, _ . . a . ~ . . _ . ~ . . u . . _ . . _ - . a V . . . . — .x.....3...:..,,:: . .2. .... . _. .7 , . . 3 . ... _;._r_.:.a....:x an 9‘ EC NE A . SAT-l F} FUR! . 3a.“. 5. I. . . yr... x. :2. fin. 9.... «PL... ”.11: ./p o. a: 4.. . r «W? .p. .Abrnflafdn My; 55...”: 41-! If”! ! .2...:........... 2.. 2.4.... . ' "._. . . L; “VHF! a amrmm v.1.vaflmwn . u .GEROL E LY a. ER _ ruvt {€33}. ...... . .,....A:t..,.u17.1....K . . It .a.» ms..§._.4:-;z‘. ......tnu.. . . ...... . e. . .. . . . . .r w 9. Z . . L ..~4§H1‘l\‘~!p 1. n. 35.3%.”. 2.. 5.. if. LE3 £50...”an .uWQQEQEimR? . .. “Mum . 3....“ A n . I. .’ i'u” \ ..l«’. (‘3 U :uvers: is; ‘ ———-—— wait—7‘“ This is to certify that the thesis entitled t Purification and Characterization of Pig Liver L-OPGcherol PhOSphate Dehydrogenase presented by James Nelson Behnke has been accepted towards fulfillment of the requirements for Ph- D- degreeinBiochemiatLy flucwgt Major professor Data/fig 0-7639 ABSTRACT PURIFICATION AND CHARACTERIZATION OF PIG LIVER L-a-GLYCEROL PHOSPHATE DEHYDROGENASE BY James Nelson Behnke The enzyme L-a-glycerOphosphate dehydrogenase (L-Glycerol-3-phosphate:NAD oxidoreductase, E.C. 1.1.1.8) from pig liver was purified 400—fold to homogeneity (speci- fic activity = 100 umoles/min/mg) using a six-step proce- dure that involved calcium phosphate gel adsorption, DEAE cellulose chromatography, phosphocellulose chromatography, preparative sucrose gradient centrifugation and preparative disc gel electrophoresis. The enzyme was judged to be pure on the basis of analytical disc gel electrOphoresis, SDS gel electrophoresis, sedimentation velocity and equilibrium experiments in the analytical ultracentrifuge and assays for the enzymes phosphoglycerate kinase, enolase, trisephosphate isomerase, glyceraldehyde phosphate dehydrogenase, aldolase and phosphofructokinase. The native enzyme has a molecular weight of approximately 70,000, a sedimentation coefficient 0.5% Of 320 ,W = electric point is 6.8 and the absorption spectrum has a 4.89 S and is composed of two subunits. The iso- broad maximum near 273 nm and a shoulder near 259 nm. The Km values for the substrates are: NADH, 8 0!; dihydroxyacetone James Nelson Behnke phosphate, 0.27 mg; L-G—glycerol phOSphate, 0.20 mg; NAD, 57 HQ. The enzyme was inhibited by ATP (Ki = 6 mg), ADP (Ki = 17 mg) and S'-AMP (Ki = 18 mg). Activity stains of disc gels of crude homogenates of pig liver showed one main band and several minor bands of G-glycerOphosphate dehydro— genase activity; our highly purified enzyme corresponded to the enzyme in the main band of the crude homogenates. Several important effects of NAD and NADH on the enzyme were observed. In the range of 0.01 mg to 0.5 mg, the pyridine nucleotides: (1) protect the enzyme against heat inactiva— tion at 49°, (2) prevent dissociation of the enzyme at low protein concentration, (3) maintain the activity of the enzyme during disc gel electrOphoresis and for pure enzyme yield a single band instead of the diffuse smear observed in its absence, (4) maintain the enzyme activity during ion- exchange chromatography on DEAE and phosphocellulose and modify the binding of the enzyme to the ion exchangers and (5) prevent the inhibition of the enzyme by ATP and other adenine nucleotides. These are the first reports of effects of NADH and NAD on column chromatography and disc gel electrOphoresis of the enzyme. PURIFICATION AND CHARACTERIZATION OF PIG LIVER L-a-GLYCEROL PHOSPHATE DEHYDROGENASE BY James Nelson Behnke A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1973 DEDICATION To Helen, Marty and Stevie ’11 ACKNOWLEDGMENT S The author wishes to express his appreciation to Dr. W. C. Deal for his guidance and suggestions throughout this research. The author also wishes to thank Drs. Massey and Stancel for their many times of discussion and argument and many suggestions. The author also wishes to thank Dr. Bieber for supplying the liver samples from young pigs. The technical assistance of Mrs. Diana Ersfeld in the amino acid composi- tion studies and of Mr. Rick North in the general studies of the glycolytic enzymes is also gratefully acknowledged. iii VITA James N. Behnke was born on May 1, 1943, in Alpena, Michigan. After spending his childhood in several mid- western and western states, he returned to Michigan (Hart) in 1955 and graduated from Hart High School in 1961. He participated in Forensics, attended Michigan Boys State, was elected president of the Hart Chapter of the National Honor Society and graduated second in his class. After one year at the University of Michigan, he moved to Flint, Michigan, where he worked for the Chevrolet Division of General Motors and attended Flint Junior College. He then transferred to Central Michigan University where he received his B.S. in Chemistry, with a Biology minor, in 1967. While studying at Central Michigan, he was involved in research on the synthesis and decomposition of two furoyl peroxides, under the direction of Dr. Fred Kabbe. He was also involved in teaching Freshman, Analytical and Organic chemistry laboratories and helped organize the CMU Chapter of Student Affiliates of the American Chemical Society. Mr. Behnke received the 1967 Outstanding Chemistry Senior Award, presented by the Midland, Michigan, Section of the American Chemical Society. Mr. Behnke then began his graduate program in the Department of Biochemistry at Michigan State University. iv Under the guidance of Dr. William C. Deal, Mr. Behnke was able to purify and study the enzyme L-o-Glycerol Phosphate Dehydrogenase from pig liver. Mr. Behnke intends to continue his research career in the Department of Microbiology and Public Health at Michigan State University, in the laboratory of Dr. Leland Velicer. Dr. Velicer has been studying the proteins of the Feline Leukemia Virus grown in cell culture. Mr. Behnke will be involved in characterizing these proteins and study- ing their arrangement within the virus particle. Mr. Behnke was married on August 25, 1967, to Helen, and they have two sons, Martin (November 6, 1969) and Stephen (September 4, 1972). TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . Chapter I. LITERATURE REVIEW . . . . . . . . II. Purification of a-Glycerol Phosphate Dehydrogenase from Various Sources . PrOperties of Various o-Glycerol Phos— phate Dehydrogenases . . . . . Kinetic Properties . . . . . . . Mechanism . . . . . . . . . Physical PrOperties . . . . . . Chemical Properties .~ . . . . . Liver a-Glycerol Phosphate Dehydro- genases . . . . The a-Glycerol Phosphate Shuttle and. Other Roles of the Enzyme . . . Other a-Glycerol Phosphate Dehydro- genases . . . . . . . . . . MATERIALS AND METHODS . . . . . . . Materials . . . . . . . . . . Enzyme Assays . . . Enzyme Assays-~Other Glycolytic Enzymes Protein Determination . . . . . . Conductivity and pH . . . . .1 . . Electrophoresis . . . . . . . Analytical Ultracentrifugation . . . Amino Acid Composition . . . . . . Isoelectric Focusing . . . . . . Analytical Sucrose Gradient .Centrifu- gation . . . . . . . . . . Cellular Fractionation . . . . . . Ammonium Sulfate Fractionation . . . Heat Stability Studies . . . . . . Ion Exchange Celluloses . . . .. . . Calcium Phosphate Preparation . . . . vi Page b.) oooommm 10 ll 12 14 14 15 15 18 18 19 22 24 25 25 26 27 28 29 29 III. IV. V. VI. Enzyme Purification . . . . . . Homogenization . . . Calcium Phosphate Gel Adsorption . DEAE Cellulose Pad . . . . . Phosphocellulose Chromatography . Sucrose Density Centrifugation . . Preparative Disc Gel Electrophoresis RESULTS PRELIMINARY FRACTIONATION STUDIES ON TEN GLYCOLYTIC ENZYMES FROM PIG LIVER . . . . . . . . Cellular Fractionation . . . . Sucrose Density Gradient Ultracentri— fugation . . . . . . . . Ammonium Sulfate Fractionation . . PRELIMINARY ENZYME PURIFICATION STUDIES Multiple Forms of the Enzyme . . . Key Findings Enabling Purification of a-Glycerol Phosphate Dehydrogenase Protection of a-Glycerol PhOSphate Dehydrogenase Against Heat Denaturation by NAD and NADH . Analysis of the Binding of NAD to a-Glycerol Phosphate Dehydro- genase . Protection of a- Glycerol Phosphate Dehydrogenase by MES, Sucrose and Ammonium Sulfate . . . . FINAL PURIFICATION‘ PROCEDURE . . . Homogenization . . . . . Calcium Phosphate Adsorption . . . DEAE Cellulose Pad . . . . PhOSphocellulose Chromatography . . Sucrose Density Gradient Centrifu- gation . . . . . . . . Preparative Disc Gel ElectrOphoresis Other Steps Tested But Not Adopted . Purity of the Enzyme . . . . . . Storage of the Enzyme . . . . . STUDIES ON THE PURIFIED ENZYME . . . Chemical and Physical Characteristics Molecular Weight . . . . . . Sedimentation Coefficient . . . Isoelectric Point . . . . . . Absorption Spectrum . . . . . Amino Acid Composition . . . . Effects of NADH on Disc Gel Electro— phoresis of Pig Liver a-Glycerol Phosphate Dehydrogenase . . . vii 30 31 31 32 33 34 35 37 37 39 44 47 47 55 60 66 67 67 69 69 7O 71 76 80 81 84 88 88 88 94 100 100 100 106 VII. KINETIC CHARACTERIZATION . . . . . Kinetic Mechanism . . . . . . . Michaelis Constants . . . . . Inhibition by Adenine Nucleotides . pH Optima . . . . . . . . . VIII. DISCUSSION 0 O O O O O O O O 0 Physical and Chemical Characterization Kinetic Characterization . . . Effects of NAD and NADH on the Enzyme Effect of NAD on Heat Inactivation Effect of NADH on Dissociation at Low Protein Concentration . . Effects of NADH on ElectrOphoretic Behavior . . . . . Effects of NAD on Ion Exchange Chromatography . . . . . Effect of NAD on Inhibition By Adenine Nucleotides . . . . Implications for Further Research . Multiple Forms of Pig Liver o-Glycerol Phosphate Dehydro- genase . . . . . . . . Binding of NAD and NADH to Pig Liver a-Glycerol Phosphate Dehydro- genase O O C O O O O 0 APPENDIX . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . viii 110 110 114 117 126 130 130 131 134 135 137 138 139 141 141 141 143 144 146 Bis BSA DHAP DTNB ENOL mS NBT NEM PCMB PGI PGK PGM Pipes PK ABBREVIATIONS aldolase N,N'-methylenebisacrylamide bovine serum albumin dihydroxyacetone phosphate 5,5'-dithiobis-(2-nitrobenzoic acid) enolase fructose-1,6—diphosphate glyceraldehyde phosphate dehydrogenase a-glycerol phosphate dehydrogenase glucose-G-phosphate L-Gtglycerol phosphate guanidine HCl lactate dehydrogenase N-(N-morpholino)ethane sulfonic acid milli-Siemens nitroblue tetrazolium N-ethyl maleimide p-chloromercuribenzoate phosphoglucoisomerase phosphoglycerate kinase phosphoglycerate mutase piperazine-N,N'-§i§(2—ethane sulfonic acid) pyruvate kinase ix PMS phenazine methosulfate SDS sodium dodecyl sulfate TEMED N,N,N',N'-tetramethylethylenediamine TPI triose phosphate isomerase LIST OF TABLES Table Page I. Summary of Kinetic Properties of Various o-Glycerol Phosphate Dehydrogenases . . . 6 II. Summary of Physical PrOperties of Various a-Glycerol Phosphate Dehydrogenases . . . 9 III. Cellular Fractionation . . . . . . . . 38 IV. Ammonium Sulfate Fractionation . . . . . 45 V. Summary of Purification Procedure . . . . 68 VI. Amino Acid Composition . . . . . . . . 105 xi LIST OF FIGURES Figure Page 1. Sucrose Gradient Centrifugation of Pig Liver Glycolytic Enzymes . . . . . . 41 2. Sucrose Gradient Centrifugation of Pig Liver Glycolytic Enzymes . . . . . . 43 3. a-Glycerol Phosphate Dehydrogenase Activity in Disc Gels of Pig Liver Homogenate Supernatant . . . . . . 49 4. Ferguson Plot for a-Glycerol Phosphate Dehydrogenase Activity Bands . . . . 52 5. a-Glycerol Phosphate Dehydrogenase Acti- vity in Disc Gels of Pig Liver Homo- genate Supernatant . . . . . . . . 54 6. Protection of a-Glycerol Phosphate Dehy- drogenase by NAD Against Heat Denatura— tion at 49° . . . . . . . . . . 57 7. Protection of a-Glycerol Phosphate Dehy- drogenase by NADH Against Heat Denatura- tion at 49° . . . . . . . . . . 59 8. Kinetic Analysis of Heat Denaturation Data I O O O O O O C O 0 O O 62 9. Theoretical Treatment of Heat Denatura- tion Data 0 O O O O O O O O O O 65 10. Purification of o—Glycerol Phosphate Dehydrogenase by Phosphocellulose Chromatography . . . . . . . . . 73 ll. Preparative Sucrose Density Ultracentri- fugation O O O C O C O I O O O 75 12. Preparative Disc Gel ElectrOphoresis . . 78 13. Analytical Gel Electrophoresis of Purified o-Glycerol Phosphate Dehydro- genase . . . . . . . . . . . . 83 xii 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. Determination of Km for Substrates for the crude enzyme . . . . . . . High Speed Sedimentation Equilibrium Centrifugation of a-Glycerol Phosphate Dehydroqenase . . . . . . . . High Speed Sedimentation Equilibrium Centrifugation of a—Glycerol PhOSphate Dehydrogenase . . . . . . . . a—Glycerol Phosphate Dehydrogenase Subunit Molecular Weight by SDS Gel Electro- phoresis . . . . . . . . . . Sedimentation Velocity Patterns for a—Glycerol Phosphate Dehydrogenase Isoelectric Focusing of a-Glycerol Phos- phate Dehydrogenase . . . . . . Absorption Spectrum of a—Glycerol Phos- phate Dehydrogenase . . . . . . Effects of NADH on Gel ElectrOphoresis of a-Glycerol Phosphate Dehydrogenase . Effect of Substrate Concentration on Initial Velocity of a-Glycerol Phos- phate Dehydrogenase . . . . . . Inhibition of G-Glycerol Phosphate Dehy- drogenase by Products . . . . . . Determination of Km for Substrates in the Reverse Assay of a-Glycerol Phosphate Dehydrogenase . . . . . . . . Inhibition of a-Glycerol Phosphate Dehy— drogenase by ATP . . . . . . . Inhibition of a—Glycerol Phosphate Dehy- drogenase by ADP . . . . . . . Inhibition of a-Glycerol Phosphate Dehy- drogenase by 5 . —AMP o o o o o 0 pH Optima for a-Glycerol Phosphate Dehy- drogenase . . . . . . . . . . xiii 86 91 93 99 102 104 10.8 112 116 119 121 125 129 INTRODUCTION The enzyme a—glycerol phosphate dehydrogenase (L-Glycer0143-Phosphate:NAD oxidoreductase, EC 1.1.1.8) has been purified and studied from a variety of sources, but most of the detailed physical and chemical studies have been with the rabbit muscle enzyme. For various reasons, it is important to characterize the enzyme from liver sources. For example, the enzyme is positioned at a branch point in metabolism between: (1) glycolysis, (2) lipid syntheSis through a-glycerol phosphate and (3) energy transport into the mitochondria via the glycerophosphate shuttle. The limited studies conducted thus far have not shown any evidence of significant regulation of the enzyme, but if such control does exist, it should be detectable with enzyme from liver. In contrast to muscle, which has virtually no fatty acid synthesis, liver is a major tissue for fatty acid synthesis. In order to study any possible control of a liver o—glycerol phosphate dehydrogenase, it was necessary to purify the enzyme. The pig was chosen as the source of liver for several reasons. Primarily, the pig is quite similar to the human in many respects, especially in the tendency toward obesity. For this reason, this laboratory is involved in a systematic purification of, and studies on, the glycolytic and lipid synthesizing enzymes from pig liver. Large amounts of the liver are easily available at low cost from local slaughter- houses. The aims of the present study were then to: (1) purify pig liver a-glycerol phosphate dehydrogenase to homo- geneity, (2) to characterize it physically, chemically and kinetically, and (3) to investigate the possible control mechanisms of the enzyme. Since this was the initial pro- ject in a much broader study, the first studies involved brief characterization of virtually all the glycolytic enzymes from pig liver. CHAPTER I LI TERATURE REVI EW Since the work presented in this thesis concerns the cytOplasmic, NADH linked, g1ycerol-3-phosphate dehydrogenase, only the literature concerning that class of enzymes will be reviewed in detail. That concerning other o-glycerol phos— phate dehydrogenases will be only briefly discussed. A sum- mary of the literature concerning the rabbit muscle enzyme is also available in the previous work from this laboratory (Holleman, 1966). Purification of a-Glycerol Phosphate Dehydrogenase from Various Sources The rabbit muscle enzyme has been purified by a variety of procedures. Beizenherz gt_al. (1955) utilized a series of ammonium sulfate precipitation steps and repeated crystallization from ammonium sulfate solution to give enzyme with a specific activity of 75 units/mg. Modi- fications of this procedure include a final differential centrifugation step (Telegdi, 1964) and a final DEAE cellu- lose column (van Eys gt_31., 1959; Otto gt_al., 1972). Fondy gt_al., (1968) purified the rat skeletal muscle enzyme to a specific activity of 115 using ammonium sulfate precipitation, DEAE cellulose chromatography, gel 3 filtration and crystallization. They also purified the rat liver enzyme (Fondy et_21., 1971) using a similar procedure that included ammonium sulfate fractionation, DEAE chromato- graphy and gel filtration. The purified enzyme had a specific activity of 100. The honeybee thorax enzyme (Marquardt and Brosemer, 1966) was purified using ammonium sulfate fractionation and crystallization from ammonium sulfate solution; the purified enzyme had a specific activity of 170. Lehmann and Pfleiderer (1968) purified the human liver enzyme by using calcium phosphate adsorption, ammonium sulfate fractionation, a heat step, DEAE Sephadex chromato- graphy and crystallization from ammonium sulfate. The pure enzyme had a specific activity of 288. The chicken liver and muscle enzymes were purified by White and Kaplan (1969). The muscle enzyme was purified using ammonium sulfate fractionation, a heat step, DEAE and phosphocellulose chromatography and crystallization from ammonium sulfate solution. The liver enzyme was purified by homogenizing in ammonium sulfate, chromatography on DEAE and phosphocellulose and gel filtration. The muscle enzyme had a specific activity of 145 and the liver enzyme 90. The rabbit liver enzyme was purified by Otto et_§1., (1972) to a specific activity of 142 using a series of heat and ammonium sulfate fractionation steps, gel chromatography and crystallization from ammonium sulfate solution. Properties of Various d-Glycerol Phosphate Dehydrogenases Kinetic PrOperties The kinetic prOperties of some of the a-glycerol phosphate dehydrogenases are given in Table I. Mechanism Some work has been done on the mechanism of the rabbit muscle enzyme. Black (1966) did extensive initial velocity and product inhibition studies and concluded that the enzyme had an Ordered Bi Bi mechanism with NADH (or NAD) binding first and NAD (or NADH) leaving last, with the possi— bility of either an inactive dead-end NADH-enzyme complex or an isomerized NADH—enzyme complex. From further studies Telegdi and Keleti (1968) concluded that abortive glycero- phOSphate—enzyme complexes were formed and concurred with Black that NAD or NADH is bound first to the enzyme. Apitz- Castro and Suarez (1969) reported that the observed substrate inhibition by NADH, a-glycerol phosphate, and dihydroxace- tone phosphate could be overcome by incubating the enzyme with N-ethylmaleimide, giving Michaelis-Menten Kinetics. Presumably the maleimide reacts with a sulfhydryl group which is essential for binding of these compounds. In another interesting study of the rabbit muscle enzyme, Telegdi (1968) reported that the product of o-glycerol phos- phate oxidation 1J5 either dihydroxyacetone phosphate or glyceraldehyde phosphate, depending on the pH. The pH ouooooum no mocz can «can can mucouooou no adz one moo a nauseoua In at: can was was nucuuonou on not: can Adana Asa“ ..Hu uo anon oouuooa uo>eq yam ooau .uouuoaoaum on. savaged nzao .nxum an S.“ um>wq cues: mama .auflaaz on. cues: c-uoc n. v.~ ou uo>aq :mxoano Aha” .ocfiauo can ooa can on . cod o‘finfl «baa ..H~ uo ouuo .nzaa .ouauooaooou co." o~h a mo~ um>ea uannam was” ..Hn uo socou coca cod OH wanna: use nomad .uoeuooun can uvuuavuat 2&2 .nloa onn m.o omnamcoz ooo.c~ mead .caaanx can «use: locum oma o com edema: cmxoazu mama .uoanoeada sonata mama ..Ha ac "and: an» .muauaflo m:4.n .ax¢.~ .maaz mead .uocoazom can nacho: d 00 Hauuadom coma ..Ha no sauce axon .aa mead .xoadm mood .oooa can mean» ow~ om m.o oma o.o~ «.5 edema: uannmm OOGOHOHOK OuouwndgnuH mUld ED :32 22 b01mu0>0¢ MUHflEOh no cuouo>auo¢ (fl; (wax oEwuoo mo mwmdzmwcmnwmma NB‘mmmOmm 40mm0>qwld mDOH¢4> b0 mmHBmmmomm UHBmZHx m0 >mNq pom 00mm .Hmnmpwmamm pom camenmq mow 000.00 um>wq cmEnm Hmmm .mmmm .munss mms 000.00 cm>mq :mxomgo Nmmm ..mmuwm ouuo mms 000.00 um>nq panama gamma .mmmm ..mm1mm mecca mmw N 000.m0 maomsz umm m0mN .nmEmmoum N 000.0m Hmma .mpmnz 000.05 mmn>mcom ahma .mmmH .mumnz mm» 000.00 maomsz cmxownu 000m .cmsmmmom was N 000.00 m0mm ..mm:mm meson mm» N 000.00 mama .0000; 000.00 m0mN .nmEmmosm N 000.00 00mm .0000 cam memo» 000.N0N 00mm .amMIwm mmm 00> mm» N 000.00 mmomsz ofinnmm mocmummmm cemummomEoo mumcsnsm unmflmz U004 OCH84 unasomaoz mmmflzmwomowmmo m9 ho mMHBmmmomm AflUHmwmm m0 wmdzzbm HH mqmda 10 contains a similar prosthetic group, since Lehmann and Pfleiderer (1968) report that the enzyme has an absorption maximum near 277 nm, but they did not study it further. Pfleiderer and Auricchio (1964),using the Sephadex column equilibrium technique, found that the rabbit muscle enzyme binds 2 moles of NADH per mole of enzyme, indicating the presence of two active sites per molecule. No work has been done on the amino acid sequence of the enzyme, but two studies have found 2 moles of C-terminal methionine per molecule of enzyme (Holleman, 1966; Brosemer and Kuhn, 1969). Holleman (1966) also did extensive studies on the reversal of the denatured enzyme from guanidine-HCl solu- tions and found, under the optimal conditions of 0.1 M Tris- HCl, at pH 7.42.0.001 M EDTA, 0.2 M B-mercaptoethanol, and a final enzyme concentration of 0.025 mg/ml, that 90 percent of the enzymatic activity could be recovered. Liver o-Glycerol Phosphate Dehydrogenases The liver a-glycerol phosphate dehydrogenases pre- viously isolated are generally similar to the muscle a-glycerol phosphate dehydrogenases. They all have molecular weights of 65,000 to 75,000 and are composed of two subunits, and their kinetic parameters are fairly similar. The liver enzymes are less stable than the muscle enzymes, however. They usually require the presence of reducing agent during purification and the chicken (White and Kaplan, 1969) and 11 human (Lehmann and Pfleiderer, 1968) liver enzymes require the presence of NADH and NAD, respectively, for stabilizing the enzyme during purification, whereas the muscle enzymes generally do not require either during purification. It has also been reported that the rabbit liver enzyme is allosteric (Lee and Craine, 1971) and that there are multiple forms of the rat liver enzyme (Ross §E_§1,, 1971). The a-Glycerol Phosphate Shuttle and Other Roles of the Enzyme The enzyme is thought to play at least two roles in the cell: (1) reoxidation of NADH to NAD for continued glycolysis by participation along with the mitochondrial glycerophosphate oxidase in the glycerophosphate shuttle, and (2) synthesis of a-glycerol phosphate for lipid synthesis. The operation of the shuttle was discussed in detail by Boxer and Devlin (1961) and it has been shown to be operative in insect flight muscle (Marquardt and Brosemer, 1966), in human adipose tissue (Galton and Bray, 1967) and in Ehrlich ascites tumor cells (Dionisi g£_a1., 1970). However, two recent studies have concluded that the shuttle is only minimally Operative in perfused rat liver (Williamson §t_gl., 1971; Canicero gt_§1., 1972). White and Kaplan (1969) preposed that the chicken liver and muscle enzymes were primarily involved in lipid synthesis and regeneration of NAD, reSpec- tively. 12 Since much of an animal's lipid is made in the liver and transported to other parts of the body (O'hea and Leveille, 1969), it is quite important that there be a con- trolled supply of a-glycerol phosphate for lipid synthesis. There are other mechanisms for synthesis of glycerol phos— phate, such as through glycerol kinase, or synthesis of lipid via the acyl dihydroxacetone phosphate pathway (Hajra and Agranoff, 1968), but most likely, synthesis of a-glycerol phOSphate through a—glycerol phosphate dehydrogenase is the most important source for lipid synthesis. Interestingly, the enzyme is generally absent in tumor cells, leaving lactate dehydrogenase as the chief, if not the only, method of regenerating NAD to allow continua- tion of the abnormally high rate of glycolysis. This has led to attempts to inhibit tumor growth by use of inhibitors of lactate dehydrogenase, but these attempts have been generally unsuccessful (e.g., see Chasin, 1967). Other a-Glycerol Phosphate Dehydrogenase A second type of a-glycerol phosphate dehydrogenase has been partially purified and studied from E. coli B and and E. coli K-10 strain 8 (Kito and Pizer, 1968). The enzyme is apparently biosynthetic in nature since it utilizes NADPH instead of NADH and it is strongly inhibited by palmitoyl Co-A and stearoly Co-A, the Ki's being approxi- mately 25 uM. 13 A third type of a-glycerol phosphate dehydrogenase is the mitodiondrial glycerophosphate oxidase which is involved in the glycerophophosphate shuttle. The enzyme contains a flavin prosthetic group and is linked to the electron transport system. Since the enzyme is part of the mitochondrial membrane, its purification has been difficult, but two procedures for partial purification of the enzyme from pig brain mitochondria have been published (Ringler, 1961; Dawson and Thorne, 1969). Other pyridine nucleotide independent a-glycerol phosphate dehydrogenases have also been found in E. coli K 12 and Streptococcus faecuim F 24 (Kistler gt_§1., 1969; Koditschek and Umbreit, 1969), but have not been studied in detail. CHAPTER II MATERIALS AND METHODS Materials Fresh hog livers were obtained from the Michigan State University Meats Department slaughterhouse and frozen within 2 hours of killing or from the Peet Packing Co., Chesaning, Michigan, and frozen within 15 minutes of killing. NAD and NADH were obtained from P-L Biochemicals (Milwaukee). DL-o-glycerol phosphate, NBT, PMS, ATP, ADP, 5'-AMP, DHAP dimethyl ketal and BSA were obtained from Sigma (St. Louis, Mo.). The DHAP was converted to the H+ salt with Dovex 50 and hydrolyzed at 40° for 4 hours and stored frozen at -20°. Phosphocellulose and DEAE cellulose were obtained from Schleicher and Schuell (Keene, N.H.). Ampholine brand ampholyte was obtained from LKB (LKB Produkter AB, Fack, 5-161 25, Bromma 1, Sweden). Thioglycerol was obtained from Aldrich (Milwaukee) and MES from Calbiochem (La Jolla, Cal.). The coupling enzymes glyceraldehyde phosphate dehydrogenase, lactate dehydrogenase, pyruvate kinase, enolase, aldolase, triose phosphate isomerase, glycerophOSphate dehydrogenase and glucose-6—phosphate dehydrogenase were obtained from Boehringer (N.Y.). All other chemicals were of reagent grade,and deionized distilled water was used throughout. l4 15 Enzyme Assays aGlycerol phosphate dehydrogenase was routinely assayed by following the decrease in absorbance at340 nm due to oxidation of NADH with a Beckman DU monochrometer equipped with a Gilford 2000 multiple sample absorbance recorder at 25°. A stock assay mix containing 0.7 mM_DHAP, 0.25 mM NADH, 0.1 M Tris-Cl, 0.001 M EDTA, pH 7.6, was pre— pared daily: 1-2 mg of enzyme in 5 ml was added to 0.395 ml of assay mix to start the reaction. For kinetic studies, the buffer also contained 0.1 M_KC1 to maintain essentially constant ionic strength. One unit of enzyme will convert 1 umole of NADH to NAD per minute at 25°; specific activity is eXpressed as units per mg protein. The reverse assay involves following the increase in absorbance at 340 nm due to reduction of NAD. For kinetic studies of the enzyme in the reverse direction, 0.1 M_glycine, 0.001 M EDTA, 0.1 M KCl, pH 10.0 or 0.1 M_Tris-Cl, 0.001 M EDTA, 0.1 M KCl, pH 7.6 was used. For conditions where acti- vity was low, fluorometric assays were performed with an Aminco-Bowman spectrofluorometer equipped with a Sargent recorder, exciting at 340 nm and observing the change in the fluorescence at 465 nm. Enzyme Assays--Other Glycolytic Enzymes The assay conditions of all the other glycolytic enzymes were based on those given by Boxer and Shonk (1960), 16 assaying either directly or coupling to a NAD linked dehydro- genase and observing the reaction at 340 nm. The buffer used for all assays was 0.1 M Tris-Cl, 0.005 M EDTA, pH 7.6. The stock assay mixes used are as follows: Glyceraldehyde Phosphate Phosphoglycerate Kinase, Dehydrogenase Phosphoglycerate Mutase 0.375 ml 2 M KCl 0.375 ml 2 M KCl 0.30 ml 0.2—M sodium arsenate 0.25 ml 0.06 M cysteine .0.25 ml 0.06—M cysteine 0. 062 m1 0. 2 M ATP 3.28 ml buffe? 0. 045 m1 2 M MgClz . *0.80 ml 0.01 M NAD 4.15 ml buffer *0.13 m1 0. 01 1:4. NADH Phosphofructokinase Hexokinase 0.13 ml 0.01 M MADH 0. 20 ml 0. 01 M NADP 0.045 ml 2 M MgCl 0. 045 ml 2 M MgClz 0.062 ml 0.2 M AT 0. 062 ml 0. 2 M ATP 4.78 ml buffe? 4. 71 ml buffer Pyruvate Kinase, a-Glycerol Phosphate Enolase Dehydrogenase, Triose Phosphate Isomerase, 0.13 ml 0. 01 M NADH Aldolase, Lactate 0.03 ml 2 M MgC12 Dehydrogenase 0.13 ml 2 M KCl 0. 062 ml 0. 2 M ADP 0.13 ml 0.01 M NADH 4.65 ml buffeF 4.87 ml buffe? The stock assay mixes were prepared fresh daily. The starred components were added immediately before a mix was to be used. The assays were performed at 25° in the Gilford spectrophoto- meter by the sequential addition of the following: Glyceraldehyde Phosphate Dehydrogenase 0.385 ml assay mix, 5 p1 of enzyme solution, 50 pl of 0.02 M glyceraldehyde- 3-phosphate. 17 Phosphoglycerate Kinase 0.385 ml of assay mix, 5 ul of glyceraldehyde phosphate dehydrogenase (10 ul of mg/ml in 0.04 ml buffer), 5 ul of enzyme solution and 5 pl of 0.2 M_3-ph03phoglyceric acid. ' 3 Phosphoglycerate Mutase 0.38 ml of assay mix, 10 ul of coupling enzymes (10 pl of 50 mg/ml lactate dehydro- genase, 20 ul of 10 mg/ml pyruvate kinase and 10 ul of 10 mg/ml enolase in 0.46 ml buffer), 5 ul of enzyme solution and 5 ul of 0.2 M 3-phosphoglyceric acid. Phosphofructokinase 0.37 ml of assay mix, 10 ul of aldo- lase (10 mg/ml stock centrifuged and the pellet dis— solved in buffer equal to the initial volume), 10 ul of a—glycerol phosphate dehydrogenase-triose phosphate isomerase (20 ul of 10 mg/ml in 0.18 ml buffer), 5 ul of enzyme solution and 5 ul of 0.2 M fructose-G—phosphate. Hexokinase 0.38 ml of assay mix, 10 ul of glucose-6- phosphate dehydrogenase (10 ul of mg/ml in 0.24 ml buffer), 5 pl of enzyme solution and 5 ul of 0.2 M glucose. Pyruvate Kinase 0.375 ml of assay mix, 10 ul of lactate dehydrogenase (10 ul of 50 mg/ml in 0.49 ml buffer), 5 ul of enzyme solution and 10 ul of 0.2 M phosphoenol. pyruvate. — Enolase 0.365 ml of assay mix, 10 ul of coupling enzymes (10 ul of 50 mg/ml lactate dehydrogenase and 20 ul of 10 mg/ml pyruvate kinase in 0.47 ml buffer), 5 ul of enzyme solution and 20 ul of 0.2 M_2-phosphoglyceric acid. a-Glycerol Phosphate Dehydrogenase 0.385 ml of assay mix, 5 ul of enzyme solution and 10 ul of 0.03 M DHAP. Triose Phosphate Isomerase 0.365 ml of assay mix, 10 H1 of a-glycerol phosphate dehydrogenase (0.1 ml of 10 mg/ml in 0.9 ml buffer), 1 ul of enzyme solutions and 25 ul of 0.02 M g1yceraldehyde-3-phosphate. Aldolase 0.38 ml of assay mix, 10 ul of coupling enzymes (0.1 m1 of 10 mg/ml o-glycerol phosphate dehydrogenase- triose phosphate isomerase in 0.9 ml buffer), 5 ul of enzyme solution and 5 ul of fructose-1,6-diph03phate. Lactate Dehydrogenase 0.385 ml of assay mix, 5 ul of enzyme solution and 10 ul of 0.2 M sodium pyruvate. 18 Protein Determination Protein concentration was determined by the modified tannic acid-turbidimetric method of Majbaum-Katzenellenbogen and Dobryszycka (1959). Tannic Acid Solution Gum Arabic Solution 98 ml 1 N HCl 0.100 g gum arabic 2 m1 phenol 100 ml water 10 g tannic acid It was usually necessary to filter the tannic acid solution before it could be used. Both solutions were stored in the cold and warmed to room temperature before use. To 0.50 ml of protein solution containing 5 to 50 pg of protein was added 0.50 ml of tannic acid solution. This was thoroughly mixed and allowed to stand at room temperature for 20.0 minutes, when 1.00 ml of gum arabic solution was added and thoroughly mixed in. After 5 to 10 minutes, the absorbance at 650 nm was read. A standard curve using BSA was con— structed for each set of protein determinations. Conductivity and pH The conductivities of solutions were measured with a Radiometer CDM 2e conductivity meter and type 114 cell (cell constant 0.56). The pH's were measured with a Radio- meter PM-4 pH meter equipped with a GK2321C electrode at the desired temperature. 19 Electrophoresis Disc gel electrophoresis was performed by the general method of Ornstein (1964) and Davis (1964). The ,standard Tris-Cl buffered gel solutions for the running (pH 8.9) and stacking (pH 6.7) gels were used, along with the Tris-glycine (pH 8.3) running buffer. Solution A Solution B 30 g Tris 5.98 g Tris 0.23 ml TEMED 0.46 ml TEMED approx. 40 m1 1 N HCl approx. 48 ml 1 N HCl (pH 8.9 - 9.0) (pH 6.6 - 6.8) water to 100 ml water to 100 ml Solution C Solution D 28 9 acrylamide 10 g acrylamide 0.735 g Bis 2.5 g Bis water to 100 ml water to 100 ml Solution E Catalyst 4.0 ml riboflavin 14 mg ammonium persulfate water to 100 ml water to 10 ml The 8.0 cm long x 0.7 cm o.d. gel tubes were rinsed with Photoflow (Eastman), dried, marked at 5.5 cm with a fine point marking pen and supported verticallyin an apparatus made from Vacutainer tops and a wooden block. They were filled to the mark with the lower gel solution (1 part A: 2 parts C:1 part H20:4 parts catalyst), overlayed with water and allowed to stand. After polymerization, the water over— lay was removed and 0.2 m1 of the upper gel solution added 20 (1 part B:2 parts D:l part E:2 parts H20), overlayed with water and photopolymerized by placing near a fluorescent lamp. The gels were supported in the electrOphoresis apparatus and the reservoirs filled with running buffer (14.4 g glycine, 3.0 g Tris in 1 liter of water with a small amount of bromphenol blue added). Five to 50 pg of protein in 5 to 10 percent sucrose and dilute buffer was layered on each gel and electrophoresed at 4 mA per tube at 4°. If the gels were to be stained for protein, they were removed from the tubes by injecting water between the gel and the tube wall, fixed in 10 percent TCA for 1/2 hour, stained with Coomassie Brilliant Blue R for 1 hour at room temperature (1.25 g Coomassie Blue, 454 ml 50 percent methanol, 46 ml acetic acid) and destained in 7 1/2 percent acetic acid at 40° to 50°. If the gels were to be stained for activity, they were stained with an activity staining solution (Dietz and Lubrano, 1967), which contained, per gel, 150 mg DL-a-GP, 0.1 mg PMS, 0.3 ml of 0.01 M NAD, 0.9 ml of 2 mM_NBT and 2.0 ml of 0.3 M Tris-Cl, pH 8.0, for 1 hour in the dark at room temperature. The staining was stOpped by placing the gels in 7 1/2 percent acetic acid at 4°; they were also "destained" and stored in 7 1/2 percent acetic acid in the dark at 4°. When protein was electrOphoresed in the presence of NADH, 5 ml of 0.25 mM NADH, 2 percent sucrose in the Tris- glycine buffer was placed in the subreservoir above each gel (Massey and Deal, 1973). Before staining for activity, it 21 was necessary to soak the gels in 0.02 M MES, 0.001 M EDTA, pH 5.9, for 1/2 hour to remove excess NADH which would react with the staining reagents. Control gels were stained in the absence of aGP and showed no activity bands, although the NADH band, which moved with the tracking dye, stained very heavily. SDS gel electrOphoresis was performed with a modi— ification of the method of Weber and Osborn (1969). The fol- lowing stock solutions were used: Solution A Solution B 0.8 9 SDS 20 9 acrylamide 11.0 g NaH2P04-H20 0.735 g Bis . 0.35 ml TEMED water to 100 m1 adjust to pH 7.1 water to 100 m1 Catalyst 14 mg ammonium persulfate water to 10 ml The 10.0 x 0.7 cm tubes were supported vertically, filled to a mark at 8.5 cm with the gel solution (1 part A:2 parts B: 1 part 20 percent glycerol:4 parts catalyst), overlayed with water and allowed to polymerize. The protein was prepared for electrophoresis by dissolving in 1 percent SDS, 1 per- cent B-mercaptoethanol, 10 percent glycerol and 0.1 M sodium phOSphate, pH 7.1 and incubating at 100° for 15 minutes. Five to 20 pg of this protein in 5 to 20 pl was applied to each gel. The electrophoresis buffer was 0.1 M sodium phos- phate, 0.1 percent SDS, pH 7.1 and the tracking dye, bromphenol blue, was added to the protein solutions. The 22 gels were electrophoresed at 8 mA per tube at room tempera- ture, removed from the tubes and fixed in 10 percent TCA for 1/2 hour. They were stained with Coomassie Brilliant Blue R for a minimum of 2 hours at room temperature and destained in 7 1/2 percent acetic acid at 40° to 50°. The molecular weight of the subunit of the enzyme was estimated by the method of Shapiro et_gl., (1967) using the molecular weights for the standards given by Weber and Osborn (1969). Analytical Ultracentrifugation The molecular weight of the native enzyme was deter- mined by the thantis high speed equilibrium method in a Model E analytical ultracentrifuge equipped with interfer- ence Optics. Five samples were run simultaneously in inter- ference double sector cells with sapphire windows and 12 mm aluminum-filledEpon centerpieces in an AnG rotor, using only the Evapatrol for temperature control (15°). The sectors of the cells were filled using Hamilton syringes; 0.13 ml of buffer was placed in the left sector (screw ring in front) and 0.01 ml of FC-43 Fluorocarbon oil (3-M, St. Paul, Minn.) plus 0.12 ml of protein solution (approximately 0.5 mg/ml) in the right sector. The protein samples had been dialyzed overnight at 4° against 0.02 M MES, 0.1 M KCl, 0.001 M EDTA, pH 5.9, before use. After 24 hours at speed, (23,150 RPM) photographs were taken using Spectrographic II-G plates. The plates were developed in D-l9 develOper 23 and after fixing and drying were read with a Bausch and Lomb comparator (van Holde, 1967). The calculations were done manually using standard techniques based on the equation 2 RT 2.303 d(log C) (l-Vplw2 d r2. M: where R is the gas constant, T is the absolute temperature, 5 is the partial specific volume of the protein, p is the density of the solvent, w is the angular velocity of the rotor, d (log C)/d r2 is the slope of the plot of 3 + log fringe displacement versus r2 and r is the radial position; wz is equal to 1.097 x 10'2 RPMZ. The sedimentation velocity experiments were per- formed using standard single sector sedimentation velocity cells with plane and wedge quartz windows and 12 mm aluminum-I filled Epon centerpieces in an AnD rotor and Schlieren optics. The cells were filled with 0.5 m1 of a 5 mg/ml solution which had been desalted into 0.02 M MES, 0.001 M_EDTA, 0.01 M ammonium sulfate, pH 5.9. Also, NADH had been added to one of the samples to make it 0.5 mM in NADH. The samples were centrifuged at 56,100 RPM at 8°. Photographs were taken at 8 minute intervals using Metallographic plates; they were developed as previously described. The peak positions were read using a Bausch and Lomb comparator. The calculations‘ were performed manually using standard techniques based on the equation 1: s = ____l___ 2.303 log EE—L—L— 24 where wis the angular velocity in radians per second and where rp (to) is the radial position of the peak at the time, to, of the first picture, and r (t) is the radial P position of the peak at a later time t. Time is expressed in seconds. Amino Acid Composition Three mg of enzyme was reduced in 6 m1 of 7.5 M_ Guanidine HCl in 1.4 M Tris-Cl 0.01M EDTA, pH 8.6, and 0.1 M dithiothreitol under nitrogen for 4 hours at room tempera- ture and carboxymethylated by addition of 213 mg of iodo- acetate dissolved in 1 ml of 1 N NaOH under nitrogen in the dark according to the method of Crestfield gt_gl. (1963), followed by dialysis against 0.02 M ammonium bicarbonate and 1y0philization. The dried, reduced, carboxymethylated pro- tein was taken up in 2.0 m1 of 6 M GuHCl for determination of tryptophan by the spectrophotometric method of Edelhoch (1967) and again exhaustively dialyzed against 0.02 M ammonium bicarbonate, 1y0philized and dried to constant weight at 105°. Samples of 0.5 mg were hydrolyzed in con- stant boiling HCl at 110° for 24, 48, and 72 hours and aliquots equivalent to 25 ug were chromatographed on a non— commercial, ultrasensitive analyzer (Robertson g£_31., 1971). The amount of each amino acid present was determined by com- paring to the added internal standard, norleucine. 25 The partial specific volume of the enzyme was cal- culated from the amino acid composition by the method of Cohn and Edsall (1943). Isoelectric Focusing The isoelectric point of the enzyme was determined using an apparatus first described by Massey and Deal (1973), but in a slightly modified form. The column was prepared from 10 mm glass tubing. A slight bulb was blown in one end of the tubing. A 5 percent acrylamide plug was formed in the enlarged end using standard polyacrylamide gel solu- tions, and soaked in 3 percent sulfuric acid overnight. After the tube was rinsed, a sucrose gradient containing 1 percent ampholyte of the apprOpriate range and 100 to 200 ug of protein was formed in the column. The column was then sup- ported in an electrophoresis apparatus containing 3 percent sulfuric acid in the lower reservoir (+) and 3 percent ethy- lene diamine in the upper reservoir (-); this was focused at 200 V for up to 8 hours in the cold. Fractions of 0.2 ml were collected by pumping 50 percent sucrose under the gradi- ent, forcing the gradient out of the t0p through capillary polyethylene tubing. The pH of each fraction was measured at 0° and the appropriate fractions assayed for enzyme activity. Analytical Sucrose Gradient Centrifugation Frozen liver was homogenized in one volume of 0.25 M sucrose, 0.02 M Tris-Cl, pH 7.6. This was then centrifuged 26 in a SW 50.1 swinging bucket rotor for 1 hour at 30,000 RPM to remove the particulate material and the supernatant was ~diluted 1:1 with 0.02 M Tris-Cl, pH 7.6, before applying to the gradients. Gradients were prepared from 2.3 ml each of 5 percent and 20 percent (w/v) sucrose in 0.02 M Tris-Cl, pH 7.6, in 5 m1 cellulose nitrate tubes using a small plexi— glass gradient maker. The gradients were allowed to stand at 4° for several hours and 0.2 ml of the diluted homogenate supernatant was layered on top of each gradient. The gradi- ents were then centrifuged in the SW 50.1 rotor for 17 hours at 40,000 RPM at 0° to 4° in a Spinco Model L ultracentri- fuge (Martin and Ames, 1961). The centrifuge was then allowed to coast to a stop and the gradients were fractionated using a device that punctured the bottom of the tube, allow- ing the gradient to flow out by gravity. Usually 30 frac— tions of 10 drops each were collected and the appropriate fractions assayed for each of the glycolytic enzymes, except phosphofructokinase and hexokinase. These enzymes could not be analyzed in the homogenate supernatant, presumably because of technical problems with the assay (interfering enzymatic reactions). Cellular Fractionation Frozen liver was homogenized in four volumes of 0.25 M sucrose, 0.02 M Tris-Cl, pH 7.6. This was centrifuged at 800 x g (2500 RPM in the Sorvall SS 34 rotor) for 15 27 minutes. The pellet was then washed twice in the Tris- sucrose buffer by repeatedly drawing into a pasteur pipette and centrifugation, finally suspending in the Tris-sucrose buffer, giving the "nuclear“ fraction. The supernatant from the 800 x g centrifugation was centrifuged at 8,000 x 9 (10,000 RPM) for 15 minutes. The pellet was again washed twice in and resuspended in the Tris-sucrose buffer, giving the "mitochondrial" fraction. The supernatant from this centrifugation was then centrifuged at 24,000 x g (17,000 RPM) for 1 hour. The supernatant from this step was called the "supernatant" fraction and the washed pellet called the "microsomal" fraction. The fractions were then assayed for each of the enzymes. Ammonium Sulfate Fractionation Frozen liver was homogenized in 4 volumes of 0.25 M sucrose, 0.02 M Tris-Cl, pH 7.6, and centrifuged at 40,000 RPM in a Spinco 50 fixed angle rotor for 1 hour. Solid ammonium sulfate was added to the supernatant to bring it to 20 percent saturation (10.6 g per 100 ml); it was allowed to stand in ice for 10 to 15 minutes and centrifuged at 15,000 RPM in the SS 34 rotor for 10 minutes. The pellet was resus— pended in sufficient buffer to give a volume equal to the initial volume of homogenate supernatant used and this was called the "20 percent pellet." Additional solid ammonium sulfate was added to the supernatant to bring it to 40 28 percent saturation (11.3 g per 100 ml), followed by standing in ice and centrifugation. The pellet was resuspended in buffer, giving the "40 percent pellet" and the supernatant brought to 60 percent saturation (12.0 g per 100 ml), again followed by standing in ice and centrifugation. The pellet was dissolved in buffer, giving the "60 percent pellet," and the supernatant was brought to 80 percent saturation (12.9 g per 100 ml) and centrifuged, giving the "supernatant" and the "80 percent pellet." Heat Stability Studies Liver was homogenized manually in 2 volumes of 0.05 M potassium phosphate, pH 7.8, and centrifuged at 13,000 RPM in the SS 34 rotor for 15 minutes. Saturated ammonium sul- fate was then added to bring the solution to 35 percent saturation and the suspension recentrifuged at 13,000 RPM for 15 minutes. The supernatant, which contained the enzyme, was stored in the cold until use. The ammonium sulfate fractionation removed the particulate material which inter- fered with desalting and the enzyme was relatively stable when stored in this manner. Just before a sample was to be used, it was desalted on a Sephadex G-25 column into 0.05 M potassium phosphate, pH 7.8. Aliquots of 0.5 ml were used for the kinetic studies. The apprOpriate amount of 10 mM NAD or NADH was added before use and the solution was allowed to stand in the ice for one or two minutes; the sample 29 was then assayed for the zero time activity. The sample was then placed in a 49° water bath and the clock started. The samples for assay were withdrawn 10 seconds before the indi- cated times and immediately assayed. Ion Exchange Celluloses The ion exchange celluloses (phosphocellulose and DEAE cellulose) were prepared for use by washing once in ethanol, three times (or until the filtrate was colorless) in 0.2 M NaOH, twice in water, three times in 0.2 M_HC1, twice in water and fined in water. The cellulose was then suspended in the apprOpriate buffer, the pH adjusted to the desired pH with 6 N NaOH, and the cellulose washed twice more with buffer. It was then stored in the cold (4°) over chloroform until use. Following use, the cellulose was regenerated by the same method, except the initial ethanol wash was omitted. Calcium Phosphate Preparation Calcium phosphate gel was prepared according to the method of Keilin and Hartree (1938). 141 g of CaC12'2H20 was dissolved in 19 liters of water in a 38 liter chromato- graphy jar equipped with a heavy duty mechanical stirrer. 240 g of Na3PO4~12 H20 was dissolved Q1750 m1 of boiling water and this was mixed with 12 liters of water. The warm sodium phosphate solution was added to the calcium chloride 30 solution over a period of 2 to 3 hours using a 2 liter separatory funnel and rapid stirring. After the mixing was complete, the gel was allowed to settle overnight and the supernatant was siphoned off. Additional water was added to bring it back to the original volume, the gel was stirred for a brief time (5 to 15 minutes) and allowed to resettle. This washing was repeated 7 times over a period of 10 days; the final 2 or 3 washes contained no Cl' as determined by addition of Ag+ to a small aliquot. The gel was harvested by centrifugation at 1500 RPM in the Sorvall GSA rotor for 10 minutes, resuspended in a minimum volume of water and additional water was added to bring the total weight to 3850 g. The suspension thus prepared contained approximately 25 mg of calcium phosphate (dry weight) per ml of mixture and was stored at 4°. Enzyme Purification Our purification procedure has been shown to repro— duceibly yield pure enzyme with a specific activity of 100 with a 5 percent yield (3 to 5 mg of enzyme from 400 g of liver). The procedure consists of six steps: homogeniza— tion, calcium phosphate gel adsorption, DEAE cellulose chromatography, phosphocellulose chromatography, preparative sucrose gradient centrifugation, and preparative disc gel electrophoresis. A11 centrifugation steps were performed in a Sorvall RC-ZB centrifuge at 0° to 4°. All purification 31 steps were performed at 0° to 4°, except where indicated otherwise. Homogenization The frozen liver was broken apart with a hammer and screwdriver and 400 g was homogenized in a Waring blender in two 200 g batches in 800 ml of buffer. The homogenization buffer (buffer A) was 0.02 M MES, 0.001 M EDTA, 0.005 M KCl, pH 5.9; to ensure reproducibility in later steps using this buffer, it was adjusted to a specific conductivity of 1.34 mS at room temperature with 4 M KCl. The homogenate was then centrifuged in the Sorvall GSA rotor at 13,000 RPM for 30 minutes and the supernatant taken off with an aspirator. About 750 to 800 m1 of supernatant was usually obtained. Calcium Phosphate Gel Adsorption To the homogenate supernatant, 1.5 ml of cold calcium phOSphate gel suspension was added per ml of homogenate supernatant. The pH of this mixture was measured directly and adjusted to 6.8 to 6.9, with 1 N NaOH if necessary. The mixture was then centrifuged at 3000 RPM in the Sorvall GS3 rotor for 10 minutes. The supernatant was decanted and saved, and the gel pellet was washed with 1 liter of 0.01 M potassium phosphate buffer, pH 6.8. The gel was suspended in the buffer and then "plumped" (homogenized) in a 1 liter polyethylene graduated cylinder with a loosely fitted Lucite pestle. This suspension was again centrifuged for 10 minutes 32 at 3000 RPM and the supernatant was decanted and combined with the first supernatant. The pH of the solution was adjusted to 7.8 with l N NaOH, NAD was added to 0.1 mM (10 m1 of 0.01 M per liter), thioglycerol was added to 10 mM (0.84 ml per liter). The protein was then stored in the cold until the next step. DEAE Cellulose Pad A 4.0 cm x 18.5 cm pad of DEAE cellulose was formed in an 18.5 cm Buchner funnel and washed by suction twice with 350 m1 aliquots of 0.01 M potassium phosphate, 0.1 mM_NAD, 10 mM thioglycerol, pH 7.8. The protein solution from the calcium phosphate step was then filtered through the pad followed by three 0.01 M_ph05phate washes. The enzyme was then eluted with eight washes (1-8) of a solution of 0.05 M potassium phOSphate, 0.1 mM NAD, 10 mM thioglycerol, pH 7.8. The wash fractions were assayed and those with the highest activity were combined, usually fractions 2 through 8. The enzyme was precipitated from solution by adding solid ammonium to 80 percent saturation (516 g per liter) and cen- trifuged at 9,000 RPM for 30 minutes in the GSB rotor. The supernatant was discarded and the pellet dissolved in 70 m1 of buffer A, giving about 110 m1 of protein solution. This was centrifuged at 14,000 RPM for 10 minutes in the SS 34 rotor to remove insoluble protein and desalted in two ali— quots at room temperature by passing through a 3.5 cm x 80 cm column of Sephadex G-25 (coarse) equilibrated with 33 buffer A. The conductivity of each fraction collected was checked to ensure complete desalting. NAD was added to give 0.3 mM (3 ml of 0.01 M per 100 m1) and thioglycerol was added to 10 mM (1 ml of l M_per 100 ml) and the solution was stored in the cold overnight. Phosphocellulose Chromatography A 2.7 cm x 22 cm column of phosphocellulose was poured and equilibrated with buffer A containing 0.3 mM NAD and 10 mM thioglycerol (buffer B). The protein solution was centrifuged at 14,000 RPM in the SS 34 rotor for 10 minutes to remove the precipitate which had formed and the clear, green solution (approximately 200 ml) was applied to the column at room temperature at a rate of 2 to 3 ml per minute. This was followed by an additional 75 m1 wash with the initial buffer, and then with a 225 m1 wash with buffer B which had been adjusted to a specific conductivity of 5.35 mS with 4 M KCl. The enzyme was then eluted from the column with a 750 ml KCl gradient formed from equal parts of buffer B to which 4 M KCl had been added to give solutions with conductivities of 5.35 mS and 15.2 mS. The protein elution pattern was followed with a Gilson column photometer and a Texas Instruments recorder. The fractions were assayed for enzymatic activity and those with the highest activity were combined, concentrated to 20 ml by diafiltration and stored overnight in the cold. 34 Sucrose Density Centrifugation The preparative sucrose density centrifugation was performed with a zonal modified International B-50 ultracen- trifuge and a B-30 titanium zonal rotor. A 5 percent to 20 percent (w/w) gradient was formed in 29 equal 20 m1 steps by pumping the lightest fraction (20 m1 of 5 percent) in from the rim at 15 to 20 ml per minute while the rotor was spinning at 2500 RPM, followed in order of increasing density by the other fractions (19.28 ml of 5 percent plus 0.71 ml of 20 percent, 18.57 ml of 5 percent plus 1.43 ml of 20 percent, through 20 ml of 20 percent). The protein solution was tested to ensure that it floated on 5 percent sucrose and if it did not, it was diluted appropriately. It was then pumped into the core followed by a 30 m1 overlay of buffer only. The buffer used for making the sucrose solu- tions and the overlay was 0.02 M MES, 0.01 M ammonium sul- fate, 0.001 M EDTA, 0.3 mM_NAD and 10 mM thioglycerol, pH 5.9. The centrifuge was pumped down and the protein centri- fuged for 12 hours at 48,000 RPM at 10°. The centrifuge was then slowed with the brake to 2500 RPM and the gradient unloaded from the core by pumping 30 percent sucrose in the rim. The fractions were then assayed for enzyme and protein and those with the highest specific activity were combined and concentrated to 50 ml by diafiltration. 35 Preparative Disc Gel Electrophoresis The preparative disc electrOphoresis was performed with a Canalco Prep Disc apparatus equipped with a PD 2/70 upper column and a Model 200 constant current power supply. The temperature of the apparatus was regulated at 20° with a Lauda type K2 thermostat. The upper column, with its bottom covered with Saran Wrap, was supported vertically in the clamp supplied with the apparatus. Sufficient 5.5 percent acrylamide gel solution was added to give a 3.0 cm lower gel and this was overlayed with water. The standard polyacrylamide gel solutions were used, but the relative amounts of solution C and water were changed to give 5.5 percent acrylamide. After the polymerization was complete, the water was drawn off, sufficient stacking gel solution was added to give a 1.0 cm stacking gel and this was over- layed with water. The gel was allowed to stand overnight before use. The unit was completely assembled, the lower reservoir was filled with the standard Tris-glycine running buffer and the upper with standard Tris—glycine buffer con- taining 0.25 mM NADH. The protein sample for electrophoresis (up to 50 mg) was concentrated to l to 2 ml by diafiltration, bromphenol blue was added and the mixture was layered on the stacking gel under the buffer. The cooling water and the elution buffer (Tris-glycine running buffer diluted 1:4 with water) were started and the slit was adjusted to l to 2 mm. Finally, the electrodes were connected, the current was 36 turned on and adjusted to 6 mA until stacking was complete and the protein band had entered the upper gel (approxi- mately 1 hour). The current was then increased to 8 mA. The elution buffer was pumped through at approximately 1 ml per minute; 8 ml fractions were collected until just before the tracking dye came off the gel (approximately fraction 20) when the rate was increased to 3 ml per minute and 3 m1 fractions were collected. It was usually necessary to adjust the slit during the run to maintain it at l to 2 mm. The appropriate fractions were assayed for enzyme and pro— tein, and those with the highest specific activity were combined and concentrated by suction filtration through a dialysis bag. The specific activity after this step was approximately 100 umoles per minute per mg protein. CHAPTER III RESULTS PRELIMINARY FRACTIONATION STUDIES ON TEN GLYCOLYTIC ENZYMES FROM PIG LIVER This research marked the beginning of a broad study by this laboratory of the glycolytic enzymes from pig liver. Accordingly, the project began with some general studies on the fractionation properties of almost all of the glycolytic enzymes. This included: (1) a cellular fractionation study, which showed the fractional activity of each enzyme in each cellular fraction; (2) sucrose density gradient ultracentri— fugation, which gave the approximate molecular weight of each of the enzymes; and (3) ammonium sulfate fractionation. These studies provided a starting point for purification of all these glycolytic enzymes from pig liver. In the three following sections, activities are expressed as units per g of liver (wet weight) for the crude homogenate and then as percent of the homogenate level for the various fractions. Cellular Fractionation The results of cellular fractionation studies are shown in Table III. It appears that the glycolytic enzymes are generally 37 38 Nm 0m N0 NNN mm mam 0m 00 mm .mm 100 unnumcumosm 0H I m HH 0 0 0 M 0m ma Amy cofiuomum HmEOmouon 0 N 0 0 m 0 0 m I 0 A00 comuomum Homuoconooumz 0 v 0 NH 0 0 0 m 0 0 A00 coHaomuh unmaooz 0H NN NN 0N N0 mm mN 000 0.N 00 10\00 mumcmmosom moo mm Noam sea son 0000 mood Has 0H4 Hum ZOHBdZOHBO¢mm mdqbqqmu HHH mqmdB 39 found in the supernatant fraction, but there are some excep- tions. Phosphoglucoisomerase, triosephosphate isomerase, phosphoglycerate kinase, phosphoglycerate mutase and pyruvate kinase were found in all of the fractions and signi- ficant amounts of aldolase and lactate dehydrogenase were found in the "microsomal fraction." There are at least two possible explanations for these enzymes being in the dif- ferent fractions: (1) these enzymes may be actually associ- ated with these organelles or structures in the cell and remain with them during fractionation, or (2) these observa- tions may be artifacts of the isolation procedure due to nonspecific association of these enzymes with these struc- tures during fractionation. An example of the first case has been described (Wilson, 1968) and it was suggested that this binding may play an important role in the control of the activity of that enzyme. Sucrose Density Gradient Ultracentrifugation The samples were prepared and centrifuged as described in Materials and Methods. It was necessary to dilute the high-speed supernatant fraction with buffer so that its density was lower than that of 5 percent sucrose. The results of two gradients from separate centrifuge runs are shown in Figures 1 and 2. Lactate dehydrogenase was used as a standard in both gradients. Its sedimentation coefficient was assumed to be 7.5 S (Kaplan, 1964). 40 .omumoflocm mmewncw one How ooSMmmm ocm omuumaaoo mumz mnemuomum mouplcms .Houou H.0m 3m m :0 av on 00 um muses ha How ommommuucoo mucmmpmum mnu pom ucmmpmum mmouoom useouma 0N oulucwoumm m o co emummmm was as N.0 cam .0.0 m0 .Noumaue 2 No.0 0003 Hum oouoawp own» mm3 ucmumcuwQSm owe .ZEm 000.0m We use: a gemsmmuucmo new .0.0 mm .Noummoe 2 No.0 .mmososm z mN.0 mo mEsHo> H CH ownmcmmofioc mm3 uw>mq mmfixucm Umumaoowaw um>mq 00m we seapomowmuucmu ucmmomuw mmonosmll.H ouommm 41 Tls 302:2 20.55: a 0 N. 2 0N 0N 0N d d d 1 u d a .e x o ......13 ....N a .222 . ’ ... I 4 .af/w/ . 0&(0IW /. m . 4 Ox \d‘IXGE ’ a . .. o/ \ _O.T\.’\V« 0 /ucm omuwaoomflw um>mq 00m 00 cemummSMHuucmu ucemomuo mmouoomun.m muomwm 43 TIINV 30532 20.55... 0 0 N. 2 0N «N 20920325530 #220410 wm03u3m F N All/“13V iAllV'l 38 n 44 It appears that there are two general size classes of the glycolytic enzymes assayed. The class of small enzymes consists of enolase, triose phosphate isomerase, phosphate glycerate mutase, a-glycerol phosphate dehydro— genase and phosphoglycerate kinase. They have sedimentation coefficients in the range of 3.5 S to 5.5 S. The other class consists of aldolase, lactate dehydrogenase, phosphoglu- coisomerase, glyceraldehyde phosphate dehydrogenase and pyruvate kinase with sedimentation coefficients in the range of 7.0 S to 9.5 S. Those in the first class appear to be monomeric or dimeric enzymes with molecular weights of 50,000 to 80,000 and those in the second class appear to be tetrameric enzymes with molecular weights of 120,000 to 220,000 (Holleman, 1966; Darnall and Klotz, 1972). Ammonium Sulfate Fractionation The ammonium sulfate fractionation results are shown in Table IV. It appears that ammonium sulfate frac- tionation would be a suitable purification step for most of the enzymes, since they are found primarily in one or two fractions. However, the enzymes phosphoglycerate kinase, phosphoglycerate mutase and pyruvate kinase seem to preci- pitate in several of the fractions, which would make ammonium sulfate fractionation quite inefficient as a purification step. The activities of several enzymes in the homogenate 45 .umHHmm ucmoumm 00 .umHHmm ucmoumm 00 exp mp3 umHHmm umumm moan on» mp3 umHHmm umufiw ecu Hamxmu 00: mp3 coHuomum unmoumm 0N mafia 0 0 0 N NN HN 0 00 0 100 ucmomcumasm unmoumo 00 N0 0H 00 0N 00 mmH Hm 0H N0 100 mumuHmHomum unmoumm 00 00 mm 0H 00 Hm NH N0 0 NN 100 mumuHmHomuo unmouma 00 m m HmHomHQ cmoumm 0 mm a m NN 0H 2 n 0 1&0 u u. . u 00 n m n 0 NH NH n n m 100 mumuHmHomum unwound 0N 0 0N NH NH NH NH mH 000 0.H Hm\00 ucmumcumosm mum:mmo&om moo em Hose :00 mom came 000-0 Has 0H¢ ZOHBflZOHBUth maflmADm SDHZOEZd >H mqmdfi 46 supernatant of the ammonium sulfate fractionation study were lower than the corresponding activities in the homogen- ate in the cellular fractionation study. This is probably due to the length of time required to finish the ammonium sulfate fractionation (2 days). I The high recoveries of glyceraldehyde phOSphate dehydrogenase and phosphoglycerate kinase in the ammonium sulfate study (200 percent) could be due to erroneous results in the original assays of the homogenate superna- tant; the enzymes were probably inhibited by competing enzyme reactions. Alternatively, the crude homogenate could contain inhibitors of the enzymes under study and the inhi- bitors were removed by the fractionation. Also, ammonium sulfate might activate these enzymes, yielding greater acti- vity in the solutions with ammonium sulfate. CHAPTER IV PRELIMINARY ENZYME PURIFICATION STUDIES Multiple Forms of the Enzyme In beginning the process of working out a purifica- tion procedure for the enzyme, it was desirable to determine how many different active forms of o-glycerol phosphate dehydrogenase were present in the homogenate supernatant. To do this, protein samples were electrOphoresed on poly- acrylamide disc gels as usual and stained for enzymatic activity. A gel of a sample from a typical crude homogenate electrophoresed in the absence of NADH is shown in Figure 3. Bands 1, 2, 3, and 4 are numbered in the Figure. We normally observed one major band of activity accounting for about 90 percent of the total activity (band 1) and several minor bands of activity. There were always at least three minor bands (bands 2, 3, and 4), and often as many as six or seven. More bands of activity were observed when samples were electrophoresed in the presence of NADH than in the absence of NADH. This effect is discussed in detail in a later section (page106). Most of the activity was in band 1, with small amounts in bands, 2, 3 and 4 (in order of decreasing mobility; band 4 is the more mobile of the two heavy bands at the tOp of the gel). 47 48 .pmHoHucmpH mum 0 0:0 N .N .H mpcmm .uxmu opp cH ponmuommp mm pmchum pom pmmmuosmouuooam cozu mp3 Hem one .Hmm map on pmmamam mmB ucmumcquSm ecu 00 H; mm “ego: H acmlzam 000.00 um ommsoHnucmo 0:0 m.m ma .Hos 2 000.0 .aeom z H00.0 .mmz : N0.0 0H omNHcmmosoe was Hm>HH .ucmumcuwmsm mumcwmoeom Hm>Hq mHm mo mama omHD CH mum>fluod ommcmmoupwcmo mumnmmocm Houmoxaouaul.m wusmHm 50 To partially characterize these isomers, and see if they were molecular weight or charge isomers, their migration was studied on gels of various acrylamide concentrations. The mobilities were calculated and plotted according to Ferguson (1964). A Ferguson plot of the data is shown in Figure 4. The fact that the lines for bands 2, 3, and 4 are parallel indicates that they are primarily charge isomers, with approximately equal molecular weights. Band 1 is a smaller protein, since its retardation coefficient (the slope of the line) is smaller than the retardation coeffi- cients of bands 2, 3 and 4. Part A of Figure 5 shows a densitometer tracing of a typical activity pattern obtained with an electrophoresis gel without NADH using the crude homogenate from an adult pig liver; only one major activity band is seen. Generally, the relative intensities of the various activity bands were similar from various livers; livers from a number of young pigs (0-3 weeks) gave patterns similar to the normal adult pattern, with only one major band. However, in two livers from 3-week pigs the intensities of the normally minor bands were much greater (Figure 5B), two of the minor bands were approximately as intense as the major band, band 1. 51 .0H x HHHHHnoz mH mumcHouo mg» .0 vamp Hos .mpocomz pcm mHmHumumz cm pmnHHomwp mm pwcmmum pom pommHonmouuomHm ones pom mcoHuwuucoocoo mpHEoH>Hom pmumoflpcm mnu m>Hm 00 Hope: one o coHuDHom mo mucsoam wumflumouomm mammo mo pound (mum mum3 mama one .Hmm some ou pwmammm mp3 ucmumcquSm was CO H; mN “Ema 000.00 we smog H How 0mmsHHuusmo 00m .0.0 mm .mpmnomonm Edmmmmuom z m0.0 :0 one: >9 poNHcmooeon mmS Hw>mq .0 can m .N .H mpcmm >ua>muo¢ mmmcmooupmzmo eumoomonm HOHmUSHoua H00 uon cowomummll.v muommm 52 T—Irl w0_£(._>uu( ...—323.— 0 0 h o m 1 d d d 4 «.mmsoqmmOthmam ..m0 020 kl «.0 0.0 «.0 m .0 0.0 5.0 0.0 0.0 0.— All'llflOW 53 .0wHoHucmoH mum 0 can N .N .H mocmm .uomH men on was cofiumumme “map momxomuu ecu mo coHuHmom ecu mwumomch B .0002 H 000 2mm 000.00 um ommsoHHucmo new 0.0 mm .mHue a no.0 .mmouoom m m~.0 cw pmmflcmmoson mm3 Hm>HH one .umHmHm 0H0 Noplmm o Scum cumuumm UmumuosmonuomHm Hmomomum so we memomuu Heumeoummcmo .m .OHHsz Hopes .HmumEouHmcmpouon mchuoomm oHumEouod Homoqlmomon m mamm: cmxmu mewomuu Houofioummcwp map can pmmouocoouuomHm me3 mHmu .cuouumm oHumHonoouuomHm uHsom Hmoflmmu m 00 mcwomuu HmumEOuHmcmo .< .ucmumcuwosm wumcwmoeom um>HH mam mo mama omHo cm >0H>Huo< mmmcmmoupmceo mumcmwocm Houmo>H0I0||.m «HomHm 54 0E ><0 on “.0 ..m0 030 a ¢m>= .530< “.0 .30 020 < 55 Key Findings Enabling Purification of a-Glycerol PhOsphate Dehydrogenase Protection of a-Glycerol Phosphate Dehydrogenase Against Heat Denaturation by NAD fiand’NADH Heat denaturation studies were performed to deter- mine whether a heat step could be used in the purification procedure and also to study the stabilizing effects of NAD and NADH on the enzyme. In order to study the binding to and stabilization of the enzyme by the nucleotides, it was necessary to have the enzyme in a controlled environment, removed from all the small molecules present in a normal homogenate supernatant. To accomplish this, simple ammonium sulfate precipitation and desalting steps were used. The studies were performed in phosphate buffer at 49°. This temperature was quite critical; at a temperature of 1 or 2 degrees lower, the denaturation slowed drastically (the half—time for the reaction increased 5 to 10 fold); at a temperature 1° or 2° higher, the half-time decreased four- fold. This indicates a high activation energy for the denaturation process. The results for NAD and NADH as stabilizers are shown in Figure 6 and Figure 7, respectively. There appears to be an initial activation of the enzyme by the NADH, followed by the heat denaturation. NADH protects the enzyme against heat denaturation somewhat better than NAD does; 0.25 mM NADH given nearly complete protection, but 0.5 mM NAD is required to give equivalent protection. NAD was used for stabilizing the enzyme during three of the 56 .wueueupeEEH nememme one mesuu peueeupcu ecu encuec mpcooem 0H eHoEem ecu Eouu pe>OEeH eHeB muoovHHm wemme Hecuo ecu “ecu :0 mp3 um eHuc3 eHoEem ecu Eoum cexeu mmB xmmme eEuu oueu ece .coflueuuceocoe peueouch ecu e>um ou peppm me3 ocz 86 0H 00 uCDOEe eueuumoumme ecu ppm .0.0 mm .euecmmocm Ecummeuom m m0.0 oucH mmuw xepecmem co peuHemeo me3 eHmEem e .emo eHOMem .pHoo ecu CH peuoum one pe>oEeH we: uceuecuemom ecu .coHuemsmHHuceo Heuum .cOHu neuouem uceeuem mm Cu peppm euMMHom EoucoEEe one .0.0 mm .eumcmmoca Ecummeuoo m m0.0 cm peNHcemOEoc mez He>ua mum .000 um coHueusueceo umem umcHemd adz mo emecemoupmceo euecamocm Homeowuwua mo scuuoeuoumll.0 eusmum . I/ is Bel/ulna 0’0 . .9. I. .. is 3.0 Ilv../ / .I I .. .I. O ... «I. z I I. / 0 4/ ..... E . . . ... O0 202.0th?— 0N “13383:! AllAllDV 58 .ocz mo peeumcu peppe euez $042 SE 0H 00 mucooEe eueHHmoummm uoeoxe .m euomuh me eEem mcoNuupcou .000 um coHumusueceo ueem umcuemfl mocz wc emecemoupmceo euecmmocm HOHeUSHOIo mo coNueeuoumI|.s euomum 59 0.2.0 02: on ) mN 0N T o.— ... H 4gb? 1.x ...... o .. o’o woo... V. fl oz ’0 coco. . I 0’. no I. SE mN—dlIYo/ 4 0’. I.’ O 1 ...—0‘2 ./ ....4 I ..Plis N o a .x. .. ’/ // I. I I . / I! w, I U o. / I 1.... I I ..II. . ... mum... .II... ZO_.—<~_:.—<2m0 ...o h0 20:02.05. 0N 0Q 00 00 00— AllAllDV 1N3)!“ 60 purification steps primarily because it is less expensive than NADH, because it is more stable at pH 5.9 than NADH (Lowry gt_gl., 1961) and because it adequately protects the enzyme. For the DEAE cellulose pad, 0.1 mM NAD was suffi— cient to protect the enzyme, but the phosphocellulose column and the preparative sucrose gradient required the presence of 0.3 mM NAD to protect the enzyme. NADH was required in the preparative disc gel electrophoresis step, because NAD gave no protection during the procedure. The NAD may not migrate into the gel, or it may migrate with the enzyme, but not protect it. Analysis of the Binding of NAD to o-Glycerol Phosphate Defiydrogenase Because of the importance of the protection of the enzyme by NAD and NADH in the purification of the enzyme, it was of great interest to analyze further the binding of the nucleotides to the enzyme. The initial activation of the enzyme by NADH makes it difficult to analyze further the NADH protection data. No such activation is observed with NAD and the analysis is straightforward. The data were replotted in Figure 8 as In percent activity remaining (0 time = 100) versus time to test whether they fit first-order kinetics. As can be seen by inspecting Figure 8 virtually all the data fit nicely on a straight line. If the rate constant k for the inactivation of the enzyme is defined as the slope of the first order plot of the data then the plot of l/k versus (NAD)2 shown in Figure 61 .meHuecHx Hepuo umuum How umeu ou eEuu msmue> >uu>muee uceonem CH me peuuoam one penmamce eue3 0 eusmum Eoum pump ecB .eueo coHueHsumcea umem mo mmmhaecm euuecHMII.0 eupwum 62 .5... .02: 0n mu 0N n— 0.. m 0 u q . H . a 1 0 O ....... / ollocoz n ./ 0 0...... I/ ......‘III in... No.0 2.... nN—dllY/I . ._ / o . o I/ ’ II/ I .... o/ m... // .... o .I Alli:— nud ... ‘ alcla’n’ /./ 000/. l ..4I.I ll. /.../0 .I:Ql . o. 20...... 4% I I/ ,o-dI. p . I] I’ll. I Iolnldmmpl 0(2 >0 20:02.0”: AllAllDV IN 3383:! “'l 63 9A is obtained. This plot shows that: (l) the rate con- stant for inactivation is inversely proportional to the square of the NAD concentration, and (2) the rate constant approaches zero as the NAD concentration increases. This first observation suggests that the binding of two molecules of NAD to each enzyme molecule is involved in the protection of the enzyme. To test this hypothesis, the data was fur- ther treated as follows (Kuczenski and Suelter, 1970, 1971). For each concentration of NAD used, a corresponding value of a was calculated (a is the fraction of binding sites occupied) using the following equation ko - kx ko - k0° In this equation, kx is the denaturation rate constant (slope of the semilog plot) for the NAD concentration of interest, k0 is the rate constant in the absence of NAD and km is the rate at high NAD concentration, assumed to be zero for this purpose. The values of a versus log (NAD) are shown in Figure BB. The line drawn between the points is the theoretical line calculated for the case of n = 2 and an average binding constant (Kav) of 126 UM, using the equation n[pKav - p (NAD)] = log I—g—H (l) where p (NAD) is the negative logarithm of the NAD concen- tration and pKav is 4.100 (see appendix for the derivation of this equation). As can be seen, the experimental points fit the theoretical curve quite well. This is additional evidence that two NAD molecules per enzyme molecule are 64 .m: 0~H u >ex pcm m u c we emeo ecu Hem ecHH Hep -Hueuomcu ecu mH mcHH was .czonm mucHom ecu mcH>H0 .10420 moH momHe> peuuoHo one .uxeu ecu CH pecuuemep we peueHoeHeo mez 0 coHueHuceocoo acz come now .m .mcoézv momue> x\a Snow ecu cw oeuuon cecu mm3 mace .c0uueuuceecoo oflz uecu How x we ceceu mp3 m eusmum :0 ecua come 00 emon ece .4 .euea coHu announcea ueem mo uceEueeHB Hepaueuoecall.m euommm 65 Hi ... 32 0.... NHSEE<2 OI” no" I o I 4 . o N m. N.0 «0| ... \N N0- .N0 v.0 . - v.0 . ._ \_ 0.0 r 1 0.0 ...u \. q— l \ j cap 0 66 involved in protecting the enzyme, with an average binding constant of approximately 125 pg. Eggtection of a-Glycerol Phosphate Dehydrogenase by MES, Sucrose and Ammonium Sulfate If the crude enzyme is stored in MES buffer at pH 6, it retains essentially all of the activity after 24 hours at 4°. In phosphate buffer it retains 90 percent of the zero time activity, but in cacodylate buffer it loses almost all the activity. In the heat denaturation studies, if MES buffer was used instead of phosphate buffer, an increase of 6° in the temperature was necessary to obtain the same denaturation rate. In preliminary studies on stabilizing the enzyme in crude solution, 10 mM ammonium sulfate was found to maintain the enzyme activity for over a week at 4° (MES buffer, pH 5.9), while the sample without ammonium sulfate lost 40 percent of its activity. In a separate study using Tris (pH 7.8) as the buffer, the enzyme stored in 10 percent sucrose had approximately twice the activity of the control sample without sucrose after 18 hours. CHAPTER V FINAL PURIFICATION PROCEDURE A summary of the major steps in the purification procedure in shown in Table V. Enzyme with a specific acti- vity of between 95 and 100 umoles/min/mg protein was obtained each time the purification procedure was run. The preparation takes three days, with the homogenization, calcium phosphate gel absorption, DEAE cellulose pad and ammonium sulfate concentration steps on the first day, the phosphocellulose column on the second, the sucrose density gradient during the night of the second and the preparative disc gel electrophoresis on the final day. This preparation uses only 400 g of liver instead of a whole liver (1200 g) primarily because of logistical problems associated with the centringation and desalting steps. Even so, one can obtain approximately 5 mg of pure enzyme quite easily from 400 g of liver. Homogenization Generally, the liver was still partially frozen when it was homogenized. This did not interfere with the homo- genization and helped ensure against overheating during homo- genization. Initially, homogenization was in 0.05 M potassium phosphate, pH 7.8, but use of the MES buffer gave 67 68 .om .aufimcmp mmouosm “omlom .CEdHoo mmoHsHHmoonmmonm “mH.o .ucmumcummsm mumcmmofion "Hmowmwu muoe mum mmcam> mufl>fluom UHMflommm mcfl3oHaow wsu was» Umuoc on pasonm pH .coflumoHMflusm ucwomu uwoe on» mo muasmwu anew ooa mmm o.m NH mammuonmouuomam omfia om omo.n mm om sunmcmo mmonosm HH oop.m cum cam casaoo mmoHsHHmooammosm mm.o oov.m 005.5 omp.m 6mm mmoHsHHmo mama wm.o oom.n _ ooo.- mam.m Hwo mumsmmosm eaaoflmo om.o oom.m ooo.mm omn ucmumcumdsm mumcmmosom muw>fluo¢ muflco m5 HE camwowmm Hobos .cwmuonm .wEsHo> Hmnoe MWMDQmoomm ZOHBdUHhHmDm m0 wm<22Dm > mqmde 69 the proper homogenate conductivity and pH for the calcium phosphate gel step which followed, without any dialysis or adjustment of the pH. The pellet obtained after centrifuga- tion was not tightly packed, so some of it was usually removed during the aspiration of the supernatant. Calcium Phosphate Adsorption This step usually gave a two-fold purification, and removed most of the hemoglobin and some of the contami- nating particulate material which was not removed by the initial centrifugation. The step is easy to perform and takes less than an hour. Only one-third of the enzyme binds to the calcium phosphate gel, and most of that which does bind is washed off with the 0.01 M phoSphate wash. It is quite important that the pH after the addition of the gel be 6.8 to 6.9, since if it is too low, the enzyme binds too tightly to the gel and cannot be washed off. DEAE Cellulose Pad The pad step was devised after it was found that the enzyme was unstable when bound to DEAE cellulose without NAD but was stable for about 45 minutes when bound in the presence of NAD. Using the DEAE cellulose in the form of a pad allows the enzyme to be adsorbed and then eluted in less than 30 minutes. Also, the pad removes the remaining parti- culate material present, giving a clear, greenish-colored 70 solution. Usually, a two-to four-fold purification is obtained with this step. Because the protein solution was so dilute, it was necessary that the solution be brought to 80 percent ammonium sulfate saturation to precipitate the enzyme. Under the usual conditions, where the protein is at higher concentrations, only 55 to 60 percent saturation was required to precipitate the enzyme. It was critical that the enzyme be completely desalted following the ammonium sulfate precipitation step, because any residual ammonium sulfate interferes with the binding of the enzyme to the phOSphocellulose column. Consequently,any desalted enzyme fractions which had a conductivity greater than 1.8 mS were not used. Phosphocellulose Chromatography This.step usually took from four to five hours to run and gave a 20- to 30-fold purification, with a greater than 50 percent yield in enzyme. Very little contaminating protein binds to the column under the conditions used, and if there is no ammonium sulfate present, all of the a-glycero~ phosphate dehydrogenase binds to the column. Generally, there is no problem with column capacity. The column can be checked visually for overloading, since a-glycerol phosphate dehydrogenase is bound to the column in a region between the very tightly bound hemoglobin (or another red protein) right 71 at the top of the column, and a less tightly bound green protein which, because of competition with stronger binding proteins, finds adsorption sites only near, or below, the middle of the column. If the green protein washes off the column initially, the column is overloaded. An initial step elution with 5.35 mS KCl is necessary to elute the green protein ahead of the a-glycerol phosphate dehydro- genase; if only a linear gradient was used, they were not separated well, giving a lower purification. The results of a typical run are shown in Figure 10. Sucrose Density Gradient Centrifugation This purification step was used instead of a gel filtration step. It is faster and requires less buffer than gel chromatography, and the zonal rotor is superior to a standard swinging bucket rotor, since larger samples can be used and much higher forces are obtained, significantly shortening the centrifugation time. This step is the only purification step in the procedure which is based on mole— cular size. It is relatively simple to run, and usually yields enzyme with a specific activity of 50, but has given a sample with a specific activity of 90. It removes a yellow-colored protein which is a major contaminant, giving a clear, colorless enzyme solution. The results of a typi- cal run are shown in Figure 11. The only problem generally encountered was loss of a portion of the gradient, due to leakage around the plug in the rotor. This decreases the 72 .nHE\QO< mo mafia: CH we >uw>wuom mfiwunm .pmumuunmonoo paw pmnfineoo mum3 NMH nmsounu mm mcofluomum .unmfipmum Hum we «.ma on we mm.m m an wmonHOM .smmz me mm.m Hmfluflcn no nuaB nEsHoo mnu Eoum concam mm3 wexunm one .mE vm.a sun>nuosocoo mnmnommm .m.m mm .Houmomamoesu we on .ouw>fluom oExunm .pououunoonoo pnm poanEoo ouo3 0H nmsounu H mnofluooum .uouou Esflnonwu «Hum o nufl3 oOSMHuunooouuH: omlm Honoflumn nuounH poAMHUoE HonoN o nfi ooa no mason NH now 2mm ooo.mv pm 30533ch new mm mm .Hououfimofiu we 0a .902 as 0.0 .oumonsm asecoesm 2 no.0 .aeom z H00.0 .mmz 2 No.0 aw pnoflpoum omouosm unoouom om ou unoouom m m on poflammo mm3 coum nasaoo omoHoHHoooanonm onn Eouw omonomoup Imnop ouonmmonm Houoowamuo pouohunoonou one .noHuomsmHHu Inoomuuao muflmnoo omouosm o>wumuomoumll.aa onsmfim 75 M . zmuo :A m. Vomo «J u A . ”.ka IA mo?— mu.— T' 5 amniaz ZOZU<~I I mu ON 2 O— m 4 *.‘.Ou - 4 q ... r1... .... .J.\ //I. ..w... d ‘1‘ 5205.0 303:... .... 02:23.3... ...... 0. V. °£ O O O --- (Iw/Bué Nlnoud ‘9 0 °. 76 resolution at the top of the gradient and since the enzyme generally sediments only about 1/3 of the way through the gradient, this can interfere with the results. On one of five runs, 100 ml was lost, making it necessary to repeat the centrifugation. greparative Disc Gel Electrophoresis The results from a preparative disc gel electro- phoresis run are shown in Figure 12. Two things are important for the success of this step: the use of NADH in the running buffer and the use of a 5.5 percent acrylamide gel. When purified enzyme is electrophoresed in the absence of NADH, a smear of protein is obtained, with few or no distinct bands. However, if NADH is present, distinct bands are obtained, with much greater enzymatic activity than in the absence of NADH. The NADH also protects the enzyme after it elutes from the column, since under normal condi- tions, the enzyme is relatively unstable in Tris buffer at high pH. If a higher gel concentration is used, the enzyme migrates too slowly and loses activity. If a lower gel con— centration is used, the enzyme migrates with the tracking dye. After the preparative sucrose gradient step, most of the enzymatic activity present electrOphoreses as a single band, with perhaps 25 percent of the total activity in the minor bands (Figure 12, also see page 47). This major band of activity runs right behind the 77 .cns\oo< no open: 2% we mufl>fluoo ofiwunm .poumuunoonoo pno ponflnEoo ono3 om amneunu 00 mconuomum .umnm3 nuns qua emusanc ran mommv nowmsn onwoaaolmflue Havauwaono on» moz Homwsn mnflnnsu one .nEsHoo Momma on\m om o nuHB msuouommo omen mono ooaonou o ca pomouonmonuooao Uno nEsHoo How opfleoamuoo “noouom m.m o On poflaamo paw HE N on H on pououpnoonoo moz moum ucoflpoum omOHUSm onu Eoum omonomoupmnop ouonmmonm Houoowawno .wfimouonmoupooam How omen o>wumuomoumul.ma ousmflm 78 3 .o All/“13V JWAZ N 3 Q “'2 F 0...... Q N v ”30232 202.05“— n an on m.mu¢01mOu.—Umam m>.h(¢Nno poHMHusm onu mo on ma maoumeflxoummo mo memouonmouuooao Hom omen ”maom muH>Huoo pno neououm .moum mwmouonm nouuooao How omen o>HuoHomoum onu nmsounu pofiuuoo oezuno mo mammaonn .ommnomoupwnoo oumnmmonm Houoomaolo poHMflusm mo memouonmouuooam How Hmowumamnnll.ma ousmflm 83 wow >=>.:u< Sonora ., 84 some of the other glycolytic enzymes. Phosphofructokinase, phosphoglycerate kinase, aldolase, trisephosphate isomerase and enolase were all assayed; the activity of each of these enzymes was less than 0.0001 percent of the a-glycerol phos— phate dehydrogenase activity present. This is especially satisfying in the case of aldolase, because with one purifi— cation procedure the rabbit muscle a-glycerol phosphate dehydrogenase and aldolase partially co-purify (Beizenherz §£_al., 1955). Several tests indicate that no modification of the enzyme occurs during the purification procedure. The enzyme purified by this procedure migrates the same as Band 1 observed on activity stains of gels of the crude homo- genate (see page 47). Secondly, the purified enzyme has a sedimentation coefficient determined in the Model E of 4.9 to 5.1 S, while the enzyme activity in crude solutions is accounted for in a peak with a sedimentation coefficient of 5.2 S as determined from the sucrose density experiments (see page 39). Finally, the Km values for the crude enzyme (Figure 14), 0.1 mg (DHAP) and 2.5 uM (NADH), were the same order of magnitude as the values for purified enzyme, 0.27 m! (DHAP) and 8 pg (NADH). Storage of the Enzyme Although extensive studies on the Optimal storage conditions for the purified enzyme have not been carried out, two methods were used. To store the enzyme in soluble 85 Figure l4.—-Determination of Km for Substrates for the Crude Enzyme. A. Liver was homogenized manually in 0.05 M potassium phOSphate, pH 7.8 and centrifuged for 1 hour at 45,000 RPM. The assays were performed spectrophoto— metrically in 0.1 M Tris—Cl, 0.001 M EDTA, pH 7.6 at 25°. The NADH concentration was 0.25 mM.- The velocity is in units of AOD/min. B. Km for NADH. Homogenization and assay conditions same as Figure 13, except the DHAP concen— tration was 0.7 mM. The velocity is in units of relative activity. _ 86 o/ HOMOGENATE ~ / SUPERNATANT 2.5. 0 K... DHAP . / 2.0 '- / o b O ..5- ./ \> * / I.o* / 0.5 / KM=OJ MM 1 J 4 l— A l A J A l A L o Is 30 45 60 Minnow) HOMGENAI’E SUPERNAIANI K... NADH 25" 21»- 1.5- \> / / - I. 0 ° / °T ./° 0/ o.s~ / / ° KarisiM/ 0.3 (.6 0.9 1.2 I/mum (pm) 87 form, buffer containing 10 percent sucrose, 0.02 M MES, 0.001 M EDTA, 0.01 M ammonium sulfate, 0.3 mM NAD and 10 mM thioglycerol, pH 5.9, was used and the protein concentration was made greater than 1 mg/ml. Otherwise, it was stored as the ammonium sulfate precipitate (70 percent saturated) under the same buffer. CHAPTER VI STUDIES ON THE PURIFIED ENZYME Chemical and Physical Characterization As indicated earlier, NADH and NAD profoundly affect the activity and heat stability of a-glycerol phOSphate dehydrogenase. Because of this, we were interested in investigating the possibility of any effects of NADH and NAD on the structure of the enzyme. Molecular Weight The molecular weight of the native enzyme was deter— mined using the Model E analytical ultracentrifuge and the meniscus depletion method (thantis, 1964) of high speed sedimentation equilibrium. A protein sample of approximately 1 mg/ml which had been purified through the disc gel electro- phoresis step (specific activity = 100) was prepared by dialysis against 0.02 M MES, 0.001 M EDTA, 0.1 M_KC1, pH 5.9, overnight at 4°. Samples of 0.36, 0.55 and 0.76 mg/ml were prepared by diluting the stock enzyme with the appro- priate amount of dialysis buffer. A 0.55 mg/ml sample which contained 0.5 mM_NADH was also prepared by adding the appro- priate amount of 10 mM_NADH to both the sample and solvent sectors of the ultracentrifuge cell. The results of the experiments for the 0.55 mg/ml samples in the absence and 88 89 presence of NADH are shown in Figures 15 and 16 (see Materials and Methods). Clearly, the sample with no NADH is heterogeneous since the plot is curved. This could be due to either the presence of an impurity or to dissociation or association of the protein. Since the other sample, which was identical except for the presence of 0.5 mM'NADH, shows no evidence of heterogeneity, the curved line must be due to the presence of either dissociated or associated protein. From the slope of the line in Figure 16, a molecular weight of 70,400 is calculated for the enzyme in the presence of 0.5 mg NADH. From the limiting slope at the bottom of the cell in Figure 15, a value of MW = 68,200 was obtained for the high molecular weight material in the sample without NADH. This value agrees quite well with the molecular weight of the enzyme in the presence of NADH, after correction for the presence of 2 moles of NADH bound per mole of enzyme in the presence of NADH. We conclude that the heterogeneity observed in the absence of NADH is due to dissociation of the enzyme at the low protein concentration near the meniscus and that the native enzyme has a molecular weight of 68,000 to 69,000. A value of MW = 53,300 was calculated for the dissociated material near the meniscus; hence, the enzyme is only partially dissociated. Assuming that the native enzyme is a dimer (see next section), then the mix- ture consists of approximately 50 percent monomer and 50 9O .mponuoz pnm mamwuouoz nfl confluomop mo poEHOMHom unoEfluomxo on» pno pouomoum mos onEom one .omonomoupwnoo oponmmonm Houoomaolo mo Esflunfiaflsqm nofiuounoeflpom poomm nmfimnn.ma ousmflm 91 ..m Om ov me 30(2 02 (3.80353 20:<._.Zu<=Oum any.“ 20.... INSWEDV'HSIO JON”! 901 + 8 92 .unomoum moz mo¢z_m5 m.o nonu umooxo .ma onsmflm mo oEom .omonomoupmnoa ouonmmonm Houoowamlo mo Edflunfiaflsqm nofluounoeflpom poomm nmflmll.ma onsmam 93 Iosu>m Amy .nflfisnam Eduom onH>on Amy .omopfimouomaomum AHV "ouo3 poms monopnoum one .nHouonm m: om on OH ponflounoo Hom noom .mponuoz pno maofiuouoz n0 confluomop mo can maom onu pno pouomonm ouo3 moamfiom one .mnmmuonmounomnm Hoe mom 0n unmflmz “masomnoz nflcsnsm omonomoupmnoo oumnmmonm Honoomaololl.na ousmflm 96 1 U T I U >p330£ on 0.0 0/0 N . m.mu¢01._0¢.—Umam .30 man v.0 ..Om 10h ..Ow roe .00— lHOIiM zmnzmow 2-0; 97 purified to a specific activity of 90 by purification through the preparative sucrose density gradient step. The protein (5 mg/ml) was prepared for centrifugation by equili- brating with the buffer on a small Sephadex G-25 column equilibrated with the centrifugation buffer (see Methods). The sample which contained 0.5 mM_NADH (upper pattern) was prepared by addition of the appropriate amount of 10 mM NADH to an aliquot of the equilibrated protein. The photo- graph taken 56 minutes after reaching speed (56,100 RPM) is shown in Figure 18. The average sedimentation coefficient 0.5% in the presence of 0.5 mM NADH, the value is 5.09 S. These for the enzyme in the absence of NADH is = 4.89 S and values are quite reasonable for an enzyme with a molecular weight of 69,000 as estimated from a plot of log 5 versus log molecular weight for several proteins of known molecular weight (Holleman, 1966). From this plot a sedimentation coefficient of 5.0 S is equivalent to a molecular weight of approximately 76,000. From the appearance of both of the patterns in Figure 18, the enzyme seems to be quite pure, even though the enzyme was not purified through the final step. Both peaks are symmetrical and there is no indication of higher and lower molecular weight material. (The rise in the base— line near the bottom of the cell was much the same in the zero time photograph, so it would seem to be an artifact in the Optical system.) Hence, these schlieren patterns are consistent with, although not proof of, purity. 98 .omm mo3 oamno ouoam ommnm on» cam HE\mE m mo3 nofiuouunoonoo nHouOHm one .modz SE m.o ponflmunoo onEom “one: one .poomm manoooH noumm mononHE mm noxou mwz nmoumouonm one .mponuoz pno maofluouoz CH ponflnomop mo poEMOMHom unoEfiuomxo onu Uno pouom Iona moB onEom one .omonomoupwnom ouonmmonm Hono0%H0|o now mnuouuom muflooHo> nofluounoeepomul.ma ousmflh 1" In. 9 ’aa5 8" Q -..t- ‘It‘ '0‘" i. ‘I’ ., (0‘. ‘ .2. ‘11" Itltlv| '1 ‘ ‘ -.V \‘0' I < 100 IsOelectric Point The isoelectric point of partially purified enzyme (purified through the preparative sucrose gradient, specific activity = 55) was determined by isoelectric focusing using a small, mini-column technique developed in this laboratory (Massey e£_31., manuscript in preparation). The enzyme was focused for 7 hours at 4° in a pH 5 to 8 gradient and the various fractions collected and assayed for enzymatic acti- vity. The results shown in Figure 19 indicate an isoelectric point of 6.8. This agrees quite well with the value of 6.7 we obtained earlier by isoelectric focusing in a large, 110 ml LKB isoelectric apparatus. Absorption Spectrum The absorption spectrum of the purified enzyme (Specific activity = 100) was determined manually with a Beckman DU monochrometer equipped with a Gilford absorbance indicator, using a sample which had been dialyzed against 0.02 M MES, 0.001 M EDTA, 0.1 M KCl, pH 5.9. The results are shown in Figure 20. There is a broad absorption maximum centered around 273 nm, and perhaps a small shoulder at 259 nm, indicating the possibility of a bound nucleotide; the 280/260 ratio is 1.1. Amino Acid Composition The amino acid composition of the purified, carboxy- methylated enzyme (specific activity = 100) is shown in 101 .HE\mpens m0 opens an we muw>fluom oexmno.one .00 no pousmmoE mm onp pno mufl>fluoo MOM pomommo ouo3 pouooaaoo wnofluoonm one .unoflpoum oueaonmfio mum mm o mnfims muoon 0 now moonnoz pno maofluouos nfl ponfluomop mo nEsHoo mnHmSUOM owuuooHoOmfl HmHoHoEEoonon .HHoEm o n0 pomSUOM mo3 oewuno one .ommnomoupwnon oumnmmonm HonoomHouo mo mnflmdoom oeuuooaoomHuu.mH ousmfim 102 On amnSDZ 20.._.U<~_n_ mN ON 2 O— m o q u Op . 5! q x. d d 1 023300“. o.d.—UmamOm. O— ALIAILDV JWAZNJ o—o 103 I .xcoankonu on cow: was “woman on» can .0.m mm .Hox z H.0 .000m 2 H00.0 .mmz z no.0 umneomo poneaoep noon won onEom nflououm one .nouoE oononHOmno UHOwHHU o nue3 commesqo uouoEounoonoE Do :oExoom o mnflms waamsnoe ponflfiuouop mmz oamfiom HE\mE mm.o o mo Eduuoomm noflumuomno one .ommnomonpmnoa oumnmmonm Houoowawno mo Eduuoomm neeumHOmnnll.om ousmflm 104 00m 1 CON - 050 1523053 owN CNN OON OmN u OVN I Saahugm 20..—._¢0mn< N.0 0. s. O O SDNVSUOSBV “2 o O.— 105 Table VI, along with the results for the rabbit muscle enzyme as reported by Holleman (1966). The compositions of the two proteins are quite similar. However, the rabbit muscle enzyme contains significantly more isoleucine, phenyla- lanine and lysine and less serine, proline and arginine. TABLE VI AMINO ACID COMPOSITION pig Liver Rabbit Muscle Residues Residues Residues Per Per a Per a 69,000 76,000 69,000 Lysine 32 55 50 Histidine 16 18 16 Arginine 18 16 15 Aspartate 50 S6 51 Threonine 24 28 25 Serine 28 21 19 Glutamate 64 82 74 Proline 36 31 28 Glycine 60 77 70 Alanine 50 S7 52 Valine 48 62 56 Methionine 12 15 14 Isoleucine 36 57 52 Leucine 50 58 53 Tyrosine 10 10 9 Half-Cystine 18 21 19 Tryptophan 5 4 4 Phenylalanine 24 32 29 aValues taken from Holleman (1966) 106 Using this amino acid composition for the pig liver c-glycerol phosphate dehydrOgenase, a partial specific volume of 0.737 cc/g was calculated. Effects of NADH on Disc Gel Electrophoresis of Pig Liver a—Glycerol Phosphate DehydEOgenase The effects of NADH on the electrophoretic behavior of the enzyme are shown in Figure 21. The two major effects are: (1) Nearly pure or pure enzyme electrophoreses as distinct bands in the presence of NADH, but as smears of indistinct bands in the absence of NADH; (2) With crude homogenate, which contains multiple forms of a-glycerol phosphate dehydrogenase, some bands which do not poSsess enzymatic activity when electrOphoresed in the absence of NADH do exhibit activity when run in the presence of NADH. To do the electrophoresis in the presence of NADH, small subreservoirs filled with the Tris-glycine buffer contain- ing sucrose and 0.25 mM_NADH were used. Since the NADH is anionic at pH 8.3, it migrates through the gel ahead of the protein; much of it is concentrated at the buffer discon- tinuity interface, giving rise to the heavily stained band at the tracking dye position on gels stained for activity. The first effect is most clearly seen with pure enzyme (specific activity = 100, right set). On the gel that was electrophoresed in the absence of NADH, there was only a smear of protein. In contrast, the gel run in the presence of NADH had one major band of enzyme, with several minor areas of possibly contaminating protein. This effect 107 .monz m5 mm.o mo oonomoum onu n0 pomononm -onnomnm mum; =mo¢z+= omnmnma mnmo .cnmnoum now 060nmum AO0H u >ue>euoo cemeoommv ofimuno ousm mo memouonmounooam .pom unmwm .Nufi>wuoo new ponfloum Rom u >ua>fiuoo ofiwfloommv oENNno ousm wauoon Mo memouonmouuooam .uom uounou .Nufl>fluoo MOM ponfioum unouonquSm ouonomOEon uo>fla uHSUM mo memouonm nouuooam .uom umoq .ommnomoupxnoo ouonmmonm Houoowauno mo memononmouuooam H00 no manz mo muoomwmll.am ousmem 108 In=>=u< 109 is probably due to stabilization of the native dimer form of the enzyme during electrophoresis by NADH, giving a sharp band, while in the absence of NADH, an equilibrium between the monomer and dimer is established, giving rise to the observed smear of protein. The other effect of NADH on electrophoresis of the enzyme is seen on the left and center sets. Electrophoresis of the crude homogenate (left set) and partially purified enzyme (specific activity = 90, center set) in the absence of NADH gives one and two major activity bands, respectively. But if these same samples are electrophoresed in the presence of NADH, one additional major and one additional minor band appear on the gel of the crude homogenate and four addi- tional minor bands of activity appear on the gel of the partially purified enzyme. CHAPTER VII KINETIC CHARACTERIZATION Kinetic Mechanism From his extensive studies on the rabbit muscle enzyme, Black (1966) concluded that its mechanism was Ordered Bi Bi, with NADH adding first and NAD leaving last, as shown below: NADH DHAP a-GP NAD l i I I E E-NADH . E-NADH-DHAP E—NAD E E-NAD-a-GP The mechanistic studies reported here were limited in sc0pe, but detailed enough to confirm that the pig liver enzyme has the same mechanism. Initial velocity studies were performed spectro— photometrically with both NADH and DHAP as the variable sub- strates (Figure 22). The data obtained were plotted in the form 1/v versus 1/5. The enzyme used had been purified to a specific activity of 65. Each point shown on the graph is the average of two assays. Although the lines on the graphs have not been extended to show the points of intersection, in both cases they intersect below the abscissa, in the third quadrant. These results are consistent with a 110 111 Figure 22.--Effect of Substrate Concentration on Initial Velocity of a-Glycerol Phosphate Dehydrogenase. A. Effect of NADH Concentration. Assays were performed in duplicate spectrophotometrically at 25°, with the indi- cated DHAP and NADH concentrations. The buffer was 0.1 M Tris—Cl, 0.001 M EDTA, 0.1 M KCl, pH 7.6. The velocity is in units of EDD/minute. _’B. Effect of DHAP concentra~ tion. Conditions same as Part A. 112 INITIAL VELOCITY A / 20i- o/O ,, DHAP L o / 0"”0036 mM- ‘0’ '-\> '- 0’0/ Iv' .". ." ’ ’ . l 25 50 75 100 mm“ vuocnv a 40 0.0064 IIIM> ° .’.. NADH ’I’ 0’. 0". 30 0 ”96 III 'D-I’ ‘- O .l' "1’ .-\> o/,o-’ ‘ ° .’ o ’ 20 n I A . .L A l A l A l A l A l o 10 20 30 40 /DHAP (mm) 113 sequential mechanism, but give no indication as to whether the reaction is ordered or random, and if ordered, which substrate binds first. To answer these questions, product inhibition studies were performed. Using the Ordered Bi Bi mechanism of the rabbit muscle enzyme as a model, the following product inhibition pattern would be predicted (Plowman, (1972): Inhibitory Product Type of Inhibition NADH Variable DHAP Variable a-glycerol phosphate non-competitive non—competitive NAD competitive non-competitive From this pattern, it was decided to use NADH as the vari— able substrate for our studies, since the results would probably be less ambiguous than those with DHAP. Since product inhibition is always non-competitive for the Random Bi Bi mechanism (Plowman, 1972), this case should be ruled out by the NADH-variable studies. These assays were per- formed fluorometrically because of the low NADH concentra- tions required. Because of large fluctuations in the intensity of the xenon lamp of the fluorometer, it was necessary to correct the assay slopes for these changes. A graph of fluorometer response as a function of NADH concen- tration using repeated measurements was constructed. A line fitting all the points was then taken as the average fluoro- meter response. The correction for an assay was obtained 114 by determining the ratio of the fluorometer response before addition of enzyme to the standard response for that NADH concentration and dividing the assay slepe by that ratio. The appropriate response data were not available for the studies using NAD as inhibitor, so this correction could not be made. The results of these studies are shown in Figure 23. In spite of the scatter in the data of the inhi— bition by NAD, the lines clearly intersect on the ordinate (competitive inhibition). It would be very difficult or impossible to draw the lines such that they intersect on the abscissa or were parallel. However, the data for the a-glycerol phosphate inhi- bition is more ambiguous. The lines are drawn as inter- secting on the abscissa (non-competitive inhibition), as one would expect for an Ordered Bi Bi mechanism. They could be drawn as parallel, however (uncompetitive inhibition). If one assumes that this ambiguity arises from the DHAP con- centration being too high, then either type of inhibition fits the Ordered Bi Bi mechanism. Michaelis Constants Using the data from the previous section (Figure 22) the apparent Michaelis constants for NADH and DHAP for the reaction in the forward direction were calculated. To obtain data for calculation of Michaelis constants for NAD and L-a-glycerol phosphate in the reverse direction, 115 Figure 23.--Inhibition of a-Glycerol Phosphate Dehydrogenase by Products. A. Inhibition by NAD. The assays were performed in duplicate fluorometrically at 25°. The assays contained the indicated amounts of NAD and NADH; the DHAP concentration was 0.071 mM. The buffer was 0.1 M Tris-C1, 0.001 M EDTA, 0.1 M KCl, pH 7.6. The velocity is in units of r§1ative actiVity. B. Inhibi- tion by L-a-Glycerol PhOSphate. Conditions same as Part A, except the data was treated to remove scatter due to fluctuation of the light source (see text). 116 PRODUCT o A INHIBITION . BY NAD raooucv _ a a" mmamou o / BY 5" I- 1 ouceaoruosruns / I" I 600 ' 900 1200 I/NADHImM) 117 experiments were performed spectrOphotometrically at pH 10.0. The results obtained are shown in Figure 24. The apparent Km values obtained for each of the substrates are: NADH, 8 uM; DHAP, 0.27 mM; L-a-glycerol phosphate, 0.20 mM; NAD, 57 uM. The apparent Km for NAD at pH 7.6 of 80 0M was estimated from the data on the inhibition of the enzyme by ATP, ADP andSYAMP in the reverse direction (see next_ section). Inhibition by Adenine Nucleotides The inhibition of the enzyme by adenine nucleotides was studied using an enzyme sample which had been purified through the sucrose gradient step (specific activity = 55). Spectrophotometric assays were performed at pH 7.6 in both the forward and reverse directions. The data were plotted in the form l/v versus inhibitor concentration (Dixon plot). The results for the various nucleotides (ATP, ADP and AMP) are shown in Figures 25 through 27. From the intersection points of the lines, the inhibition constant (Ki) for each nucleotide can be calculated. The values obtained for each nucleotide are: 5'-AMP, 18 mM (forward) and 18 mM (reverse); ADP, 17 mM (f) and 16 mM_(r); ATP, 6 mM (f) and 8 mM (r). The inhibition in each case is competitive with respect to the pyridine nucleotide, as might be eXpected, since a por— tion of the pyridine nucleotide molecule is identical to an ADP molecule. 118 Figure 24.-—Determination of for Substrates in the Reverse Assay of a-Glycerol Phosphate Dehydrogenase. A. Km for NAD. Assays were performed spectrophotometri- cally in duplicate at 25° in 0.1 M glycine, 0.001 M EDTA, 0.1 M KCl, pH 10.0. The L-a-GP c5hcentration was 8'mM. The velocity is in units of AOD/minute. B. Km for L-a- Glycerol Phosphate. Conditions same as Part A, except 5 mM NAD was used. 119 nevus: ASSAY Km NAD 1 1 . 1 1 0 7.5 I5.0 22.5 30.0 JmAnbnm) REVERSE ASSAY K... lot L GlYCEROPI-IOSPHATE 5 I- «I 4 "' / cl 0 O > 3 _ /°/ D / o 2 P O K...-0.2 mM 1 1 L . 1 P 1 1 1 1 1 0 3 6 9 12 120 Figure 25.--Inhibition of a-Glycerol Phosphate Dehydrogenase by ATP. A. Forward Assay. Assays were performed spectrophotometrically in duplicate in 0.1 M Tris-C1, 0.001 M EDTA, 0.1 M KCl, pH 7.6. The indicaEed amounts of ATP and NADH werE added; the DHAP concentra- tion was 0.71 mM. The data were plotted in the form of the Dixon plot,—l/v versus i. The velocity is in units of ADD/minute. B. Reverse Assay. Conditions same as Part A, except the indicated amounts of NAD were added; the L-a-glycerol phosphate concentration was 0.5 mM. —__ 121 mmsmou BY ATP 80 - 0 /’ ..\> / 60" o/o o/ o/ \005 MM NAD Kl'a MM . ..... I ————— I ..... . ._._I.._.— - - ................ \QS ”M ° 25 5.0 7.5 .0 ATP (MM) 122 Figure 26.--Inhibition of a-Glycerol Phosphate Dehydrogenase by ADP. A. Forward Assay. Conditions same as Figure 25, except ADP was used. B. Reverse Assay. Conditions same as Part A. 123 INHIBITION A BY ADP * . 2O " .. > —\ ’- / I O/o/o Ki'" mM o/°¢ 0.0067 mM NADH ._ _,_..I--"’ ‘ .. --—-----"-'=i0.027 mm In -------- — ..... ...-I- m‘ ...-...“...me W... ...... q 0067 mM . A l A 1 A A l A I A l A l A 0 2.5 5.0 7.5 IO ADP (mm) REVERSE ASSAY _ B ‘ INHIBITION BY ADP I00 - . BO - .. \> '— 60 r- q , ...—.— K5316 MM 40 I. . . ...—...— .--—- ""' o<-0.05 mM / ‘- -““'" - .1162 . / .9 ------- ' ------- I — - ':0.I mM — ————————————— - 29 g A ‘- .. a .. \ 0.5 mM . l A l_ A l l l A 1 A 0 2.5 7.5 IO 124 Figure 27.--Inhibition of a-Glycerol Phosphate Dehydrogenase by 5'-AMP. A. Forward Assay. Conditions same as Figure 25, except 5'-AMP was used. B. Reverse Assay. Conditions same as Part A. 125 INHIBITION BY A 5|-AM' b q 20 I J ’ q ..\> . . h ...-’0" ./ ' v- , . KI'IBmM IO _....——-""""""o a“(9.0067 MM NAM .// ..-“..- q ..-.-.oo-O' .-'-. __._.._.- ------ -° ---""" ‘0.027 IIIM . _._._--—'- ‘ ‘ Am‘m ‘m ‘0067 um . l A l A A I A l A l A l A O 2.5 5.0 7.5 IO 5'-AMR (mM) REVERSE ASSAY . ' I INHIBITION av s’-Am - 1 .0 " I 0 .- — " g. 0.05 M d Ki-IB MM 40 .. - -0 __ _° 0 MAE” an." ’.-.‘ ......ooo-O. ”....”- . ””””” . 1 —°‘— ... .....- .‘W' - "" ““‘ "' OI mM 0 0 . —— 0 o o \005 MM 1 0 25 ‘ 5.0 7.5 IO 126 We also tested the following compounds for inhibi- tion in the forward direction at low levels of substrate (0.01 mg NADH and 0.36 mg DHAP): 3',5'-cylic AMP (5 mg), citrate (20 mg), fructose-1,6-diphosphate (10 mg), lactate (10 m!) and acetyl Co A (0.31 mg). No appreciable inhibi— tion (less than 15 percent) was observed under these condi- tions with these compounds. ATP (10 mg) and ADP (10 mg) inhibited approximately 60 percent and 40 percent, respec— tively, under these conditions. Several of these compounds were also tested for inhibition in the reverse direction under low substrate conditions (0.05 mg NAD and 0.5 mg L-a—glycerol phosphate). No inhibition was observed with 3',5'-cyclic AMP (5 mg) and lactate (10 mg), but an inhi- bition was observed with citrate (20 mg) and with FOP (10 m!) both of which seemed to affect the extent of the assay reaction along with the velocity of the reaction. pH Optima The pH optima for the reactions in the forward and reverse directions were determined with an enzyme sample which had been purified through the phosphocellulose column step (specific activity = 15). A stock assay mix containing the apprOpriate buffer (0.05 g), 0.1 §_KC1, and substrates (0.7 mg DHAP and 0.25 mg NADH or 8 m!_ L-a-glycerol phos- phate and 5 m! NAD) was prepared and adjusted to the desired pH values by addition of l N NaOH or 1 N HCl with a 127 microliter syringe. For each pH, an aliquot of the assay mix was removed and assays were performed in duplicate immediately. The results obtained are shown in Figure g8. The pH Optimum for the reaction in the forward direction is approximately pH 7.5 and that for the reverse direction is approximately pH 10.0. The sharp decrease in enzymatic activity above pH 10.2 was due to inactivation of the enzyme during the assay due to the high pH. 128 .muscfls\oo< mo mafia: cfl ma >ua>fluom mahwcm wse .mumoflamso CH xaquflUmEEfi UmEHOwqu mum3 mammmm one .Hom zH Ho momz 2H mo coHuMppm xn m.mm pmufimmv may ou Umumsmmm can Umummwmm mm3 AQ¢Z SE m new mumnmmond Houmomamnsua zs.m so momm as m~.o cam ammo ZS 5.0V mmumuumnsm cam .Hux z H.o .Az mo.ov umwmsn mum nfludoummm wsu mcflcflmusoo xflE womwm xooum 4 .mmmcwmoupxnmo mumnmmozm Houmowawld MOM mEHumo mall.mm musmfim 129 w a u: ENZYME ACTIVITY, FORWARD to I .... 03.3) menial \/ .\ 2/ .... / ”232. Q\. N s bx ...-O. o O ... .3 AA. \ O 0‘ 0b. / o.Io 0:230 ... .. .. _. .. .. .. .. p .. l / \ olo ......u ...fl... .....- .Eu . I .0... °.Nm /. ....... /.P u .. 10 oh I. .. — .. . A ohm Li A A — p — p N w o .0 EN ZYME ACTIVITY, REVERSE CHAPTER VIII DISCUSSION Physical and Chemical Characterization The molecular weight value of approximately 69,000 of the native enzyme is similar to the molecular weights reported for the a-glycerol phosphate dehydrogenases from other sources, which range from 63,000 for the rat muscle and liver enzymes (Fondy §£_al., 1968, 1971a) to 77,000 for the rabbit muscle (Brosemer, 1969), rabbit liver (Otto gt al., 1972) and honeybee (Brosemer, 1969) enzymes. All of the a—glycerol phosphate dehydrogenases for which a subunit structure has been reported are dimers, so the pig liver enzyme is typical in this respect, also. The pig liver enzyme sediments as a single symmetri- cal peak (Figure 18) and yields a sedimentation coefficient of $265% = 4.89 S, which agrees quite well with the value of ,w o = 4.86 S reported for the rabbit muscle enzyme S20 w (Holleman, 1966). However, it is higher than the 520,w values of 3.0 S and 3.6 S reported for the rat liver enzyme (Ross §£_gl., 1971). and the 320,w value of 2.8 S value for the rat muscle enzyme (Fondy et_3l., 1968). The absorption spectrum observed with the pig liver a-glycerol phosphate dehydrogenase is similar to that of 130 131 a-glycerol phosphate dehydrogenases isolated from various sources. The rabbit-muscle enzyme has an absorption maximum near 270 nm (van Eys, §E_gl., 1959) and a 280/260 ratio of 1.1 and the rabbit liver enzyme has an absorption maximum near 273 nm and 280/260 ratio of 1.2 (Otto gt_31., 1972). The rabbit muscle enzyme is known to have a bound nucleotide, but its structure is presently unknown. The partial specific volume of 0.737 cc/g calculated for the pig liver a—glycerol phosphate dehydrogenase is within the range of 0.70 to 0.76 cc/g normally observed for proteins. However, it is lower than the value of 0.747 cc/g reported for the rabbit and rat muscle enzymes (Holleman, 1966; Fondy g£_31., 1968). Ross gt_§1. (1971) reported values of 0.744 and 0.737 cc/g for the two forms of the rat liver enzyme. Kinetic Characterization While our studies seem to indicate that the enzyme has an Ordered Bi Bi mechanism, they have not ruled out the Theorell—Chance mechanism. It would be necessary to do product inhibition studies with DHAP as the variable sub- strate to definitely prove this. However, since this mechanism was ruled out for the rabbit muscle enzyme, we can probably rule it out for the liver enzyme, also. Kim and Anderson (1968) studied the inhibition of the rabbit muscle enzyme by several adenine nucleotides. 132 They studied the inhibition of NAD binding at pH 7.85 and found the following inhibition constants: adenosine, 4.3 mg; 5'—AMP, 2.1 mM; ADP, 0.7 mM and ADP-ribose, 0.2 mM. They did not mention inhibition by ATP and did not discuss any possible in vivo significance of their findings, since they were primarily interested in studying the binding of NAD to the enzyme molecule. The inhibition of pig liver a-glycerol phosphate dehydrogenase by ADP and 55AMP is much different. The inhibition constants for ADP and 5'-AMP are 16 mM and 18 mM, respectively, for the pig liver enzyme. The finding in our studies that the inhibition con— stant in the forward reaction for each nucleotide is approxi- mately equal to that in the reverse reaction would seem to indicate a simple, competitive inhibition where the Ki is actually the equilibrium constant for the binding of the nucleotide to the enzyme. Since the adenine nucleotides inhibit competitively with the pyridine nucleotides, it is likely that they are binding to the same site on the enzyme. Hence, the adenosine moiety in the pyridine nucleotide plays an important role in binding the nucleotide to the enzyme. However, the "adenine-phosphate" portion of the binding site is not the major factor, since the pyridine nucleotides are bound 100 to 1000 times as tightly as are the adenine nucleotides. Further, since cyclic AMP does not significantly inhibit the enzyme, it appears that the adenine ring itself plays only a minor role in the binding of the nucleotide to the enzyme. 133 This inhibition of the enzyme by ATP, ADP and AMP appears to be quite significant in playing a role in the control of the enzyme in vivo. Its importance is dependent on the levels of NADH and the adenine nucleotides in the cell, however. There are major problems involved in estima- tion of the level of free NADH in a cell, but it appears that the concentration is in the range of l to 10 umoles per kg (1 to 10 pg) (Biochemists' Handbook; Krebs, 1967; Willms gt_31., 1970) for liver from fed animals. The levels of the adenine nucleotides are easier to estimate and are approximately 3 umoles per g for ATP, 0.6 umoles per g for ADP and 0.1 umoles per g for AMP (mg), (Willms gt_§1., 1970; Williamson and Corkey, 1969). The levels of ADP and AMP are not high enough to give significant inhibition and control the enzymatic activity. However, the ATP concen— tration is high enough to be important if the NADH level is near the low end of the concentration range mentioned (1 to 4 UM). One can calculate the inhibition eXpected using the equation: V: V 1 + Km/s (17+ 17Ki) =6mM_andATP=3mM_, Assuming Km = 8 UM, NADH = 2 uM, Ki an inhibition of 40 percent would be observed. Any inhibi- tion present could be easily overcome by an increase in the NADH concentration however, as might occur during an increased glycolytic flow. The pH optima of 7.5 and 10.0 observed are similar 134 to those reported for a-glycerol phosphate dehydrogenase from other sources. Young and Pace (1958) reported Optima of 7.4 and 10.0 for the rabbit muscle enzyme, Lehmann and Pfleiderer (1968) reported an Optimum of 7.8 for the human liver enzyme in the forward direction and Marquardt and Brosemer (1966) reported an Optimum of 6.5 for the forward reaction of the honeybee enzyme. Effects of NAD and NADH on the Enzyme Several important effects of NAD and NADH on the pig liver enzyme have been observed during these studies and some have been discussed earlier in this thesis. They pro- teCt the enzyme against heat inactivation, prevent dissoci- ation at low protein concentration, affect both the electro- phoretic pattern and activity of the enzyme on disc gel electrophoresis, affect both the stability and chromato- graphic behavior of the enzyme on ion exchange chromatography and control the inhibition of the enzyme by ATP and other adenine nucleotides. These effects are quite significant and indicate the importance of the pyridine nucleotides in maintaining the structure and activity of the enzyme. While this report of stabilization of a-glycerol phosphate dehydrogenase is not unique, the effects of NAD on the binding of the enzyme during ion exchange chromato- graphy and the effects of NADH on electrOphoresis of the enzyme have not been reported before. During our work on 135 the purification of pig liver a-glycerol phosphate dehydro- genase, the reports of the use of NAD during the heat step in the human liver enzyme and the use of NADH during the heat step in the purification of the chicken liver enzyme were published. However, none of the published purification procedures utilize NAD during column chromatography steps. Effect of NAD on Heat Inactivation Protection against heat denaturation may merely be a laboratory phenomenon, but it is quite possible that this protection is physiologically significant. The heat dena- turation studies were performed at 49° instead of 37° pri— marily because the rate of denaturation of the enzyme at 49° is such that the protection could be studied easily. Even though the rate at 37° would be lower, it should still be significant. From interpolation between the denaturation rate at 4° (half-life = 48 hours) and at 49° (half-life = 2.5 minutes), we estimate that the half-life of denaturation at 37° of a-glycerol phosphate dehydrogenase in the absence of NAD or NADH would be in the range of 20 minutes under our.experimenta1 conditions (phosphate buffer, pH 7.8). The levels of NADH in vivo are probably too low to protect the enzyme against heat denaturation, but the levels of NAD are high enough that they would protect the enzyme against simple heat denaturation quite well. The NAD concentration in vivo in rat and guinea pig liver is approximatey 0.5 mg (Williamson and Corkey, 1969; Willms et al., 1970); under 136 these conditions the protection would be nearly complete, maintaining a high level of active enzyme. But if the NAD ‘COncentration were to drOp below 0.2 mM for any reason, the loss of enzyme in vivo due to heat inactivation could become quite significant. Although studies on protection against proteolytic degradation by the nucleotides have not been performed, it is also quite possible that high levels Of NAD and NADH would protect the enzyme against this type of denaturation. Protection against these two types of dena- turation could play a major role in determining the level of active enzyme in the cell. It is difficult to decide whether the "protective" nucleotide binding sites are identical to the catalytic binding sites or not. The only pertinent evidence available is the data for the binding constants from the kinetic data and the denaturation data. The Km for NAD was determined at pH 10.0, so it is of little use in this comparison. How- ever, a Km of 80 mM_ at pH 7.6 was estimated:flnmnthe adenine nucleotide inhibition data. Also, one can determine the Ki for NAD at pH 7.6 from the product inhibition studies (page ), by replotting the data in the form of l/v versus NAD at the various levels of NADH (not shown). This is the standard Dixon plot, and it gives a value of approximately 0.5 mM. This is about four fold higher than the binding constant determined from the heat denaturation data, 0.125 mM. These values are somewhat different, but this could be 137 due to differences in pH (7.6 for kinetics and 7.8 for heat Adenaturation), buffer species and ionic strength. However, there is a large difference inthe binding constant for NADH as determined by the two methods. The binding constant (Kia) for NADH from the kinetic data is approximately 2 “M (Plowman, 1972, page 139). This value is the point at which the lines intersect in the initial velocity studies (Figure 22). Even though the heat denaturation data for NADH could not be analyzed because of complications, we estimate a binding constant of 0.05 to 0.15 mM from inspection of the data shown in Figure 7. This is 100 fold greater than the binding constant from the kinetic data and this difference is much too large to be due to differences in.the buffer or pH. A reasonable eXplanation for this large difference is that the large difference in temperature causes a rather drastic change in the conformation of the enzyme, which in turn changes the affinity of the enzyme for the nucleotide. To test this hypothesis, however, further binding eXperi- ments would be necessary. Effect of NADH on Dissociation at Low Protein Concentration This effect was observed during the high speed sedi— mentation equilibrium eXperiments. In all three samples without NADH, there was clearly low molecular weight material present (Figure 15), but in the presence of 0.5 mg NADH, there was no evidence of low molecular weight material. 138 Also, an earlier eXperiment using material purified through the sucrose gradient step(specific activity = 90) indicated that 0.5 mM NAD also prevented the dissociation at low protein concentration. Nucleotides may protect the enzyme from heat inactivation by preventing dissociation. It is quite possible that the monomer is less stable to heat than the native dimer. If so, the binding of the nucleotide to the enzyme would prevent the stable dimer from dissociating to unstable monomer, thereby greatly slowing down the inactivation of the enzyme. Since the intercellular a-glycerol phosphate dehy- drogenase concentration is on the order of 0.1 to 0.3 mg/ml (calculated from the crude homogenate levels observed and an assumed specific activity of 100 units/mg), the dissociation could occur in vivo but would be somewhat limited. Effects of NADH on Electrophoretic Behavior The effects of NADH on the enzyme in disc gel elec— trOphoresis eXperiments were quite important in the purifi— cation of a-glycerol phosphate dehydrogenase. The use of NADH in the running buffer was critical to the success of the preparative disc gel electrophoresis step. The finding that electrophoresis of crude or par— tially purified enzyme in the presence of NADH increases the number of activity bands was quite interesting. An enzyme sample of high apparent purity (purified through the sucrose density gradient; specific activity = 90 and one 139 peak in the Model B) electrophoresed in the absence of NADH showed at least 8 protein bands, but only 2 activity bands (Figure 21. When the sample was electrophoresed in the presence of NADH, 6 of the bands had activity. Several electrophoretic forms of the enzyme are not active unless electrOphoresed in the presence of NADH. Apparently, the NADH binds to these forms during electrophoresis and pre- vents their denaturation. NADH also protects those forms of the enzyme which retain activity when electrOphoresed in the absence of NADH, since it is necessary to stain gels electrophoresed in the absence of NADH at least twice as long as gels electrophoresed in the presence of NADH to obtain activity bands of equal intensity. Ross §£_31. (1971) reported the existence of several forms of a-glycerol phosphate dehydrogenase in rat liver. The forms were separated by isoelectric focusing on polyacrylamide gels, and the enzymatic activity was visual— ized by using an activity stain similar to the one used in this thesis. They Observed five different forms of the enzyme using this technique and were able to purify the two major forms. They found that the enzymes were similar kine- tically and had similar molecular weights, but differed in their amino acid compositions. Effects of NAD on Ion Exchange Chromatography Early attempts to purify the enzyme using ion exchange chromatography were futile because no NAD vnus used in the 140 buffers. DEAE, sulfoethyl, carboxymethyl and phOSphate ion exchangers were tried, and in every case, enzymatic acti- vity could be recovered only if the enzyme did not bind to the column. If the enzyme did bind, no activity could be recovered. Later, the protective effect of NAD was found and the two ion exchange steps finally incorporated into the purification procedure were worked out. The simplest explanation for the instability of the enzyme toward ion exchange chromatography is that there is a portion of the native enzyme molecule-~either a small molecule, such as a nucleotide, or a small peptide—-which is necessary for the enzyme to be active that is removed from the enzyme during chromatography. In the presence of NAD, this removable portion is bound to the enzyme more tightly and the enzyme can be eluted from the column in an active form. A second explanation is that there is a reactive group on the enzyme that is modified while the enzyme is bound to the column; when NAD is bound to the enzyme, this group is masked and protected. Another effect of NAD on the chromatographic beha- vior was a change in the binding of the enzyme to the phos- phocellulose with a change in the NAD concentration. In chromatography experiments with 0.1 mM_NAD present, the enzyme bound to the column in the presence Of 0.02 M_KC1 (3.7 m8) and eluted from the column at approximately 0.10 M KCl (13 m8). However, if 0.3 mM_NAD was present, then it 141 was necessary to use 5 mM KCl (1.34 mS) initially and the enzyme eluted near 0.065 M KCl (8.8 m8). This different behavior of the enzyme in the two systems was quite repro— ducible. Approximately two-fold higher yields were obtained in the presence of 0.3 mM NAD, and this required that the conditions for chromatography be redefined for the new system. Effect of NAD on Inhibition by Adenine Nucleotides Since the Km for NADH is in the range of NADH con- centrations observed in liver, a major controlling factor in the inhibition of the enzyme by ATP in vivo is the NADH level. Earlier in this thesis, it was calculated that under well-fed conditions (NADH = 2 pM_and ATP = 3 mM), the enzyme would be inhibited 40 percent. However, if the NADH level increased to 10 pM, as might occur under fasting or fat—fed conditions (Krebs, 1967; Willms gt_31,, 1970) then the inhi- bition would be decreased to 20 percent. Accompanying this decrease in inhibition would be a three-fold increase in the overall rate of the enzyme reaction, due only to the increase in the NADH concentration near the Km. These two effects would combine to play an important role in the modu- lation of the enzymatic activity in vivo. Implications for Further Research Multiple Forms of Pig Liver a-Glycerol Phosphate Dehydrogenase There are several interesting areas of study involv- ing the multiple forms of the enzyme that should be 142 investigated. The different forms may be produced during different stages of development, but we have not been able to show this in our brief eXperiments with livers from pigs of different ages. More extensive studies with fetal and newborn pigs should show if they are different developmental forms. Whatever the source of the various forms, it would be interesting to investigate which forms are present in one or more hepatomas. Such a study would show whether the decrease in the a-glycerol phosphate dehydrogenase activity in the tumors is due to a decrease in the levels of several or all the forms or to a decrease in or disappearance Of only the major forms. One of the extra forms is partially purified during the preparative disc gel electrophoresis step (Figure 12, fractions 36 through 38). While this enzyme fraction does not have a high specific activity (fraction 36 has a speci- fic activity of 23) it may be quite pure. Analytical disc gel electrOphoresis should show its purity; if it is grossly impure, a second preparative disc gel electrophoresis step using a lower gel concentration should increase the purity. This form of the enzyme could then be studied to see how it differs from the major formisolated and characterized in this thesis. If this second form can be obtained in pure form, experiments should be performed to see if it can be converted to another form; no conversion of the purified major form to any of the other forms was Observed in our studies. 143 Binding of NAD and NADH to Pig Liver a-Glycerol Phosphate‘Dehydrogenase Earlier, the question of the identity of the "cata- lytic" nucleotide binding sites and the "protective" nucleotide binding sites was discussed. This question could be answered using fluorescence titration of the enzyme with NADH. If there are two types of NADH binding sites with binding constants of 2 uM and 75 uM as suggested by the kinetic and heat denaturation data, they should be clearly observable with this technique. However, if there is only one type of site, and the weak binding of NADH to the enzyme Observed in the heat denaturation studies is due to the high temperature, only one type of site should be observed with the fluorescence studies. The titrations could best be studied following the fluorescence of bound NADH activated by energy transfer from the protein molecule or the change in the fluorescence of the protein upon binding of the nucleotide (Velick, 1958). An alternative method of studying the question is performing both the heat denaturation studies and the kinetic studies at the same temperature, such as 37°. APPENDIX 144 APPENDIX Derivation Of Equation 1 (Patterned after Kuczenski and Suelter, 1970, 1971) Assuming the successive equilibria E E + NAD‘Z.E—NAD K = I 1 [NAD] I [E-NAD] E-N E-NAD + NAD'Z E—NA02 KII— I AD] [NAD] [E-NADz] through [E-NAD NAD] E-NAD —1 + NAD'Z.E-NADn Kn = .n-1] [ n [E—NADn] If KI > KII > Kn’ then _ E-N D _ KI+KII+ - - +Kn=[NAD]$-—LEl—— + _i§_E§Bl + — — - + [ A n 1]) [E—NAD] [E—NADZI [E-NADn] ) . ( [E] [E—NAD] [E—NADn_1] )_ 1 - a Slnce ( [E-NAD]+[E-NAD2] +’ ’ ’ +TE—NADn] )- then taking logarithms O. pKI + PKII + - - - + pKn = p [NAD] log I—:—fi or, rearranging n(DKav - p INAD]) = log 1 a 145 REFERENCES 146 REFERENCES Alfonzo, M., and Apitz-Castro, R. (1971), FEBS Letters 19, 235. Anderson, B. M., Kim, S. J., and Wang, C. N. (1970), Arch. Biochem. Biophys. 138, 66. Ankel, H , Bucher, T., and Czok, R. (1960), Biochem. Zeit. 2, 315. we Apitz, R., Ryder, E., and Gaede, K. (1965), Biochem. Zeit. l, 518. Apitz-Castro, R., and Suarez, Z. (1969), Biochim. BiOphys. Acta 185, 258. Apitz-Castro, R., and Suarez, Z. (1970), Biochim. BiOphys. Acta 198, 176. Beizenherz, G., Bucher, T., and Garbade, K. (1955), Methods in Enzymology I, 391. Black, W. J. (1966), Can. J. Biochem. 44, 1301. Blanchear, M. C. (1965), Can. J. Biochem. 43, 17. Boxer, G. E., and Devlin, T. M. (1961), Science 134, 1495. Boxer, G. E., and Shonk, C. E. (1960), Can. Res. 20, 85. Brosemer, R. W., and Kuhn, R. W. (1969), Biochemistry 8, 2095. Brosemer, R. W., and Marquardt, R. R. (1966), Biochim. Biophys. Acta 128, 464. Carnicero, H. H., Moore, C. L., and Hoberman, H. D. (1972), J. Biol. Chem. 247, 418. Celliers, P. G., Stock, A., and Pfleiderer, G. (1963), Biochim. Biophys. Acta 77, 577. Chasin, M. (1967), Ph.D. Thesis, Michigan State University. Cohn, E. H., and Edsall, J. T. (1943), Proteins, Amino Acids, and Peptides as Ions and Dipolar Ions, Rein- hold Publ. Co., New York, N. Y. 147 148 Crestfield, A. M., Moore, S., and Stein, W. H. (1963), J; Biol. Chem.” 238, 622. Darnall, D. W.,and Klotz, I. M. (1972), Arch. Biochem. Bio— phys. 149, l. Davis, B. J. (1964), Annals N. Y. Acad. Sci. 121, 404. Dawson, A. P., and Thorne, C. J. R. (1969), Biochem. Jour. 111, 27. Dietz, A. A., and Lubrano, T. (1967), Anal. Biochem. 20, 246. Dionisi, 0., Cittadini, A., Gelmuzzi, G., Galeotti, T., and Terranova, T. (1970), Biochim. Biophys. Acta 216, 71. Edelhock, H. (1967), Biochemistry 6, 1968. Fondy, T. P., LeVin, L., Sollohub, S. J., and Ross, C. R. (1968), J. Biol. Chem. 243, 3148. Fondy, T. P., Ross, C. R., and Sollohub, S. J. (1969), J; Biol. Chem. 244, 1631. Fondy, T. P., Herwig, K. J., Sollohub, S. J., and Rutherford, D. B. (1971a), Arch. Biochem. Bi0phys. 145, 583. Fondy, T. P., Solomon, J., and Ross, C. R. (1971b), Arch. Biochem. Biophys. 145, 604. Furgeson, K. A. (1964), Metabolism 13, 985. Furthmayer, H., and Timpl, R. (1971), Anal. Biochem. 41, 510. Galton, D. J., and Bray, G. A. (1967), J. Clin. Endocrin. Metab. 27, 1573. Hajara, A. K., and Agranoff, B. W. (1968), J. Biol. Chem. 243, 1617. Holleman, W. H., (1966), Ph.D. Thesis, Michigan State Univer- sity. Kaplan, N. O. (1964), Brookhaven Symp. Biol. 17, 131. Keilin, D., and Hartree, E. F. (1938), Proc. Royal Soc. (London) B124, 397. Kemp, R. G., and Mauk, M. R. (1971), FEBS Letters 16, 51. Kim, S. J., and Anderson, B. M. (1968a), Biochem. Pharm. 11, 2413.’ ' 149 Kim, S. J., and Anderson, B. M. (1968b), J. Biol. Chem. 243, 3351. Kim, S. J., and Anderson, B. M. (1969a), J. Biol. Chem. 244, 231. Kim, S. J., and Anderson, B. M. (1969b), J. Biol. Chem. 244, 1547. Kistler, W. S., Hirsch, C. A., Cozzarelli, N. R. and Lin, E. C. C. (1969), J. Bacteriol. 100, 1133. Kito, M., and Pizer, L. I. (1968), Biochem. BiOphys. Res. Comm. 32,408. Koditschek, L. K., and Umbreit, W. W. (1969), J. Bacteriol. 28, 1063. Krebs, H. A. (1967), Adv. Enz. Reg. 5, 409. Kuczenski, R. T., and Suelter, C. H. (1970), Biochemistry 9, 939. Kuczenski, R. T., and Suelter, C. H. (1971), Biochemistry 10, 2867. Lee, Y. P. and Craine, J. E. (1971), J. Biol. Chem. 246, 7616. Lehmann, F., and Pfleiderer, G. (1968), Hoppe-Seyler's Zeit. Physiol. Chem. 349, 1777. Lowry, O. H., Passonneau, J. V., and Rock, M. K. (1961), J. Biol. Chem. 236, 2756. Marquardt, R. R., and Brosemer, R. W. (1966), Biochim. Biophys. Acta 128, 454. Martin, R. G., and Ames, B. N. (1961), J. Biol. Chem. 236, 1372. Massey, T. H., and Deal, W. C. (1973), J. Biol. Chem. 248, 56. Mejbaum-Katzenellenbogen, W. and Dobryszycka, W. A. (1959), Clin. Chim. Acta 4, 515. O'hea, R. K. and Leveille,G. A. (1969), J. Nutr. 99, 338. Ornstein, L. (1964), Annals N. Y. Acad. Sci. 121, 321. Otto, J., Raggi, A., Machleidt, W., and Bucher, T. (1972), Hoppe-Seyler's Zeit. Physiol. Chem. 353, 332. 150 Pfleiderer, G., and Auricchio, F. (1964), Biochem. BiOphys. Res. Comm. 16, 53. Plowman, K. (1972), Enzyme Kinetics. McGraw-Hill, New York: NOY. Ringler, R. L. (1961), Jour. Biol. Chem. 236, 1192. Robertson, D. C., Hammerstedt, R. H., and Wood, W. A. (1971), J. Biol. Chem. 246, 2075. Segrest, J. P., Jackson, R. L.,Andrews, E. P., and Marchesi, V. T. (1971), Biochem. Biophys. Res. Comm. 44, 390. Shapiro, A. L., Vinuela, E., and Maizel, J. V. (1967), Biochem. Biophys. Res. Comm. 28, 815. Suarez, Z., and Apitz-Castro, R. (1972), Biochim. Biophys. Acta 258, 339. Telegdi, M. (1964), Acta Physiol. Acad. Sci. Hung. 25, 181. Telegdi, M. (1968), Biochim. BiophyA, Acta 159, 227. Telegdi, M., and Keleti, T. (1968), Acta Biochim. et Bio- phys. Acad. Sci. Hung, 3, 1311 van Eys, J., Nuenke, B. J., and Patterson, M. K., Jr. (1959), J. Biol. Chem. 234, 2308. van Eys, J. (1960), Fed. Proc. 19, 26. van Eys, J., Judd, J., Ford, J., and Womack, W. B. (1964) Biochemistry 3, 1755. Van Holde, K. E. (1967), Fractions, 1. Velick, S. F. (1958), J. Biol. Chem. 233, 1455. Weber, K., and Osborn, M. (1969), J. Biol. Chem. 244, 4406. Werner, V. S., and Schwandt, P. (1969), Hoppe-Seyler's Zeit. Physiol. Chem. 350, 286. White, H. B., and Kaplan, N. O. (1969), J. Biol. Chem. 244, 6031. White, H. B. (1971), Arch. Biochem. Biophys. 147, 123. Williamson, J. R., and Corkey, B. E. (1969), Methods in Enzymology XIII, 434. Williamson, J. R., Jakob, A., and Refino, C. (1971), J; Biol. Chem. 246, 7632. 151 Willms, B., Kleineke, J., and Soling, H. (1970), Biochim. Biophys. Acta 215, 438. Wilson, J. E. (1968), J. Biol. Chem. 243, 3640. Young, H. L., and Pace, N. (1958), Arch. Biochem. BiOphys. 7_5, 125. thantis, D. (1964), Biochemistry 3, 297. Ila Ilfllu LI I1..II!II“.II u, "'Tlfi'l‘fifllfilifijflllyfi!fliflmjflfiflflifljfiflflfiu‘mES