1 and Km values obtained with the oligomer induced by high protein concentration were those characteristic of low activity dehydrase; that is, the Km for L-threonine was very high (225 mM) and the‘Vmax was near that expected. Thus, oligomerization by itself does not account for allosteric activation. Nonetheless, allosteric activation could require binding of AMP to the dehydrase, as well as, oligomerization. When AMP is added to the dehydrase subsequent to the substrate, the activation of the enzyme proceeds as a rela- tively slow process which can be followed in the Spectro- photometric assay at dilute enzyme concentration. From this, the molecularity of the activation process with respect to dehydrase concentration may be determined. 164 Both the preliminary studies, which have been reported here, and the more detailed characterization con- ducted subsequently in this laboratory (109) indicate that the rate limiting step of the activation process is second order with reSpect to the enzyme. AS a result, both AMP binding and oligomerization must be considered as obliga- tory steps in the allosteric activation under these condi- tions. The question next arises as to whether AMP binds before or after the oligomerization step. Even though the AMP-induced catalytic activation is a relatively Slow process, addition of AMP to the reaction cuvette has an immediate effect upon the kinetic behavior of threonine dehydrase (at least at high L-threonine concentrations): the rate of threonine dehydration noticeably decreases upon addition of AMP with the rate of this initial "inactivation" being too fast to follow in the coupled-Spectrophotometric assay (Figure 24). This observation strongly indicates that AMP binds to the dehydrase prior to the oligomeriza— tion. At this point it must be emphasized that while oligo- merization has been demonstrated to be an obligatory step in the allosteric activation with threonine preincubated dehydrase, whether oligomerization is obligatory for acti- vation with.AMP preincubated dehydrase has not been estab- lished. Under the latter conditions the rate of activation 165 is too fast to follow with an ordinary Spectrophotometer. It is plausible that oligomer formation is required for activation, also, under the latter conditions. It has been shown that velocity versus L-threonine and velocity versus AMP plots are hyperbolic under condi- tions where AMP is added to the dehydrase prior to 1, threonine; however, when L-threonine is included with the dehydrase prior to AMP the velocity versus L-threonine plot remains hyperbolic while the velocity versus AMP plot is sigmoidal. Although a slope of 1.84 was obtained from a Hill plot of the latter data, the atypical shape of this Hill plot suggests that the Hill coefficient does not reflect cOOperative binding, but rather may reflect the obligatory oligomerization step necessary for kinetic acti- vation. The maximal slope inthe Hill plot was not attained at the mid-point of the curve, but rather, the Slope increased continually with rising AMP concentration. If AMP binding is truely non-cOOperative then direct mea- surement of AMP binding with L-threonine-preincubated enzyme Should reveal hyperbolic binding. One additional ligand-associated conformational alteration Should be considered at this point. As has already been mentioned, when AMP and L-threonine are added to threonine dehydrase either simultaneously, or with AMP preceeding L-threonine (Figure 19) the dehydration of L- threonine commences at a rate characteristic of low Km 166 ("activated") dehydrase: that is, no lag is detected before the final reaction velocity is attained. However, when the dehydrase is preincubated in 20 mM L-threonine, kinetic activation occurs at a Slow rate after addition of AMP. If this altered behavior in the nature of threonine dehydrase activation was due entirely to L-threonine binding to the catalytic Site, then at 20 mM L-threonine concentration only a small percentage of the enzyme should exhibit the Slow activation behavior. This is due to the fact that with the Km for L-threonine in the absence of AMP being 50-70 mM. only 25% of the catalytic Sites should be occupied with threonine at 20 mM concentration. With 75% of the enzyme free of L-threonine, the shape of the activa- tion curve detected upon addition of AMP should closely resemble curve A of Figure 193 that is, activation should be quite rapid for a very large percentage of the dehydrase. This is clearly not the case. Hence, L—threonine may bind to an alternative non-catalytic Site on the dehydrase, having an affinity for L—threonine considerably higher than the catalytic site. Physical chemical studies would appear warranted in the future to more fully evaluate the nature of the high affinity for L-threonine than can be deduced from this line of reasoning. Comparison to Theoretical Models The reported observations on biodegradative threonine dehydrase may now be considered in relation to the theoretical 167 models prOposed for allosteric behavior. It is clear that any model used to explain the behavior of threonine dehydrase must account for the following phenomena associ- ated with dehydrase catalysis and activation: (a) the lack of substrate c00perative homotrOpic interactions, which result in sigmoidal rate versus substrate concentra- tion plots; (b) a possible lack of effector cooperative homotrOpic interactions; (0) the protomer-oligomer equilib- rium of the dehydrase; (d) the ability to form a non- activated oligomer and (e) an obligatory requirement for oligomerization in the activation process. For numerous reasons the observed behavior of threo- nine dehydrase is inconsistent with the concerted transi- tion model pr0posed by Monod, Wyman, and Changeux (23) (see literature review). This model assumes that all forms of a regulatory enzyme are accessible even in the complete absence of ligands: symmetry must, however, be maintained in each oligomer at all times. The develOpment of the concerted transition model has led to a number of predictions which include a require- ment (a) that all regulatory enzymes exhibiting altered Km upon allosteric activation or inhibition Show coopera- tive homotropic substrate binding interactions. Threonine dehydrase does not under any conditions exhibit Substrate cooperative behavior. At least under the conditions where AMP is incubated with the dehydrase prior to Substrate, the 168 enzyme also fails to reveal effector cooperativity. In addition, the concerted transition model was intended to explain the behavior of regulatory enzymes for which the degree of oligomerization does not change upon allosteric modification. This is not, however, necessarily the case with threonine dehydrase. The sequential induced fit model was advanced by Koshland, Nemethy and.Pilmer (21) as an extension of induced fit theory (see Literature Review). The hetero- tropic effector molecule is assumed in this model to bind to the regulatory enzyme and as a consequence of this asso- ciation induce the allosteric alteration. Molecular interactions between subunits are then used as a basis to explain both positive and negative cooperativity. However, neither the substrate nor effector cooperative behavior need be present to account for heterotropic activation or inhibition. AS may be recalled, homotropic behavior is not always associated with.AMP activation of threonine dehydrase. While this model more satisfactorily accounts for threonine dehydrase behavior than the concerted transi- tion model, the Sequential model, as described by Koshland, Nemethy and Filmer (21) is also limited to isomerization rather than association-dissociation processes. Thus, in a sense this model is still too restrictive to include threonine dehydrase. Several theoretical papers have appeared which mathe- matically treat allosteric models on the basis of association- 169 dissociation equilibria (111-116). These treatments all assume pre-existing equilibria and primarily eXplain homo- tr0pic behavior. Therefore, they do not adequately con- sider the situation for threonine dehydrase. Proposed Model for Activation of Threonine Dehydrase A model to account for all the observed molecular changes in threonine dehydrase is presented in Figure 25. For the sake of clarity Species higher than dimer are not shown in this figure; nonetheless, equilibria between dimer and higher oligomer forms (tetramer) are assumed to exist. Each of the three branches in Figure 25 will be considered in turn in the following description: Branch A: At low enzyme concentration in the absence of AMP, the enzyme exists as a monomer (molecular weight 37,000-40.000). Increasing the dehydrase concentration causes association of this enzyme Species; however, the kinetic prOperties of the resulting oligomer are the same as those of the AMP-free monomer, i.e., a relatively high Km for L-threonine is observed. Branch.B: Addition of L-threonine to the AMP-free monomer causes a conformational transition resulting in a new monomeric Species. This monomer exhibits catalytic activity and a Km for Lmthreonine of 50-70 mM (the "non- activated" state of the dehydrase). This monomeric Species differs from the unliganded monomer in the longer time 170 nodpmbapod capopmoaa< one mnoamaoeaoonopaH Hoaowaaonnoaocoz you Homo: oomoaonm .mm ohsmam 171 A230 0.25 9 u .. L if @fi /.. .50. ....o... 2...... 0:040 E2000 :9: $24 1; A. .300 302. 23:00 E... + Ex—, 30. Sou .60: Ex :2: m Es .L.r llly \@ IL 4. V .2: Em All. AIII. Pu Leap—m" . a2... Agave 30.5 < 05:5: 38 :24. 172 interval required for its conversion to the low Km Species (see Branch C below); that is, AMP-induced activation of this monomer proceeds as a Slow rather than a rapid process. Kinetic evidence already discussed has suggested that L- threonine may be bound to a non-catalytic Site, as well as, to the catalytic site. Upon addition of AMP, the threonine-monomer complex is converted to a third monomer form, diSplaying even less catalytic activity than the former Species (as noted in Figure 22). Subsequently, two threonine-AMP-monomer com- plexes combine to yield a dimerized dehydrase which exhibits the low Km for L-threonine characteristic of activated dehydrase. This dimerization is the rate limiting step of Branch B. Branch C: Addition of AMP to non-liganded monomeric dehydrase causes rapid alteration to a form of the dehydrase which exhibits high affinity (low Km) for L-threonine and oligomerization of the dehydrase. However, it has not as yet been established whether in Branch C, as in Branch B, oligomerization must precede formation of the low Km Species. The activated dimer Species of Branch C and Branch B are symbolized differently so as to distinquish these Species from each other. While both oligomers exhibit the same Km and Similar Vmaxv it has not yet been established whether these Species are otherwise identical. It is of interest to consider whether the association- /— 173 dissociation process is necessarily involved in the activa- tion mechanism for threonine dehydrase under conditions which apply in the organism. Estimation of dehydrase con- tent in cells of'g.lggli indicates that this enzyme is present in rather high concentrations, possible as high as 0.5 mg/ml. At such a concentration the dehydrase may exist as an oligomer even when the AMP level decreases very appreciably. Under such conditions AMP promoted activation may involve only isomerization of the already oligomerized forms of the dehydrase (as shown on the far right of Figure 25). The exact activation mechanism which applies under physiological conditions, however, remains to be elucidated. Other Regulatory Enzymes Diaplaying Association- Dissociation Equilibria Some mention Should be made of the other regulatory enzymes which have been shown to undergo association-disso- ciation reactions. A list of these enzymes is presented in Table 11. Numerous factors have been shown to affect the state of oligomerization of these enzymes including modifier type (61, 63, 64, 69, 112, 117-131, 133-135). substrate (118, 120, 132), coenzyme (135). pH (118), protein concentration (136, 137, 64) and buffer type (69). In certain cases the association-dissociation process has been shown not to be 174 .11) [I] s) Soapmeapom Mom sodpmao an Bondsman manopmwaano Iommm momsmo Aafloo mazoahonommv names» mane ma zodpmuaaoaowdao neopm>d00< mam ommaphnom enasoohne Soapmdo Ihfisammoaoq Iommde chemo 10pm» azaoanpmoaov mo .mm whopm>a00< mm< ommaeanom osdsooase soapmaoom oopmowwSm undo omsmo mz< Aoaonsa pannmhv mHH cosmoaoaop pooado who0m>Ap0< .omoosHu a ommamnonamonm sodpmaoom undo omsmo hHHcomhoao ++w2 . ++CZ GHHOHUNHOOPMGHQV oopmommSm opmhpaooma opmapdo ondSoMOHohsom mad ooaoosoaoo pooaam .opmhpao .m24 cesspfioomH Uopmomw5m HopanassA Soapmaoommdp you soapmdoommao mmmam0 odow Anemone snowmoasozv maa no commemoaoo pooadm HopflnassH cadasmHSpn< ommposuSmm madampaao noapmaswom SA mmoooam soapmaoomdm opmpm cocoaomom Isoapmaoomm< Sodpmmonmm< mao00>apo¢ whopdnasaH mahusm E E mo pzoaobaobsH SA mowamso mawpaaaamw Sodpmdummndmnaoapmaoommw madmandswm moahuam whopmaswom AH mamQB 175 sodpmapsooSoo adopoaa swan pm :odpmao nonmao omzmo 0:00Soaod hapoonao wpoommo Aambaa osd>omv omdsow maa 0Haoapoaos nopanassH whopandssH mm< moo .meo nosomzom mamampsao I. .lkwwcomhoflw eoaeoao oaaoaemaoopooamv Iceman omzmo +Q¢2 mew ommcow oma maopapaeeH .maa .ma4 .mao .aae -oaoasoa measopaao :oApmao nomon omsmo mmauwma HopHQASSH mace aged omegaamom mace Anaemmap Soapmao Hmaasm msoaambv Iommm momsmo pace ommamxon nuanmma nonopapoq oaaaao name sou aspoo< soapmao Apnoea ondbomv towns momdmo enscow HNH nopoeapoq age :94 uoaoaaoo opoapaoomH opmnpnnSu use mnopm>apom an commoboh mass baseman ooemomwsm gonads soaps uoaauoaia asaaanammooosmv manna you coaumaoommm uaoommm 00500 endowed ommsowoachaon end no consummate pooaam maopanassH noman czasooasalu osdnomoaom sodpmdo AmopaonSQmao zmaamv nommm omsmo omdaowoapmnoa mma maopseapoa mzae .zma opoaamosmwouomooaao . AAHOO .mv coapmao composwom momma mmamo m990 .m900 macspo opmgamonaam ems meopapaaea .maqo .aa< mafia .aaao ooapooaosaonam soapm>dpom use sodpmdoommm mac .mmop I codenaasea neon woe omaeo maop .maoo .maeo AaHoo .mv UopmmwwSm abapmaoommm smpapom one .mm4o .mzup wasp ommzam fiw mma no oesoosoeoo pooaAo maopanasaH .mooo .aaoo .maeo .meao oaaoaaaaphxooo 1 soapmao nommm 00:00 I ++ma as: .ea« Aafloo .mv and opoeponsm mac omopozpasm :90 Soapmdo AHHoo .MV nomwm omsmo ommpospsam ems .oma eeooeoaoo eoz oaopoeaeoe .mzH as: opesamonm_flasonnoo soapmaswom SH mmooonm Sodpmaoommam mumpm monoaomom uaoapmaoomm< codpmmoamww mHOpmbapo< neonapassH oahwzm mo psoaobaosz SA mowsmso Aooseapeoov .HH mamas 177 involved in the allosteric regulatory mechanism (131). For example, IMP and ornithine can cause both oligomerization and activation of carbamyl phOSphate synthetase. However, at low enzyme concentration the monomeric Species is always present regardless of modifier. Thus, oligomer formation is not necessary for either catalytic activity or manifes- tation of the positive allosteric effects. Frieden and Colman (112) have suggested that the association-dissociation reaction of glutamate dehydrogen- ase may be important in the regulatory mechanism of this enzyme. They demonstrated with both kinetic and direct binding studies that homotropic interactions (but not heterotropic interactions) of the allosteric inhibitors GDP and GTP are dependent upon the monomer-polymer equilib- rium. For the biodegradative threonine dehydrase the regulatory (heterotrOpic) behavior of the enzyme is obliga- torily dependent upon association. Iwatsuki and Okazaki (133) have suggested that oligomer formation may be required for both allosteric activation and inhibition of deoxythymidine kinase of _E_. 233$. However, the evidence obtained by these authors did not eliminate the possibility that oligomerization may occur subsequent to allosteric modification. 178 Effect of the Allosteric Alteration on the Catalytic Mechanism The most significant finding of the current investi- gation was the determination of the precise step in the catalytic mechanism which is altered as a consequence of activator binding. This was done by observing the effect of AMP on (a) "Spectral" and "dichroic" intermediates in the catalytic mechanism, and (b) dehydrase affinity toward competitive inhibitors which cannot undergo all of the partial steps of catalysis. Evidence from Circular Dichroism Measurements - Threonine dehydrase exhibits positive circular dichroism with a maxi- mum near 415 mu (59, 71); this circular dichroism disappears when L-threonine is added and returns when threonine is exhausted. The loss of circular dichroism is attributed to transaldimination between the internal lysyl-pyridoxal phOSphate aldimine and dehydrase bound L-threonine. This is suggested from the observation that competitive inhib- itors which possess an d—amino group are able to evoke Similar loss in enzyme circular dichroism at 415 mu, whereas, competitive inhibitors which lack the d-amino group cause no change in dichroism. AMP causes a marked increase in substrate-induced loss of circular dichroism. This finding suggests that AMP enhances a step before the return of the lysyl-pyridoxal 179 phoSphate aldimine in the catalytic mechanism (the second transaldimination of Figure 1), i.e., non-covalent threo- nine binding, transaldimination with threonine, or dehy- dration of threonine-pyridoxal phOSphate aldimine. Evidencaafrom.Absorption Measurements at 455 mu - During catalysis the absorption maximum of threonine dehydrase at 413 mu is Shifted to 455 mm (60, 71); this peak subse- quently returns to 413 mu as substrate becomes depleted. From the following lines of evidence we suggest that this absorbancy band reflects the accumulation of a pyridoxal phOSphate-bound dehydrated intermediate, and that this intermediate is the pyridoxal Schiff-base of d-aminocroton- ate: (a) the competitive inhibitors D-threonine, DL-allo- threonine, and L-a-aminobutyrate which are not dehydrated, do cause loss in circular dichroism and these do not cause the appearance of any absorption at 455 mu; (b) an absorbance shift to longer wavelength would be expected from extension of the conjugated double bond structure of the Schiff-base by dehydration; (c) Goldberg and Baldwin (99) have shown that the intermediate absorbing at 468 mu in the tryptophan synthe- tase reaction is probably the pyridoxal phoSphate Schiff- base of a-aminoacrylate, the three carbon analogue of d-aminocrotonate. The finding that AMP markedly increases the extent 180 to which the absorptive intermediate at 455 mu is accumu- lated indicates that AMP enhances a step in the dehydration mechanism involved in formation of the 0-8 elimination product, i.e., one or more of the first four partial catalytic steps. Affinity£gf Competitive Inhibitors - AMP was observed to cause a marked increase in dehydrase affinity for a number of substrate analogues, including Dips-hydroxybutyrate, DL-B-hydroxybutyrate, y-hydroxybutyrate, L-lactate, propi- onate and butyrate. These inhibitors are incapable of undergoing the second step of catalysis (transaldimination with pyridoxal phoSphate of the dehydrase). Thus, the major affect of AMP in allosteric activation of threonine dehydrase must be exerted on non-covalent enzyme-substrate binding. It is not possible at this time to establish whether the partial reactions of transaldimination and dehydration are to some degree also affected by AMP. However, an increase in non-covalent binding alone would be sufficient to enhance the extent of transaldimination and the effective rate of dehydration and to cause the changes in absorbancy and dichroism observed. Further, as equations 1 and 3 reveal, an increase in non-covalent binding of the order noted with the competitive inhibitors is sufficient to elicit the observed 25-35 fold change in Km for L-threonine. 181 Structural Requisite for Inhibitor Binding to the Catalytic Site of Threonineagehzdrase The investigations undertaken with analogues of L-threonine revealed that two distinct classes of competi- tive inhibitors exist for threonine dehydrase. Inhibitors of the first class diSplay a marked increase in dehydrase affinity mediated by the allosteric activator AMP. In all cases where separated enantiomers were available for experiment, only the L-conformers were found to belong to this category of inhibitors. Specificity requirements of this class of inhibitors are not otherwise very restrictive. Analogues of L-threo- nine and L-Serine, in which either the c-amino group or the B-hydroxyl group is removed, or in which the hydroxyl group is either on the a or Y carbon still diSplay appreciable dehydrase affinity in the presence of AMP. Even pr0pionate and butyrate bind reasonably well to the dehydrase. The affinity exhibited toward the second class of inhibitors is uneffected by AMP. The Specificity require- ments of these inhibitors is much more restrictive than for those of the first class. All of these inhibitors were of the D-configuration, and only those D-analogues possessing an a-amino group, as well as, a hydroxyl substituent in the B or y position (or a substituted hydroxyl in the 8 position) showed appreciable affinity toward the dehydrase. The high affinity observed with this class of 182 inhibitors suggests that the affinity exhibited toward these analogues is always that of the "activated" state, regardless of the presence of AMP. It would appear that in the non-activated state (high Km) the binding affinity exhibited toward substrate and substrate analogues of the L-configuration is greatly impeded by a structural constrant of the enzyme. This constrant does not restrict the dehydrase in regard to D-analogue binding. The end result of the allosteric transition mediated by AMP is to remove this restrant upon substrate and L—analogue affinity and to permit the substrate and L-analogue to bind with the same facility as the D-analogue. It is also possible that some of the inhibition detected with the substrate analogues, eSpecially with the D-analogues, may occur at the non-catalytic threonine Site which has been postulated above. However, further investi- gation will be necessary before any such definitive assign- ments can be made. SUMMARY The kinetic prOpertieS and protomer-oligomer inter- conversions of highly purified biodegradative L-threonine dehydrase were examined. AMP caused a 25—35 fold decrease in the Km for L-threonine and a 3-4 fold stimulation of 'Vmax- The Km values for L-threonine were 2 mM and 50-70 mM. in the presence and absence of AMP, reSpectively. The Ka for AMP was 5 x 10'4'M. The dehydrase diSplayed neither substrate nor effector homotrOpic (cooperative) interac- tions. Hill coefficients of near 1 were obtained in velocity measurements with both L-threonine and AMP. From sucrose gradient and molecular sieve chromato- graphy the dehydrase was found to exist as a monomer of 3.1-3.6 S, or 40,000 approximate molecular weight, at both low protein concentration and in the absence of AMP. AMP. as well as, increased dehydrase concentration caused oligo- merization of the dehydrase. Species as large as 8 S, correSponding to tetramer, were observed. In sucrose gradients, which approximated the conditions of the coupled- SpectrOphotometric assay in both composition and dehydrase level, the 820 values were characteristic of the monomer form in the absence of AMP and the tetramer form with.AMP. Oxidation yielded an inactive monomer of 3.2 S; neither AMP nor increased enzyme concentration caused 183 184 association of this Species. Reduction with dithiothreitol reactivated the dehydrase and restored the monomer-oligomer equilibrium. A number of experiments were performed for the pur- pose of elucidating the mechanistic step(s) altered by the allosteric activation. AMP was found to cause; (1) a marked increase in substrate-induced loss of circular dichroism at 415 mu; this loss of circular dichroism is attributed to transaldimination of the substrate with pyri- doxal phOSphate of the dehydrase; (2) a marked enhancement in the accumulation of an enzyme-substrate intermediate which absorbs at 455 mu, and considered to be d-aminocroton- ate-pyridoxal phOSphate azomethine; and (3) an increase in the affinity of the dehydrase for competitive inhibitors which are unable to undergo dehydration, as well as, several inhibitors which lack a—amino substituents and, therefore, are incapable of undergoing transaldimination. 0n the basis of the above observations it was estab- lished that the major effect of AMP in alteration of the catalytic mechanism occurs on the first step of catalysis; i.e., on non-covalent enzyme-substrate complex formation. In order to evaluate the role of oligomer formation in the allosteric activation the K1 for certain competitive inhibitors, as well as, the Km for L-threonine and Vmax were determined for oligomer induced by high dehydrase concentration. The K1. Km and Vmax values obtained were 185 similar to those for the AMP-free monomer. On this basis, it was concluded that oligomer formation cannot, by itself, account for activation. Through examination of the kinetics of activation the molecularity of this process was found to be second order with reSpect to enzyme concentration. Hence, oligo- merization is necessary but not sufficient for activation. An investigation with analogues of AMP revealed that considerable alteration in the base moiety of the nucleotide could be tolerated; however, the requisites of the phoSpho- ribosyl moiety were very Specific. A large number of L-threonine analogues were found to function effectively as competitive inhibitors of the de- hydrase. Analogues of the L-Series diSplayed appreciable dehydrase affinity only in the presence of AMP, whereas, for those inhibitors of the D-Series binding affinity was not deleteriously affected by the removal of AMP. Structural requirements for D-analogue binding were much more restrictive than for the Lpanalogues. No one position on the L-analogues was found to be all important. A model to account for the alterations in quaternary structure and the allosteric activation is proposed. APPENDIX APPENDIX Derivation of Km and'Vmax for the Threonine “—— ————7 Dehydrase Reaction The overall mechanism of threonine dehydration may be represented by the following scheme: K1 K2 K3 K4 K5 K6 E + S.;=£:E81 ;:2:ESZ ;=£?ES3 ;=£:ESu,;=£:ESS ;=£:E + P (Scheme I) where K1 through K6 are the dissociation constants corres- ponding to the partial reactions: (Ef)(sf) (E81) . (332) K a: 3 K: = __....__ ; I = 3 1 (351) 2 (E82) X T355) (ES ) (ESu) K = ; K A 4 (334) 5 (E35) The relationship of each of the above to intermediates in Figure 1, is given on page 92 , and summarized here: E81, the non-covalent enzyme substrate complex; E82 the threonine- pyridoxal phOSphate azomethine; E83 the conjugate base of threonine-pyridoxal phoSphate azomethine; E84, aminocrotonate- pyridoxal phoSphate-azomethinq;and E85, non-covalently bound aminocrotonate and any other enzyme bound rearrangement product of aminocrotonate, such as iminobutyrate. Ef is the 186 18? concentration of free enzyme, and Sr the concentration of free threonine. The kinetic functions of Km and‘Vmax are derived here from an equilibrium assumption for the two cases where the rate limiting step of dehydration is assumed to be either the second transaldimination, or some step asso- a. ciated with product dissociation. Arguments for the rate limiting step occurring at either of these phases of F dehydration are presented on page 114: I The case in which transaldimination is the rate E; limiting step will be considered first. In this Situation the velocity of the overall dehydration reaction may be ‘ expressed as assuming that there is no appreciable back reaction from E85 to E84, and that k5 is sufficiently slower than all other forward rate constants. As long as the concentration of Substrate is in suf- ficient excess over the concentration of total enzyme, then (8,) = (S) The following relationship between Ef and E0, the initial enzyme concentration holds providing that enzyme- substrate complexes which lie subsequent to the rate limiting step do not accumulate to any appreciable extent: 188 (Ef) = E0 - E81 - 332 - £33 - E84 Then, (Ef)(3f) (Eo ' ES1 ' ES2 - E33 - E54)S K = = (A2) 1 (E81) ES1 or (E31)K1 = EO(S) - (E81)S - (E82)S - (E83)S - (ESu)S Since E53 = Ku‘ESu); E32 = K3(E33) = K3Kh(E84); (E81) = K2K3K4(E8u) K1K2K3K4(E34) = EOS - K2K3K4(E34)S - K3K4(E34)S - K4>(EIX) , and E = Eo - (EIX) 190 [E0 - (EIXU (I) EOI K = (EIX) (EIx) i 1 K 1 WI; = mo + E"; (A12) The loss in circular dichroism at 415 mu is propor- ‘tional to the concentration of enzyme inhibitor complexes, that is, (AAL,RO - AAL,R) = C(EI)X then, 1/(AAL.RO .. AALR) ... K1/0E0(I) + i/oEo (A13) *‘J Alternative Derivation - AS has been discussed on page 123 , enzyme-inhibitor complexes beyond the amino acid-pyridoxal phOSphate aldimine (E12, in relation to Figure 1) are not considered to accumulate appreciably. The interaction of inhibitor with enzyme may thus be represented as follows: E + I :2 (E11) : (E12) (Scheme IIIb) where E I K1 ... {EIUJL (A14) K2 = (Ell) (A15) (E12) and E = Eo - EIl - EI2 Substituting in equation A14, 191 30(1) - (E11)(I) - (BIZ)(I) _ (E11) EO(I) - (EI1)(I) - (EI2)(I) = K1(EII) (I)(Eo - EIZ) = (EIl)(K1 + I) (I)(Eo - E12) (E11) K1 + I (A16) (P and substituting equation (A15) into (A16) (I)(Eo ‘ E12) (K1 + I)(EIZ) ‘? (I)(Eo) = (EI2)(I + KIKZ + 1K2) 1 K 1 K K . FIE = ‘23": * 101327. (A17) E K K 0 1 2 Q = (K2 + 1) + (Alb) (E12) (I) The initial enzyme concentration is proportional to the initial circular dichroism, i.e.. AAL'RO =2 CEO (A19) If the non-covalent complex of inhibitor with enzyme (EIl), like native enzyme, exhibits positive circular dichroism at 415 mu, then only accumulation of E12, the amino acid pyridoxal phOSphate azomethine would cause a decrease in the circular dichroism at 415 me. Then, AAL’RO . AIME = c(EI2) (A20) Substituting (A6) and (A7) in (A5) 192 13A 14,130 K = (K2 + 1) + Kle ALLRO " AL,R (I) (A21) However, K K' , x, ... 1 3 (A22) 1 + K2 Then fig AAL R K1K2 K K2 L— - 2 (AALJRO - AAL,R) " Kl' + (A 3) from which K1 may be evaluated. Treatment of Threonine Dehydrase with g. SulfaydrylgReagents A preliminary effort was made to establish the role of sulfhydryl groups on the enzyme. Since these studies utilized impure dehydrase and were not carried to comple- tion, the observations which are of some importance are included in the Appendix. Further, these data do not relate well to the theme of the thesis. Of the sulfhydryl reagents tested, which included N-ethyl maleimide (NEM), iodoacetate, dithioglycolate, and p-hydroxymercuribenzoate (pHMB), only p-hydroxymercuribenzo- ate was found to inhibit the enzyme. AMP afforded partial protection from the pHMB inhibition, as Shown in Appendix Figure 1. In the absence of AMP, a 10 fold molar excess of pHMB almost totally inactivated the dehydrase in 30 minutes; in the presence of.AMP, no inactivation was observed at all. However, AMP could not prevent inactivation when higher 193 ratios of pHMB were used. Under these conditions, AMP could only Slow the rate of pHMB inactivation. The ability of AMP to protect against pHMB inhibi- tion is also demonstrated in Appendix Figure 2. With a dehydrase preparation of Specific activity, 2000 units/mg protein, a five fold excess of pHMB was required to cause inactivation of the dehydrase in the absence of AMP. while a ten to fifteen fold excess of pHMB was required in the presence of AMP. At pHMB to enzyme ratios above 15:1 no AMP protection was apparent. While AMP could protect the dehydrase from inhibi- tion by pHMB, the nucleotide could not reverse the inacti- vation of enzyme already treated with the sulfhydryl inhi- bitor. IpThreonine, to a lesser degree than AMP, was found to be capable of affording some protection from sulfhydryl group inactivation. Glutathione was found to be capable of reversing pHMB inhibition at low pHMB to enzyme ratios, as Shown in Appendix Table 1. However, as the pHMB to enzyme ratio increased, the sulfhydryl reducing agent became ineffective as a reactivator of the dehydrase. Finally, it should be mentioned that since pHMB is known to attack other amino acids besides cysteine, i.e.. histidine and tyrosine, the possibility exists that inac- tivation may result from modification of an amino acid other than cysteine. 194 .mz< 25 m use mama and: oHSPan Soap0350SA : mafia coupon ”mama and: ospSNAa coapmpsoca I enaa ooppoeieoSwmo wsdpmSHopHm nmz< as m .mama on .Hoapaoo n mafia ooamoe “mama no age on .Hohpaoo u made eaaom .Huma .oApmH maausoumzma .m H enemam "HAOH .oapma osmwaoumzma .< H oasmam .ommhomsmo one how ooo.owH mo unmask HmH50oHoa m wSAESmmm 0000H50Hmo was oapmn mamasoumzma one .oHSpHAE ahead on» anm oopoaop was enema wsdoscon HoAnp m was» soap iaooxo on» and; oomoaaao 003 means cahpoEOposaonpooamlumaasoo on» “menus you Scamp one; moHSpNAE godpSDSOQH 0:0 mo mposuAH< .HE mmo.o mm: madaob Hazau on» “mama mo coapmapsoonoo opmawaowaam as mean .o.m ma .hommdn opmnamosa asAmmMpoa a 00.0 .ooamaooam muons mz< as m Andopoua ma\mpA:: ooom mpabapom camaooamv osmwso no we m~.o no woumdmsoo onapre noapmp:0:A one .oo 00 oopososoo one: msodpm950sH opmouconah50hoaawoaommna an ommaomsoo czacoohse mo soapmpauomaH .H ohswdm Ndosoanw 195 ON. 00 SEE m2: N. o q 0 (“’w/Ob‘ivv) Ail/\IIDV BWAZNB 0. O md 196 .mz< pSOSpAz 00009502A ommaomsoe u mmhmsvm “m:< spas housewoaa onshemnoo I moaoaao .H ohswfim adenoaaw SA confluence mm 0903 msoHpAoaoo ommapmsom enasoohse mo SoapmbapomzH endosconahsonoamxonpmmia aohm mzw an :oApoopoam .N oaswdm Nacsoaa< 197 0N Aoehcmumaxe oeom ON 0. O. m N. 0 V. o (U!w/'0'0) 0103 so JhOH l 10 AigAgiov <0. 0 m. 0 198 APPENDIX TABLE 1 Reversal of pHMB Inhibition by Glutathione Reaction mixture contained 0.023 mg threonine dehydrase (approximate Specific activity 1,000) in 0.06 M potassium phOSphate buffer, pH 7.0; the final volume was 10 ul. Glutathione was added to 0.2 M 30 minutes after the addition of pHMB. Incubations were conducted at 0°. '_T pHMB Ratio Percent Inhibition Percent Activity x 10-5 M pHMBxENZ 25 Minutes Regaineda 1 10:1 99% 87% 2 20:1 99% 61.7% 3 30:1 100% 0% 840 minutes after glutathione addition. LIST OF REFERENCES (l) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) LIST OF REFERENCES Stadtman, E. R., Advan. Enaymol., 38, 41 (1966). Atkinson, D. E., Ann. Rev. Biochem.. 25, 85 (1966). Cohen, G. N.. Ann. Rev. Microbiol., 12, 105 (1965). We 5 Koshland, D. E., and Neet, K. E., Ann. Rev. Biocagm.. Atkinson, IL E., Ann.a§gv. Micrabiol., 22. 47 (1969). Koshland, D. E., Current Tapics in Cellular Regulation, 1.1 (1969). .. Umbarger, H. E., Science, 123, 848 (1956). Yates, R. A., and Pardee , A. E., J. Biol. Chem., 221, 757 (1956). Umbarger, H. E., Cald Springagarbor Symp. Quant. 8191.. fig, 301 (1961). Gerhart. J. C.. and Pardee, A. B., J. Biol. Chem.. 22 . 89 (1962). Gerharg, J. C.. and Pardee, A. B., Cold SpringgHarbor §ymp._guant. Biol.. 28. 491 (1963). Hill, A. J.. Biochem. J.. 2, 471 (1913). Adair, G. S., J. Biol. Chem., 6;, 529 (1925). Changeux, J. P.. Cold Spring Harbor Symp. Quant. Biol.. Monod, J.. Chan eux, J. P.. and Jacob, F., J. Mol. Biol.. g. 30 (1963). Gerhargg)J. C.. and Schachman, Biochemistry, 4, 1054 9 . - Changeux, J. P.. Colngpring Harbor Symp. Quant. Biol.. 199 (18) (19) (220) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (1322) (13:3 ) (33l+-) (13:5,) 200 Mansour, T. E., and Mansour, J. M.. J. Biol. Chem.. 237. 629 (1962). Hathaway, J. A., and Atkinson, D. E., J. Biol. Chem.. 2 8, 2875 (1963). Wyman, J., Cold Spaing Harbor Symp. Quant. Biol.. 28. 483 (1963). Koshland, D. E., Nemethy, G. and Filmer, D.. Biochemistry, Changeux, J. P.. Brookhaven Syap. Biol.. 12, 232 (1964). Monod, J., Wyman, J., and Changeux, J. P.. J. Mol. Biol.. 13, 88 (1965). Atkinson, D. E., Hathaway, J. A., and Smith, E. C.. Q J. Biol. Chem.. 240, 2682 (1965). ‘ Atkinson, D. E., Hathaway, J. A., and Smith, E. C.. Biochem. Biophys. Res. Commun.. 18. 1 (1965). Changeux, J. P.. Gerhart, J. C.. and Schachman, H. K., Biochemistry, Z, 53 (1968). Changeux, J. P.. and Gerhart, J. C.. The Regulation of Enz me Activity_and Allosteric Interactions, (OsIo: UnIversitetsfortaget (1968)), p. 13. Weber, K., Nature, 218, 1116 (1968). Wiley, D. C.. and Lipscomb, W. N.. Nature, 218, 1119 (1968). Gerharg,)J. P.. and Schachman, H. K., Biochemistry, 7,538 (19 8 . ’ McClintock, D. K., and Markus, G., J. Biol. Chem.. 244. 2855 (1968). McClintock. D. K., and Markus, 0., J. Biol. Chem., 244, 36 (1969) . —- Changeux, J. P. . and Rubin, Biochem. . _7_. 553 (1968). Levitzki, A., and Koshland, D. E., Proc. Natl. Acad. Sci. U.S.. 62, 1121 (1969). T“ Conyayé7A.. and Koshland, D. E., Biochemistry, 7, 4011 19 . ’ (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) (Lee) (47) (48) (459:) (5(3)) (5&1 ) (QSZR ) 201 Koshland, D. E., Conway, A., and Kirtley, M. E., The Regulation of Enzyme Activity and Allosteric "Interactions (Oslo: Universitetsfortaget,1968) p.TI43. Kirschner,Kg, Eigen. M., Bittman, E., and Voight, B., Proc. Natl. Acad. Sci# U. _S., 2a. 810 (196) Kirschner, K., The Regalation of Enzyme Activit and Allosteric Interactions (Oslo: Universifets¥ortaget, p. ‘1 Wood A., Current To ice in Cellular Re julation, 1, 161w (1969) Umbarger, H. E., Current Topics in Cellular Regulation, 1. 57 (1969). Gale, E. F., and Stephenson, M.. Biochem. J.. 32. 392 (1938). e Brinkley, F.. J. Biol. Chem.. 150, 261 (1943). Lichstein, H. C.. and Umbreit, W. W., J. Biol. Chem.. 120. 423 (1947). Lichstein, H. C.. and Christman. J. P.. J. Biol; Chem.. __z5_. 649 (1948). Wood, W. A., and Gunsalus, I. C.. J. Biol. Chem.. 181, 171 (1949). Nishimura, J. S., and Greenberg, D. M.. J. Biol. Chem.. 226, 2684 (1961). Umbarger, H. E., and Brown, 8.. J. Bacteriol.. 73.105 (1957) Dawes, E. A., J. Bacteriol.. 62, 647 (1952). Boyd, W. L., and Lichstein, H. C.. J. Bacteriol.. _2. 545 (1955). Pardee, A. E., and Prestidge, L. S.. J. Bacteriol.. 22, 667 (1955). ‘“‘ Charfigff E., and Sprinson, D. B.. J. Biol. Chem.. 148, (1943). Chargaff, E., and Sprinson, D. E., J. Biol. Chem.. 151, 273 (1943). 202 (53) Metzler, D. E., Ikawa, M.. and Snell, E. E., J. Am. Chem. Soc., 6, 648 (1954)- (54) Braunstein, A. E. , and Shemyakin, M. M. . Biokhimiya, 18, 393 (1953). (55) Phillips, A. T., and wood, w. A., J. Biol. Chem.. gig. 4703 (1965). (56) Krongelb, M.. Smith, T. A., and Abeles, R. H. . Biochim. B10phys. Acta, 167, 493 (1963). (57) Flav1n6, I;I., and Slaughter, J.. J. Biol. Chem. . 2111+, 11131; (19 9 . "’ “— (58) Reissig, J. L., Arch. Biochem. and Biopliys” 26, 234 (1952). — " (59) Nakazawa, A., Tokushige, M.. and Hayaishi, 0., Biochem. Biophys. Res. Commun., 22, 184 (1967). (60) Tokushige, M.. Nakazawa, A., Shizuta, Y.. Okada, Y., and Hayaishi, 0., in Symposium on ‘Pyridoxal Enzymes, K. Yamada, Katunuma, NT. and Wada, H., Eds. (Tokyo: Maruzen, 1968) p. 105. (61) Phllllps. A. T., and Wood, W. A. , Biochem.‘ Biophys. Res. Commun.. $20 530 (196“). (62) Gilbert, G. A., Proc. Royal Soc. A., 250. 377 (1959). (63) Whiteley, H. R., J. Biol. Ch_e_m., 2111. 4890 (1966). (61+) Whanger, P. D., Phillips, A. T. . Rabinowitz, K. w.. Piperno, J. H., Shada, J. D., and Wood, W. A., J. Biol. Chem.. 2&3, 167 (1968). (65) Shizuta, J. . Nakazawa, A. , Tokushige, 191. , and Hayaishi, 0., J. Biol. Chem.. 244, 1883 (1969). (66) Hayaishi, o.. Gefter, M.. and Weissbach, H., J. Biol. Chem.. 228, Zone (1963). (6?) Nakazawa, A., and Hayaishi, 0., J. Biol. Chem.. 2‘12. 11116 (1967). '— (58 ) Whiteley, H. H., and Tahara, M.. J. Biol. Chem.. 21A. #881 (1966). “' (69) Vanquickenborne, A., and Phillips, A. T. . J. Biol. Chem.. 2a}. 1312 (1968). (70) (71) (72) (73) (74) (75) (76) (77) (78) (79) (80) (81) (82) (83) (84) (85) (86) (87) 203 Vanquickenborne, A., Vidra, J. D., and Phillips, A. T., J. Biol. Chem.. 2““, 4808 (1969). Nicderman, R. A., Rabinowitz, K. W.. and Wood, W. A., Biochem. Biophys. Res. Commun.. 6, 951 (1969). Rottman, P.. Ibershof, M. L., and Guarino, A. J., Biochim. Biophys. Acta, 26, 181 (1963). Griffin, B. E., and Reese, C. 3.. Biochim. Biophy . Acta, 68, 185 (1963). Rabinowitz, K. W.. Shade, J. D., and Wood, W. A.. J. Biol. Chem.. 242, 321“ (1968). Nirata, K., Biochim. Biophys. Acta, 0, 352 (1958). Greenstein, J.. and Winitz, M.. Chemistr of the Amino Acids (New York: John Wiley_and Sons. I951) p. 2250. Birnbaum, S. M.. Lovintow, L., Kingsley, R. B., and Greenstein, J. P.. JLgBiol. Chem.. 12#, #55 (1952). Mix: Hé£)and Wilcke, R. W.. Z.‘Physiol. Chem.. 221. #0 19 . Flake. c. H., and SubbaRow, 2.. J. Biol. Chem. . 66. 375 (1925). Lowry. O. H., Rosebrough, N. H., Farr. A. L., and Randall, R. J.. J. Biol. Chem.. 122. 265 (1951). Martin. R. C.. and Ames. B. N.. J. Biol. Chem.. 236, .1372 (1961). —' “" Rutter, W. J., Federation Proc., 32, 1248 (1969). Cecil. R.. and Ogston, A. C.. Biochem. J..‘32, 105 (1951)- Samejima, T., Kamata, M.. and Shibata, K., Biochem. J.. 21. 181 (1962). Noll, H., Nature, 212, 360 (1967). Dixon. M.. and Webb, E. C.. Enzymes (New York: Academic Press, 1964)‘p. 328. Dixon, M. and Webb, E. C.. Enzymes (New York: Academic Press.'196#) p. 327. (88) (89) (90) (91) (92) (93) (94) (95) (96) (97) (98) (99) (100) (101) (102) (103) (10#) (105) 20a Baronowski, T., and Niederland, T. R.. J. Biol. Chem.. 180. 593 (1949). Maehley, A. C.. and Chance, B.. Methods of Biochemical Anal sis, Vol. I. Click. D., ed.. (New York: InterscIence PuBlishers. Inc. 195h) p. 357. Tokushige, M.. J.‘Vitaminol.(Kyoto), 12, 165 (1967). Levin, 0.. Methods in Enz molo , Colowick, S. P.. and Kaplan, N. 5.. eds. (New Yorfi: Academic Press, 1962) Vol. 2. p. 27. "1 Burns, G. A., and Christian, J. E., Anal. Chem.. 22. 1216 (1961). Dowd J. E., and Riggs, D. 3.. J. Biol. Chem.. 339. 863 (1965). Hirata. M.. Tokushige, M.. Inagawa, A., and Hayaishi. ti 0.. J: Biol. Chem.. 240. 1171 (1965). Holleggn, W.. Ph.D. Thesis. Michigan State University, 19 . Andrews. P.. Biochem. J.. 6. 595 (1965). Frieden, C.. J. Biol. Chem..‘§22. 3522 (1969). Dyer, J. H., A lications of Absor tion Spectrosco of Dr ic Co ounds (Englewood s, N.J.: Prenéic - Hag; Inc., 5955). Goldberg, M. E., and Baldwin, R. L., iochemistry, 6. 2113 (1967). ' Jenkins, W. T., Federation.Proc.. 22, 978 (1961). Jenkins, W. T., J. Biol. Chem.. 224. 1121 (1961). Schirc2,)L., and Mason. M.. J. Biol. Chem.. 228, 1032 19 3 . Schirch L. and Jenkins W. T., J. Biol Chem. 2 2 3801'(196u). ' £7 ' ' Marinoé §., and Snell, E. E., Federation_Proc.. 2h, 530 19 5 . ’- Marino, Y.. and Snell, E. E., J. Biol. Chem.. 2h2, 2800 (196?). (106) (107) (108) (109) (110) (111) (112) (11.3) c1114) (115) (116) (117) (118) (119) (120) (121) (122) (123) 205 .. Hudgins. W. R.. Roszell. J. A., J. Biol. 4“, 2809 (1969). Brown, F. C Chem.. 2 Marino, 1.. and Snell. E. E., J. Biol. Chem.. 292. 2793 (1967). Piperno. J. H., and Wood. W. A., unpublished observations. Gerlt, J..A.. and Rabinowitz. K. W.. Federation Proc.. 32. 400 Abs. (1970). Babinowitz. K. W.. Piperno, J. H., and Niederman, R. A., F1 Egderation.Proc.. 328. 3&1 (1969). Frieden. C.. J.»Biol. Chem..lgfig. “0&5 (1967). Frieden, C.. and Colman. R. P.. J. Biol. Chem.. 2&2, - 1705 (1967). i Nichol. L. W.. Jackson. H. J. H., and Winzor, D. J.. ;, Biochemistrz.'6, 2h#9 (i967). ’ Kurganov. B. 1.. Molekulzarnaya Biologiza. g. 166 (1968). Kurganov. B. 1.. Surin, 8. A., Sugrobova, N. P.. Molekulzarnaza Biologiza, g. 180 (1968). Xurganov. B. 1.. Molekulzarnaza Biologiza.‘§, 430 (i968). Kappox, M.. Bray. D. F.. and Ward, G. W.. Archives of Biochem.. 128. 810 (1968). Le'John. H. B.. McCrea. B. E., Suzuki. 1.. Jackson. 8., J. Biol. Chem.. 24“, 248W (1969). Wang. J. H., Shonko. M. L., and Graves, D. J.. Biochem- istrz, g, 2296 (1965). Datta, P.. Gest, H., Segal . L., Proc. Natl. Acad. Sci.J_U.S.. 5;. 125 (1954), Chen. R. P.. Brown. D. M.. Plant, G. W. E., Biochemistrz, Vagelos, P. P.. Alberta. A. W.. and Martin, D. M.. J. Biol. Chem.. 228. 533 (1963). Gregolin. C.. Ryder, E., Kleinschmidt. A. K., Warner. R. C.. lane. M. D., Proc. Natl. Acad. Sci. U.S.,‘2§, 1&8 (i966). (212340 (125) (126) (127) (128) (129) (1:30) (1:31) (132) (133) (13a) (135) (136) (137) 206 Matsuhashi, M.. Matsuhashi, 8.. and Lynen, P.. Biochem. Z.. 240, 263 (196a). Numa. S.. Bortz. W. M.. and Lynen. P.. in Advances in Enz e Re lation ed.. G. Weber (New Yoffié Pergamon gress. I965) Vol. 2. p. (+07. Maley.6fi5 P.. and Maley, H., J. Biol. Chem.. 2 2. 1168 (19 . Maley. G. P.. and Maley. R.,.g._§1o1. Chem.. zuo, pc 3226 (1965). —— .... Maley. G. P.. and Maley. H., J. Biol. Chem.. 35;, 4513 (1968). Le'John. H. B.. and Jackson, S., J. Biol. Chem.. 322. 3#47 (1968). Anderson P. and Marvin 8. M. Biochem. BiOphzs. Re . Commnn.. '2. 928 (1968). ' Anderson; P.. and Marvin. S. M.. Biochemistrz. 2, 171 (1970 . Long. C. W.. Levitzki. A., and Koshland, D. E., J. Biol. Qhem.. ZGQ. 80 (1970). ‘ Iwatsugii N.. and Okazaki, H., J. Mol. Biol.. 32. 139 19 7 . Brown. N. C.. and Reichard, P.. J. M01. Biol.. fig, 25 (1969). Tsutsui. E. A., and Marks. P. A., Biochem. Bioghzs. Res. Commun.. 8, 338 (1962). Ling. K. H., Marcus. F.. and Lardy, H. A., J. Biol. Mansour. R. E., Wakid, N. W.. and Sprouse. H. M.. Biochem. 3192215. Res. Commun.. 12, 721 (1965). IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII Ill(IIINJIHUIMMHIflbmlllljllfllnflllylljlfllIIIIH