" i“ a..— ' \‘ .. .0—‘wq -- . . - ._.. -. , p- Q-¢....-.Q>«‘-u-. -.¢—....-oo..d .ol - .oo-uq--- --- . . - ‘Vv'fi‘v-me 0--...- .--- ‘ ‘¢ IO\Q.QQ09«¢or-' ’ x ...\.'.l- ." RAT BRAIN HEXOKINASE: A KINETlC COMPARISON - ‘ . 0F SOLUBLE AND PARTICULATE FORMS ‘ Thesis for the Degree of M. S. MICHKGAN STATE UNIVERSITY JOHN P. 'TUTTLE _ ‘ 1969 . . . . .. . . . I . - _ . .. . . - . . . . . . . . -4. . . . . ' " ' ‘ o o. .. . . . . I . . -. ¢ .._ I - - V. .. . . - a .e . . ' . op . . 4 ‘ - .. ,, . . . . n .. .. . , ' p .. - - ..- ... , . ‘ . . - . .. 4 . . . ' . . .. . - . . . .. . . ..1. . . . .o .o . _‘ ‘ .. . - - . .. , .., - . . . . ~ - ' ’ . . ., . , , -4 .. a ‘ — v .o -l . . . ‘ . _ . . _ . . - - - .. . -.- —. ._.. . . . ._ ‘ - . r . . .- a1.o.o- - r,’ -. .-'. . .va. . . . _ . . ' , . . -. v - o. .“ p. o . .o.,. . - - .. . . -. - _ o .,,. .' 1' -‘ v ’ v . -...' .. - ... . ' _, , e -- 9 ., . _ 2c. . _.oQ --,' C’I‘JOOP.“.O --ov’;0 ' ' n: . _ - — t - I. 7 VP _ - . r -- : ~ ... .- —. . . . . ; . . . . . V 'I ' . . . - 1: , o _. . .. o , _ . — . ' ‘ _" v o ....o '00.]. ! e r - . .Iop ' g .. ., a r -“ .p,’....- - . . , . . ' -' ' ' - ' I _ .-' - - nu - or Ja'- .p. p . .- . .. .. . . ' ' - ' ‘ - > .".. ‘r“"' . : —r .- . . o I l o 'y'c' ,. as. c < a . o o . . . a I! . 1‘ . .uv'upo. . - . .v ... . . . ‘ 1° . -o- - ..-:.Al. 0 .. .. ,., .. , _, r o I ‘ " ' ... . . e . ‘o a - , . . ' _ ‘ ' .. ,.'1- r'- I .._,.. ,. I' u ' o- . u . I . o - . - . ’ . , ' ‘ l" '0'! " ..‘ ' ‘.“"' . . - - I. .. p . '7 ' . ' . o I -- e. .. I ,. .oo. 7-, ...,.,_ ..' .- --.. . ‘ .,' I . - ’ . . '. ..— .. . ., ».,... . . . . o .A.,,.~. . p'. f. .. -’.. I r. . , o 9 .o . . I - I ' b ‘ _ . ' ' - ~ ..n ., .. 'o-e... - .o.- f. . - - '; . , .p .-u . .1 . . . I . . e. - ‘ -. . , , . A . - . . - a . l'l ‘ -. ‘ ‘ . . . , a ..0 . ‘1‘ ' l ' —....b a ‘ .. u e u ‘ ‘ o l - . . . . . ' O . l ' c .. , . . ' o . . . .- , _ . . . .- . . I u. - .'I - . . .. .. . .— .. L- . ‘ . . . , . , . o ’ ' "' ‘ — 4.1: . . . , . ‘ . -. a . . as- n .. __ ‘ .- ‘ ol , _ .A . _ I ‘ . - .1 - DO -' 4. ave.-- A ' . ' ' - “ - .,' . ‘1 . ‘. I u . ’ , ~4. -. . .. r o . D. O _ ' vv,-I»o... . . ‘- ..‘ 0 cu, . . ’ .01.: .‘ ._' . _ ' " _' ‘ v. o ¢-.'O'. .u- --o 4", ‘ r" . - ‘ ' . ' l . -o-’ ...-.. _‘ . ' . . . ' .o ._ ' .I- “.I - I ' ' .... . . I.._(p 0--.. ... _ .. . " .‘ ‘—r-- A-" . ' cnyv l .-y g o - '.>'.'.""" ""’ '9 -.-v- .. .. ,. .o . - I ’u. I. ,_'. _, ' ."_’... ’. .. 'I.--'..io .a‘¢._,( ‘ a ".v-o.- -, "D' A0.) THESIS mnmmmmnnmmrrmn ,,, WW” / 31293 01090 9244 _ Michigan State University ABSTRACT RAT BRAIN HEXOKINASE: A KINETIC COMPARISON OF SOLUBLE AND PARTICULATE FORMS by John P. Tuttle Most animal and some plant tissues studied contain a soluble and particle-bound hexokinase. In brain 70-90 percent of the hexokinase present in crude homogenates is associated with mitochondria. Particulate hexokinase can be solubilized by physiological concentrations of glucose—6-phosphate and ATP. The soluble-particulate distribution of hexokinase is also influenced by pH, ionic strength, and monovalent cations. Solu— bilization by glucose—6—ph03phate is reversed by physiological concentrations of orthophOSphate. The Specificity with which hexokinase is solubilized has suggested a possible regulatory mechanism based on the distribution of soluble and particulate hexokinase. The conditions under which particulate hexokinase is solubilized would provide glycolytic control if particulate hexokinase were more active than soluble hexokinase. Support for this proposed mechanism has come from reported kinetic l John P. Tuttle differences between soluble and particulate hexokinase from bovine brain and frog skeletal muscle. Solubilization of rat brain hexokinase has been well characterized but no kinetic comparison of the particulate and soluble forms is available. This thesis was initiated to study the kinetics of soluble and particulate rat brain hexokinase in an effort to evaluate the function of mitochondrial hexokinase relative to glycolytic control. Data is presented that indicates the soluble and particulate forms of rat brain hexokinase are similar with respect to ATP and glucose Michaelis constants, and inhibition constants for N—acetylglucosamine and 1,5— anhydroglucitol—6—phosphate. Although both forms of hexokinase were inhibited to nearly the same extent by g1ucose—6-phosphate, the particulate enzyme was more sensitive to reversal of glu- cose-6-phosphate inhibition by orthophosphate. This was the only kinetic difference observed between soluble and particu- late hexokinase. The functional significance of mitochondrial hexokinase is discussed. RAT BRAIN HEXOKINASE: A KINETIC COMPARISON OF SOLUBLE AND PARTICULATE FORMS BY John P. Tuttle A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1969 (3,5 530 ACKNOWLEDGEMENTS The author wishes to thank Dr. John Wilson under whose guidance this research was conducted. The author also wished to thank Dr. Steven Aust, Dr. Clarence Suelter, and Dr. William Wells for their suggestions and assistance. ii TABLE OF CONTENTS ACKNOWLEDGEMENTS . . . . . . LIST OF FIGURES. . . . . . . LIST OF TABLES . . . . . . . LIST OF ABBREVIATIONS. . . . INTRODUCTION 0 O O O O C O I Soluble and Particulate Hexokinase Hexokinase Isozymes, Characterization and Tissue Distribution. . . Hexokinase Kinetics: A Comparison of and Particulate. . . . . Brain. . . . . . . . Ascites Tumor. . . . Frog Skeletal Muscle Heart Muscle . . . . Mammary Gland. . . . Soluble Intestinal Mucosa and Adrenal Medulla. . Mitochondrial Hexokinase Binding Specificity iii Page ii vi vii viii ll 12 l3 l3 l4 Mitochondrial Hexokinase, Relationship to Oxidative Phosphorylation. . . . . . . . . . MATERIALS AND METHODS. . . . . . . . . . . . . . Chemicals. . . . . . . . . . . . . . . . . . Animals. . . . . . . . . . . . . . . . . . . Soluble and Particulate Enzyme Preparations. Hexokinase Assays. . . . . . . . . . . . . . C—l4-Glucose Assay . . . . . . . . . . . Glucose—6—Phosphate Dehydrogenase Assay. Lactate Dehydrogenase and Pyruvate Kinase Assay . . . . . . . . . . . . . . Michaelis—Menten Kinetics. . . . . . . . . . Synthesis of Anhydroglucitol-6—Phosphate . . DEAE-Cellulose Chromatography. . . . . . . . Solubilization of Particulate Hexokinase . . RESULTS. . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . Proposed Regulation of Hexokinase and Glyco 1Y8 is o o o o o o o o o o o o o o o o 0 Comparison of Apparent Michaelis Constants for Soluble and Particulate Hexokinase . . . Comparison of Inhibition by Glucose-6- Phosphate and Anhydroglucitol—6-Phosphate. . iv Page 16 18 18 18 19 20 21 21 21 22 23 26 27 28 62 63 65 67 Page Oxidative Phosphorylation and Bound HeXOkinase O O O O O O O O O O O O O O O O O 71 Kinetically Similar Soluble and Particulate HexokinaseS. . . . . . . . . . . . . . . . . 72 Latent Particulate Hexokinase in Brain . . . 76 Speculation on the Function of Soluble and Bound Hexokinase . . . . . . . . . . . . . . 77 REFERENCES 0 O O O O O O O O O O O O O O O O O O 82 10. LIST OF FIGURES Linearity of glucose—6—ph03phate assay for soluble and particulate hexokinase. Inhibition of soluble hexokinase by N-acetylglucosamine . . . . . . . . . . Inhibition of particulate hexokinase by N-acetylglucosamine . . . . . . . . . . Inhibition of soluble hexokinase by 1,5— anhydroglucitol-6-phosphate (An G6P). . Inhibition of particulate hexokinase by l,5-anhydroglucitol-6-phosphate (An G6P) Inhibition of DEAE—cellulose purified soluble hexokinase by glucose-6-phosphate and 1,S-anhydroglucitol-6—phosphate (AnG6P)................ Inhibition of soluble hexokinase by glucose-6—phosphate and reversal by phosphate. . . . . . . . . . . . . . . . Inhibition of particulate hexokinase by glucose-6—phosphate and reversal by phosphate. . . . . . . . . . . . . . . . Soluble and solubilized hexokinase apparent Michaelis constants for ATP . , Solubilization by 1,5-anhydroglucitol- 6-ph03phate. o o o o o o o o o o o o o o Page 30 34 36 38 40 45 49 51 59 61 Table II. III. IV. VI. VII. LIST OF TABLES A Comparison of Glucose and ATP Michaelis Constants Reported in the Literature for Soluble and Particulate Hexokinase. . . . . . . . . . . . . . . A Comparison of G1ucose-6-PhOSphate Inhibition Constants Reported in the Literature for Soluble and Particulate Hexokinase. . . . . . . . . . . . . . . Apparent Michaelis Constants for Soluble and Particulate Hexokinase. . . Inhibition Constants for Soluble and Particulate Hexokinase. . . . . . . . . Reversal of Glucose—6-Phosphate Solubilization by Orthophosphate. . . . Inhibition of Soluble and Particulate Hexokinase by Glucose-6—Phosphate and Reversal of Inhibition by Phosphate . . Solubilization of Particulate Hexokinase vii Page 31 42 46 53 57 LIST OF ABBREVIATIONS ATP, ADP, 5'AMP CTP, GTP, ITP, UTP DEAE POPOP PPO NADP+ (H) NAD+ (H) EDTA G—6-P An G6P viii adenosine tri—, di—, and mono— phosphates cytidine, guanosine, inosine, and uridine triphOSphates diethylaminoethyl (l,4—bis(2—(5-phenyloxazolyl))—benzene; phenyl-oxazolylphenyl—oxazolyphenyl) 2,5—diphenyloxazole nicotinamide adenine dinucleotide phosphate (reduced) nicotinamide adenine dinucleotide (reduced) (ethylenedinitrilo) tetraacetic acid glucose—6—ph05phate l,5-anhydroglucitol-6-phosphate INT RODUCT ION Of all the mammalian tissues that metabolize glucose only the brain, retina of the eye, and red blood cells are completely dependent on the oxidation of glucose to supply energy for endergonic processes (1). Because the brain must perform a unique and critical function, the metabolism of glucose and the regulation of this metabolism in brain has become an impor- tant biochemical problem. Evidence has accumulated that both phosphofructokinase and hexokinase are involved in the regulation of cerebral glycolysis (2). Control at the level of phospho- fructokinase has been well studied and characterized (3). Inhibition of phosphofructokinase by ATP and citrate is antagonized by 5'—AMP, 3',5'—cyclic AMP, ADP, orthophOSphate, and fructose—6-phosphate. The regulation of hexokinase activity is not as well understood. Glucose—6-phosphate efficiently inhibits mammalian hexokinase by binding presumably to an allosteric site on the enzyme (1). Adenosine triphosphate and orthophosphate may have a modulating effect on the inhibition by glucose—6-phosphate. In contrast to the mammalian enzyme, yeast hexokinase does not have an allosteric site for glucose— 6-phosphate and is relatively insensitive to inhibition by this metabolite (2). There is some evidence that yeast hexokinase is controlled by mannose—6—phosphate inhibition. Most yeast contain an active mannose-6—phosphate isomerase (4). Soluble and Particulate Hexokinase Another important characteristic of brain hexokinase is that it is found in three forms in crude homogenates (5, 6). The enzyme is found in the supernatant in a soluble form, bound to the mitochondria in an exposed form, and bound to the mitochondria in a latent form. The latent form can be exposed by disruption of the mitochondrial membrane (6, 7, 8). The soluble and particulate (nonlatent) forms from numerous tissues have been shown to be in dynamic equilibrium (9, 10, ll, 12). This equilibrium is shifted by the metabolites glucose-6- phOSphate and ATP and is also effected by pH, ionic strength, monovalent and divalent cations, and inorganic phosphate. The unique binding of hexokinase to mitochondria and the effect of metabolites and cations on this binding has stimulated specula- tion that the regulation of hexokinase activity may involve association and disassociation from mitochondria (7). The concentrations at which glucose-6—ph03phate and ATP are effective in eluting half the maximal amount of particulate hexokinase in preparations from rat brain and the concentra- tion of phOSphate effective in reversing solubilization by glucose-6-ph03phate correspond closely to the concentrations of these compounds normally found in the brain (7). The specificity with which glucose-6—phosphate solubilizes bound hexokinase and inhibits the reaction and the similarity between the kinetics of both processes suggests that glucose— 6—phosphate is acting allosterically as a regulator rather than a product inhibitor (7, 9). As a substrate and antagonist of glucose—6—phosphate inhibition, ATP is relatively Specific, but as a solubilizing agent it is only slightly better than the same concentrations of CTP, GTP, ITP or UTP, which are poor substrates in the reaction (9). Also pyrophOSphate at 10 mM is as effective in releasing hexokinase as ATP at 1.6 mM (9). These compounds may be acting as chelating agents that are binding a metal which is involved either in the direct inter- action of hexokinase with its binding site or in the structural integrity of the binding site itself. Other chelating agents have been shown to promote solubilization (7). Magnesium ion suppresses the solubilization by glucose-6—phosphate and aids in the rebinding of hexokinase to eluted mitochondria (7, 9). The effects of pH, monovalent cations, nucleotides, and mag— nesium ion all indicate that the membrane—hexokinase interaction may involve electrostatic bonding (9). That other forces are also important is suggested by the characteristics of solubil— ization by glucose-6-ph03phate (7). Hexokinase Isozymes, Characteristics, and Tissue Distribution Isozymes of hexokinase have been reported. Grossbard and Schimke (13) have characterized four ATP: D—hexose—6—phOSpho- transferases on the basis of starch gel electrophoresis, DEAE- cellulose chromatography, heat stability, proteolytic inacti- vation, and kinetics. The different forms have been designated types I—IV corresponding to their increasing mobility on starch gel electrophoresis. Hexokinase types I, II, and III have a low Kb for glucose, while type IV is specific for glucose and has a high Km (14). The Michaelis constants for ATP and inhib- ition constants for glucose—6—ph03phate and ADP also differ slightly among the various hexokinase types. Types I, II, and III are similar with respect to apparent molecular weight, pH optimum, and substrate specificity. Type IV glucokinase has been found only in the liver. Other tissues contain different proportions of hexokinase types I—III (14). Brain and kidney contain primarily type I, while both types I and II have been isolated from fat pad and muscle. Despite the potential com- plexities posed by the three intracellular forms of rat brain hexokinase (soluble, exposed particulate, latent particulate), the isozyme content appears to be uniformly type I (15). In guinea—pig cerebral cortex all three types were observed in the soluble fraction, but only types I and II were eluted from the particulate fraction (16). Whether type III is also capable of binding to the mitochondrial membrane has not yet been determined. None of the four types isolated from liver will bind to particles. Hexokinase Kinetics: A Comparison 2; Soluble and Particulate In pig heart, ascites tumor, and rat brain, hexokinase lg _1itrg has been shown to become associated and dissociated from the mitochondria in reSponse to the metabolic regulators, ATP, glucose—6-ph03phate, and orthophosphate (9, 10, 11). The spe- cificity and efficiency with which these metabolites are effective in solubilizing bound hexokinase suggest that the distribution of enzyme between a soluble and insoluble form could have some physiological significance. The most likely place to look for such a function is in the regulation of hex- okinase activity. The conditions under which bound hexokinase is solubilized would provide a mechanism of regulation if the particulate enzyme is more active or less inhibited than the soluble enzyme. The proposed correlation between soluble— particulate distribution and hexokinase activity has been stim- ulated by several reported kinetic differences between soluble and particulate hexokinase (7). The kinetics of soluble and bound hexokinase from several tissues are compared below. Table I summarizes literature values given for substrate apparent Michaelis constants, while Table II compares soluble and particulate inhibition constants for glucose—6-phosphate. .BEEEB — Solubilized bovine brain hexokinase has been isolated and purified by Schwartz and Basford (17). The kinetics of this enzyme have been extensively studied and found to be slightly different than the mitochondrial enzyme. The Michaelis constant for glucose was the same for the two forms, but the Michaelis constant for ATP was approximately ten times higher for the soluble form (18, 19). Also the Ki values indicated that the soluble bovine hexokinase was approximately six times more sensitive to inhibition by g1ucose-6-phosphate than was the particulate form. The inhibition constant re- ported for the purified soluble enzyme is the lowest reported for any mammalian hexokinase. Using guinea-pig cerebral cortex, Bachelard (16) has re— ported that the mitochondrial enzyme solubilized by freezing and thawing had a different Michaelis constant for glucose than the soluble enzyme at high ATP concentrations, 5 mM. No dif- ference was found at low ATP concentrations (0.2 mM), and the Michaelis constant for ATP at 100 mM glucose was the same for the soluble and solubilized enzymes. Newsholme t al. (20) used NH vm wm mm mm ow ma hm.© 0H mama m>mmm¢ pcmummman n * pmflmfiusm u m mozuwcflomamamosflo n we TCHEMHOGMSUOHHB fl QWB Homumflue n He o.v omm mesuo omumm o.m “a sac. man>om MHHSOOE H mcwhmvé OH ma omm oom as mass m om o.m «me Emma. HH mama mmuaoma new . ems mm m om m.s we zmo. mmuflome mnmnmus * W.w * N.“ mm mm mango mm q.s He zmo. mmuflome mum HOESB m.ma m.~ m mm m.s He zmo. Hmumsmxm moan o.m OHH m an o.m He smo. Hmumamxm “mm m.» omm as mass m mm a.» He who. Hmumamxm umm m o.m o.m me me mcsuo mm A.s «we 2H. pummm “mm @HUmDE o.¢ o.a we we mango Am o.m we zmH. mamummcnsw m.m m.~ ow mm mango mm v.n He zmo. pom o.v me omflxm Am v.> Me Sveo. umm m.mv om mNHme om o.n me see. mca>om AH mm mNHme mm o.h “a zmmo. was>om ¢.m so omxm mm o.» as Smo. mcn>om Cw mum manssom mumHsonuumm mansaom mumaponuumm Zenoa x ucmumcou z 10H x mucmumcoo mahucm 00 mm mousom maammnoflz.mem mWHmmnoAz mmoosflo mo musumz .QEme + “mumsm nomads .mmmcflxoxmm mamasofluumm 0cm mHQDHom you musumuwqu on» Ca omuHommm mucmumcou mflawmcoflz mam cam mmoosaw mo comHummEoo m .H manna Table II. A Comparison of G1ucose-6—Phosphate Inhibition Constants Reported in the Literature for Soluble and Particulate Hexokinase. Type of G1ucose—6-Phosphate Tissue Buffer + Temp. Inhibition Inhibition Constant Source pH 0C. with respect Soluble Particulate Ref. of ATP ( HM) ( pM) Brain 4 Bovine .032M Tr 7.6 28 comp. 11.7 18 Bovine .08M Tr 7.6 28 comp. 67 19 Guinea-Pig ..lM TEA 7.4 28 comp. 23** 46** 20 Rat .074M Tr 7.4 37 comp.+ 26 13 Rat .05M Tr 7.4 25 comp. 41 43 Table IV Muscle - - Rat Heart .lM TEA 7.7 25 mixed 8O 40 26, 27 Rat Skeletal .08M Tr 8.0 29 non—comp. 42 40 Rat Skeletal .074M Tr 7.4 37 comp. 21 13 Frog Skeletal .05M Tr 7.5 23 comp.+ 26 80 25 Tumor - + ELD Ascites* .125M TEA 8.0 25 mixed (B)+ 3O 70 24 comp. (U) Krebs—2 Ascites 0.03M Tr 7.5 30 comp. 28 34 lELD Ascites 5 mM Tr 7.4 25 mixed 65 6O 24 Tr = Tris—HCL TEA = Triethanolamine + = An G6P i = Solubilized * = Purified type II U Unbound B = Bound ** = Concentrations giving 50% inhibition under assay conditions (See text) guinea-pig cerebral cortex to study hexokinase inhibition by glucose—6—phosphate and relief of this inhibition by ATP, a—glycerol phOSphate, and orthophosphate. Essentially equal activities of hexokinase were found in the soluble and particu— late fractions. At 0.17 mM glucose, 1.0 mM ATP, and 1.0 mM Mg++, 50 percent inhibition was obtained with 0.023 mM, 0.046 mM, and 0.068 mM glucose—6—phosphate for the soluble, particulate, and whole homogenate respectively. Inorganic phosphate and a -glycerol phOSphate decreased the inhibition of hexokinase by glucose-6—phosphate in whole homogenates. The concentration of cz-glycerol phosphate required to nOticeably reduce the in— hibition was too high (20 mM) to be of physiological significance. Ascites tumor — Rose and Warms (9, 21) found that the maximum velocity of hexokinase from sarcoma 37 ascites tumor was inde— pendent of its association or dissociation from mitochondria. Both free and bound enzyme also showed the same dependence on pH from 6.3 to 9.4. Li and Ch'ien (22) have reported that bound ascites tumor hexokinase exhibits a higher affinity for ATP and is less sensitive to glucose—6—phosphate inhibition than the soluble enzyme. They observed that the activity of hexokinase was closely related to the state of mitochondria. Compounds that normally inhibit shrinkage of mitochondria (sucrose, azide, insulin, and thyroxine) inhibited the activity of the bound 10 hexokinase by as much as 50 percent. Sauer (23) has studied the kinetics of bound and soluble hexokinase from ELD ascites tumor. The two hexokinases were essentially identical with respect to Michaelis constants and inhibition constant for glucose-6—phosphate. The mitochondrial enzyme used, however, only accounted for 5 percent of the hexokinase present in the original homogenate. Kosow and Rose (24) isolated hexokinase types I and II from ELD ascites tumor cells and partially purified the enzymes by streptomycin and ammonium sulfate precipitation. Hexokinase type II was separated from type I on DEAE cellulose and then put through a hydroxyapatite column. The partially purified type II enzyme was bound to sarcoma 37 ascites tumor mitochon- dria from which the hexokinase has been eluted by salt. The only significant difference observed between the bound and soluble enzyme was in the type of inhibition by 1,5-anhydro- glucitol—6-phosphate* with respect to ATP and an increased sensitivity of the unbound enzyme to this inhibitor.' 1,5—Anhy- droglucitol—6-phosphate was competitive with ATP for the soluble and non—competitive for the bound enzyme. Kosow and Rose also compared the combined isozymes of ascites tumor mitochondria with the salt eluted enzyme and observed a 2.5 *See MATERIALS AND METHODS 11 fold difference in the inhibition constant for glucose-6- phosphate with the soluble form again being more sensitive. Frog Skeletal Muscle - The difference between soluble and particulate hexokinase from frog skeletal muscle has been studied by Karpatkin (25). The distribution of the two forms was 50 percent soluble and 50 percent bound. From the particulate fraction Karpatkin isolated a sarcoplasmic vesicle fraction that contained half of the bound activity and 25 percent of the original homogenate activity. The particulate (sarcoplasmic vesicle) enzyme, but not the soluble enzyme, was activated by orthophosphate and appeared to be more stable to heat. The soluble enzyme was purified 2 fold by ammonium sulfate pre- cipitation, and this purified enzyme was used for kinetic studies. Soluble purified enzyme had a Michaelis constant for ATP (1.2 mM) approximately 4 fold higher than the bound enzyme (0.3 mM), and the particulate enzyme was more sensitive to reversal of l,5—anhydrog1ucitol—6-phosphate inhibition by ortho- phosphate at high ATP concentration. At 11.5 mM ATP, 1 mM 1,5—anhydroglucitol—6—ph05phate, and 20 mM orthophosphate the activity of the particulate enzyme increased 2.5 fold over the activity without orthophosphate. At these same concentrations orthophosphate had no effect on the soluble enzyme. 12 Heart Muscle - Pig heart muscle homogenates contain a soluble and mitochondrial hexokinase with approximately 80—90 percent of the total enzyme in the bound form (10). The particulate enzyme can be solubilized with salts of monovalent cations, but the extent of solubilization is pH and ionic strength dependent. The shape of the pH dependent solubilization curve lead Hernandez and Crane (10) to propose that an imidizole group was being titrated and that an imidizole group was in— volved in the binding of hexokinase to the mitochondrial membrane. Pig heart preparations, however, showed no difference in kinetic properties between the soluble and particulate forms of hexo— kinase. England and Randle (26, 27) have studied the soluble and particulate hexokinases from rat heart muscle and found that the Michaelis constants for the two forms were the same. Insignifi— cant differences were also found for the inhibition constants of the two forms of the enzyme. Starch gel electrophoresis indicated that rat heart muscle contains type I and type II hexokinase with type I in excess. However, the inhibition characteristics of the muscle enzyme are quite different than the brain enzyme which is almost entirely type I (l3, l4). Hexokinase from brain exhibits mixed inhibition by ADP and AMP with reSpect to ATP and competitive inhibition by glucose-6- 13 phOSphate with respect to ATP (18). Heart Hexokinase appears to have just the Opposite kinetics with ADP and AMP being com- petitive and glucose-6—phosphate mixed with respect to ATP (27). Both the soluble and solubilized enzyme from rat heart muscle produced identical patterns after starch gel electrophoresis and staining for hexokinase activity. Soluble and particulate enzyme also diSplayed the same pH optimum. Mammary Gland - Rat mammary gland hexokinase exists in a solu- ble and particulate form. Walters and McLean (28) found 46 percent of the total tissue hexokinase in normal lactating rat mammary gland was associated with the particulate fraction. The percentage of bound enzyme was reduced to 11 percent of the total hexokinase in alloxan—diabetic lactating rats. The acti- vity of the soluble fraction remained essentially unchanged, but the ratio of type II to type I hexokinase was decreased. These results suggested that insulin may indirectly influence the binding of hexokinase to the mitochondrial membrane. Intestinal Mucosa and Adrenal Medulla - Shakespeare gt a1. (29) observed that hexokinase in the small intestinal mucosa of rat and guinea-pig was distributed approximately equally between the soluble and particulate fractions. Starch gel electrophor— esis indicated that type I and type II were present. Hexokinase from bovine adrenal medulla was also found equally distributed 14 between mitochondria and cytoplasm. The amount of hexokinase associated with mitochondria was dependent on the concentra— tion of hexokinase in solution suggesting an equilibrium dis- tribution between the bound and soluble enzyme (12). No kinetic comparison of the two forms of the enzyme in intestinal mucosa or adrenal medulla has been reported. Mitochondrial Hexokinase Binding_§pecificity Rose and Warms (9) were able to show that the number of binding sites on ascites tumor mitochondria roughly corresponded to the total amount of hexokinase in the cell. They also tested the possibility that the enzyme dissociated into subunits upon solubilization by determining the apparent dissociation constant over a range of concentrations of the enzyme and particulate binding sites. The results support a simple dissociation with— out dissaggregation into subunits. The addition of insulin (6 units/m1), glutathione (2 mM), sodium succinate (10 mM), or glucose (1 mM) did not change the apparent dissociation constant. Liver mitochondria isolated from rats, rabbits, and guinea— pigs contain very little or no bound hexokinase activity. Mouse liver has been found to contain from 3 to 30 percent bound hexo— kinase depending on the strain (9). Liver mitochondria will bind solubilized ascites tumor in the presence of magnesium ion; however, soluble hexokinase from liver homogenates will 15 not bind to liver or ascites tumor mitochondria (9). Rose and warms (9) also observed that the binding ability of solubilized ascites tumor hexokinase was lost when the enzyme was incubated with cx-chymotrypsin. No activity was lost during incubation. These results suggest that the peptide cleaved by chymotrypsin is required for the binding of hexokinase to mitochondria but does not effect the active site (9). The inner and outer membranes isolated from guinea—pig liver are both able to bind tumor hexokinase, but the outer membrane had approximately four times the number of binding sites. Both membranes bind more hexokinase activity than the frozen, thawed, and washed mitochondria from which they were isolated indicating that binding sites were uncovered (9). Solubilized rat brain hexokinase will also bind to both the inner and outer membrane of isolated rat liver mitochondria with more activity being bound to the outer membrane (11). In the presence of magnesium ion, bovine brain hexokinase can be bound to tumor mitochondria previously freed of hexo- kinase. Likewise, the particles from bovine brain freed of its hexokinase by glucose-6-phosphate elution are able to bind the solubilized enzyme from either tumor or brain (9, 11). Rose and Warms (9) were unable to get yeast hexokinase to bind to rat liver mitochondria while Siekievitz and Potter (30) re— ported that yeast hexokinase would bind to liver mitochondria 16 and that the bound enzyme was many times more active than the soluble form. Mitochondrial Hexokinase, Relationship 59 Oxidative Phosphorylation The location of ascites tumor hexokinase on mitochondria has been reported to facilitate the interaction between enzyme and ATP generated by oxidative phOSphorylation (22). However, Rose and Warms (9) could find no such relationship between oxidative phosphorylation and hexokinase in ascites tumor. Mitochondrial hexokinase was as effective in competing for ATP as added glycerol and soluble glycerol kinase whether the ATP was generated externally by creatine—phosphate and creatine kinase or internally by oxidative phosphorylation. Using atractyloside they found that ascites tumor hexokinase and oxidative phOSphorylation are on Opposite sides of the ATP transport barrier. Sauer (23) also studied the possibility that bound ascites tumor hexokinase could have a higher activity due to the formation of ATP through mitochondrial oxidative phOSphorylation. No difference in the rate of glucose—6-ph05phate formation could be detected whether ATP was generated outside the mitochondria by phosphoenolpyruvate and pyruvate kinase under conditions that inhibit and uncouple oxidative phosphor- ylation or was generated by internal oxidation of succinate. The purpose of this thesis was to investigate particulate 17 and soluble rat brain hexokinase relative to kinetic differences in order to determine the possible significance of an in vivo regulatory function for this distribution. A detailed kinetic study of the two forms of hexokinase has not previously been attempted, although solubilization of rat brain hexokinase has been well characterized (7). MATERIALS AND METHODS Chemicals Glucose—6—phosphate dehydrogenase, lactate dehydrogenase, pyruvate kinase, yeast hexokinase type V, tg-D-glucose penta- acetate, ATP, ADP, AMP, NADH, N—acetylglucosamine, glucose—6- phOSphate, Tris (hydroxymethyl) aminomethane (trizma base), and phosphoenolpyruvate were obtained frOm Sigma Chemical Company, St. Louis, Missouri. D-glucose—UL—C—l4 was obtained from Mallinckrodt/Nuclear, St. Louis, Missouri. Lithium alumin— um hydride was obtained through Metal Hydrides, Incorporated, Beverly, Massachusetts. Eastman Kodak's 30 percent hydrogen bromide in glacial acetic acid was a gift of Dr. W. Wood. Exchange resins AG 3 x 4 and 50 w x 8 were purchased from Bio- Rad Laboratories, Richmond, California. Animals Adult male rats Of the Sprague—Dawley strain were used for all experiments. The rats were fed a stock laboratory animal diet ag_l£b. Brains from 175—250 gram unstarved rats were re- moved immediately after decapitation and placed on ice. All subsequent Operations were performed at 0—4 degrees. 18 19 Soluble and Particulate Enzyme Preparations One or two brains were weighed and homogenized in a volume of 0.25 M sucrose correSponding to ten times the brain weight. Homogenization was accomplished with fifteen passes of a teflon pestle in a .004" to .006" clearance, 45 m1 glass vessel. The crude homogenate was centrifuged at 1,000 x g in a Lourdes Beta-fuge for twenty minutes. The supernatant from the 1,000 x g Spin was centrifuged again at 40,000 x g for twenty minutes. The supernatant from the second Spin was designated the "soluble fraction" and contained soluble hexokinase. The pellet was re- suSpended in a volume of 0.25 M sucrose equivalent to four times the original brain weight. This "particulate fraction" was either used directly or washed one or more times with 0.25 M sucrose by repeating the last centrifugation and resuspension procedure. Assay of the supernatants from these washes indicated that negligible hexokinase activity was eluted from the particles by this procedure. Purified mitochondrial particles were Obtained by centri- fuging 10 m1 of the crude preparation, layered over 15 ml of 1.2 M sucrose, at 40,000 x g for one and a half hours. The pellet was resuSpended in a minimum volume Of 0.25 M sucrose. 20 Hexokinase Assays C—l4—qlucose assay — The reaction was typically performed in a volume of 20 “1 containing 1 mM glucose, 5 mM magnesium chloride, 20 mM Tris—HCl pH 7.4, 0.75 mM ATP, and 0.11;;C of D-glucose—UL-C—l4. The assay was performed at room temperature, approximately 22 degrees. Reaction was initiated by addition of enzyme and stopped by either adding 20 pl Of absolute ethanol or heating the reaction mixture in a boiling water bath for twenty seconds. Heating was generally preferred since it elimi— nated precipitation of salts and protein. After the reaction was stepped, 4 pl Of 30 mM carrier glucose-6—phosphate was added and 20 p1 aliquots were spotted on Whatman No. l, 46 cm x 57 cm, filter paper sheets. The spots were dried and placed in a Gilson High Voltage ElectrOphorator, model D, containing pyri- dine acetic acid buffer at pH 6.5. Electrophoresis was continued for one and a half hours at 2,000 volts. The paper was then dried at 170 degrees in a chromatographic drying oven. After approximately one hour in the oven, glucose and glucose—6—phos— phate spots could be seen under ultra—violet (U. V.) light. The reaction that produced the U. V. visible Spots was time and temperature dependent. The location of radioactive Spots was determined using a Packard Radiochromatogram Scanner or by cutting the paper into sections and counting these in a liquid scintillation counter. The location of 6-phOSphogluconate 21 which did not produce a U. V. visible Spot was determined by both methods. The radioactive spots were cut out and placed in scintil- lation vials. Liquid scintillation solution containing 0.3 grams POPOP and 5 grams PPO per liter Of toluene was added and the vials were counted in a Packard Liquid Scintillation Counter for ten minutes. Channels ratio was used to determine the ex— tent of quenching. This assay was used when inhibition by glucose—6—phosphate was studied with crude preparations. Glucose-6-phosphate dehydrogenase assay — The assay solu— tion contained, in a total volume of 1.0 ml, 2.0 or 25 mM glucose, 5 mM ATP, 10 mM MgCl 50 mM Tris-HCl buffer, pH 7.4, 0.3 mM 2. NADP+, and 1.0 international unit Of glucose-6-phOSphate dehy- drogenase, The assay solution was incubated at 25 degrees for five minutes prior to initiation of the reaction with enzyme. After enzyme was added, the absorbance at 340nm was recorded using a Coleman Hitachi 124 double beam spectrophotometer and Sargent SRL recorder. The reaction was kept at 25 degrees by circulating water from a constant temperature bath through a water jacketed cell housing. Lactate dehydrogenase and pyruvate kinase assay — This assay was used to study glucose—6-phosphate inhibition Of DEAE 22 cellulose purified soluble hexokinase. The presence of excess ATPase activity in crude soluble and particulate fractions prevented the use of this assay with these preparations. The reaction solution contained in a total volume of 1.0 ml, 2 mM glucose, 5 mM ATP, 10 mM MgC12, 50 mM Tris-HCl buffer, pH 7.4, .76 I units of pyruvate kinase, 2 pg lactate dehydrogenase, 1 mM phOSphoenOlpyruvate, 0.15 mM NADH, and 50 mM potassium chloride. Reaction was initiated by the addition of enzyme and the progress of the reaction was recorded as the decrease in absorbance at 340nm in the same manner as described for the glucose—6—phosphate dehydrogenase assay. Michaelis-Menten Kinetics The glucose-6—phOSphate dehydrogenase and lactate dehydro- genase—pyruvate kinase assays for hexokinase activity were used to determine apparent Km's and inhibition constants. All re- actions were run at 250 unless stated otherwise. Kinetic constants were determined from Lineweaver—Burk reciprocal plots of (V0)-1 as a function Of (substrate)-1. Eadie—Hofstee and Wolf plots gave constants identical to those Obtained from Lineweaver-Burk plots. The data when plotted in reciprocal form.was linear enough to eliminate the need tO use least squares for slope and intercept determination. Although experimental error was hard to evaluate, the error inherent in Lineweaver—Burk plots was estimated 23 to be lower than the experimental error. The values reported for Michaelis constants and inhibition constants are estimated to be accurate within 10 percent. All concentrations are expressed as moles per liter unless otherwise stated. Synthesis 9: Anhydroglucitol—6—phosphate The procedure described by Ness, Fletcher, and Hudson (31) was used to synthesize l,5-anhydroglucitol—6-phosphate. H OH 1,5—Anhydroglucitol Tetraacetyl-cx—D-glucopyranosyl bromide was formed by adding 25 ml of 30 percent hydrogen bromide in glacial acetic acid to 23.75 grams Of B—D-glucose pentaacetate. The solid was dis— solved within thirty minutes. Ether was added after two hours at room temperature and the organic layer washed once with ice water and once with cold, saturated sodium bicarbonate solution. The tetraacetyl—cx-D—glucopyranosyl bromide was dried over so- dium sulfate and then slowly (sixty minutes) and with stirring added to 350 m1 of 1.8 M (23.8 gram) lithium aluminum hydride solution in ether. The reaction was protected from atmospheric moisture with Drierite tubes. Thirty minutes after addition 24 was complete, 400 m1 of water was very cautiously added (almost two hours was required). This solution was stored overnight. Aluminum hydroxide was removed from the aqueous phase by vacuum filtration. This solution was deionized by passing it through columns of Bio-Rad 50 w -x 8 (150 gm) and Bio-Rad AG 3x —4 (125 gm). The deionized solution was evapo- rated in_yagug at 650 to a thick syrup and dissolved in 100 ml of hot ethanol. The product did not crystallize from this solution so some ethanol was boiled off until the volume was approximately 75 ml. This solution was crystallized in the cold overnight to yield 4.25 grams (42.5% yield). The product had a melting point of 143—144, while the authentic material is reported to melt at 142-143 (31). Reaction: 1. LiAlH4 2. H20 Glucose HBr Tetraacetyl- 1,5-Anhydro- Pentaactate glucopyranosyl— gluc1tol bromide The isolated 1,5—anhydroglucitol was phosphorylated using a modification of the procedure described by Ferrari, Hendelstam, and Crane (32). The reaction solution adjusted to pH 8.2 contained 7.5 mmoles Of 1,5—anhydroglucitol (1.23 gm), 7.8 mmoles ATP, 6 mmoles magnesium chloride, and 5 mmoles sodium ' 25 bicarbonate in a final volume of 150 ml. After addition of 50 mg of yeast hexokinase (Sigma type V, 50 units/mg), the solu- tion was incubated overnight at 37°, Using the lactate dehydrogenase—pyruvate kinase assay without glucose, the con- centration of ADP was determined and with the glucose-6—phosphate dehydrogenase assay, ATP was determined. The reaction was only 60 percent complete after twelve hours. An additional 10 mg of hexokinase was added and incubatiOn continued for six hours, after which the reaction was 80 percent complete. The nucleo— tides were separated from l,5—anhydroclucitol—6-phosphate by adding 40 ml of 50 percent trichloroacetic acid fOllowed by 10 g Of acid washed Norit A charcoal. After vacuum filtration, the filtrate was treated three more times with charcoal adjusting the pH to 0.8 with trichloroacetic acid before adding charcoal to prevent absorption Of l,5-anhydroglucitol—6—phosphate. The nucleotide free filtrate was neutralized to pH 8.2 with 0.3 N barium hydroxide. The insoluble residue was removed by vacuum filtration and 3 volumes (1500 ml) of 95 percent ethanol was added. The precipitate that formed after standing overnight at 40 was collected by vacuum filtration and washed with ethanol and then ether and dried_ig_yaggg. The crude barium salt of 1,5—anhydroglucitol—6-phosphate (2.81 grams) was dissolved in 25 ml of water. The barium.was exchanged for hydrogen using 26 Dowex 50 hydrogen form. Cyclohexylamine was used to neutralize this solution to pH 8.2. The solution was evaporated in yagug until a precipitate formed. Most of the water was removed by repeated evaporation using absolute ethanol. The precipitate was finally dissolved in a minimum of ethanol and allowed to crystallize at 40. The product was recrystallized three times from absolute ethanol and dried over P205 to give 1.66 grams of crystals. Theoretical yield for phOSphorylation was 3.3 grams (50% yield). The product gave a single spot positive to periodate and acid molybdate Spray after silica gel chromatogra— phy using a methanol-formic acid—water (80:15:5 v/v) solvent system. The product had Rf slightly smaller than glucose—6— phosphate but greater than glucose. DEAE-Cellulose Chromatography Soluble fractions (100 m1),which had been accumulated from previous enzyme preparations and stored at -20°, were thawed and applied to a 3 x 35 cm DEAE—cellulose column packed under 1 lb/in2. The column was washed with 400 ml, 0.01 M potassium phosphate buffer, pH 7.0, containing 0.5 mM EDTA and 10 mM glu— cose. Hexokinase was eluted with 500 ml of a linear gradient from 0 to 0.6 M KCl in phosphate buffer containing 0.5 mM EDTA and 10 mM glucose. Forty 10.5 ml fractions were collected and assayed for protein and hexokinase activity. Hexokinase was 27 eluted at an ionic strength equivalent to 850—1480 mhos. The fractions containing the majority of hexokinase activity were pooled and dialyzed for six hours against 0.01 M phosphate buffer, pH 7.0, 0.5 mM EDTA, and 0.1 M glucose. The enzyme was then concentrated using Aquacide III. Solubilization 9f Particulate Hexokinase Rat brain particles were isolated using the standard pro- cedure outlined above. The solubilizing agent in a volume of 1 ml was added to 3 m1 centrifuge tubes containing 1 ml of particles. In some experiments 0.8 ml of particles and 0.2 ml of solubilizing agent was used. The solutions were mixed by slow inversion and then incubated at 25°C for thirty minutes followed by centrifugation for twenty minutes at 40,000 x g. The supernatants were removed and assayed for hexokinase activi— ty. Percent solubilization was based on the units of solubil- ized hexokinase in the supernatant compared to the units of particulate enzyme added. The distinction between soluble and solubilized enzyme must be emphasized. The soluble enzyme is Obtained after centrifugation of the crude rat brain homogenate, while solubilized enzyme is Obtained by incubation of particulate hexokinase with a solubilizing agent. RESULTS The hexokinase activity of both soluble and particulate fractions was found to be directly proportional to protein concentration as shown by Figure l. Negligible particle swelling or settling was observed with most particle prepara: tions when either glucose or ATP was omitted from the assay solution. With purified mitochondrial preparations, particle swelling became a significant problem. The apparent Michaelis constants were determined for ATP and glucose using soluble and particulate hexokinase. Origi- nally 370 was chosen as the reaction temperature, since the rate would be faster and less enzyme could be used. The soluble enzyme gave linear rates at this temperature for at least five minutes when either glucose of ATP was the varied substrate. If ATP was varied and glucose held constant at 25 mM, the particulate enzyme also gave linear kinetics. However, nonlinear continuously decreasing rates were Obtained if ATP was held constant at 4 mM and glucose was varied. When the temperature was reduced to 25°, both enzyme fractions gave linear rates for a minimum of three minutes under most assay conditions. Table III summarizes the results Obtained at 370 and 25°, while Figure 2—5 illustrate typical Lineweaver—Burk 28 29 Figure l.——Linearity of glucose—6—phosphate assay for soluble and particulate hexokinase. The hexokinase assay contained in 1.0 ml, 25 mM glucose, 4 mM ATP, 7.5 mM MgC12, 75 mM Tris-HCl buffer, pH 7.4, 0.3 mM NADP+, and 1.0 I. U. of glucose-6—phesphate. Par— ticulate and soluble hexokinase was isolated as described under METHODS. Protein was determined by the method of Lowry gt a}. (42). A 2x _E\mEV coinicoucou 520.5 H wusmflm «I od own 0.... o..n o..« a... so.“ :O.¢ 0; a :5 AV. .mu. L.O.0 D .6... 0* O — D U m *5 Cl 3 00°—v .nb— x .:_E\»o_o=:: 33:: 31 Table III. Apparent Michaelis Constants for Soluble and Particulate Hexokinase. Varied Constant Apparent Michaelis Constant Tem} Exp. No. Substrate Substrate Soluble Particulate (ATP) mM (glucose) mM ATP (mM) 1 .02—1 25 0.29 0.23 37° 2 .02-.4 25 0.20 37° 3 .02—1 25 0.15 37° 4 .04-.4 25 0.17 37° 5 .04-.4 25 0.25 25° 6 .04-.4 25 0.16 25° 7 .08—.8 25 0.24 25° 8 .02—0.1 25 0.24 25° 9 .02—0.1 25 0.19 25° 10 .02—0.2 .05 0.28 25° 11 .02—0.2 .05 0.68 25° 12 .2—1 2 0.25 250 13 .2-1 2 0.25 25° 14 .2—1 2 0.25 25° average at 25 mM 0.24 0.19 250 average at 2 mM 0.25 0.25 25° (Glucose) mM (ATP) mM Glucose (mM), 0 15 .02—.2 4 .069 37 16 .04—.2 4 .036 37° 17 .04-.2 4 .054 37g 18 .04—.4 4 .020 .038 25 19 .04—.4 4 .059 25° 20 .04—4 4 .057 250 21 .04-.2 4 .050 250 22 .025-.1 4 .025 25° 23 .025—.2 4 .038 25° average .040 .052 The apparent Michaelis constants were determined as described under METHODS. 32 plots used for the determination of apparent Michaelis constants for ATP and glucose for both soluble and particulate hexokinase. Because there has been debate concerning the best way to plot kinetic data to Obtain apparent Km and Vmax’ the values obtained from Eadie—Hofstree (Vo vs Vo/S), Wolf (S/Vo vs S), and Lineweaver- Burk plots were compared (33). Since the constants Obtained were essentially identical, Lineweaver—Burk plots were routinely used for calculating Michaelis constants. Although the Km values occasionally varied from one pre— paration to another, the average values for particulate and soluble enzymes are approximately the same. Even the extreme values did not Show a difference greater than two fold between bound and unbound hexokinase. The values obtained at 370 were Similar to those found at 250 suggesting that if the Km repre— sents a true binding constant the equilibrium is relatively unaffected by temperature and the enthalpy of binding must be small. Changing the concentration of glucose from 25 mM to 5011M also did not seem to effect the binding of ATP. As shown in Table III the apparent Km values for ATP are essentially the same at 25 mM and 2 mM glucose for both soluble and particu- late enzyme. The data from Table III indicates that there is no difference in binding of ATP or glucose to soluble hexokinase or bound mitochondrial hexokinase. 33 Figure 2.—-Inhibition Of soluble hexokinase by N—acetylgluco- samine. Hexokinase activity was determined using the glucose-6-phOSphate dehydrogenase assay. Sol- uble enzyme was isolated as described under METHODS. 25° K = 2.5 x 10"5 M m _ —4 K0 _ 1.7 X 10 M 1 34 o— e .ZS.16 .2:.Q6 0:02 .0:::u-o«:_u_>ucou.zv x1 fl 4 mucus—0v — I) .. on .00 .N' o—x o> 35 Figure 3.-—Inhibition of particulate hexokinase by N-acetyl- glucosamine. Hexokinase activity was determined using the glucose—6—phosphate dehydrogenase assay. Particulate enzyme was isolated as described under METHODS. _ -5 m K. = 1.5 x 10‘4 M 1 36 #0:. a m ¢<:.0Au ¢<:.¢Av 0:02 “GEE—encozuzosunze a D ‘ ‘ Aeneas—0v m musmflm :2 ..Q« 37 Figure 4.-—Inhibition of soluble hexokinase by 1,5-anhydroglu— citOl-6—phosphate (An G6P). Hexokinase activity was determined using the glucose-6-phosphate dehy— drogenase assay. Soluble hexokinase was isolated as described under METHODS. K = 2.5 x 10—4 M m K. = 4.1 x 10‘5 M 1 38 3:: v mnsmfim 9x I n _ m e m a l. a. 1...- ..o .8. 30.2 2: an 4 .305 21¢“ a .v P .30.; oz 0 a :QN eon I 4 5.0” .80 o> .fiOQ O—uulfll 39 Figure 5.-—Inhibition of particulate hexokinase by 1,5- anhydroglucitol-6-phosphate (An G6P). Hexokinase activity was determined using the glucose—6-phos- phate dehydrogenase assay. Particulate hexokinase was isolated as described under METHODS. 4 K 1.9 x 10' M m K. 1 4.6 x 10’5 M 40 «In (Fe m Tasman F I up“ . I XE. 90— . m m m .395 2:3 4 .3 .30: 2: on a .30: oz 0 .. Lvmp :VN :0” 1.00 0) opxllu 41 Inhibition by N-acetylglucosamine and 1,5—anhydroglucitol— 6—phosphate was studied to determine any difference in inhibi- tion characteristics between the two forms of enzyme. N-acetyl— glucosamine is a competitive inhibitOr with glucose and 1,5- anhydroglucitol—6-phosphate is competitive with ATP. These inhibitors could indicate a difference in the binding site for ATP or glucose that was not apparent from the Km's for these substrates. Figures 2 and 3 show that the inhibition of both soluble and particulate forms by N—acetylglucosamine was com- petitive with glucose, with virtually identical Ki values. The inhibition constants Obtained with 1.5—anhydroglucitol—6— phosphate were also the same for both soluble and particulate hexokinase (Figures 4 and 5). The inhibition data summarized in Table IV and the apparent Michaelis constants all suggest that the binding sites for ATP and glucose are similar, whether hexokinase is associated with mitochondria or is free in solution. l,5—Anhydroglucitol—6—phOSphate did not inhibit rat brain hexokinase with exactly the same characteristics as glucose-6- phOSphate, although the Ki was similar to that reported for glucose—6—phosphate using purified rat (13) and bovine brain hexokinases (34). Orthophosphate is an antagonist Of inhibition and solubilization by glucose-6—phosphate. Inhibition of rat 42 Table IV. Inhibition Constants for Soluble and Particulate Hexokinase. Apparent Inhibition Constant DEAE- Inhibitor Type of Purified Plot Soluble Particulate Soluble Enzyme N—acetylglucosamine L—B 0.17 mM 0.15 mM Dixon 0.28 mM 0.20 mM 1,5—Anhydroglucitol- 6-phOSphate L—B 41 11M 46 11M 23 11M 41 11M Glucose—6—phosphate L—B 7.3 MM L—B = Lineweaver—Burk 43 brain hexokinase by 1,5—anhydrog1ucitol-6—phosphate is com— pletely unaffected by orthophosphate. Figure 6 shows the results of an experiment with a purified soluble enzyme, and similar experiments using particulate and crude soluble pre- parations and the g1ucose—6—phosphate dehydrogenase assay confirmed this insensitivity to phosphate reversal of anhydro- glucitOl-6—phosphate inhibition. Solubilization by 1,5- anhydroglucitol—6-phosphate is also independent of phosphate (Table V). This difference made it somewhat tenuous to draw conclusions about glucose-6—phOSphate inhibition from these studies with its nonphysiological analog. Owing to the problems posed by ATPase activity in the preparations, the inhibition by glucose—6—phosphate was studied using a C—14 glucose assay. True initial velocities were impossible to obtain since the rate was not linear under the conditions necessary for this asSay. A high enzyme activity was required to get significant incorporation of relatively low Specific activity C-l4 glucose into glucose—6—phOSphate. The C-l4 glucose was contaminated with approximately 0.01 percent glucose-6-phosphate which pro— duced a substantial blank rate. Product inhibition by g1ucose—6—phosphate was responsible for the nonlinear rates Observed using the C—14 glucose assay, as shown by the fact that this inhibition could be removed by 44 Figure 6.—-Inhibition of DEAE-cellulose purified soluble hexokinase by g1ucose-6-phosphate and 1,5-anhydro— glucitol—6-phosphate.(Afi G6P). Hexokinase activity was determined using the lactate dehydrogenase- pyruvate kinase assay. Soluble hexokinase was purified by DEAE—cellulose chromatography as des- cribed under METHODS. K = 3.3 x 10'4 M m Ki (An G6P) = 2.3 x 10'5 M Ki (G6P) = 7.3 x 10‘6 M 45 m musmflm N n N p w: «a 211:: n — :2 .30 .2: an s E 2E2 . + .30 E: 3 i: E 262 ... .392 2: on ... 30.2 S: on :‘N O .232... oz 6 ..on m at on ... q I e: I +3 o> I 2..an 46 Table V. Reversal of Glucose—6—Phosphate Solubilization by Orthophosphate. Solubilizing Conditions .3 mM G6P .3 mM G6P + 5 mM Pi .3 mM An G6P .3 mM An G6P + 5 mM Pi Percent Solubilized 65 33 60 55 47 adding glucose-6—phosphate dehydrogenase and NADP+. When g1ucose-6—phosphate was removed eight minutes after the initiation of the hexokinase reaction, the rate increased and became linear for at least five minutes. Orthophosphate was also effective in partially removing the product inhibition. Both soluble and particulate hexokinase reSponded similarly to reversal Of this inhibition by oxidation of glucose—6- phosphate and addition of phosphate. The relative sensitivity Of soluble and particulate hexo— kinase toward phosphate reversal of glucose—6—phosphate inhi— bition was studied using a known concentration of glucose-6- phosphate in the reaction solubiOn. The results are Shown in Figures 7 and 8. Since 5 mM ATP was required in the assay due to a high ATPase activity, 0.25 mM glucose—6—phosphate was necessary to achieve significant inhibition. Under these con- ditions 0.5 mM phosphate was effective in lowering the inhibi- tion. This reversal was increased substantially by 2.5 mM phosphate. Initial rates were determined by taking a tangent to the curve at time zero. Although there is a high degree Of uncertainty in these rates, they do suggest a slight difference in the sensitivity of soluble and particulate hexokinase to reversal of glucose-6—phosphate inhibition by phOSphate. Par- ticulate and soluble hexokinase were 81 and 89 percent inhibited 48 Figure 7.-—Inhibition Of soluble hexokinase by glucose-6- phosphate and reversal by phosphate. Hexokinase activity was determined asing the C—l4-glucose assay. The reaction was performed at room temper- ature (approximately 22°). Soluble enzyme was isolated as described under METHODS. O..- “1-.....- A L1. 4 m . w m e m a 1 7235.505: . . . 1 b musmb 1r” .39 2.53.0 E ten. + :0 153.0 :0— ... 56.3 + .36 283. nvnF as x .:_E\£c:ou 38:23 02 50 Figure 8.—-Inhibition Of particulate hexokinase by glucose- 6—phOSphate and reversal by phosphate. Hexokinase activity was determined using the C-l4-glucose assay. The reaction was performed at room tempera- ture (approximately 22°). Particulate enzyme was isolated as described under METHODS. 51 on ‘00 so: an. I iEn. .1 $00 is. an. : SEQ.” .1 £00 2:. «fl. acct-11¢ 02 oh 6 m AmosfievwwE: 15' 1 p .n A e o— 1 no. x.¢_E\:caou m musmfi 52 respectively by .25 mM glucose—6—phOSphate in the absence of phOSphate. .When 0.5 mM phosphate is added, the inhibition of soluble enzyme is reversed 5 percent, while the particulate is 17 percent reversed. The initial velocity data in Table VI indi— cates that the particulate enzyme is approximately 3.5 times more susceptible to reversal of glucose—6—inhibition by ortho— phOSphate. The total incorporation of C-14 glucose into glucose—6- phOSphate after eight minutes Of reaction indicates no Obvious difference between the inhibition characteristics of both forms of hexokinase. Both forms were inhibited approximately the same percent by .25 mM glucose-6-phosphate and were reversed about 10 and 30 percent by 0.5 mM and 2.5 mM phosphate respectively. The data is summarized in Table VI. The most physiologically significant way to consider the results from Table VI is on the basis of initial velocity since the instantaneous change of glucose—6—phOSphate concentration represents most accurately the activity of hexokinase under the conditions of the assay. Since ATPase is also producing ADP, an inhibitor, and consuming ATP, an antagonist of glucose- 6-phOSphate inhibition, the data on total incorporation of C—l4-glucose into glucose-6-phOSphate is impossible to inter— pret. These factors should, however, have had minimal effect 53 Table VI. Inhibition Of Soluble and Particulate Hexokinase by Glucose—6—Phosphate and Reversal of Inhibition by Phosphate. Initial Velocity Incorporation after 8 min. Conditions (Percent) (Percent) . Soluble Particulate Soluble Particulate No Additions 100 100 100 100 0.250 mM G6P ll' 19 27 40 0.25 mM G6P + .5 mM Pi 16 36 36 53 0.25 mM G6P + 2.5 mM Pi 25 65 57 77 See Figures 7 and 8 for details. 54 on the initial velocities observed. Both ways of evaluating the results confirm the kinetic studies with 1,5—anhydroglu- citOl-6-phosphate that indicate the glucose—6—phOSphate binding site is not appreciably affected by the binding of hexokinase to a membrane. Bound hexokinase, however, appears to be more sensitive than soluble enzyme to reversal of glucose—6—phOSphate inhibition by phosphate. It was not possible to Obtain kinetic constants for glu— cose—6—phosphate inhibition with crude soluble and particulate preparations since initial rates could not be obtained using the C—14 assay and ATPase activity eliminated using the lactate dehydrogenase—pyruvate kinase assay. Soluble hexokinase was purified by DEAE-cellulose chromatography and found to be free Of ATPase activity. This purified enzyme was used to study inhibition by l,5—anhydroglucitol—6—phosphate and glucose—6— phosphate and the reversal of this inhibition by phosphate using the lactate dehydrogeanse—pyruvate kinase assay. The results from this experiment are shown in Figure 8. The enzyme was eluted from DEAE-cellulose in one major peak of activity. Soluble hexokinase was purified approximately forty times by this procedure. The Km for ATP was relatively unaffected by purification, but the sensitivity to glucose—6— phosphate was approximately ten times that reported by Katzen and Schimke (14) for rat brain hexokinase. However, the 55 inhibition constant for l,5—anhydroglucitol-6—phosphate was only one half that obtained for crude soluble enzyme. The inhibition by g1ucose—6—phosphate was reversed by phOSphate, but inhibition by l,5-anhydroglucitol-6-phosphate was inde- pendent of phosphate. These results obtained using the same assay procedure confirm the difference in inhibition charac— teristics Observed using two separate assay procedures. G1ucose—6—phosphate appears to be approximately three times more efficient than l,S—anhydroglucitol-6-phosphate as an in— hibitor Of rat brain hexokinase. These results indicate a significant function Of the C-1 hydroxyl in the binding char— acteristics Of glucose—6—phosphate with brain hexokinase. The C-l hydroxyl may also be directly involved in the mechanism of phOSphate reversal. Bound hexokinase can be solubilized using salts, ATP, or glucose—6—phosphate. The latent activity once exposed by de- tergent can also be solubilized with these agents. The apparent Michaelis constants for ATP were determined for the enzyme solu- bilized before and after exposure of the latent hexokinase to establish if the enzyme obtained in the soluble fraction after homogenization was the same as the solubilized enzyme. Sodium chloride (0.9 M) was used to solubilize the enzyme to avoid the complications that would result from solubilization by ATP or 56 glucose-6—phosphate. After thirty minute incubation at 25°, only ADP is present due to ATPase. Table VII summarizes the results of the solubilization and Figure 9 shows the kinetics. Half of mitochondrial rat brain hexokinase is present in a la— tent form. The salt was effective in solubilizing virtually all of the exposed particulate enzyme, and salt plus Tritoh X-100 solubilized both the latent activity and the exposed activity. The Michaelis constants for ATP were not significantly different for soluble and both solubilized preparations. These results are consistent with the results obtained from ascites tumor and heart muscle hexokinase that suggest the enzyme is in equilibrium between a soluble and particulate form (9, 10). The enzyme solubilized by glucose-6-phosphate and 1,5- anhydroglucitol—6—phosphate was unstable and lost activity even at 00. When the incubation time was varied from zero to thirty minutes, maximum supernatant activity was Obtained at ten min- utes, after which the supernatant activity decreased. Figure 10 shows the solubilization Observed by 1,5-anhydroglucitol-6- phOSphate as both a function of time and concentration. Similar results were observed using glucose-6—phosphate. The superna— tant activity was maximum at 2 mM and decreased at higher con- centrations. It appears that the mechanism by which glucose- 6-phOSphate and l,5-anhydroglucitol-6—phosphate solubilizes also causes the enzyme to be less stable. 57 Table VII. Solubilization of Particulate Hexokinase. Solubilizing Conc. Percent Agent Solubilization G6P 3 mM 106 G6P + 0.5% Triton 3 mM 202 An G6P 2 mM 106 ATP 20 mM 113 ATP + 0.5% Triton 20 mM 206 NaCl 0.9 mM 98 NaCl + 0.5% Triton 0.9 mM 222 0.5% Triton 61 None 20 58 Figure 9.--Soluble and solubilized hexokinase apparent Michaelis constants for ATP. Hexokinase was deter— mined using the glucose-6—phosphate dehydrogenase assay. Particulate hexokinase was isolated and solubilized as described under METHODS. Particulate hexokinase was solubilized with 0.9 M NaCl and 0.9 M NaCl + 0.5% Triton X-100. Km (soluble) = 2.5 x 10—4 M Km (solubilized NaCl) = 4.3 x 10'4 M Kh (solubilized, NaCl + Triton) = 3.7 x 10"4 M 59 m Tasman q 1 C I - ton—.32....“ 1 2.9-0.— d 10u___aa_om I 0.33.0.» 0 0> 8...... 60 Figure 10.--Solubilization by l,5—anhydroglucitol-6—phOSphate. Hexokinase was determined using the glucose-6- phosphate dehydrogenase assay. Particulate hexo- kinase was solubilized as described under METHODS. a 2 x 1226356126562are: 61 e a a. 1. .2; Wu 0 o .— 3 0.33.1 .55 n. O O catamatzcou .2: 05:. 0 an O 0 en 0 12V 0 SW 3 R «.0... 2.0.0503.» E 22.: 30:30.3: DISCUSSION Most tissues studied except for liver have been found to contain a soluble and particle-bound hexokinase. Existing evidence indicates that the enzyme does not bind to liver be— cause either ip yiyp or during homogenization proteolytic enzymes cleave a peptide segment required for binding but not necessary for catalytic activity (9). The conditions under which the bound enzyme can be solubilized would provide a unique regulatory mechanism if the particulate enzyme were more active than the soluble enzyme under the same conditions. Glucose—6—phosphate and ATP solubilize considerable amounts of bound hexokinase at concentrations that have been determined to be physiological in the brain. Phosphate antagonizes the solubilization by glucose—6—phOSphate also at physiological concentrations (7). The Specific activity of hexokinase in— creases as much as twenty times after solubilization indicating that few other proteins are solubilized (7). Under conditions of energy demand, the concentrations of ATP and glucose—6- phosphate would be low, and inorganic phosphate would be high, providing a medium in which hexokinase (at least_ip_yippp) binds to mitochondria (7, 9, 11). If particulate hexokinase 62 63 had more catalytic potential than soluble enzyme, glucose phOSphorylation would be accelerated. This proposed correl— ation between soluble and particulate hexokinase would explain the observed difference in the rate of glucose utilization under conditions of annoxia and energy demand. Proposed Regulation_pf Hexokinase and Glycolysis Currently there are two schools of thought on the regula— tion of hexokinase activity in response to glycolytic potential. It is generally accepted that the primary control point is at phosphofructokinase, which is regulated by the concentrations of 5'-AMP, 3',5'-cyclic AMP, ADP, ATP, and citrate. The acti— vity of phosphofructokinase is roughly proportional to the ADP/ATP ratio (4). An increase in glucose phosphorylation must follow an increased rate of fructose-6—phosphate phOSphoryla— tion. One school proposes that the rate of hexokinase is controlled by glucose-6—phosphate product inhibition, and the other suggests that regulation is more complicated and involves the particulate-soluble distribution of enzyme. If glucose—6— phOSphate were the controlling factor, hexokinase activity would directly follow phOSphofructOkinase activity. Since the equilibrium for the phOSphoglucoisomerase reaction lies in favor of glucose—6—phOSphate, when phOSphofructOkinase is inac— tive, g1ucose—6—phosphate would build up and turn off hexokinase. 64 Glucose phOSphorylation would continue when the ADP/ATP ratio increased enough to activate phosphofructokinase. Racker has studied the regulation of glycolysis in re- constituted systems using purified mammalian enzymes. The formation Of triose—phosphate was followed Spectrophotometrically eliminating the direct participation of phosphate. The produc- tion of triosephosphate was determined by oxidation of NADH in the presence of triosephosphate isomerase and a—glycerolphos- phate dehydrogenase. The complete glycolytic system was also studied by following lactate formation. The results indicated a coordinated stimulation of hexokinase and phosphofructokinase by orthophosphate. Only Slight changes in the level of glu— cose-6-phOSphate were Observed compared to large changes in the rate of glucose conversion to triosephosphate and lactate. When mammalian hexokinase was replaced by yeast hexokinase, the coordinated control was lost. (Yeast hexokinase is not inhibi- ted by glucose-6—phosphate). The coordinated control by orthOphos— phate can be interpreted as an antagonism of the glucose—6— phosphate inhibition of hexokinase and ATP inhibition of phosphofructokinase. Under this type Of control the two kinase reactions of glycolysis would function in unison in response to the concentration of orthophosphate. A third reaction, the oxidation Of glyceraldehyde—3-phosphate, also requires ortho- 65 phosphate. This coordinated regulation of glycolysis by orthophOSphate would be influenced by the competition Of oxidative phOSphorylation for ADP and orthophosphate (4). Comparison_p§ Apparent ATP Michaelis Constants for Soluble and Particulate Hexokinases A more complicated regulation Of hexokinase is supported by recent reports that indicate the apparent Michaelis con— stants for ATP and inhibition constants for g1ucose-6—phosphate are different for soluble and mitochondrial hexokinase. Tables I and II compare literature kinetic constants reported for soluble and particulate hexokinase from a variety of tissues and sources. From Table I it becomes Obvious that except for bovine brain and frog skeletal muscle the apparent Michaelis constants for glucose and ATP are not changed drastically by association and dissociation from a membrane. Ascites tumor hexokinase has been reported to have different kinetics depen- ding On its location (22), but these Observations have not been confirmed by other workers (23, 24, 35). The apparent Michaelis constant for ATP observed with particulate hexokinase is rela— tively constant for most tissues studied. The differences reported between particulate and soluble enzymes are a result of an unusually high Km for ATP Obtained with soluble hexokinase. Solubilized and purified bovine hexokinase is reported to have 66 a Km for ATP of 1.7 mM and 4.9 mM, or 5.0 and 14.4 fold higher, respectively, than particulate hexokinase (0.34 mM). The kin- etics were performed on 1129 fold purified solubilized enzyme. In contrast to the reported high Km for purified bovine brain hexokinase, crude rat brain hexokinase has a Km for ATP of about 0.4 mM for the salt—solubilized enzyme in agreement with the Km (.4 mM) Observed by Grossbard and Schimke (13) for 150- fold purified rat brain hexokinase. The average ATP K.m Ob- tained for soluble and particulate rat brain hexokinase was 0.25 mM. Although crude and purified rat brain hexokinase have similar Michaelis constants for ATP, it is possible that manipulation during purification could have caused an increase in the apparent ATP binding constants for the bovine enzyme. Substrate binding constants have been reported to in- crease after the purification of other enzymes (36). Also the apparent Michaelis constant for ATP and glucose—6—phosphate inhibition for dog heart hexokinase has been reported to in- crease on aging of the enzyme (37). The apparent binding constants for ATP were also different for soluble and particulate frog muscle hexokinase. Soluble hexokinase was approximately four fold less effective in bind— ing ATP compared to the particulate enzyme. Again the difference is due to a relatively high ATP Km for soluble enzyme. The 67 enzyme was purified but not to the same extent as solubil- ized bovine hexokinase. However, the possibility remains that the difference is a result Of the purification procedure. Comparison pf Inhibition py Glucose-6—Phosphate and lLS—Anhydroglucitol-6-Phosphate Literature values for glucose-6—phosphate inhibition con— stants are compared in Table II. Except for rat heart and rat brain the soluble enzyme appears to be slightly more sen- sitive to glucose—6—phosphate or l,5—anhydroglucitol—6-phos- phate inhibition. However, the maximum differences between the two forms do not generally exceed two fold. Bovine brain is an exception. The apparent glucose-6-phosphate dissociation constant reported for purified bovine hexokinase (11.7 MM) was almost six times lower than the value reported for particulate bovine enzyme (18, 19). Unlike bovine hexokinase no difference in inhibition of soluble and particulate forms could be seen for rat brain enzyme. The inhibition constants for 1,5-anhy— droglucitol—6-phosphate were approximately 40 pM for both soluble and bound rat brain hexokinase, but soluble enzyme purified forty fold by DEAE cellulose chromatography was found to have an apparent inhibition constant of 23 31M. This in- dicates that l,S—anhydroglucitol-6-phOSphate binding may be sensitive to purification. 2K1 values for g1ucose—6-phosphate 68 inhibitors of crude particulate and soluble enzymes could not be obtained due to ATPase activity and the nonlinearity of the C-14 assay. The C-14 assay did, however, indicate that both soluble and particulate enzyme were inhibited by glucose—6- phOSphate to nearly the same extent. The glucose-6-phOSphate Ki determined for 40 fold purified hexokinase was 7.3 “M. It is significant to note that the inhibition constants determined for DEAE-cellulose purified enzyme were approximately three times lower for glucose-6—phOSphate than for 1,5-anhydroglucitol— 6—phosphate. Thus glucose—6—phosphate is a somewhat more potent inhibitor than 1,5—anhydroglucitol—6—phOSphate. Another difference between glucose—6—phOSphate and 1,5- anhydroglucitol—6—phosphate inhibition of rat brain hexokinase was discovered when the reversal by phosphate was studied. Al- though phosPhate is able to decrease the inhibition produced by glucose—6—phosphate, it has no effect on 1.5-anhydroglucitol- 6-phosphate inhibition (Figure 6). In contrast, Karpatkin (25) has reported that orthOphosphate will decrease inhibition Of frog skeletal muscle hexokinase caused by 1,5-anhydroglucitol- 6-phosphate for both soluble and particulate hexokinase. The particulate enzyme (sarcoplasmic vesicles) was significantly more susceptible to the reversal by orthophOSphate. At 1 mM 1,5—anhydroglucitol—6—phosphate, 20 mM phOSphate, and 11.5 mM ATP the particulate enzyme showed an increase of 253 percent 69 over the rate without phosphate, while inhibition of the soluble enzyme was not reversed. Again in contrast to the results Observed with rat brain hexokinase, the soluble and particulate hexokinases of frog Skeletal muscle have been reported to differ in their sensi— tivity to inhibition by 1,5-anhydroglucitol-6—phosphate (25). Soluble (two fold purified) frog skeletal muscle hexokinase was approximately three times more sensitive to 1,5—anhydro- g1ucitol—6—phosphate inhibition with 8 K1 of 26 MM compared to particulate hexokinase with a Ki of 80 pM. The apparent l,5—anhydroglucitol-6—phosphate Ki for soluble frog hexokinase is similar to the value obtained for crude soluble and par— ticulate rat brain hexokinase and identical to the glucose—6- phOSphate Ki reported for 150 fold purified rat brain hexokinase (13). Besides the particulate enzyme being less sensitive to l,5-anhydroglucitol—6-phosphate and having a higher affinity for ATP, frog skeletal muscle hexokinase was also Observed to be more stable to heat denaturation than the soluble enzyme. Karpatkin (38) has recently purified soluble frog skeletal muscle hexokinase 180 fold and identified by Sephadex G—200 gel filtration a "heavier" molecular form and "lighter" molecular form. The molecular Species could be interconverted by changing ionic strength and appeared to be in equilibrium. 70 The "heavier" hexokinase displayed many of the kinetic char- acteristics of the particulate frog skeletal muscle hexokinase. Karpatkin prOposes that association and dissociation of hexo- kinase from the sarcoplasmic reticulum may involve the aggregation and disaggregation of "lighter" hexokinase and that this interconversion represents an ip.yiyp regulatory mechanism. This is the first time multiple interconvertible molecular forms of animal hexokinase have been reported. This may point to a distinct difference between amphibian and mammal— ian hexokinases, or the "heavier" aggregated form from mammalian sources may only exist when associated with the mitochondrial membrane. Wen-Yuu Li and JO-Lan Ch'ien (22) have reported that par- ticulate ascites tumor hexokinase had a higher affinity for ATP and a lower sensitivity to glucose—6—phOSphate inhibition. They also stated that the "functional state of mitochondria" influenced hexokinase activity. The location Of hexokinase on mitochondria is suggested to "facilitate the enzyme's taking up the ATP synthesized by mitochondrial oxidative phosphory— lation" (22). The same group found that insulin and thyroxine inhibited hexokinase activity. This is the first time any— one has observed these characteristics with ascites tumor hexo— kinase. Sauer (23) has studied both soluble and particulate 71 ascites tumor hexokinase and found them to be similar. Kosow and Rose (24) studied purified hexokinase II in soluble and particulate form and found only slight differences with respect to type of inhibition and inhibition constants for 1,5—anhy- droglucitOl—6—phosphate. NO differences in the binding affinity for ATP was Observed between the two forms of enzyme. Except for the report by'Wen-Yuu Li and Jo-Lan Ch'ien the studies with ascites tumor suggest that the tumor enzyme is similar to rat brain hexokinase in that the kinetics of both are not dramatically affected by binding to mitochondria. Oxidative Phosphorylation and Bound Hexokinase Some other workers have attempted to correlate the loca- tion of hexokinase on mitochondria with the functioning of oxidative phOSphorylation. Rose and Warms (9) Observed that the competition for ATP was relatively the same for bound ascites tumor hexokinase and added soluble glycerol kinase whether the ATP was generated externally by creatine—phOSphate and creatine kinase or internally by oxidative phosphorylation. They also found that bound hexokinase was external to the ATP transport barrier that is blocked by atractyloside. Sauer (23) also studied the interaction between bound hexokinase and oxi— dative phosphorylation in ascites tumor. Glucose-6—phOSphate was produced at the same rate whether ATP was generated exter— 72 nally by the addition of pyruvate kinase, phOSphoenOlpyruvate, and ADP or intramitochondrially by oxidative phosphorylation. The oxidative and phosphorylative activity of the mitochon— dria from rat brain has been reported to be unaffected by the presence or absence of bound hexokinase (7). The rate of oxygen consumption was measured (using ADP and succinate as substrates) before and after elution of hexokinase with glu— cose—6—phosphate or ATP. Oxidative phosphorylation did not cause solubilization and addition of an uncoupler (2,4—dinitro- phenol) or inhibitor (KCN) had no effect on hexokinase. The reported differences between soluble and bound ascites tumor hexokinase must remain tentative until other experiments con- firm these results. Kinetically Similar Soluble and Particulate Hexokinases Soluble and particulate rat heart hexokinase have Similar kinetic characteristics. The Michaelis constants for glucose and ATP appear to be exactly the same for both forms of the enzyme, 0.045 and 0.5 mM respectively. The inhibition constant for ADP with reSpect to ATP was the same, but the particulate was found to be twice as sensitive to glucose—6—phosphate in— hibition as the soluble enzyme (26, 27). This is in contrast to the reported differences in frog skeletal muscle, ascites 11 an an 73 tumor, and bovine brain where the particulate is less suscepti- ble to glucose—6—phosphate inhibition. However, the inhibition of soluble rat heart hexokinase was studied by following the formation of ADP, and the results may be inaccurate due to ATPase activity. Inhibition of particulate enzyme was assayed using a C—14 glucose assay. The difference reported for the two forms of enzyme could also be due to the difference in assay procedures. The kinetics of soluble and particulate hexokinase from rat heart were used to predict the rate of glucose phosphory- lation in perfused heart under varying conditions. When the concentrations Of glucose—6—phOSphate, ADP, AMP, orthOphOSphate, and ATP were determined in perfused hearts, the rates of glu— cose phOSphorylation in these hearts could be accurately pre- dicted from the kinetic prOperties of hexokinase. England and Randle (27) concluded from these studies that the activity of hexokinase in rat heart is controlled by glucose-6—phosphate product inhibition. This control is modulated by ATP and inorganic phOSphate. Guinea—pig cerebral cortex soluble and bound hexokinases are also quite Similar with reSpect to ATP binding constants, 0.4 mM (16). The average glucose binding constant for soluble and bound hexokinase was .074 and 0.14 respectively. This bl 81‘. th the mit Exp Slg 74 difference can be attributed to the presence of hexokinase type III (which has a low glucose Km) in the soluble fraction. Hexokinase types I and II were present in the particulate fraction, while types I, II, and III were found in the soluble fraction. Soluble and particulate rat brain hexokinase con— sists predominantly Of type I (7). The soluble enzyme from guinea-pig cerebral cortex appears to be more sensitive to glucose—6—phOSphate inhibition (16, 20). At 0.046 and 0.023 mM glucose-6—phosphate the particulate and soluble enzymes are 50 percent inhibited respectively. The results from studies on guinea-pig cerebral cortex hexokinase also suggest that glucose phOSphorylation is controlled primarily by glucose— 6—phosphate inhibition which is modified by ATP and orthophos— phate levels. Since the soluble enzyme is more sensitive to glucose-6—phosphate inhibition, the soluble-particulate distri— bution of hexokinase can not be eliminated as a possible influ- ence on regulation. Kinetic studies on bound and soluble hexokinase indicate that aside from bovine brain and frog Skeletal muscle, where the enzyme is found in the sarcoplasmic vesicle instead Of mitochondria, the two forms are not dramatically dissimilar. Experiments with rat brain hexokinase failed to indicate a significant difference in the reaction characteristics between 75 mitochondrial and soluble enzyme. The average Km for ATP was .25 mM at 2 mM glucose whether the enzyme was associated or dissociated. The average Km for glucose was 40 “M for soluble enzyme compared to 51 MM for particulate. Similarly, the ap— parent inhibition constants for N—acetylglucosamine and 1,5— anhydroglucitol-6-phosphate for both forms Of enzyme were the same within experimental error (see Table IV). Relative rates and percent inhibition at .25 mM glucose—6—phosphate and 5 mM ATP indicated that both soluble and bound enzyme were equally sensitive to glucose-6—phosphate inhibition. The only possible difference observed was in the reversal Of glucose—6—phosphate inhibition by orthOphOSphate. Particulate hexokinase may be more susceptible to orthophOSphate antagonism of glucose-6— phosphate inhibition. The percent reversal of particulate hexokinase with 0.5 mM and 2.5 mM orthophosphate was approxi- mately 3.5 fold greater than for soluble enzyme on the basis Of initial rates. Particulate frog skeletal muscle hexokinase has also been shown to be more sensitive to reversal of 1,5- anhydroglucitol-6-phosphate inhibition than the soluble enzyme (25). Since orthophosphate has been Shown to play a major role in the regulation of glucose-phosphorylation (4), the difference in response to orthophosphate between soluble and mitochondrial hexokinase would lend support to proposed regu— In...‘ It me 76 lation based on the distribution of particulate and soluble hexokinase. The concentration dependence and Specificity of inhibition and solubilization of bound hexokinase by glucose-6—phosphate suggest that the same mechanism is involved. If glucose—6- phOSphate binds at an allosteric regulatory site that causes both solubilization and inhibition, the antagonism of both processes by phOSphate may be Significant. Phosphate would tend to keep the enzyme in a bound form that is less susceptible to glucose-6—phosphate inhibition. However, the significance of the difference in reSponse to phOSphate is hard to evaluate, Since the particulate-soluble distribution under different physiological conditions_ip.yiyp is at present impossible to determine due to the time required to separate soluble and particulate fractions and the rapid equilibrium establised be— tween soluble and particulate forms. Latent Particulate Hexokinase-lg Brain In brain the distribution between soluble and particulate hexokinase may not be as important as in other tissues, since brain contains at least 50 percent of the bound hexokinase in a latent form that is exposed only when the membrane is dis— rupted. The latent activity can only be solubilized after the membrane has been broken. The kinetics Of solubilized latent 77 hexokinase are Similar to soluble and solubilized externally bound hexokinase (see Figure 9). Starch gel electrophoresis has indicated that soluble and latent solubilized hexokinase are both the same and correSpond to isozyme type I (15). The functional significance for the multiple locations Of hexokin- ase in brain is not known. Since exogenous substrates are not accesible to latent hexokinase or more likely glucose-6—phosphate is not transported across the mitochondrial membrane, the re- lationship between intra- and extra—mitochondrial hexokinase is uncertain. It is possible that cytoplasmic glucose phos— phorylation is completely independent of intra—mitochondrial glucose-phosphorylation. It is also possible that the par- ticulate—soluble distribution may have a significant cytoplasmic function. Speculation.gp the Function_pf Soluble and Bound Hexokinase If the kinetic properties of hexokinase are not affected by its association with mitochondria or the distribution of enzyme is not regulatory, the question of why this enzyme binds so specifically to mitochondria remains to be answered. The function of hexokinase bound to the mitochondrial membrane can only at present be the subject Of speculation. An interesting possibility is that mitochondrial hexokinase provides for re- cycling ADP in oxidative phosphorylation. The flow of ATP to 78 glycolysis and ADP back to oxidative phOSphorylation might be facilitated by the association of hexokinase with mitochondria. This may be a function of "latent" internal hexokinase, but the permeability barrier to nucleotide translocation would make ATP transport slower than diffusion, and exposed bound hexokinase and oxidative phosphorylation are on Opposite sides of the membrane barrier, i. e., translocation would be rate limiting and location would be unimportant. Another question that arises is the fate Of intramitochondrial glucose-6—phOSphate. The association of many enzymes with membrane structures has stimulated interest in binding characteristics and the sig— nificance of this association. Enzymes normally thought of as soluble are being discovered in particulate fractions. Phos- phofructokinase is found in both a soluble and insoluble form in rat muscle (39). The distribution between the two forms changes with pH and age Of the tissue. Although phosphofructo— kinase has been reported to be associated with brain mitochondria, the total amount bound is insignificant compared to hexokinase (4). Balézs has studied glycolysis in rat brain mitochondria and de— termined that at least 10 percent Of brain glycolysis occurs in mitochondria (40). Conditions under which many glycolytic enzymes bind to mitochondria will probably soon be discovered. 79 Coordinated interaction between two bound enzymes provides a unique mechanism by which metabolites can be directed into a specific pathway especially if the metabolite can undergo a variety of reactions. A high adenylate kinase activity is associated with brain mitochondria (35). Hexokinase and adenylate kinase could inter— act on the membrane to produce ATP and AMP from the ADP generated by hexokinase. In this process two phOSphate esters would be produced for every nucleotide transported across the atractylo- side barrier. Since translocation is slower than diffusion, this system would be faster and more efficient than the transfer of two ADP molecules. Rat brain hexokinase solubilized with glucose-6—phosphate was less stable than the soluble or ATP solubilized enzyme (Figure 10). Inhibition and solubilization by glucose-6-phos— phate are similar and may be a result Of the same interaction between glucose-6—phosphate and hexokinase. The characteristics of glucose-6-phosphate inhibition and solubilization suggest that it is binding at an allosteric regulator site. Binding of glucose—6—phosphate to the allosteric site could cause a conformational change that both solubilizes and inhibits the enzyme. The data in Figure 10 indicates that if a conformational change does occur (in this case with 1,5-anhydroglucitOl-6- 8O phOSphate), the altered conformation appears less stable and leads to rapid loss of activity. Solubilization by ATP and high ionic strength represent a different process in which the conformation of hexokinase does not change and the enzyme remains stable. If the decreasedsstability of glucose-6- phosphate solubilized hexokinase is physiologically significant, the soluble-particulate distribution may involve turnover of the enzyme. Karpatkin (25) also found that particulate frog skeletal muscle hexokinase was considerably more stable to heat than soluble enzyme. Although this thesis has not substantiated that rat brain hexokinase activity is regulated by association and dissociation from mitochondria, the specificity with which this enzyme binds to mitochondria suggests that it has some functional Signifi- cance. An explanation Of why 80 percent of brain hexokinase is associated with subcellular particles must still be sought. The answer may involve association of hexokinase with other enzymes, the recycling of oxidative phosphorylation intermedi- ates, or the stability and turnover of the enzyme. These are at least the directions that one must look. As for control of hexokinase activity, the majority Of evidence suggests that glucose—6—phosphate inhibition and reversal Of this inhibition by orthophosphate are of primary importance. That control of 81 both kinases and overall glycolytic activity can be coordinated by orthophosphate has been shown in reconstituted mammalian enzyme systems (4). 10. REFERENCES The Nature and Function Of Hexokinase in Animal Tissues, walker, D. G., Essays in Biochemistry, Vol. 2, 1966, 33. Effect of Ischemia on Known Substrates and Cofactors Of the Glycolytic Pathway in the Brain, Lowry, O. H., Passonneau, J. V., Hasselberger, F. V., and Schulz, D. W., J. Biol. Chem., 1964, 232, 18. Allosteric Regulation of Enzyme Activity, Stadtman, E. R., Adv. in Enzymology, 1966, 28, 41. - Regulatory Mechanisms in Carbohydrate Metabolism, Uyeda, K.' and Racker, E., J. Biol. Chem., 1965, 240, 4689. Regulation of Glycolysis in Brain, In Situ, During Convulsions, Sacktor, B., Wilson, J. E., and Tiekert, C. G., J. Biol. Chem., 1966, 241, 5071. Properties of Membrane-bound Hexokinase in Rat Brain, Teichgrgbee, P. and Biesold, D., J. Neurochemistry, 1968, 15, 979. A Proposed Relationship between Soluble and Particulate Brain Hexokinase, Wilson, J. E., J. Biol. Chem., 1968, 243, 3640. Solubilization Of Mitochondrial Hexokinase from Ox Cerebral Cortex with Triton-X—lOO, Thompson, M. F. and Bachelard, H. 8., Biochemistry J., 1969, 111 (3), l8. Mitochondrial Hexokinase: Release, Rebinding and Location, Rose, I. A. and warms, J. V. B., J. Biol. Chem., 1967, 242, 1635. Association of Heart Hexokinase with Subcellular Structure, Hernandez, A. and Crane, R., Arch. Biochem. Biophys., 1966, 113, 223. 82 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 83 E. Kropp, Unpublished Data. Intracellular Distribution and Some Properties of Hexokinase of Bovine Adrenal Medulla, Toews, C. J., Canadian J. Biochem., 1966, 44, 879. Multiple Hexokinases of Rat Tissues: Purification and Comparison Of Soluble Forms, Grossbard, L. and Schimke, R. T., J. Biol. Chem., 1966, 241, 3546. Multiple Forms of Hexokinase in the Rat: Tissue Distribution, Age Dependency, and Properties, Katzen, H. M. and Schimke, R. T., Proc. Natl. Acad. Sci. U. S., 1965, 54, 1218. The Latent Hexokinase Activity Of Rat Brain, Wilson, J. E.,. Biochem. Biophys. Res. Commun., 1967, 28, 123. The Subcellular Distribution and Properties of Hexokinases in the Guinea-Pig Cerebral Cortex, Bachelard, H. 8., Biochem. J., 1967, 104, 286. The Isolation and Purification of Solubilized Hexokinase from Bovine Brain, Schwartz, G. P. and Basford, R. E., Biochem., 1967, g, 1070. Kinetic Studies of the Brain Hexokinase Reaction, A Reinvestigation with the Solubilized Bovine Enzyme, Copley, M. and Fromm, H. J., Biochem., 1967, p, 3503. Kinetic Studies of the Brain Hexokinase Reaction, Fromm, H. J. and Zewe, V., J. Biol. Chem., 1962, 237, 1661. Factors Affecting the G1ucose-6-P Inhibition Of Hexokinase from Cerebral Cortex Tissue Of the Guinea-Pig, Newsholme, E. A., Rolleston, F. S., and Taylor, K., Biochem. J., 1968, .106, 193. Reversible Release Of Mitochondrial Hexokinase, Rose, I. A. and Warms, J. V. B., Fed. Proc., 1965, 24, 297. Physiological Functions of the Mitochondrial-bound Hexokinase, Li, W. V. and Ch'ien, J. L., Shih Yen Wu Hsueh Pao, 1965, 10, 112: Chem. Abstr., 1966, p5, 7498 a. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 84 Hexokinase Inhibition after the Addition Of Glucose to Ehrlich-Lettré Ascites Tumor Cells under Aerobiosis, Sauer, L. A., J. Biol. Chem., 1968, 243, 3623. Ascites Tumor Mitochondria Hexokinase II. Effect of Binding on Kinetic Properties, Kosow, D. P. and Rose, I. A., J. Biol. Chem., 1968, 243, 3623. Soluble and Particulate Hexokinase of Frog Skeletal Muscle, Karpatkin, S., J. Biol. Chem., 1967, 242, 3525. Metabolic Effectors of Rat—Heart Muscle Hexokinase, England, P. J. and Randle, P. J., Biochemistry J., 1967, 105, 10P. Effectors of Rat-Heart Hexokinase and the Control of Rates of Glucose PhosPhorylation in the Perfused Rat Heart, England, P. J. and Randle, P. J., Biochem. J., 1967, _195, 907. The Effect of Anti-Insulin Serum and Alloxan—Diabetes on the Distribution and Multiple Forms of Hexokinase in Lactating Rat Mammary Gland, Walters, E. and McLean, P., Biochem. J., l968,_192, 737. Hexokinase in the Mucosa of the Small Intestine of the Rat and Guinea-Pig, Shakespeare, P., Srivastava, L. M., and Hfibscher, G., Biochem. J., 1967, 103, 10P. Biochemical Structure of Mitochondria, Siekevitz, P. and Potter, V. R., J. Biol. Chem., 1955, 215, 237. The Reduction of Acetylated Glycopyranosyl Bromides to 1,5—Anhydroglucitol with Lithium Aluminum Hydride, Ness, R. K., Fletcher, H. G., and Hudson, C. S., J. Am. Chem. Soc., 1950, 13, 4547. 1,S-Anhydroglucitol-6—Phosphate and Its Use for the Specific Inhibition of the Hexokinase Reastion in Tissue Homogenates, Ferrari, R. A., Mandelstam, P., and Crane, R. K., Arch. Biochem. Biophys., 1959,.89, 372. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 85 A Comparison of Estimates of Michaelis-Menten Kinetics Constants from Various Linear Transformations, Dowd, J. E. and Riggs, D. S., J. Biol. Chem., 1965, 240, 863. Animal Tissue Hexokinases, Crane, R. K. and $018, A., in Methods in Enzymology, Vol. 1, 277-286. Properties of Particulate Hexokinase Of the Krebs—2 Ascites Tumor, McComb, R. B. and Yushok, W. D., Biochim. Biophys. Acta, 1959, 34, 515. Phosphoenolpyruvate—Dependent Formation of D—Fructose-l— Phosphate by a Four—Component Phosphotransferase System, Hanson, T. E. and Anderson, R. L., Proc. N. A. S., 1968, .él, 269. Heart Muscle Hexokinase: Subcellular Distribution and Inhibition by Glucose-6—Phosphate, Mayer, S. E., Mayfield, A. C., and Haas, J. A., Molecular Pharmacology, 1966, 2, 393. Interconvertible Molecular Forms of Soluble Hexokinase in Frog Skeletal Muscle, Karpatkin, S., J. Biol. Chem., 1968, 243, 3841. Soluble and Insoluble Phosphofructokinase in Rat Muscle, Poon, W. M. and Wood, J., Biochem. J., 1968, 110, 792. The Point Of the Aerobic Inhibition of Glycolytic Activity Associated with Brain Mitochondria, Balézs, R., Biochem. J., 1959, 7_2, 561. Kinetic Studies with Skeletal Muscle Hexokinase, Toews, C. J., Biochem. J., 1966, 100, 739. Protein Measurement with the Folin Phenol Reagent, Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J., J. Biol. Chem., 1951, 193, 265. "I111111.111111111155