ABSTRACT ELECTRON TRANSPORT FLAVOPROTEINS IN THE LACTATE FERMENTATION OF PEPTOSTREPTOCOCCUS ELSDENII BY Howard L. Brockman, Jr. The components of the electron transport system between B—lactate and a,B-unsaturated acyl CoA in g. elsdenii were found to be Q(-)-lactic dehydrogenase, electron-transferring flaVOprotein and acyl CoA dehydro- genase. The experimental evidence which supports this conclusion includes: (a) Qflactate will reduce Q(-)- lactic dehydrogenase, (b) gelactate will reduce a sub- strate level of electron-transferring flaVOprotein in the presence of a catalytic level of 2(—)-lactic dehydro- genase, (c) Q-lactate will reduce acyl CoA dehydrogenase in the presence of a catalytic amount of Qfl-)-lactic dehydrogenase and a catalytic amount of electron- transferring flaVOprotein and (d) in the presence of E(-)-lactic dehydrogenase, electron-transferring flavo- protein and acyl CoA dehydrogenase, Belactate will reduce crotonyl CoA to butyryl CoA. No additional protein Howard L. Brockman, Jr. components were found and neither Coenzyme Q, dithio- threitol, 1,3-dimercaptopropanol, flavin adenine dinucleotide, flavin mononucleotide nor boiled cell extract would stimulate the rate of electron transport. NADH will reduce electron-transferring flavo- protein and will reduce acyl CoA dehydrogenase in the presence of a catalytic amount of electron-transferring flavoprotein. 2(-)-lactic dehydrogenase is not reduced by NADH, even in the presence of a catalytic amount of electron-transferring flaVOprotein. Thus, electrons from NADH enter the electron transport chain via electron- transferring flavoprotein. Kinetic experiments suggest that binding is involved in the transfer of electrons between protein components of the system. The apparent Km for reduced Q(-)-lactic dehydrogenase in its reaction with electron- transferring flaVOprotein is 9.5 x 10-8 g and the apparent Km for acyl CoA dehydrogenase in its reaction with reduced electron-transferring flavoprotein is 1.5 x 10-5%, Phosphate buffer (pH 7) is a non-competitive inhibitor of electron transport with a Ki of 0.23 g, Phosphorylation studies with AMP and ADP, as well as experiments with uncouplers and inhibitors of oxida- tive phosphorylation, suggest that the electron trans- port system alone is unable to catalyze phOSphorylation concomitant with electron transport. Howard L. Brockman, Jr. E(-)-lactic dehydrogenase has been purified to near homogeneity. It is specific for Eflactate and Q,Ef a-hydroxybutyrate as electron donors, but will utilize a variety of acceptors, including dyes, oxygen, and cyto- chrome c. However, pyridine nucleotides will not function as either acceptors or donors. The apparent Km for Eflactate is 0.24 g and the binding of Qflactate is negatively c00perative; the Hill number is 0.65. The flavin cofactor is flavin adenine dinucleotide. Inactivation of the enzyme by orthOphenanthroline, but not by metaphenanthroline, indicates the presence of a divalent metal cofactor. Divalent zinc was superior to either cobalt or manganese in reactivating orthophen- anthroline-inactivated enzyme. The electron-transferring flaVOprotein was puri- fied to 50 per cent homogeneity. The flavin cofactor is either flavin adenine dinucleotide or an unknown flavin cofactor, but not flavin adenine mononucleotide. Acyl CoA dehydrogenase was purified to 30 per cent homogeneity. The oxidized enzyme exists in two colored forms, green and yellow. The green form can be irre- versibly converted to the yellow form by reduction by dithionite and subsequent reoxidation by oxygen. ELECTRON TRANSPORT FLAVOPROTEINS IN THE LACTATE FERMENTATION OF PEPTOSTREPTOCOCCUS ELSDENII BY Howard L} Brockman, Jr. A THESIS Submitted to Michigan State University in partial fulfillment fo the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1971 ACKNOWLEDGMENTS The author wishes to express his appreciation to Dr. W. A. Wood for his guidance and encouragement during the course of this research and in the preparation of this thesis. The author would also like to thank Ernesto and Julio Gallo, whose fine products aided in the rapid preparation of this manuscript. ii VITA Howard L. Brockman, Jr. was born on September 5, 1944 in Olney, Illinois. In 1966 he received a B.A. in Chemistry from DePauw University, Greencastle, Indiana. Following graduation from college he entered the Depart- ment of Biochemistry at Michigan State University, East Lansing, Michigan as a graduate student. After receiving his Ph.D. in biochemistry from Michigan State University, Mr. Brockman will continue his studies under Dr. John Law in the Department of Biochemistry at the University of Chicago, Chicago, Illinois. iii TABLE OF CONTENTS Page ACKNOWLEDGMENTS . . . . . . . . . . . . . ii VITA . . . . . . . . . . . . . . . . . iii LIST OF TABLES . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . vii SECTION I. INTRODUCTION . . . . . . . . . . . . 1 II. LITERATURE REVIEW . . . . . . . . . . 3 Part I. Electron Transport and Phosphorylation in Anaerobes . . . . . . . . . . . 4 Clostridium aminobutyricum . . . . . . . 5 Clostridium glycolicum . . . . . . . . 9 Clostridium kluyveri . . . . . . . . . 12 Clostridium sticklandii . . . . . . . . 19 Streptococcus faecalis . . . . . . . . 23 Streptococcus agalactiae . . . . . . . 25 Peptostreptococcus elsdenii . . . . . 26 Summary of the Electron Transport Systems . . 36 Part II. Electron Transport Enzymes Found in Peptostreptococcus elsdenii . . . . . . 38 Acyl CoA Dehydrogenase . . . . . . . 38 Electron-Transferring Flavoprotein (ETF) . . 41 Df(- )-Lactic Dehydrogenase (LDH) . . . . . 45 III. MATERIALS AND METHODS . . . . . . . . . 51 Bacteriological . . . . . . . . . . . 51 Substrates . . . . . . . . . . . . 52 Fatty Acid Determination . . . . . . . . 53 Protein Determination . . . . . . . . . 54 LDH Assay . . . . . . . . . . . . . 54 Anaerobic Assay Procedure . . . . . . . . 54 Disc Gel Electrophoresis Procedure . . . . . 56 Curve Fitting Procedure . . . . . . . . 56 Chemicals . . . . . . . . . . . . . 57 iv SECTION IV 0 RESULTS 0 O O O I C O C O C C C O 0 Part I. Components and Properties of the Electron Transport System . . . . . . . Role of Electron-Transferring FlaVOprotein in the Reduction of a, B-Unsaturated Acyl CoA by D- Lactate . . . . . . . . . . Role of E1ectron-—Transferring Flavoprotein in the Reduction of Yellow Acyl CoA Dehydrogenase by NADH . . . . . . . . Kinetic PrOperties of the Electron Transport System . . . . . . . . . . . . . Tests for Electron Transport-Mediated Phosphorylation . . . . . . . . . . Part II. Purification and Properties of the Electron Transport Enzymes . . . . . . . D(-)-Lactic Dehydrogenase . . . . . . . Electron-Transferring Flavoprotein . . . . Acyl CoA Dehydrogenase . . . . . . . . V. DISCUSSION 0 O O O O O O O O O O O 0 Part I. Electron Transport System . . . . . Principal Findings of Electron Transport Studies . . . . . . . . . . . Role of Electron Transport in Lactate Metabolism . . . . . . . . . . Relation Between Phosphorylation and Electron Transport in Lactate Metabolism . . . . Control of Electron Transport . . . . . . Physical State of the Electron Transport System . . . . . . . . . . . . . Physiological Relevance of the Reconstituted Electron Transport System . . . . . . Part II. Properties of Electron Transport Enzymes . . . . . . . . . . . . . D(-)-Lactic Dehydrogenase . . . . . . . Electron-Transferring Flavoprotein . . . . Acyl CoA Dehydrogenase . . . . . . . . SUMMARY 0 O O C O O O O O I O O O O O 0 LITERATURE CITED . . . . . . . . . . . . . Page 60 61 62 85 91 106 113 113 144 148 155 155 155 157 160 161 162 162 164 164 165 165 167 168 Table 1. LIST OF TABLES Page Products of the lactate fermentation by Peptostreptococcus elsdenii . . . . . . . 32 The electron-transferring flavoprotein activities of protein bands eluted from a polyacrylamide disc gel . . . . . . . . . . . . . 81 Effect of inhibitors and uncouplers on the rate of electron transfer . . . . . . . . . 108 Phosphorylation and butyrate production by the electron transport system. . . . . . . . 112 Purification of Q(—)-lactic dehydrogenase from Peptostreptococcus elsdenii . . . . . . . 114 Substrate specificity of D(-)-lactic dehydro- genase . . . . . . . . . . . . . . 118 Acceptor Specificity of D(-)-1actic dehydro- genase . . . . . . . . . . . . . . 124 Activators of D(-)-lactic dehydrogenase . . . 140 Purification of electron-transferring flavo- protein from P. elsdenii. . . . . . . . 145 vi Figure l. 2. 10. ll. 12. 13. 14. LIST OF FIGURES Fermentation of y-aminobutyrate by Clostridium aminobutyricum . . . . . . . . . Fermentation of ethylene glycol by Clostridium glycolicum . . . . . . . . . . Fermentation of ethanol and acetate by Clostridium kluyveri . . . . . . . Fermentation of "excess" ethanol in the ethanol- acetate fermentation of Clostridium kluyveri . Electron transport between DPNH and glycine in Clostridium sticklandii . . . . . . Equations for the reactions of the lactate fer- mentation by Peptostreptococcus elsdenii Electron transfer in Peptostreptococcus elsdenii . . . . . . . . . . . Bleaching of green acyl CoA dehydrogenase by several electron donors . . . . . . Bleaching of yellow acyl CoA dehydrogenase by reduced lactic dehydrogenase or NADH . . Butyrate production by the reconstituted electron transport system . . . . . . . . Bleaching of electron-transferring flavoprotein by reduced lactic dehydrogenase or NADH . Absorbance tracings of a polyacrylamide disc of electron—transferring flavoprotein . Dependence of the rate of electron transport the amount of electron-transferring flavo- protein added . . . . . . . . . Dependence of the rate of electron transport D(-)-lactate dehydrogenase concentration vii gel on on Page 10 13 17 21 29 34 64 68 72 75 79 93 96 Figure Page 15. Dependence of the rate of electron transport on acyl CoA dehydrogenase concentration . . . 98 16. Dependence of the rate of electron transport on phosphate buffer concentration . . . . . . 103 17. Absorbance tracings of a polyacrylamide disc gel of a 0.08 mg sample of D(-)-1actic dehydro- genase . . . . . . . . . . . . . . 116 18. Hill plot for D—lactate with D(-)—lactic dehydrogenase . . . . . . . . . . . 121 19. Dependence of the rate of the lactic dehydro- genase assay on ferricyanide concentration . . 125 20. Spectra of oxidized and substrate-reduced D(-)-lactic dehydrogenase . . . . . . . 128 21. Inactivation of D(-)-lactic dehydrogenase by orthOphenanthroline . . . . . . . . . 132 22. Reactivation of D(-)-lactic dehydrogenase by divalent metal ions . . . . . . . . . 136 23. Dependence of the velocity of the lactic dehydrogenase assay on phosphate concentration. 142 24. Absorbance tracings of a 220 ug sample of green acyl CoA dehydrogenase . . . . . . . . 151 25. Lactate metabolism and electron transport in P. elsdenii . . . . . . . . . . . . 158 viii SECTION I INTRODUCTION Anaerobic microorganisms which utilize organic com- pounds grow in the absence of oxygen and hence, most do not possess the usual cytochrome-mediated pathway of electron transport and phosphorylation. To dispose of electrons generated in the oxidation of growth substrates, organic electron acceptors are synthesized. Although the energy from these oxidations is usually conserved by substrate level phosphorylation, in a few cases there is evidence that electron transport mediated phosphorylation occurs. Since such phosphorylation has been found in soluble systems, it does not closely resemble mitochondrial oxida- tive phosphorylation which has been extensively character* ized. Molar growth yields for Peptostreptococcus elsdenii grown on lactate show that 0.5 equivalents of ATP are syn- thesized for every mole of lactate consumed, but the known reactions of lactate metabolism in this organism predict a much lower yield of ATP. One possible explanation for this discrepancy is that electron transport mediated phosphory- lation occurs during lactate metabolism. To set the stage for detailed phosphorylation studies, the electron transport 1 system between D(-)-lactate and acrylyl CoA was character- ized and preliminary tests for electron transport phos- phorylation were performed. SECTION II LITERATURE REVIEW To obtain energy for growth and reproduction, microorganisms must rely upon the utilization of inorganic and organic compounds in exergonic reactions. To support biosynthetic processes, energy from such reactions must be made available to the cell in a chemically useful form. By far the most ubiquitous energy storage compound is adenosine triphosphate, ATP, which stores about 7 Kcal per mole in each of its phosphoanhydride bonds. ATP may be generated by any of four recognized types of reactions: substrate oxidative phosphorylation, in which ATP formation is coupled directly to an oxidative reaction; phosphorclastic systems, which involve the phospholytic splitting of C-S, C-C, or C-N bond; an intra- molecular oxidation-reduction, such as the dehydration of 2—phosphog1ycerate to phosphoenolpyruvate, and by phosphory- lation coupled to electron transport. In the context of this thesis, an electron transport system is considered to be any group of compounds whose function within the cell is to facilitate or enable the transfer of electrons from one point to another. In most aerobic systems electrons generated in oxidative reactions are transported to molecular oxygen with concomitant phosphate esterification. Anaerobic microorganisms, however, must find alternate electron acceptors if they are to carry out any oxidative reactions. Some bacteria utilize inorganic ions as acceptors, whereas a vast array of fermentative bacteria, of which Pepto- streptococcus elsdenii and the Clostridia are examples, employ linked oxidation—reduction reactions involving organic compounds. Among these organisms, the prepon- derance generate ATP by substrate level phosphorylation with no ATP being generated via electron transport. There is reason to believe, however, that in a few fermentative organisms ATP may be generated via electron transport phosphorylation. Because of this intriguing possibility, this review will describe several electron transport systems of the latter type and will then deal specifically with the components of the P. elsdenii lactate—acrylyl CoA system which is the subject of this thesis. Part I. Electron Transport and Phosphorylation in Anaerobes As this investigation is concerned with electron transport, which is thought to be part of the normal energy yielding metabolism of the obligate anaerobe, Peptostreptococcus elsdenii, only similar systems will be discussed. Also, since elucidation of the relationship between electron transport and phosphorylation is an integral part of the study, it will be reviewed here. In many of the systems to be discussed, little, if any- thing, is known of the components of the electron trans- port and phosphorylating systems involved. A. Clostridium aminobutyricum When grown with y-aminobutyrate as its sole carbon and nitrogen source, C. aminobutyricum produces one molecule of acetate, one molecule of butyrate, and two molecules of ammonia for every two molecules of sub- strate fermented (Hardman and Stadtman, 1960). In further studies with dialyzed crude extracts supplemented with y-aminobutyrate, NAD, NADP, CoASH, orthophosphate, 2-mer- captoethanol, and a-ketoglutarate, they were able to show that y-aminobutyrate is transaminated to succinic semi- aldehyde and then reduced to y-hydroxybutyrate (Hardman and Stadtman, 1963). Additional investigations with crude extracts established the existence of the reactions pre- sented in Figure 1 (Hardman and Stadtman, 1963a and 1963b). Following the reductive deamination of y-hydroxybutyrate and its subsequent dehydration and isomerization to crotonyl CoA, the product is oxidized and reduced to yield two mole- cules of acetate and one molecule of butyrate for each two molecules of substrate fermented. Figure 1.--Fermentation of y-aminobutyrate by Clostridium aminobutyricum. 2 NHa-CHZ-CHz-CHz-co; NADH+H+ NH3 NAD+ 2 HO C Hz'CHz‘CHz’CO-é ' AFHZO O T ZCHz-CE=CH-C"3-S-COA 4 NAD” r——*NADH+H+———fi ‘) NADH+H+¢A NAD~——-/ O '0 Y . CH- -c:H2 -c- S- CoA CHg-CHZ-CHz-C-S-COA CoA SHxl CoA——J CH3 -6- --S CoA CH3—CHé-CH2-co; ‘CoA CoA- SH CH3 -ZCO TCH -Cog The energy yield for the process is one molecule of ATP synthesized for each two molecules of y-amino- butyrate fermented. In anaerobes cell growth is pro- portional to the quantity of substrate fermented and, more specifically, to the amount of ATP produced by the fermen- tation (Monod, 1942 and DeMoss et_§1,, 1951). If one employs the widely accepted molar growth yield of ten grams of cells for each mole of ATP produced in an anaerobic fermentation (Bauchop and Elsden, 1960 and Gunsalus and Shuster, 1961) g. aminobutyricum should pro- duce 5 grams of cells per mole of y-aminobutyrate fer- mented. The observed yield is 7.6 grams per mole of y—hydroxybutyrate (Hardman and Stadtman, 1963a). These data indicate an additional Site of phosphorylation. Believing this phosphorylation to be oxidative in nature, the authors studied the effects of classical inhibitors and uncouplers of mitochondrial electron trans- port and oxidative phosphorylation. With dialyzed crude extracts 2.5 x 10_4M dinitrophenol inhibited butyrate production by 60 per cent, while antimycin A, cyanide, 2-heptyl-4-hydroxyquinoline-N-oxide, and sodium amytal were without effect. Also, neither ATP trapping with hexokinase nor ATP-32P exchange could be demonstrated. Although the classical inhibitors were not effective, and dinitrOphenol was only partially effective, the authors concluded that their data indicate the existence of phosphorylation coupled to electron transport (Hardman and Stadtman, 1963b). This is a reasonable conclusion as, a priori, there is no reason to predict that soluble electron transport and phosphorylation in a cytochromeless anaerobe should behave like that found in a mitochondrion. B. Clostridium glycolicum The overall fermentation of ethylene glycol by this organism in anaerobic cultures (Gaston and Stadtman, 1963) is: 2 Ethylene Glycol + Acetate + Ethanol Acetaldehyde is an intermediate in the process, but the exact mechanism of its formation in this species is not known. The reaction does not require a B coenzyme as 12 does the Aerobacter aerogenes enzyme that catalyzes the same reaction (Lee and Abeles, 1963). However, by assum- ing that this organism catalyzes reactions analagous to those known in the ethanol-acetate fermentation of g. kluyveri, the scheme shown in Figure 2 was prOposed (Stadtman, 1966). Because this dismutation of two molecules of ethylene glycol to one molecule of acetate and one mole- cule of ethanol results in the net formation of one thioester bond, growth yields of 5 grams of cells per mole of substrate fermented should be obtained. In reality, the yield is 6.8-8.7 grams. This additional 10 Figure 2.-—Fermentation of ethylene glycol by Clostridium glycolicum. 11 HOCHz-CHon CH3-002 H20 CH;CHO i r 7‘. -NADH+H+ 1 V NAD++ J 9 e CH3-C-S-C0A CH3-CH20H /Pi “(Com/ /ADP ATP/ 12 1.8-3.7 grams of cells represents an additional 0.17-0.35 equivalents of ATP produced, above the 0.5 equivalents expected. AS with C. aminobutyricum these data suggest oxidative phosphorylation coupled to the electron transfer occurring in the simultaneous oxidation and reduction of acetaldehyde (Figure 2). An alternate explanation is to postulate the existence of a substrate level phosphorylation occurring by an unknown mechanism during the fermentation. The AF' for the overall process is —11.07 Kcal per mole of ethylene glycol, more than the -7 Kcal required to synthe- size the phosphodiester bond of ATP (Gaston and Stadtman, 1963). If the additional growth is the result of sub- strate level phosphorylation, it would be expected to be 100 per cent efficient and give growth yields in multiples of 5 grams; instead, intermediate values are obtained. Thus, the phosphorylation is probably oxidative. C. Clostridium kluyveri g. kluyveri ferments acetate and ethanol to butyrate and water (Barker, 1937 and Bornstein and Barker, 1948): Ethanol + Acetate + Butyrate + H20 The reaction sequence for the fermentation is presented in Figure 3 and predicts two linked oxidation-reduction pairs. Note that the energy of the thioester bond 13 Figure 3.--Fermentation of ethanol and acetate by Clostridium kluyveri. 14 CHa-CHZOH CHa-cog O . (CH3-G-S-CoA+R- -S-CoA \- CoA-SH+ R-cog '9 Q . CHB-C-CHz-C-S-COA [—- NADH + H+——fi L +.__.2 NAD 'KHZO v , CH3-CHO CH3-CH=CH-8-S-C0A NADH+H+——fi NAD+‘-——-/ 9 v ‘ Q CH3-C-S-CoA CH3-CH2-CH2-C-S-COA ( R-‘COE K. R-g-S-COA ' - CH3-CH2-CH2-COZ 15 produced in the oxidation of ethanol to acetyl CoA is not available for formation of ATP but must be used for substrate activation. If these equations are correct, the organism should have no available energy for growth. One prOposed solution to this dilemma is electron transport phosphorylation. Since AF' for the fermentation is -1l.7 Kcal per mole, sufficient energy exists for the formation of a phosphate anhydride bond (Bornstein and Barker, 1948). A report of ATP formation coupled to the reduction of crotonyl-N-acetylthioethanolamine to butyryl- N—acetylthioethanolamine by hydrogen in crude extracts of g. kluyveri appeared in the literature (Shuster and Gunsalus, 1958) but was never substantiated (Brown and Stadtman, 1958). More recently an alternate explanation has been prOposed which does not require electron transport phosphorylation (Thauer et_al., 1968). Although past studies indicated that during the fermentation (1) ethanol consumption exceeds acetate utilization, (2) pro- tons are generated, and (3) hydrogen is produced, these observations were ignored. Careful measurements of fermentation products relative to growth yield revealed that for every "excess" molecule of ethanol consumed in the fermentation, one molecule of hydrogen and one half molecule of protons were produced. In addition, growth measurements indicated that each mole of "excess" ethanol 16 fermented yielded 5 grams of cells or one half mole of ATP. Thus, 9. kluyveri grows not on the ethanol-acetate fermentation but on the fermentation of the "excess" ethanol. Figure 4 Shows the overall fermentation as Thauer et_al. envision it. Section A describes the net ethanol fermentation and explains the generation of hydrogen gas. The final product of these reactions is butyryl CoA which provides the equivalent of acetyl CoA necessary for the ethanol-acetate fermentation shown in Section B. These reactions are identical to those pro- posed by Bornstein and Barker (1948) with the exception that the thioester bond generated is not required for acetate activation at the beginning of the sequence. It is now provided by the reactions of Section A. Section C describes the reactions by which the net thioester bond energy is converted to ATP. These reactions are also similar to those previously described except that a net proton is generated. In the overall fermentation net proton production occurs because ethanol is oxidized to an acid, acetic acid. Are the reactions of Section B necessary at all? It would appear that the thioester bond generated in Section A could be utilized directly in Section C to generate ATP. The authors state that neither the net fermentation of ethanol to acetate (AF' = +9.6 Kcal/mole) 17 Figure 4.-—Fermentation of "excess" ethanol in the ethanol-acetate fermentation of Clostridium kluyveri. 18 2E1 OH *- ZACCOA r w ‘1 ZHg‘wLo“ 4NA0H+4H+\CoA—J A BU COA " J AcAc CoA <1 Ac” / Bu'/ "AcCoA f- ' BuCoA '— EtOH /Ac" 8 . '1 L. _ Bu .1 N BUCOA‘fi ACCOA ETOH BU’K — AC- CoA-SH 19 nor to butyrate (AF' = +1.1 Kcal/mole) is an energeti- cally favorable process. However, when butyrate formation is coupled to the reactions of Section B, the overall free energy change is —8.6 Kcal per mole of butyrate. This explanation makes it reasonable to expect the reactions of Section B to cycle once for every two molecules of ethanol consumed in Section A reactions, but in reality the ethanol-acetate fermentation of Section B cycles many times for each pair of ethanol molecules fermented. Thus, while these authors have eliminated the need to postulate oxidative phosphorylation to explain the growth of g. kluyveri, they fail to explain why the cell wastes so much substrate in the reactions of Section B. D. Clostridium_stick1andii The most extensive study of anaerobic electron transport has been made with this organism. In cultures it can grow on glycine as its sole carbon and nitrogen source by coupling the oxidative deamination of one molecule of glycine to the reductive deamination of another as shown below (Nisman, 1954): . + GlyCIne + H20 + DPN + Glycolate + NH + + DPNH + H+ 4 . + + + Glyc1ne + DPNH + H + Acetate + NH4 + DPN 2 Glycine + H20 + Acetate + Glycolate + 2 NH4+ 20 The most interesting aspect of this system is that net ATP formation can be demonstrated concomitant with the reductive deamination of glycine (Stadtman and Elliott, 1956 and Stadtman et_al., 1958). Using 1,3-dimercapto- propanol as an electron donor, glycine reductase activity was resolved into two protein components and showed an absolute dependence upon orthophosphate and an adenine nucleotide. For every two equivalents of 1,3-dimercapto- propanol oxidized, one phosphodiester bond is formed (Stadtman e£_al., 1958). To further substantiate that the observed phosphorylation was electron transport mediated, instead of occurring at the substrate level, Dr. Stadtman and her coworkers attempted to trap any high energy acetyl intermediate with hydroxylamine. They were unsuccessful. Furthermore, phosphoramidate can be excluded as an intermediate because it did not give rise to ATP when incubated with the enzyme system containing ADP. Further fractionation of the crude extracts and the coupling of glycine reductase to an NADH generating system resulted in the elucidation of the electron trans- port system presented in Figure 5 (Stadtman, 1962). Initially a diaphorase catalyzes the reduction of FMN by NADH. Using reduced FMN as a donor, a mercaptan dehydrogenase then catalyzes the reduction of a dimercaptan protein in a reaction which is inhibited by 5 x lO-SM . 2:: . :3; :— :T:.:.....< 17...... _.___._ L... _.___. ?:_.__.. .1: ._.l:_ .3 __:_..1.__::.:_ .1: __._3 T..L:TC~L... .3. Ti? fist—V .__.:_:._.._... .L .._. \1_ _.:__:_ .. :2... 2.5.7.. :73 ..:;.._ 3:: . 4 .:_._:._ .7. .— 2 :_::.___.:..._L :_ ..___..\.__. 7.... 3:; 2.7.37; ....;.;.:.... .21.... 22 .33 Mb? w: 0% 1. $4 1. CPLSQ S l l _ «33 user 3 WAQVXENNQ \w\§N>QQb S \ mmemSEh§me mmEQQQ‘EQ IN\_\$<\§Q QNQQQNnt IAIINSQNQN ERNQGGAKNxAQQ A, QSQSEQQ >§Quombo memHHQDQCD ¥¥¥ .mOOSUOHQ coflumucmEuom mo mmHoEE pm>ummbo mzu moswoum on pmuflsvmu mpmuoma mo mmHoEE mo umbEsc may ma musmflm mane xx kommHv .HM #0 CQUWHM EOHW UGQHMfiQO ®H®3 MUMU COHflM¥Q®EH®W Um>H®mQO ¥ 32 mm m.oon o.m m.oau mozmmmmmHo %.o ««¥.om km.hm «0.0HH Qm>mmmmo o.mm m.oan S.Hm .4G.OHH emz m.» u .o .o .G.m ammouwxm 8.0 . m.m a .o 4N.H mumoummo o.oeu 0.0m: 0.0m Po.om mumumam> m.mo m.ma G.Hm .m.ma oumumusm G.Hm- m.mau o .m.ma mumcoHQOLm m.mm m.ma m.mH .m.ma mumnmoa m mga Noo _ mmHoEE uoscoum mmHoEE coflumucmaumm nonpoum UTDOHUTHE .flflcmpmam msoooooumouumoumwm ma coflumucmEHmm evapomH map mo muospoumul.a mamme 33 Until the fermentation data are clarified, we must consider the possible existence of electron transport linked phosphorylation in addition to phosphorylation at the substrate level. The calculated E5 for the lactate to pyruvate:crotony1 CoA to Butyryl CoA couple is 0.38 volts, enough energy to make 2 equivalents of ATP. The scheme in Figure 7 presents an interpretation of the electron transport data available for S. elsdenii. The prOperties of the acyl CoA dehydrogenase and the D-lactic dehydrogenase will be presented in a later section of this review. Fraction II is a partially purified diaphorase fraction discovered in the purifica- tion of the acyl CoA dehydrogenase. It stimulates the transfer of electrons from butyryl CoA to dichlorOphenol- indophenol (DCPIP) and allows transfer of electrons from DPNH to crotonyl CoA as catalyzed by acyl CoA dehydrogenase. This fraction is also reported to stimulate the D-lactic- SSSLS assay for lactic dehydrogenase (Baldwin and Milligan, 1964). The source of electrons to reduce NAD was investi- gated with a spectrophotometric assay for transfer from molecular hydrogen via hydrogenase NAD. When ferridoxin is removed from crude extracts, the ability to carry out this reaction is lost, but with the addition of partially purified ferridoxin from S. elsdenii or crystalline ferri- doxin from Clostridium pasturianum, this activity is 34 .uozmflansm map Ucm nocpsm mzu mo coflmmflEqu mzp :uflz Avomav cmmflaaflz paw cflzpamm Scum couscoummu ma musmflw mace .Hflcmpmam m300000bmmuumoummm CH ummmcwuu couuooamnl.n wusmflm 35 <00|4>K>H3m doou4>ZOFomo M . 342391350 <8..;o< mo : 295$: mo «8.45985 <8.._>.;mo< \‘II/ lllllllllllllll J +zao zzao < mmqmoxaso mqumoomeimzqm» I ................. 1 ~11 mqumoomo»: / ~19. E e 2mpm>m mqumwomo>xmo + o a Li .1 «8 .1 £qu Jhmfioomoxamoza “1.235.. 0.54.. .\._\..>G qde z. muumzqm» 288qu m ._.<._.U< 4 36 restored. Thus, ferridoxin appears to be an obligatory intermediate in this electron transfer. Since ferridoxin has been shown to be the immediate electron acceptor for the phosphorclastic reaction, it was concluded that electrons from the oxidation of pyruvate will reduce DPN+ using reduced ferridoxin:DPN oxidoreductase. The pathway of electrons from S—lactic dehydro- genase to the acyl CoA dehydrogenase was not elucidated. Therefore, this investigation was undertaken to determine the path of electrons from S—lactic-dehydrogenase to acyl CoA dehydrogenase and to test for concomitant phosphoryla- tion. H. Summary of the Electron Transport Systems S. glycolium and S. aminobutyricum appear to exhibit electron transport phosphorylation solely on the basis of molar growth yields. Growth yields are probably a good indicator of ATP production, but they can be affected by the composition of the media and the nature of the substrate. Low levels of phosphorylation have been observed in crude extracts of S. agalactiae, but because of the low P/O ratios and the crude nature of the system, more detailed studies should be made. In S. fecalis and S. sticklandii both electron transport and phosphorylation have been convincingly 37 demonstrated. In the presence of oxygen S. faecalis phosphorylates ADP during the oxidation of NADH by either the NADH oxidase or the NADH peroxidase found in the organism. Clostridium sticklandii contains a much more involved electron transport system than S. faecalis. The components of this system have been resolved and partially purified, and the site of phosphorylation has been definitely located at the terminus of the chain. It is interesting that in both S. faecalis and S. sticklandii a quinone appears to be involved in the phosphorylation site. Clostrudium kluyveri apparently does not exhibit electron tranSport phosphorylation. With E. elsdenii the evidence for phosphorylation is indirect and warrants further investigation. Since the phosphorylation site, if it exists, is reasoned to be located between S-lactic dehydrogenase and acyl CoA dehydrogenase, electron transport between these enzymes was investigated. The remainder of this review will be devoted to describing the prOperties of enzymes similar to those found in the E. elsdenii electron transport system during the course of this research. \ 38 Part II. Electron Transport Enzymes Found in Peptostreptococcus elsdeniI A. Acyl CoA Dehydrogenase This enzyme has been identified in pig liver (Crane §E_Elx' 1956 and Hauge EE_3£°' 1956), bovine heart (Beinert, 1962), sheep liver (Seubert and Lynen, 1953), monkey liver (Hoskins, 1966), leptospira (Baseman and Cox, 1969), mycobacteria (Gelbard and Goldman, 1961), and S. elsdenii (Baldwin and Milligan, 1964). The enzymes of pig, cattle and sheep have been obtained in high purity. In mammalian systems and mycobacteria two and three forms of acyl CoA dehydrogenase have been identified. These differ in their specificities for the acyl CoA substrate (Crane §E_2l,, 1956; Hauge §E_§£., 1956; Beinert, 1962; Green g£_gl., 1954; Seubert and Lynen, 1953; and Gelbard and Goldman, 1961. The butyryl CoA dehydrogenase most closely resembles the enzyme found in S. elsdenii and will be referred to as ACD in the following discussion. ACD is found in the mitochondria of higher protists and functions in the B-oxidation of fatty acids. Electrons derived from butyryl CoA enter the mitochondrial electron transport chain via an electron-transferring flaVOprotein (ETF) where they ultimately reduce oxygen to water (Beinert, 1963a). The physiological role of the S. elsdenii ACD is to convert acrylyl, crotonyl, pentenoyl, 39 and hexenoyl CoA to their respective saturated derivatives (Baldwin and Milligan, 1964). Thus in S. elsdenii ACD serves as the terminus of the electron transport chain, rather than its origin. The most unusual characteristic of ACD is that it exists in two colored forms, green and yellow (Green gE_§£., 1954). As isolated from mitochondria it is in the green form. Upon reduction by either substrate or dithionite and subsequent reoxidation, or by addition of crotonyl CoA, green ACD is irreversibly converted to the yellow form. The yellow pig liver enzyme has absorption maxima of 275 mp, 362 mu, and 445 mu with a ratio of absorptivities of E275:E362:E445 = 9.42:0.92:1.0, whereas the green form has maxima at 266, 357'.426' and 710 mu 'with a ratio of absorptivities of 8.1:0.83:l.00:0.35 (Steyn-Parve and Beinert, 1958). As yet the origin of the long-wavelength maximum is unexplained but may involve an unusual acyl CoA bound to the enzyme (Engle, 1970). Kinetically, the green and yellow forms exhibit almost identical Michaelis constants (Km) for butyryl CoA, i.e., 17-20 x 10-7%, but the maximal velocity (Vmax) for the yellow form is twice as great. The enzyme was assayed by following its ability to catalyze the reduction of DCPIP by butyryl CoA in the presence of excess ETF. For the yellow form, Km does not change when 4O PMS and cytochrome c are substituted for ETF and DCPIP, respectively, but Vmax doubles. Also, yellow ACD is able to find 0.5 to 1.0 moles of substrate per mole of flavin with a Km of less than lO—log (Hauge, 1956). These data suggest that ACD possesses three different sites which do not overlap, an ETF acceptor site, a low Km acyl CoA site and a higher Km acyl CoA site. Propionyl CoA and isobutyryl CoA are dehydrogenated at a rate less than one-fiftieth that of butyryl CoA while a-methylbutyryl CoA is attacked at one-tenth the rate, Isovaleryl CoA is also a substrate (Beinert, 1962), and crotonyl CoA is a strong competitive inhibitor with a KI of 10-7S (Hauge, 1956). Mammalian ACD is very specific for its electron acceptors. DCPIP, methylene blue and cytochrome c are essentially inactive, ferricyanide is slightly active, and PMS shows considerable activity. Interactions with all of these acceptors are enabled or facilitated by the physiological acceptor, ETF (Crane and Beinert, 1954). The Km for ETF in the butyryl CoA-ETF-DCPIP assay system was of the order of lO-lOS indicating a strong binding between ACD and ETF (Hauge, 1956). The molecular weight of ACD is between 120,000 and 220,000, but is most probably 220,000, and there are two molecules of the cofactor, FAD, for each enzyme molecule. Metals are found only in trace amounts (Crane et a1., 1956). 41 The ACD from S. elsdenii differs from mitochondrial ACD in that it readily reduces cytochrome c, DCPIP, methylene blue, and ferricyanide with butyryl CoA as a substrate. When propionyl CoA is the substrate, the ratios of prOpionyl CoA activity to butyryl CoA activity for the above dyes, except for ferricyanide, are 0.175, 0.71, 0.0, and 0.0, respectively. In the reverse direction acrylyl CoA is reduced to prOpionyl CoA by leucosaffanine (Baldwin, 1962). This enzyme has been purified 23 fold (Baldwin and Milligan, 1964). B. ,Electron-Transferring FIaVOproteinYETF) ETF was originally identified as an obligatory intermediate in the reduction of dyes by acyl CoA dehydrogenase and their reduced substrates. This function has been shown in pig and bovine liver (Crane and Beinert, 1956; Crane 32.3i'r 1956; and Hauge 35:31., 1956), sheep liver and bovine heart (Beinert, 1963), and mycobacteria (Gelbard and Goldman, 1961). A second flaVOprotein, sarcosine dehydrogenase, has also been shown to have an ETF requirement (Hoskins and Mackenzie, 1961; and Mell and Huennekens, 1960) and has been purified from rat liver (Frisell EE_El°I 1962; and Cronin 33431., 1967) and primate liver (Hoskins and Bjur, 1965). Since the discovery of these enzymes, it has been shown that ETF from rat liver sarcosine dehydrogenase system could 42 replace pig liver ETF in the acyl CoA dehydrogenase system and that pig liver ETF functions with the rat liver sarcosine dehydrogenase (Beinert and Frisell, 1962). In mycobacteria where there is a butyryl CoA dehydrogenase (ACD) and a long chain acyl CoA dehydrogenase, animal ETF will replace mycobacterial ETF only with the long chain dehydrogenase (Gelbard and Goldman, 1961). As the two mammalian ETF's are found in mitochondria in the same tissues and appear interchangeable, it is likely that they are the same protein. The purest ETF has been obtained from primate liver and was 140 fold purified with 15 per cent recovery (Frisell §E_El3' 1966). In the acyl CoA dehydrogenase system of pig liver, the enzyme was close to homogeneous but measured only 40 fold purified with a recovery of 0.75 per cent (Crane and Beinert, 1956). The difference in recoveries could merely reflect a difference in isolation techniques or the acyl CoA dehydrogenase may have a more specific or easily damaged binding site for ETF than does the primate liver sarcosine dehydrogenase. The half life of the primate enzyme when stored at -25°C is approximately three weeks, and activity is not restored by the addition of FAD, cysteine, or glutathione (Hoskins and Bjur, 1965). However, addition of FAD to the sarcosine system, which is composed of sarcosine, sarcosine dehydrogenase, ETF, and DCPIP or 43 mitochondrial electron transport particles, stimulates activity 30-50 per cent (Frisell §E_S£., 1966) in rat liver and 8 per cent (Hoskins and Bjur, 1965) in primate liver. AS this phenomenon is observed at all stages of purification, it is not a recovery of activity but a stimulation by FAD. The molecular weight of pig liver ETF has been estimated at 30,000 to 70,000 based on studies in the ultracentrifuge, and at 80,000 based on flavin content (Crane and Beinert, 1956). The flavin cofactor is released in acidic ammonium sulfate, and, in the case of pig liver, ETF activity is partially restored by the addition of FAD. In all cases the cofactor was identified as FAD (Crane and Beinert, 1956 and Frisell §£_§l3, 1966). The pig liver ETF exhibits absorption maxima at 270, 375, 437.5, and 460 mu (Crane and Beinert, 1956) whereas the maxima for the primate enzyme are 375, 410, and a broad shoulder at 440-480 mu (Hoskins and Bjur, 1965). Although rat liver and pig liver ETF absorb similarly at 270 mu and in the 450-460 mu region, the latter has an increased adsorption at 410 mu and the peaks at 375 and 437.5 mu are missing or masked (Frisell et a1., 1966). These differences in spectra lead one to question the common identity of these enzymes, however, they may only reflect differences in purity or isolation procedures and 44 conditions. All ETF's are fluorescent, a trait not exhibited by many flav0proteins (Beinert, 1963). The spectra of all ETF's have been shown to be partially reduced upon the addition of reduced substrate and catalytic amounts of the apprOpriate dehydrogenase. Dehydrogenase substrate alone has no effect (Crane and Beinert, 1956; Beinert and Frisell, 1962; and Hoskins and Bjur, 1965). The wide variation observed in the extent of reduction may reflect different degrees of purification and supports the above contention that spectral discrepancies may result from contaminating material. If one assays ETF with a fixed level of acceptor, saturating sarcosine or acyl CoA, and varying levels of ACD, saturation kinetics are obtained. Pig liver ACD gives a Km of about 2 x 10—7g (Hauge, 1956) as compared with 1-2 x 10.6% for the rat liver sarcosine dehydrogen- ase. In each case there is a high affinity for reduced dehydrogenase; however, it is not high enough to suggest the formation of a stable protein-protein complex. Fluoresence binding studies also support this observa- tion (Beinert, 1963). Although ETF is unique with regard to its donor specificity, it can react with a wide variety of acceptors, including DCPIP, ferricyanide, PMS, cytochrome c, and electron transport particles (Crane and Beinert, 1956; 45 and Mell and Huennekens, 1960). Reactivity toward cytochrome c is variable from one preparation to another (Crane and Beinert, 1956). The reactivity, or lack of it, of ETF with pyridine nucleotides has long been discussed in the literature. The purest preparations always contained diaphorase activity, but it was generally assumed to be a contaminant since it had properties identical with the "DT diaphorase" Ernster EE_§l" 1962) from rat liver cytOplasm. That is, it showed equal activity toward diphosphopyridine nucleotide and triphosphopyridine nucleotide, hence the name "DT diaphorase," and was 50 per cent inhibited by dicumarol (Crane and Beinert, 1956; Hoskins and Bjur, 1965; and Frisell EE_E£" 1962). Subsequently, two diaphorases were separated from rat liver ETF on DEAE cellulose, one a DPNH diaphorase and the other a DT type diaphorase as described above. ETF itself showed no diaphorase activity (Frisell §£_Elx' 1966). C. D-(-)-Lactic Dehydrogenase (LDID Pyridine nucleotide independent S—(-)-lactic dehydrogenases have been characterized from a variety of sources and are similar in many respects. The following discussion describes some of the more thoroughly studied enzymes and some bacterial enzymes which have not been 46 characterized in as much detail. The cytochrome- containing lactic dehydrogenase, cytochrome b2, from yeast, although widely studied, differs significantly from S. elsdenii LDH and does not merit discussion here. S—d-hydroxy acid dehydrogenases are very similar to LDH, differing only in substrate specificity, but they, too, will be omitted. The LDH of aerobic yeast has been purified to near homogeneity (Gregolin and Singer, 1963). When added to osmotically disrupted mitochondria it functions as a lactic oxidase system coupling electrons from lactate to the terminal electron transport chain. It has a molecular weight of 100,000, contains two moles of FAD per mole of enzyme and two to three gram atoms of Zn+2 per mole of FAD. The purified enzyme is quite labile. The FAD may be dissociated by acid ammonium sulfate without loss of zinc, and the addition of FAD restores activity (Gregolin and D'Alberton, 1964; Roy, 1964; and Gregolin EE_3£., 1964). Metal chelating agents like EDTA and oxalate cause rapid inactivation which can be reversed by diluting the enzyme., Inactivation by c- phenanthroline is relatively slower and is not reversed by either dilution or rapid passage through Sephadex G-25. An inactive enzyme-o-phenanthroline complex is formed which can be restored to active enzyme by dialysis, high temperature, or addition of Zn+2. 47 Protection against this inactivation by substrate is competitive, suggesting that the binding site for sub- strate is near or at the zinc binding site (Cremona and Singer, 1964). LDH utilizes only S—lactate and S-d-hydroxybuty- rate as substrates with Km values of 2.85 x 10_4S and 1.4 x lO-BS, respectively. Pyruvate is not a substrate and cannot function in the reverse reaction. Cytochrome c and PMS were the only active electron acceptors. The Km for cytochrome c is variable, depending on its source and its method of isolation, and estimated to be no higher than 6 x 10-6 .S (Gregolin and Singer, 1962). Nygaare (1961) isolated a similar enzyme from yeast which in the presence of polyvalent cations such as protamine increases its reactivity toward ferricyanide and decreases that toward cytochrome c. From anaerobically grown yeast another stpecific LDH has been isolated which will utilize-DCPIP and other artificial electron acceptors which are not reduced by the aerobic enzyme. Acid ammonium sulfate will resolve the FAD, but to regain activity, FAD and Zn+2 must be added simultaneously (Iwatsubo and Curdel, 1961). Zinc was shown to reactivate better than cobalt or other metals, and cells grown in a zinc-free medium supplemented with cobalt gave an LDH with a higher Km for Sflactate, but a greater total activity. Cells grown with no metal 48 Showed little LDH activity until Zn+2 was added to the enzyme preparation (Curdel, 1966). The substrate levels necessary for 50 per cent protection from inactivation by chelators was approximately 0.003 x Km (Stachiewicz §E_E£°' 1961). The bacterium, Lactobacillus arabinosis, contains S— and S—specific pyridine nucleotide dependent and Sf and S—specific pyridine nucleotide independent dehydro- genases. The S—specific pyridine nucleotide independent enzyme has been purified 560 fold, but has not been extensively characterized because it is unstable in the purified form. The only substrate found was S—lactate which has a Km of 2.3 x 10_3g, Oxalate is a strong com- petitive inhibitor with a K of 1.1 x lO-SS, whereas I S—lactate competitively inhibits with a K of only 8.6 I x 10-2%. Only dyes with an E ' of +0.1 volts or greater, 0 such as DCPIP, will serve as electron acceptors (Snoswell, 1966). Propionibacterium pentosaceum Sflactic dehydro- genase preparations utilize Sflactate three times faster than the S isomer. Since the best purification obtained was fifteen fold, it is likely that either both Sf and S—lactic dehydrogenases are present or that a lactic racemase was present. Lactate was the only substrate utilized and its Km is 5.1 x 10-5%. PMS, DCPIP, methylene blue, and 1:2-napthaquinone-4-sulphonate 49 function as electron acceptors. The Km for PMS is 6.4 x 10—4S, whereas the assay rates with the other dyes were independent of their concentrations in the range measured. Ammonium sulfate was shown to increase the activity of 2.6 fold at 0.5g. This activation is primarily due to the ammonium ion and occurs, at least in part, by lowering the Km for Selactate. In attempts to release a flavin cofactor the enzyme was treated with perchloric acid, and when redissolved the resulting white precipitate retained as much as 60 per cent of its original activity. Contrary to the other LDH'S no flavin cofactor could be identified and the enzyme is stable in 0.01 M_EDTA (Molinari and Lara, 1960). Pyridine nucleotide independent LDH's have also been reported in Aerobacter aerogenes (Pascal, 1966), Lactobacillus casei (Mizushima and Kitahara, 1962), and Butyribacterium rettgeri (Wittenberger and Haaf, 1966). This latter enzyme is stimulated more than two fold by sulfate ion in the assay with DCPIP. All these enzymes are relatively specific for S—lactate and react with a variety of artificial electron acceptors. In S. elsdenii a S—specific lactic dehydrogenase has been purified 25 fold. It is pyridine nucleotide independent and will transfer electrons to DCPIP (Baldwin, 1962; and Baldwin and Milligan, 1964). A diaphorase 50 fraction from the same organism reportedly stimulates the rate of dye reduction. Attempts to reduce pyruvate with DPNH or leucosaffranine were unsuccessful. SECTION III MATERIALS AND METHODS A. Bacteriological A culture of Peptostreptococcus elsdenii (ATCC no. 17752; strain B159) was the kind gift of Professor M. Bryant. The organism was maintained in stock culture as described by Elsden and Lewis (1953), except that 0.001 per cent resazurin was added as a redox indicator; a red color of this indicator indicates a lack of anaerobiosis (Schneider, 1969). For long term storage cultures were frozen or ly0pholyzed and frozen. From time to time Gram stains were made to determine the purity of the culture (Conn, 1957) and from these it was determiend that the cells Show a variability in size. Young cultures exhibit small cells that tend to be Gram positive, whereas cultures in the stationary phase contain primarily large, Gram negative cells. Contaminants were removed by plating on thioglycollate medium with 2 per cent sodium lactate (60% solution) in ordinary plates in a dessicator under nitrogen. Plates are prepared by cooling the medium to 45°C, followed by innoculation and immediate pouring into plates under nitrogen. S. elsdenii was grown in deep culture on corn steep liquor and lactic acid as described 51 52 by Ladd and Walker (1959), except that trace metals were added as described for a defined medium (Allison §E_E£" 1966, and Bryant and Robinson, 1961). For large cultures grown in 55 gallon drums, sodium dithionite was present. only in the innoculum. Extracts were prepared by suspending the frozen cells in an amount of distilled water in milliliters equal to the cell weight in grams. The cells were dis- rupted either by sonication at 10 Kc for 20 minutes or by two passages through a Manton—Gaulin homogenizer at 1600 psi. The latter method was used only when more than 100 g of cells were being disrupted. Both procedures were carried out at 0-4°C and DNAase was added to homogenized extracts to lower the viscosity. The extracts were centrifuged at 18,000 x g to remove cell debris and were used immediately. B. Substrates Butyryl and crotonyl CoA were prepared by the anhydride method of Stadtman (1957). Complete reaction of the SH groups was ascertained with the nitroprusside spot test (Stadtman, 1957) and if necessary for the com- pletion of the reaction additional acid anhydride was added. The reactions were carried out at 4°C and were monitered with a pH meter. The pH was maintained at 7.0-7.5 by the addition of l S KOH and when the reaction 53 was complete the pH was adjusted to 6. Although crotonyl CoA is normally prepared from crotonic acid by the mixed anhydride method (Stadtman, 1957), it was found that it could be directly synthesized using the anhydride method with crotonic anhydride. The yield of 30 per cent for this reaction is comparable to that obtained by Stadtman with the mixed anhydride method. Purity of crotonyl CoA was determined Spectrophotometrically by the method of Stern (1955) as modified by Schneider (1969). C. Fatty Acid Determination Butyrate, crotonate and other acids were determined quantitatively by gas—liquid chromatography. The apparatus was a dual column Packard Instrument Company gas chromato- graph equipped with dual hydrogen flame detectors and a dual-pen Texas Instrument Company recorder. The 2 mm- diameter columns were packed with 10 per cent FFAP on Chromosorb W, acid washed DMCS, 80/100, which is produced by Wilkins Instrument Company. The temperatures used were column 130, inlet 155, outlet 200 and dectector 165°C. The sensitivity of the system permits determination of 1 nmole of any volatile fatty acid. 10 ul samples of incubation mixture were mixed with 1 ul of concentrated H2804 as an internal standard and eliminated errors due to and 2 ul of 0.01 S sodium isobutyrate which served injection. The peaks were integrated by multiplying height by width at half height. 54 D. Protein Determination Protein was measured spectrOphotometrically using absorbance at 260 mu and 280 mu in conjunction with a nomograph distributed by the California Corporation for Biochemical Research. The nomograph is based on extinc- tions determined for nucleic acid and enolase by Warburg and Christian (1942). E. LDH Assay S(-)—lactic dehydrogenase was determined spectro— photometrically by following the disappearance of ferri- cyanide at 420 mu upon reduction by Sflactate. The pro- cedure was a modification of that described by Symons and Burgoyne (1967). Each assay consisted of 160 umoles of Selactate, 0.5 umoles of potassium ferricyanide, 10 umoles of potassium phosphate buffer (pH 7) and a sample of enzyme in a total volume of 0.2 ml. Assays were normally performed at 24°C. The extinction of ferri- cyanide was taken as 1040 liter per mole-centimeter relative to ferricyanide and to conform with IUPAC recommendations, a unit of LDH activity is defined as that amount of enzyme which forms 1 umole of ferro- cyanide per minute. F. Anaerobic Assay Procedure Since oxygen "short circuits" electron transport processes in S. elsdenii, it was necessary to remove it 55 from samples when performing experiments with the electron transport system. To equip a cuvette for anaerobiosis, a 3/4" long piece of 1/2" I.D. gum rubber tubing is fitted over the tOp 3/8" of a standard square microcuvette and secured with two lengths of 18 gauge copper wire, which was twisted and tightened with pliers. Following addition of the solution to be deaerated, a septum, which is the upper half of the stOpper from a B-D "vacutainer," is fitted into the rubber tubing and a 20 gauge syringe needle part of a deaeration manifold is inserted through the septum into the airspace above the sample. The mani- fold is an arrangement of tubing and stopcocks which allows one to alternatively evacuate the sample with a water aspirator (16 mm Hg) and to flush it with argon. This procedure is essentially that used for performing assays in a standard Thurnberg cuvette except that the equipment has been reduced in size; the sample can be as small as 0.17 ml. During deaeration the bottom half of the cuvette can be immersed in a water bath to equilibrate the sample to a defined temperature. With such small samples in a narrow cuvette, losses through foaming and bumping are inevitable, but with practice one can minimize these losses. When the assay is complete, the septum is removed and the cuvette is cleaned and dried with the rubber Sleeve still attached. This procedure was used routinely in the course of this research and is referred to as "deaeration." 56 G. Disc Gel ElectrOphoresis Procedure Standard 7% per cent polyacrylamide disc gels were prepared in quartz tubes according to the method of Davis (1964) except that spacer and stacking gels were not used. The gels were pre-electrophoresed and electro- phoresed in tris—barbital buffer (pH 7) at 5 milliamperes per gel at 5°C. Tracking dye was used with a blank gel to mark the front. Following electrophoresis, the gels were directly scanned in a Gilford Instrument Company recording spectrophotometer equipped with a Gilford Instrument Company linear transporter. H. Curve Fitting Procedure To eliminate human bias in drawing straight lines, computer curve fitting was employed. The basic program was POLFIT*** from Applied Computer Time Share, Inc., Detroit, Michigan, which was modified for use on the Burroughs 5500 computer of the Philco—Ford Corporation Time Sharing System, Detroit, Michigan. Routines were added which allowed the operator to (1) perform arithmetic operations on the dependent and independent variable, (2) list the data to be fitted, (3) plot the data to be fitted on the teletype console, and (4) delete points from the beginning and end of the data array. The curve fitting part of the program employed the method of orthogonal polynomials. The program will fit any array 57 of points to a polynomial of the form y = a f bx / cx2 f . . . l x11 where the highest power of x is selected by the operator. For example, to fit data to a straight line y = a f bx would be employed. For each polynomial to which data are fitted, the coefficients and index of determinations, a measure of how well the points fit the line, are outputed. I. Chemicals Most of the chemicals used in the course of this work are listed below according to their sources. Corn steep liquor was a gift of the A. E. Staley Manufacturing Company, Decatur, Illinois. Chemical Commercial Source bovine serum albumin Sigma Chemical Co. carbamyl phosphate cytochrome c flavin adenine dinucleotide flavin mononucleotide fructose diphosphate, Na a—glycerol phosphate, Na orthophenanthroline phosphoenolpyruvate, Na 3-phosphog1ycerate, Na trizma base uridine monOphosphate, Na uridine diphosphoglucose, Na dichlorOphenol indophenol K and K Rare Chem. Co. D L-glyceric acid " I D,L-a-hydroxybutyrate, Na " D,L-d-hydroxycaproic acid " D,L—a-hydroxyisobutyric acid " D,L-d-hydroxyvaleric acid " L-malic acid " D—malic acid " 58 Chemical 3-acetylpyridine, DPN adenosine diphosphate adenosine monOphosphate Coenzyme A, Li cytosine monophosphate cytosine triphosphate guanosine triphosphate inosine monophosphate NAD, NADH and NADP barbital glucose potassium ferricyanide potassium phosphate, mono- and 3',5'-adenosine monophosphate, adenosine triphosphate N,N'-bipyridyl hydroxypyruvic acid phosphate S—lactate, Li, grade A Sflactate, Li, grade A acrylamide ammonium persulfate N,N'-methy1enebisacrylamide N,N,N',N'-tetramethylenediamine butyric anhydride crotonic anhydride bactotryptone yeast extract 60% sodium lactate EDTA egg albumin thiamine perphosphate sodium pyrophosphate l4C-adenosinemonOphosphate, UL adenosine-8-l4C-5'-diphosphate 14 adenosine-8- C-5'-triphosphate Commercial Source P—L Biochemicals, Inc. Mallinkrodt dibasic " Na Calbiochem Canalco Eastman Organic Chemicals Difco Laboratories Phanstiehl Lab., Inc. J.T. Baker Chem. Co. Pentex, Inc. General Biochemicals Fisher Science Company Int. Chem. and Nuc. (ICN) Corp. Calatomic New England Nuclear Corp. 59 Chemical Commercial Source DEAE cellulose (DE-52) Whatman Sephadex G-25 Pharmacia Fine Chem., Inc. Sephadex G-100 " Sephadex C-25 " Sepahdex A-50 " SECTION IV RESULTS The experiments presented in this section are divided into two groups: (I) those which describe the electron transport system between S-lactate or NADH and a,8-unsaturated acyl CoA, and (II) those which describe the partial purification and some properties of each component of the system. The eXperiments of the first part will: (A) demonstrate that electron transport between S—lactate and d,8-unsaturated acyl CoA is catalyzed by a minimum of three flav0proteins, a pyridine- nucleotide independent S—lactic dehydrogenase and suggest that these components are both necessary and sufficient to catalyze this electron transport; (B) demonstrate that electron transport between NADH and a,8-unsaturated acyl CoA is catalyzed by a minimum of two components, electron-transferring flav0protein and acyl CoA dehydro- genase; (C) present kinetic parameters which describe the interactions of some of the enzymes involved, and (D) demonstrate that under the conditions employed, the electron transport system with its substrates is not sufficient to catalyze the phosphorylation of either AMP or ADP. The second part will include (A) purification 60 61 procedure, substrate Specificity, effect of inhibitors and activators, the identity of cofactors, pH optimum and Spectral prOperties of the lactic dehydrogenase; (B) purification procedure, the identity of the cofactor, stability studies and spectral properties of the electron- transferring flav0protein, and (C) purification procedure and Spectral prOperties of the acyl CoA dehydrogenase. Because the names of the enzymes involved in electron transport are frequently referred to, they will be abbreviated as follows: S(-)-lactic dehydrogenase will be called LDH, electron transferring flav0protein will be called ETF and acyl CoA dehydrogenase will be called ACD. Furthermore, if an enzyme is in a reduced state, the abbreviation for it will contain the suffix (R). For example, the addition of S—lactate to a sample of LDH will partially bleach its flavoprotein spectrum, which indicates the formation of reduced LDH, designated LDH(R). I. Components and Properties of the Electron Transport System The evidence of the degree of purity of each enzyme in the system is not discussed until part II, however, this information is necessary for the interpreta- tion of the experiments presented in this part. Conse- quently, a brief discussion of the enzyme preparations used in part I is apprOpriate here. LDH has been purified 193 fold and such preparations are nearly homogeneous. 62 The preparations used were at least 50 per cent pure. Because of its liability ETF was purified only 20 fold. However, from polyacrylamide disc gel electrOphoretic studies these fractions are estimated to be 50 per cent pure on the basis of protein components, and in addition, roughly 50 per cent of the flavin absorption is found in the band with ETF activity. ACD is estimated to be 30 per cent pure on the basis of electrOphoretic studies. As isolated from S. elsdenii ACD is green in appearance, but it can be converted to a yellow form by a procedure to be described in part II-C-l of this section. In terms of flavin content, virtually 100 per cent of the flavoprotein absorption exhibited by ACD preparations arises from ACD. Except where otherwise noted, such enzyme preparations were used in all experiments and were free of cross contamination. [1. Role of Electron-Transferring I?1avoprotein in the Reduction of (1,3-Unsaturated Acyl CoA by EN—laCtate In a previous study Baldwin and Milligan (1964) deunonstrated that S. elsdenii contains LDH and ACD, but thexy did not demonstrate electron transport from S- lachate to d,B-unsaturated acyl CoA. Since S—lactate is a Slikbstrate of LDH and d,8-unsaturated acyl CoA's are SUbEStzrates for acyl CoA dehydrogenase, it was initially assuurued that LDH and ACD constitute the terminal enzymes OJ of the electron transport chain. Thus, the problem became either to identify any constituents of S. elsdenii which catalyze the transfer of electrons from LDH(R) to ACD or to characterize the direct transfer of electrons from LDH(R) to ACD if it occurs. 1. Electron-transferring flaVOprotein-catalyzed reduction of green acyl CoA dehydrogenase by reduced lactic dehydrogenase.--Since S—lactate reduces LDH to LDH(R), —-—-—- adding a catalytic amount of this enzyme and an excess of e to a sample of ACD provides a source of electrons S~lacta‘ for the reduction of ACD. This reduction results in a bleaching of the flavoprotein spectrum of ACD which can be measured spectrOphotometrically. Furthermore, since bleaching does not occur spontaneously at low levels of LDH, electron transport components can be detected by their ability to catalyze the bleaching of the spectrum <>f ACD by 1DH(R). The ability of ETF to catalyze the kileaching of the flavoprotein Spectrum of green ACD by LEE](R) is shown by curves 1, 2, and 3 of Figure 8. Curve 1 (Figure 8) shows the spectrum of oxidized grxeen ACD which exhibits maxima at 360 mu, 438 mu and 710 mu- The addition of catalytic amounts of LDH and a sub- e amount of S—lactate to both the sample and reference r-duced curve 2 (Figure 8). This curve shows 3-10 0 (D f“ r.’ U T :6 O ts in the 360 mu and 438 mu peaks, but exhibits only a Sinaill drop in extinction at these maxima. Curve 3 .pcocomsoo uan v:b mo Goebflccm 33D bouwc nmcdzaa om popuooru QuoB subovdm age .Uovm uc mzcfiwbsfloe mp3 pcm poucumcow mmB OHQEMm comm .o>onc pmbcoflpcfl mCOADLTUc o:u msfim An :mv umewsc Tuscarora Edammmuom mo HE m.a cospmucco mamidm sumo no“ HHoo oozcuomau one .xsov m mHmEcm roumuoma (91 um mo mofioan ow was re; do mafia: mH.o pvcflmucoo Am o>usov m waiscm ”mCOHbetfic mcfi30HHOm or» msam Am ODMSOV An :QV powwsn ouczmmoxm Edfimmcbcg E H.o mo a: m.~ cw :34 compo Lo 3E mn.m pocflcbcoo cubo>so camscm :07: .mu020p ‘a V woMLnXHTo Hmuméfxwrgb ohms A JHVAIOmV >HO Hhon Afiflytm Co Halilozofl:-u. .v 9 mtzmah 65 is; Eozmdi; owk com 000 00v Ody 05V BONVSHOSBV 66 (Figure 8) was obtained with the same components as curve 2 (Figure 8) except for the addition of a catalytic amount of E-fold purified ETF to both the sample and reference cells. The large drop in extinction observed at lower wavelengths, particularly around 440 mp, indicates a significant reduction of the flavin moiety of green ACD. Although the spectral bleaching was complete within one minute, the curve shown was obtained after 60 minutes to be consistent with the other spectra shown in Figure 8. Since the complete system containing LDH(R) and ETF reduced green ACD rapidly while the cuvette lacking ETF exhibited negligible bleaching in the 440 mu region, it can be concluded that ETF catalyzes the reduction of green ACD by DLH(R). Curve 4 (Figure 8), which represents the reduction of green ACD by butyryl CoA, will be discussed in part II-C-2. Upon reoxidation by oxygen, green ACD becomes irreversibly converted to a yellow form (see Literature Review). Thus, it was feared that green ACD as isolated from E. elsdenii might be a mixture of green and yellow forms. To insure uniform preparations of ACD, the yellow form was chosen for all subsequent studies and its forma- tion from green ACD was carried out on a large scale described in part II-C-l. 67 2. Electron-transferring flavoprCLein-catalyzed reduction of yellow CoA dehydrogenase by reduced lactic dehydrogenase.--To demonstrate that the electron transport prOperties of yellow ACD are similar to those of green ACD, the ability of ETF to catalyze the reduction of yellow ACD by LDH(R) was investigated. The experiment is essentially the same as the previous experiment except hat yellow ACD was used in place of green ACD. The ability of ETF to catalyze the bleaching of the flavo- protein spectrum of yellow ACD is shown by curves 1 and 3 of Figure 9. Curve 1 (Figure 9) shows the typical flavo- protein spectrum of oxidized yellow ACD which exhibits maxima at 375 mu and 452 mu. If a catalytic amount of LDH and a substrate amount of Sflactate are present in both the sample and reference cells, the extinction is unchanged at 452 mu from that shown by curve 1 (Figure 9). Thus, LDH(R) alone is unable to bleach the flavoprotein spectrum of yellow ACD. However, if a catalytic amount of ETF is also present in both the sample and reference cells, extensive bleaching occurs, as shown by curve 3 (Figure 9). The extinction at 452 mu is decreased by 72 per cent, which is roughly twice the decrease in extinction observed in the bleaching of green ACD by LDH(R) via ETF. Curve 2 (Figure 9) represents the ETF-catalyzed reduction of ACD by RADH and will be discussed in part I-B-l. 68 .ucmcomEoo umma one mo cofloflccm map umumw mmuccfls om pmcuoovw mums mupoodm mce .Ooem um cwcflmucflma cam cmumwmmmc mm: madame comm .m>onm cmumoflpcw mcoflpeccm ecu mcam Aw :cv womwSQ mumcmmocm Ecflmmcuom_w mo.o mo HE m.o cmcflmucoo mamemm comm MOM Hamo mocmummmu mce .mem mo as wuoa x m cam mumpomHum wo mmHoEE m.o .moq mo muwcs.mno.o pmcflmpcoo Am m>wsov m madamm cam mew w0 06 enoa x m can zodz mo mmaosa mm cmcflmpcoo Am m>wsov ,m onEcm "mcofluflccw mcHBOHH0w ecu mcam AH m>wsov HE m.o mo oESHo> m cH oumuomaum mo mmaoEE m.o x As mac ummwcn wpmcdmocm Esflmmmuom mo mmHOEE mmo.o ccm Qua Boaamw m0 m8 m.m cmchucoo muum>so onEmm comm .modz mo mmmcmmowcwcmc oflpoma couscou an mmmcmmoucmcop 400 axon BoHHmm wo mcflcommamll.m mwdmflm as: Eczema; owe 0mm 09» 0mm 0.0 69 HONVSHOSSV 70 Since ETF is required.for the bleaching of the spectrum of yellow ACD by LDH(R), it functions as a catalyst in the reduction of yellow ACD by electrons derived from LDH(R). Thus, yellow ACD is similar to green ACD in that both are reduced by LDH(R) in the presence of catalytic amounts of ETF. 3. Reconstitution of the electron transport chain.-—To verify that the preceeding Spectral reduction experiments were measuring electron transport from LDH(R) to ACD, crotonyl CoA was added to the system, and the formation of butyrate was measured. This reconstructed electron transport chain contained 0.49 mg of yellow ACD, 0.006 mg of ETF, 80 umoles of S—lactate, 0.3 umoles of crotonyl CoA, 2.75 umoles of potassium phosphate buffer and 0.29 units of LDH in a total volume of 0.176 ml. This mixture minus LDH, was deaerated, equilibrated at 37°C, and LDH was added to start the reaction. At timed inter~ vals, 0.01 ml aliquots were withdrawn and added to 0.001 ml of concentrated sulfuric acid to stop the reaction and to hydrolize all thioester linkages. Following the addi— tion of 0.002 ml of 0.01 S isobutyric acid to the aliquot as an internal standard, butyrate was determined on the gas chromatograph. Controls were run in an identical manner except that each had one component of the assay replaced by water or 0.05 S potassium phosphate buffer. 71 In Figure 10 the production of butyrate is plotted as a function of time. The upper curve (Figure 10) represents butyrate formation for the complete system and the lower curve represents any of the controls. Any sample lacking S-lactate, LDH, ETF or ACD failed to pro~ duce a significant amount of butyrate, whereas, when all components were present the entire 0.3 umoles of crotonyl CoA were converted to butyrate, presumably as butyryl CoA. This experiment clearly demonstrates that S—lactate, LDH, ETF, and ACD are all required for electron transport and is consistent with the spectrophotometric evidence for ETF catalyzing the reduction of ACD by LDH(R). Furthermore, the lactate requirement shows that the source of reducing power for the conversion of crotonyl CoA to butyryl CoA must be electrons derived from Sf lactate. The only other possible sources of reducing power, the dithiothreitol in the LDH and free CoA in the crotonyl CoA, are thus eliminated. It was hoped that pyruvate production by the reconstituted system could be correlated to butyrate pro~ duction, but extremely high blanks, attributed to pyruvate in the Sflactate made this impossible. Extraction of the Sflactate solutions with 2,4-dinitrophenylhydrazine in 0.1N HCl was helpful, but did not eliminate the problem. 72 .Uohm um @QCflquASE cam commuomoc mm3 cofluommu comm .HE mna.o mo cEsHo> Amway m ch :3; mo muwcs mm.o mam An mac wowwcb oumcmmocm Ecflmmmuom mo mcao: on.~ .moo fiascoono mo mmHOEn m.o .oumuomaum mo mofiofix om .msg cc we 000.0 .i .oom soHHom WE mw.o cmcflmucoo Emummm ouofiafioo ecu CH caeshm . .EmomMm whommcmwu cowbooflm Umusuflumc ooa who mg coHuosvoLd comtxbjxun.cfi star“; 73 Ammknzzzv m2; Om Om ow ON m. /m.r<._.o< . L . c .70..- mo 8<¢E-.:3-\ 5‘ (\1 O (SB-low m EIVHAlnS '0. O 74 4. Reduction of electron-transferring flavo- protein by reduced lactic dehygrggenase.--The preceeding experiments established that ETF catalyzes the reduction of both green and yellow ACD by LDH(R), but they do not show whether ETF functions as a true intermediate in electron transport or if it catalyzes the direct transfer of electrons from LDH(R) to ACD. If ETF is a true inter- mediate, its oxidation reduction potential, probably but not necessarily, lies at an intermediate value between those of LDH and ACD. If this is true and if ETF is an intermediate, it should be possible to bleach the flavo- protein spectrum of ETF with LDH(R) in much the same manner as in the spectral reduction experiments with ACD. The results of this experiment are shown in curves 1 and 2 of Figure 11. Curve 1 (Figure 11) shows the atypical flavo- protein spectrum of 20-fold purified ETF which exhibits maxima at 372 mu and 453 mu with shoulders at 474 mp, 426 mu and 358 mu. The presence of substrate amounts of Sflactate in both the sample and reference cells does not alter this spectrum. However, if a catalytic amount of LDH is also present in both cells, a bleaching of the spectrum is observed at 452 mu (curve 2, Figure 11). Curve 3 (Figure 11) shows the bleaching of ETF by NADH and will be discussed in part I-B-2. 75 Figure 11.--Bleaching of electron-transferring flavo- protein by reduced lactic dehydrogenase or NADH . Each sample cuvette contained 0.145 mg of ETF and 25 umoles of potassium phosphate buffer (pH 6) in a total volume of 0.5 ml (curve 1) plus the following additions: sample 2 (curve 2) contained 20 umoles of S—lactate and 6.5 units of LDH and sample 3 (curve 3) contained 55 moles of NADH. The reference cell for each sample contained 0.5 ml of 0.05 S potassium phosphate buffer (pH 6) plus the additions described above. Each sample was deaerated and maintained at 24°C. The spectra were recorded 5 minutes after the addition of the last component. ABSORBANCE 76 0.4r 0.3 r 0.2 0.! T 0.0 300 460 560 WAVELENGTH ( M )1) 77 Since LDH(R) reduces the flavin moiety of ETF and since ETF catalyzes electron transfer from LDH(R) to ACD, ETF is quite probably a true intermediate in electron transport: that is, it functions as a carrier of electrons from LDH(R) to ACD. The extent of bleaching of ETF by LDH(R) is one half that observed for the reduction of green ACD by LDH(R) via ETF (curve 3, Figure 8). However, if a correction is made for the observation that only 50 per cent of the flavin absorption of ETF preparations arises from ETF itself, then the extent of bleaching is comparable to the extent of bleaching of green ACD. 5. Enzyme activities in electron—transferring flav0protein preparations.--Although the preceeding experiment indicated that only part of the flavoprotein spectrum of ETF preparations is reduced by LDH(R), it did not show that only a single electron carrier functions between LDH(R) and ACD. If only a single protein is involved it should be possible to separate the proteins of an ETF preparation by polyacrylamide disc gel electro- phoresis and to detect ETF activity in only one of the resulting bands. Alternatively, if more than one protein is required, none of the bands will exhibit ETF activity when assayed alone. Ninety pg of a fresh ETF preparation, which had been adjusted to 0.01 S dithiothreitol and incubated for 78 one hour at 4°C, was subjected to polyacrylamide disc gel electrOphoresis as described in Materials and Methods. Following electrophoresis in a quartz tube, the gel was directly scanned at 280 mu and 450 mu as described in Materials and Methods. The upper line of the tracings (Figure 12) represents absorbance by protein at 280 mu and the lower line represents absorbance by flavins at 450 mu. The protein scan shows two predominant peaks, each of which exhibits flavin absorption at 450 mu. Thus, ETF preparations consist predominantly of two flavo- proteins in roughly equal amounts. Each section of the gel which contained one of the two protein bands, designated U and L in Figure 12, was excised from the gel, but into small pieces and incu- bated for 2 hours at 4°C in 0.05 ml of 0.05 S potassium phosphate buffer (pH 6). An 0.02 ml aliquot of the eluate from each section was assayed for ETF activity. The assay used was the kinetic assay which was used to purify ETF and which will be described in detail in part I-C-l. In essence it measures the rate of bleaching of yellow ACD by LDH(R) at 450 mu. The rates observed with this assay are prOportional to the concentration of ETF in the assay. Table 2 gives the results of ETF assays of the gel eluates from bands U and L. The rates shown should be considered qualitative Since the exact amount of protein eluted from each band is not know. With 79 .mflmouocmowuomfim o» woflum Homeowcuoflcuflc w Ho.o CH Dov um woo: oco LCM coumccocfl mm3 oshwco mo mHQEmm ms mo.o 0:9 .cflououmo>mam mcfluuowmcmwp Icouuomao mo Hem omflp mcHEmahwomwaom m mo mmcfiomup mochHOmn<|I.mH ouomflm 80 ~1025 flfih-L U‘W‘ Mm 0 1w 1.2 09V .5. woz (:LD HR<;> ETF( (:M 74:)? L’RYL C61 PYRUVATE LDH( ETF::> ACD(R) (ACD(R) NAD‘x> Hmflpflcfl oce .Uonm um pocflmucflam cam coumnmmmc mmB hammm comm .cmumoflccfl mm motto mmB mBm .HE m.o wo oESHo> Hmuou m Ce mod mo muflcs v.m cam An may Howmsn mumsmmozm Esflmmmuom wo mmHOEJ m.ma .obmuomalm mo mmHOE; om .ood BoHHom wo moHOEA moo.o tocflmucoo mppm>so comm .popcm cflmuonmo>mam mcfluummmcmuuncouuomam wo ucsoEm mcu co unommcmuu conuomam mo mama mgw mo mocmpcmmooal.ma musmflm 94 $0203: Z_m._.omn_o><4m oz_mmn_mz Hmuoc m CH An :QV powwoc mumcmmocm Esflmmmpom mo mmHOE: o.m ccm hem wo @E moo.o ‘mumcomaam mo moHOE: om room BoHHmm mo moHoEn mmoo.o Umcamucoo mmmmm comm .QOHumuucmocoo mmmcmmoucmcmp mcmuomanctvm co uuommcmuu couuomam mo mumu mcu mo mocmccmmmonn.va ouomflm 97 3-9223 mm1mo o_._.o<|_ _ 09 OO. Om O - - d In ._ 3.0 8.0 ON «3 Elva I Q E? O N (OQVVV/SalnNIW) 98 .Uonm um cocflmucflme Ucm copmummoo mm: muum>so comm .HE m.o mo mESHo> Hmuou m cfl woumoflccfl mm 904 Ucm Am :mv nowmoc mumcmmocm Esflmmmuoa mo moHoEn o.m .mem mo 06 moo.o .moq mo muflcs mo.o .mumuomalm wo moHOEn om cocflmucoo opno>5o comm .GOflwmuucoocoo mmmcmmoucxcob <00 Hmom co uuommcmuu couuomam mo momu mcu mo mocmcammmoll.ma madman 99 A .0223 mqumoomoémo <8 #2 3 On. m..m o . $033504 . cm pm 0 Q .m . -wa - l - .OVB O— s \ \O .00 _ 0 Im— [10 A 0\ .ON ELLVH (09’7VV/83inNIw) 100 and a molecular weight of 125,000 are assumed,1 the apparent Km can be expressed as an LDH concentration of 9.5 x lO-BE. This value compares quite well with the apparent Km of 2 x 10.7 g_which Hauge (1956) observed for the interaction of ETF with ACD(R) using the mammalian enzymes. While the reaction of ETF with ACD(R) studied by Hauge is not identical to the reaction of LDH(R) with ETF under consideration here, it does have similarities with respect to the following: (a) the ETF-ACD(R) system is one of the few cases in which the apparent Km for protein- protein interaction has been determined, (b) it involves one of the same enzymes used in this experiment, and (c) it is the same type reaction in that it involves flavin cofactors. 3. Dependence of the rate of electron transfer upon ACD concentration.--To study the effect of ACD concentration on the rate of ACD reduction, the same general assay procedure was employed as in the previous two experiments. In this case the LDH concentration was fixed at 0.026 units and the level of ETF was 3 x 10.3 mg, while the ACD level was varied. Figure 15 shows the rate 1The specific activity of pure enzyme is taken to be that of the preparation described in part II-A-l which is nearly homogeneous. The molecular weight of 125,000 is approximated from preliminary molecular weight determinations using Sephadex G—100. 101 of electron transfer versus ACD concentration (insert) and a Lineweaver-Burke plot of the same data. Application of the computer curve fitting program to the data yielded an apparent Km of 1.5 x 10-5% for yellow ACD reacting with ETF(R). The linearity of the Lineweaver—Burke plot shows that this reaction involves binding as a part of the mechanism. This observation also supports the contention that binding is involved in the reaction of LDH(R) with ETF. It is improbable that ETF will bind to one flavo- protein substrate (ACD) but not to the other flav0protein substrate (LDH(R)). Although this apparent Km is 2-3 orders of magnitude higher than that obtained for LDH(R) binding to ACD, this does not necessarily indicate that large a difference in the binding of these two enzymes to ETF. Whereas LDH was constantly being regenerated, ACD is not regenerated and for this reason ACD may give a higher apparent Km. In any case a Km of 1.5 x 10-5% is not an unreasonably high value for the half saturation of one protein by another. Also, it compares favorably with the apparent Km of 2 x lO-7fl (Hauge, 1956) for the binding of ACD(R) to ETF, considering that ACD(R) was being con~ stantly regenerated. 4. Dependence of the rate of electron transfer upon .— - .“v-- phosphate buffer concentration.—-In the course of running assays during ETF purifications, an inhibition of electron transport was observed when the concentration of phosphate 102 buffer (pH 7) in the assay was increased. To further characterize this inhibition a series of standard ETF assays were performed at fixed LDH, ETF, and ACD con- centrations while the buffer concentration was varied. The inset in Figure 16 shows a plot of the rate data obtained at various concentrations between 0.25 and 0.8 fl_potassium phosphate (pH 7). Since inspection of these data once again suggested hyperbolic kinetics, they were replotted (Figure 16) as the reciprocal of the fraction inactivation versus the reciprocal of the phosphate buffer concentration. To obtain fraction inactivation for each buffer concentration a corrected rate was calculated by subtracting the rate at infinite buffer from each rate. Dividing this rate by the corrected rate at zero buffer concentration yielded the fraction inactivation. In effect, fraction inactivation expresses each rate change from zero buffer concentration as a fraction of the total rate change observed from zero to infinite buffer con- centration. The binding constant of 0.23 g for potassium phosphate at pH 7 which was calculated from the data on Figure 16 is rather high to be of any physiological significance to E. elsdenii, but phosphate may be acting as an analog of some other phosphorylated compound such as acetyl phosphate or ATP. If this is the case the inhibition may actually represent a control mechanism for 103 .mcoflpmasoamo Eoum cmcsaoxo mmB ucflom mace k. .coflummucoocoo ummmoc mumcflmcfl ou ouom Eoum ocmu cfl mmcmco Hmuou mcu mc Uo©H>Ht coaumnucoocoo common ouoN ou m>mummmu mcmn CH omcmco mcc mm pocmmmc mm coflom>fluomcfl cofluommm .Uonm um cmcflmuchE Ucm cmcmmmmoc mm3 muuo>5o comm .oopmoflccfl mm ummwdc mumcmmocm Eoflmmmuom Ucm mem mo 08 maoo.o .mumpomalm mo mmHoE om .mom mo mums: o.w .QU¢ BoHHmm mo moHOE wmo.o Umcflmucoo mmmmm comm .coflumuucmocoo ummmsc mcmcmmocm co unommcmuu couuooao mo mpmu oco mo oocmpcomooll.ma ousvflm 104 10.x ' 2 " o . «“3“ <1- . S": I. .95 O / E «2 ‘. 8 .o (\l 42 o co to wpfimom nomad pmchucflmE mm3 mam mom mo cowuflpmm mcu on uoflum oohm um moumummmc mm: musuxHE mca .HE voa.o mo manao> Hmqu m he Acmmmm cmcsv ammo mo maofi: o.H cam .mcfluomaosc mcflcmcmlo mo mac moa x m mcflcflmucoo mmeuomaosc mcflcmmm mo mmHoE: m.o .Ah may ummmsc mpmcmmocm Edwmmmuom mo mmHOE: va.o ~mom BoHHm> com mo mmHOE: m om x o.m .mem mo 08 m 0H x m "mo Umumflmcoo mnsuxHE cofluommu comm .pmmoamam mCOfluflpcoo mcu nomad musmoooum Emumowméoucooflmmu mcv mo uHEflH cofluomumc mcu mpcmmmummu musmflm mace % I: In mH.o mcoc Hwo *000.0 00.0 Ommw + mofi H.0 *000.0 00.0 mo¢ m0.0 *000.0 mH.0 UmMU + m2¢ 00.0 *000.0 MH.0 mZd mmaofin mmHOEn O\m poscoum poumamuocmmocm UmEuom wumumudm mcofluwmmd .Emummm uHOchmnu couuomam mcu mc coHuoSUOHQ mumumuoc 0cm coflumamuocmmocmnl.v mmmde 113 it should have been possible to phosphorylate either AMP or ADP. Although these negative experiments are not conclusive, they suggest that the electron transports system does not possess any phosphorylating ability. Part II. Purification and Properties of the Electron Transport Enzymes A. D(-)-Lactic Dehydrogenase l. Purification.--A summary of the purification of P. elsdenii LDH is presented in Table 5. A 193-fold puri- fication was obtained with a 9.4 per cent recovery. All purification steps were performed in the presence of 1 mM dithiothreitol which helped to prevent activity losses. The data in this table were obtained from the fractiona- tion of crude extract obtained from 300 g of cells grown in 200 liters of medium. The crude extract was mixed with one-fourth volume of moist DEAE cellulose, stirred for 10 minutes at 4°C and filtered under vacuum on Whatman #1 filter paper. The resin, containing the LDH, was washed twice with 500 ml of the same buffer at 0.3.M and applied to the top of a column containing the same quantity of moist DEAE cellulose, equilibrated to 0.3 M potassium phosphate buffer (pH 6). LDH was eluted with the same buffer in a gradient from 0.3-0.8 M. The enzyme was then precipitated by 70 per cent saturation with ammonium sulfate. The resulting precipitate 114 mam e.a .omm one.m is mmv.m H.o .mumsam muflumaamsxoumsm .. -- u- .. As mao_m H.o .mnmsam mmno mm mm .om ome.am woe ..umm eommxemzv Ne mm .om ome.am Am may 2 m.o-m.o .mmoasaamo ammo H OOH ee.m oom.om pomuuxm mmsuo 2.. mwwomw mmmmm mam .3. .Hficwomao mooooooummnpmoummm Eonw mmmcmmoucmcmp oHuomHIAIVm mo coHumoHMHnsmnl.m mmm<9 115 was dissolved in a minimum volume of water and desalted by passage through a 2.5 x 10 cm column of Sephadex G-25, equilibrated at 0.1 M_potassium phosphate buffer (pH 7). Following desalting, an aliquot of enzyme solution was applied to a 0.75 x 15 cm column of hydroxyappatite equilibrated to 0.1 M potassium phOSphate buffer (pH 7) and the enzyme was eluted with this same buffer. Figure 17 shows absorbance tracings of a polyacryl- amide disc gel of an 80 ug sample of 193-fold purified LDH. That this gel exhibits only one major protein band (280 mp trace) indicates that the LDH is nearly homogeneous. The flavoprotein nature of this enzyme will be more fully discussed in subsequent experiments. 2. Substrate specificity and kinetics.--Tab1e 6 shows the per cent activities relative to Q-lactate of various a-hydroxy acids as measured with the ferricyanide assay. Of the ten analogs tested, only three, Qflactate, E-lactate and 2,27d-hydroxybutyrate, showed any appreciable activity. However, the activity exhibited by Eflactate was shown to be due to Dflactate contamination of com- mercial 1.1—lactate.2 The relative lack of activity 2The amount of available substrate in commercial 2- and E-lactate was measured by modifying the standard LDH assay to make it an end point assay for lactate. The normal assay level of LDH was increased 1000 fold, the level of ferricyanide was maintained at 0.0025 M and small amounts of lactate, less than the total amount of ferri- cyanide, were added. Under these conditions two moles of 116 .mmmcmmoucmcoc oauomanfiuvm mo mHmEmm ma mo.o m mo Hmm omflp wUHEmaxuomwaom m mo mmcflomup mocm3u0m341|.ha musmflm 117 .32 Om? he. moz ays were ' X > a \Z —| .. a) 8 .J 0 /° - - o 2 O 8/ J -3 1 1 -6 -4 -2 o Loge Q-LACTATE (M/L) 123 (b) the cells possess an active transport mechanism which concentrates lactate to very high levels in the cell or (c) the enzyme requires an as yet unidentified component which decreases the Km of gelactate. Although LDH exhibits a narrow specificity for electron donors, its acceptor specificity is rather broad. Each acceptor tested, with the exception of pyridine nucleotide, was effective. Table 7 shows the relative rates obtained with assays consisting of 25 lactate, LDH and various acceptors. All dyes, oxygen, and cytochrome c were effective, indicating that the enzyme possesses a broad specificity for its electron acceptor. Further, lack of reactivity with 3-acetyl- pyridine DPN shows that this LDH is of the "NAD independent" type. An attempt to run the reaction in the reverse direction with NADH and pyruvate was also unsuc- cessful. The broad specificity exhibited by E. elsdenii LDH toward electron acceptors other than cytochrome c distinguishes it from that isolated from aerobic yeast (Gregolin and Singer, 1962 and Nygaare, 1961) but relates it to the LDH of anaerobic yeast (Iwatsubo and Curdel, 1961). An apparent Km for ferricyanide, the acceptor routinely used, was calculated from data obtained by varying its concentration in assays run at fixed Q-lactate and LDH concentrations. Figure 19 shows a computer fitted 124 .moq mo cofluflppm map an pmumauflcfl mums can Uovm um can muck manomm Ham .HOpmmoom on» was cmmmxo msocmmOpcm paw HE m.m m0 mESHo> m ou ceaunomoum CH pwmmmuocfl mums wucmmmmu mcu coHuQEdmcoo cwmmxo mo mucmsmusmmmfi Mom .HE N.o mo madao> Hmuou m ca topmoflcca mm uoummoom paw mumuomalm mo mmHoE: omH .Ah may umwmsb mumcmmocm Edwmmmuom mo mmHoE: OH "mo coumflmcoo hwmmm oaupwfiouonmouuommm comm .mpouuowam cmmmxo mmwu xumau m :uHB commfldwo ammummxo .oo ucoEduumcH mmcflumm BOHHm» m nufl3 cmHDmmmE mmz coaumasmcoo cmmhxo «a m.m mm on csmw N I Ha «*woouuowam o z Vloa x v.m cmmmxo 0 room Ha\ma m.o zmo mcfleflusmamumoasm oom omm .vaOH x m.H o oEouzooumU mmm oom .m mloa x v Hocm£m0©CH Hocmzmouoanoflo mm ems Ha\ma A meflnoHno asflHoN ImuumuahcmsmlmIawcmcmouuflc ImummnmIamcmcmocOHuwummnm mom omv z H.o mpflcmmofluumm HE\cHE\meoEJ SE , mumm cumcmq m>m3 coflumuucmocou Monaco?N .mmmcmmoucmnmc cauomaualvm mo muAOHwHommm uoummoomun.n mqmde 125 .Uovm um cam mums mammmm mca .Uopmoflpcfl mm mcflcmmofluumw Esfimmmpom cam Ah mmv ummmsn mumcmmozm Esflmmmuom mo mmHOE: OH rmuwpomalm mo mmHOE: oH pmcflmucoo muuw>50 comm .coflumuucmocoo mpflcm>0Huumw co mommm mmmcmmouomnmp owuomH on» we mama map mo mocmpcmmmonn.mH musmflm 126 \O O 0 7L0 [x O O “0 L0 _ O O 8 °\ N “l O. O. O. O. O 00 «0 <1" N All/010V (.-SllN0) BSVNEOOHCIAHBG OLLOV-l (L/M) FERRICYANIDE 127 fitted Lineweaver-Burke plot of the data which yielded an apparent Km of 2.2 x 10-45. Since this level is more than ten fold lower than the normal assay level of ferricyanide, the acceptor is saturating in LDH assays. Fitting the same data to the Hill equation yields an n value close to 1, indicating that no c00perativity is involved in the binding of ferricyanide to the enzyme. 3. Spectral properties.--Figure 20 (curve 1) shows the spectrum of oxidized, deaerated LDH. It exhibits a "typical" flavorprotein spectrum with maxima at 450 mu and 378 mu. Upon the addition of 0.2 moles of gelactate, a substantial reduction in absorbance was observed at 450 mu with some increase in absorbance around 570 mu (Figure 20, curve 2). Note also the appearance of a shoulder at 475 mu. This spectral bleaching by Qflactate indicates that substrate reduces LDH to LDH(R). The spectral properties of this LDH are similar to those obtained by Gregolin and Singer (1963) in their study of the NAD independent lactic dehydrogenase from anaerobic yeast. 4. Cofactors.-—Almost all pyridine nucleotide independent lactic dehydrogenases and a-hydroxyacid dehydrogenases studied are zinc-containing flav0proteins. To determine what cofactors are isolated as part of E. elsdenii LDH, flavin analysis and metal chelator studies were performed. 128 .mumuomalm mo mmHOEJ oom pmcflmucoo Omam m mHmEmm MOM mupm>so mocmummmu map pom Am mmv umwmsn mumcmmonm EDHmmmuom mo mmHoEE m.H Umcflmpcoo muum>do oocmumwmu comm .uovm um mmcflmucflms cam mmumummmp mmB muum>so comm .oumpomatm mo mmHoEJ.oom Umcflmucoo OmHm Am m>usov N mHmEmm .AH m>usov HE m.o mo mEdHo> Hmuou m ca Am may umwmsn mumnmmosm Esflmmmuom mo mmHOEE m.H Ucm man no me v.H Umcflmucoo muum>50 comm .mmmcmmouowcmm owuomauflnvm UmodcmnumumuquSm mam monamflxo mo munommmun.om ousmflm 129 8%. 00m SAEVIHozmqm>§3 e. 0.0 l (x! O 4% o EONVBHOSSV 130 The flavin cofactor was identified by thin layer chromatography. One half ml of a concentrated LDH prepara- tion (specific activity 105) was heated to 100°C for five minutes and the resulting white precipitate was collected by centrifugation and discarded. Aliquots of the super- natant were chromatographed in two solvent systems with authentic standards on Silica Gel F254 thin layer plates and the flavin spots were identified. The solvent systems were water-saturated collidine and tfbutanol and water in a ratio of 60:40, v/v (Kilgore gt_al., 1957). The resolution and separation with the second system were excellent. Although both of these systems were originally designed for paper chromatography, they worked well with thin layer plates. In both systems the only flavin detected was FAD. Even when an ultraviolet light was used to increase the sensitivity of detection, no FMN or any other flavin could be detected. Thus, the flavin cofactor of LDH is FAD. This finding is consistent with the identi- fication of FAD as the flavin cofactor of the LDH's from both aerobic and anaerobic yeasts (Gregolin and Singer, 1964 and Iwatsubo and Curdel, 1961). To determine if LDH contains a divalent metal cofactor, inactivation of the enzyme by metal chelators was studied. preliminary studies indicated that EDTA, orthophenanthroline, and N,N'-bipyridyl would all inacti- vate the enzyme. In more detailed experiments the 131 kinetics of inactivation of LDH by orthophenanthroline were studied. Equal samples of LDH in 0.05 g potassium phosphate buffer (pH 8) were incubated with the various additions shown in Figure 21. At various times aliquots were withdrawn and assayed for LDH activity. Figure 21 shows the relative activities remaining in each sample as a function of the time of incubation. Note that each inactivation curve is biphasic; there is a rapid inactivation during the first five minutes of incubation which is followed by a slower, pseudo first order loss of activity. Computer fitting of the data for the 1.9 mg sample (line 4, Figure 21) yields a rate constant of 0.75 sec-1 for the slow inactivation. An identical sample which also contained 0.5 §_Qflactate (line 2, Figure 21) shows a decrease of about one half in the rapid inactivation and for the slow loss the rate 1 to 0.49vsec-1 in the constant drops from 0.75 sec- presence of Q-lactate. Thus, Q—lactate partially protects the enzyme against both the rapid and the slow losses of activity. Considering the relatively small protection afforded by a concentration of Qflactate which is twice the apparent Km for gelactate, one cannot conclude that the metal is located in the lactate binding site, as was done with the LDH from anaerobically grown yeast. In the latter case a Qflactate concentration of 0.003 x Km provided 50 132 Figure 21.--Inactivation of E(-)-lactic dehydrogenase by orthOphenanthroline. Each incubation mixture contained an aliquot of LDH plus the following additions: sample 1 (line 1) contained 1.9 mg meta- phenanthroline, sample 2 (line 2) contained 1.9 mg_orth0phenanthroline plus 0.5 m§_ Q—lactate, sample 3 (line 3) contained 0.91 mg orthophenanthroline and sample 4 (line 4) contained 1.9 mfl orthophenanthroline. The incubation mixtures were maintained at 4°C and aliquots were withdrawn with time as indicated in the figure. Each assay for LDH activity contained 20 umoles of Qflactate, 0.5 umoles of potassium ferricyanide and 10 umoles of potassium phOSphate buffer (pH 8) in a total volume of 0.2 ml. 2.00 \O “k 0 ’ o \\ \o \\\\ \n O we L75- \0 \v >_ El 1: v n l:— \V\n o 1.50 20 9 v 0 3 3 l.25- 4 ° 0 2'0 4'0 6'0 TIME (MINUTES) 134 per cent protection against inactivation by orthophenan- throline (Stachiewicz et_al., 1961). Lowering the concentration of orthophenanthroline to 0.91 mg produces only a slight lessening of the rate of fast inactivation, but the rate constant for the slow loss decreases from 0.75 sec-1 to 0.48 sec-1 (curve 3, Figure 21). Thus, while the concentration of ortho- phenanthroline does not have much effect on the rapid step, the slow inactivation shows a definite orthophenanthroline dependence. On the basis of these data one would predict that orthOphenanthroline does not cause the rapid inacti- vation, but such is the case. If metaphenanthroline replaces orthOphenanthroline, the rapid inactivation is markedly decreased, but still occurs, and the slow inactivation falls to near zero (curve 1, Figure 21). Metaphenanthroline is an analog of orthophenanthroline which does not chelate metals. Thus, the rapid inactiva- tion appears to be non-specific, but dependent upon orthophenanthroline or metaphenanthroline, while the slow, pseudo first order inactivation depends on the chelating ability of orthophenanthroline. A pseudo first order inactivation of LDH from anaerobic yeast was observed by Cremona and Singer (1964). After inactivating LDH with orthophenanthroline, Cremona and Singer (1964) also found that they could pass the enzyme through a short column on Sephadex G-25 and 135 then regain activity by the addition of divalent metal ions. Zinc was the most effective reactivator. Attempts to repeat this experiment with 3. elsdenii LDH resulted in an irreversible loss of activity with passage through Sephadex G-25. However, it was observed that prior to chromatography, orthOphenanthroline-inactivated LDH could be reactivated by divalent metal ions. Thus, reactivation studies were performed to determine which metal was the most effective in reactivating LDH. A sample of enzyme was inactivated in 4.0 mg orthOphenanthroline for a period of 45 minutes as described above. Aliquots were mixed with equal volumes of divalent metal ion solutions and the mixtures were immediately assayed for LDH activity. Figure 22 shows the resulting activities as a function of various concentrations of several divalent metal ions. Activities are expressed as a percentage of the activity observed before inactivation by orthophenanthroline. At any given concentration zinc was the best reactivator, an observation consistent with its being the metal cofactor. It cannot be concluded, however, that zinc is the metal ion because orthophenan- throline was present in the reactivation mixtures and because the time of reactivation was short. What can be said is that in the presence of orthophenanthroline, divalent zinc reactivates LDH faster than either cobalt or manganese. The cofactor for the aerobic yeast enzyme 136 Figure 22.—-Reactivation of Q(-)-lactic dehydrogenase by divalent metal ions. Each assay contained 0.5 umoles of potassium ferricyanide, 160 umoles of B—lactate and 10 umoles of potassium phosphate buffer (pH 7) in a total volume of 0.2 m1. Assays were run at 24°C. %ACT|V|TY ICC 137 o * 0..-..Zn Oz 75 ' Loo Clg A---Mn Clz o u 50 - o 25 - _ O A Q I A A_r_; -e -4 -2 LOGIOMETAL CONCENTRATION 138 has been definitely identified as zinc (Cremona and Singer, 1964) and the anaerobic yeast enzyme can be reactivated with zinc and FAD following inactivation by precipitation with ammonium sulfate (Iwatsubo and Curdel, 1961). Thus, P. elsdenii LDH contains at least two cofactors, FAD and a divalent metal ion, probably zinc. 5. pH optimum.--Ferricyanide assays were per- formed at pH values ranging from 6 to 9 with a non- saturating concentration of Delactate to determine the dependence of the assay on pH. Each assay contained 0.5 umoles of potassium ferricyanide, 10 umoles of Dflactate, l70 umoles of potassium phosphate buffer and an aliquot of LDH. The enzyme exhibited a broad pH optimum at pH 7.9. Above pH 8 the assays stOpped long before all of either substrate was consumed, which suggests that the enzyme is not stable at high pH. 6. Activators.--Just as an inhibition by potassium phosphate buffer was observed in the electron transport assay, phosphate buffer stimulates the ferri- cyanide assay for LDH. In addition controls run with potassium chloride and sodium phOSphate buffer indicated that phosphate was the component responsible for the stimulation. In view of the high Km exhibited by LDH for D-lactate, the phosphate buffer inhibition of electron 139 transport and the phosphate stimulation of the LDH assay, it is believed that an activator, possibly a phosphate compound, exists for the LDH. A series of phOSphate con- taining nucleotides, carbohydrates and other compounds were tested for their ability to stimulate the ferri- cyanide assay at a concentration of Delactate which was approximately Km/S. Table 8 lists the compounds tested and the per cent activities observed relative to no additions. Unless otherwise indicated, the concentration of activator in each assay was 0.9 mg/ml. This concentra- tion is approximately 5 mg for a compound with a molecular weight of 200. The table shows that only thiamine pyrophosphate and sodium perphosphate produced any significant stimula- tion of the rate. In terms of the magnitude of stimulation expected, a change of only 50 per cent is not considered significant. The eXperiment was designed so that if a Km effector were tested which would drop the Km for D—lactate by at least several fold, a stimulation of at least several fold would be observed in the rate. If any of the compounds were truely an effector at a concentration in the mg_range, it should have a marked effect on Km, and thus, on the rate. None of the compounds tested is an effective activator of LDH and the identity of the activator, if it exists, is still unknown. 140 TABLE 8.--Activators of_D(—)-lactic dehydrogenase. Compound Per Cent Activity Nucleotides ATP, sodium salt 85 ADP, sodium salt, 0.1 M 97 AMP, potassium salt — 100 3',5'-AMP, sodium salt 77 CTP, sodium salt 89 CMP, sodium salt 84 GTP, sodium salt 89 IMP, sodium salt 74 UMP, sodium salt 84 Carbohydrates UDPG, sodium salt 75 glucose 78 glucose-l-P, sodium salt 97 fructose-1, 6-diP, sodium salt 101 a-glycerol phosphate, sodium salt 103 3-phosphoglycerate, sodium salt 108 phosphoenolpyruvate, sodium salt 85 hydroxypyruvic acid, phosphate, dimethyl ketal 84 phospholactic acid, ammonium salt, 0.1 M. 100 Other carbamyl phosphate, dilithium salt 79 thiamine pyrophosphate, chloride 140 sodium perphosphate 148 yeast extract 119 bacto tryptone 100 NAD 5 2 NADP _4 73 pyridoxal phosphate, sodium salt, 9.xlO mg/ml* 82 acetyl phosphate, 0.5 M 115 none — 100 * At a ten fold higher concentration, PLP inhibited 100%. Each assay contained: 10 umoles of D-lactate, 0.5 umoles of K3Fe(CN)6, and 10 umoles of potassium phosphate buffer (pH 7) and effector at 0.9 mg/ml or as indicated. Assays were run at 24°C and were initiated by the addition of LDH. 141 Polyvalent ion effects, such as those observed in P. elsdenii electron tranSport, are not unique to this system. The LDH from Propionibacterium pentosaceum is stimulated 2.6 fold by 0.5 M ammonium sulfate in an assay containing D—lactate and DCPIP (Molinari and Lara, 1960), and ammonium sulfate also stimulates the DCPIP assay for the LDH from Bupyribacterium rettgeri Wittinberger and Haaf, 1966). Lastly, in working with an LDH from anaerobically grown yeast, Nygaare (1961) observed a decrease in activity toward cytochrome c (the physiological electron acceptor) and an increase in activity toward ferricyanide in the presence of poly- valent cations such as protamine. The dependence of the velocity of the LDH assay on phosphate buffer concentra- tion was determined by running LDH assays in varying concentrations of phOSphate. Figure 23 shows that phosphate is not required for activity and that the rate observed at fixed E—lactate and LDH concentration is a linear function of the phosphate concentration. Mechanistically, this linear relationship is difficult to explain unless one assumes that the line is the lower part of a saturation curve for phosphate. In this case the Km for phosphate would have to be greater than 2.0.M and phosphate is probably of no physiological significance. 142 .Uovm um can ouoz mammmm Ham .Uoumoflmcfl mm An :dv common mumcmmocm mam mac mo uosqflam cm .oumuomanm wo moHOE: 0H .ocflcmxofluuow Esflmmmuom mo mmaoan m.o conflmucoo wmmmm comm .COHumuucoocoo oumcdmocd co ammmm ommcoqoupacmp oauoma ocu mo quUOHm> ocp mo mocopcommoll.mm madman 143 0.8 0 O4 PHOSPHATE (M/L) l 1 Jan 0 N) (\l‘ '- 0. O. 8 O O (Sian)AJ.|/\I13V ESVNBSOHGAHEJG OllOV'] 144 B. Electron-Transferring Flavoprotein l. Purification.--A summary of the purification of P. elsdenii ETF is presented in Table 9. A 22 fold purification was obtained with a 20 per cent recovery of activity. Routinely, ETF preparations were 17 to 22 fold purified with recoveries of 5-10 per cent. ETF was assayed using the standard ETF assay which is described in part I-C—l. A unit of ETF activity is defined as that level of ETF which gives a rate of reduction of ACD of 1.0 A per minute at 450 mu. The data in Table 9 were obtained from the fractionation of a crude extract of 30 g of cells grown in 20 liters of medium. Since the enzyme was labile, further purification steps were not routinely employed. The crude extract was applied to a 2.5 x 25 cm column of DEAE cellulose, equilibrated to 0.1 M potassium phosphate buffer (pH 6) and the enzyme was eluted with a phosphate gradient from 0.1-0.3 M (pH 6). The peak fractions were pooled and brought to 70 per cent satura- tion with ammonium sulfate. The resulting precipitate was dissolved in a minimum volume of 0.1 M_potassium phosphate buffer (pH 6) and applied to a 2.5 x 50 cm column of Sephadex G-100, equilibrated to 0.1 M potassium phosphate buffer (pH 6) and ETF was eluted with the same buffer. 145 mm om mm mmm.H 0 mm .m H.o .xmomammm ooanu I: II II I: won .wummasm Edwcoead a .. mm .. m ma .w m.o-H.o .mmoHsHHmo mama a cos a oom.m uomuuxm mmsuo mum>oomm mufl>fluo¢ muHCD macs pamo umm camsomdm Hmuoe doom .Hflcmpmam .m Scum camuonmo>mam OCHHHmmmCMHfllQOHfiomHm MO COHUMOHMHHSQII.® mqumh. 146 As judged from the absorbance tracings of the polyacrylamide disc gel shown in Figure 12, this prepara- tion is estimated to be 50 per cent pure. Only the band designated L on the figure exhibited ETF when eluted and assayed (see part I-A-S). 2. Flavin cofactor.--Since all ETF preparations were yellow and since the polyacrylamide disc gel showed flavin absorption in the peak exhibiting ETF activity, thin layer chromatography was performed to identify the flavin. A sample of ETF from Sephadex G-100 was pre- cipitated with ammonium sulfate, dissolved in a minimum volume of water, heated to 100°C for five minutes and centrifuged to remove denatured protein. Aliquots of the resulting clear yellow supernatant were taken for flavin analysis by thin layer chromatography in two solvent systems (Kilgour eE_al,, 1957). In this case the results were not as clear cut as they were with LDH. The first system was 5 per cent NaZHPO4-7HZO in water and yielded only one Spot for the unknown which co-comatographed with authentic FAD. No FMN could be detected. In the Efbutanol-water system (60:40, v/v) the results were ambiguous. FMN was not found and FAD was found, but in addition, an unidentified spot appeared on the chromatogram. This yellow, florescing spot migrated just behind FAD but was not as sharp and well defined. In all studies with mammalian 147 ETF the cofactor has been identified as FAD (Crane and Beinert, 1956 and Frisell et a1., 1966). 3. SpectralgprOperties.--Curve l in Figure 11 shows the spectrum of a deaerated sample of ETF. It exhibits the typical falvoprotein absorption maxima at 453 mu and 372 mu and has shoulders at 474 mp, 426 mu and 358 mu. While the occurrence of a shoulder at 474 is relatively common, those at 426 and 358 are not. In comparing the spectrum of this ETF preparation to that exhibited by mammalian ETF preparations, one finds that the pig liver enzyme has maxima at 375 mu, 437.5 mu, and 460 mu (Crane and Beinert, 1956) and the primate ETF shows a maximum at 375 mu and at 410 mu with a broad shoulder in the 440—480 mu region. Thus, E. elsdenii ETF is not unique in displaying a shoulder at 426 mu. In fact, an absorption maximum between the typical 370 mu and 450 mu peaks seems to characterize ETF preparations. This unusual maximum is probably caused by some unique environment around the flavin in ETF, or in the case of P. elsdenii the flavin moiety may not be FAD or FMN. 4. Stability.--ETF preparations were found to be labile during storage, either at 4°C or when frozen. At 4°C loss of activity was observed over the course of several hours and this loss was much greater than was observed with frozen enzyme. When stored at pH 6 activity 148 losses were less than at higher pH values. The enzyme is not stabilized by any of the following: 1.0 mM_ dithiothreitol, one half saturation with ammonium sulfate, bovine serum albumin at 10 mg/ml, egg albumin at 25 mg/ml, casein hydrolysate, 0.01 M_EDTA, l x lO-GM FAD, l x 10—6 3 M FMN or 2 x 10- M_1,3-dimercaptopropanol. In fact, these substances increased the rate of loss of activity. Entire purifications were run anaerobically and in the dark with no amelioration of activity losses. In testing different purification procedures the following resulted in significant losses of ETF activity: precipi- tation with ammonium sulfate, DEAE cellulose chromatography, DEAE Sephadex chromatography, CM Sephadex chromatography, G-100 Sephadex chromatography, hydroxyappatite chroma- tography, calcium phosphate adsorption and concentration by ultrafiltration. C. Acyl CoA Dehydrogenase l. Purification.--Because of its unique green color it was not necessary to use a catalytic assay to purify P. elsdenii ACD. Potassium phOSphate buffer (pH 6) was employed in all steps of the purification procedure presented below unless otherwise indicated. Crude extract from 300 g of cells was mixed with one fourth volume of moist DEAE cellulose, stirred for 10 minutes at 4°C and filtered on Whatman #1 filter paper. The 149 resin, containing ACD, was washed twice with 500 ml of 0.1 M_buffer, once with 500 ml of 0.3 M buffer and applied to the top of a column containing the same quantity of DEAE cellulose equilibrated with 0.3 M buffer. A buffer gradient from 0.3 §.t° 0.8 M buffer was continued until the greenish brown band eluted from the column. This band was precipitated with 70 per cent saturation by ammonium sulfate. The resulting pre- cipitate was dissolved in a minimum volume of water and passed through a 2.5 x 10 cm column of Sephadex G-25 equilibrated at 0.1 M buffer. Following desalting, the sample was applied to a 2.5 x 20 cm column of Sephadex A-SO, washed stepwise with 0.3, 0.5 and 0.8 M buffer. ACD was eluted by a gradient from 0.8 M_buffer to 0.8.M buffer plus 0.5 M_KC1. The dark green band which eluted from the column was brought to 70 per cent saturation with ammonium sulfate, and the resulting precipitate was dis- solved in a minimum volume of 0.05 M buffer (pH 7) and chromatographed on Sephadex G-25 to remove excess salt and bring the sample to 0.05 M_buffer (pH 7). Enzyme prepared in the above manner was two fold purified relative to crude extract when assayed with butyryl CoA and DCPIP (Baldwin and Milligan, 1964). This seemingly small net purification is observed because ETF, which markedly stimulates the DCPIP assay (Baldwin and Milligan, 1964), is present in crude extracts, but is 150 removed by chromatography on DEAE cellulose. The purity of such enzyme is better assessed by polyacrylamide disc gel electrophoresis. Figure 24 shows optical density tracings of a polyacrylamide gel containing ACD. The tracings show two major and one minor protein bands (280 mu), only one of which exhibits any flavin absorption (450 mu). Thus, while ACD is not only about 30 per cent pure on the basis of protein, its flav0protein spectrum arises solely from one protein, ACD. This makes it suit- able for use in Spectral and kinetic eXperiments with the electron transport system. In these experiments molar concentrations were determined from absorbance at 450 mu using an extinction coefficient of 12 A450 L/mmole (Engle, 1970). Correcting for the 30 per cent purity of ACD preparations, the yield from 300 grams of cells is 20-30 mg of pure ACD. ° Yellow ACD is produced from the green form by the addition of solid sodium dithionite to a preparation of green ACD until its color is bleached, followed by chromatography on Sephadex G-25 equilibrated at 0.05 M potassium phosphate buffer (pH 7). This yellow ACD was employed in most of the spectral and kinetic studies involving ACD which are presented in this thesis. The conversion of green ACD to the yellow form did not destroy the catalytic ability of the enzyme in the butyryl CoA- DCPIP assay. 151 Figure 24.--Absorbance tracings of a 220 pg sample of green acyl CoA dehydrogenase. 152 8. b 1.2 8v .3 mozto prOpionate (Baldwin, 1962 and Schneider, 1969) and for the reduction of d,B-unsaturated acyl CoA's during fatty acid synthesis (Tung, 1970). Since ETF accepts electrons from both NADH and LDH(R) it serves to funnel electrons from several sources into unsaturated acyl CoA's which, in the strict anaerobe, E. elsdenii, replace oxygen as the terminal electron acceptor. 157 B. Role of Electron Transport in Lactate Metabolism Figure 25 depicts electron transport in E. elsdenii as it relates to lactate metabolism by the organism. AInitially Q: and M—lactate are interconverted by a lactate racemase (LR) which has been partially puri— fied and characterized (Schneider, 1969). This enzyme enables the organism to grow when Eflactate is the pre— dominant energy source available, such as in starch- enriched cultures of rumen fluid (Baldwin, 1962). Eflactate is oxidized to pyruvate by LDH, generating two reducing equivalents which are transferred via the LDH flavin coenzymes directly to ETF. The pyruvate dehydro- genase complex (PDC) oxidizes pyruvate to CO2 and acetyl CoA (Peel, 1958 and Peel, 1960) and in so doing donates reducing equivalents to ferridoxin (Fd) (Baldwin and Milligan, 1964). Rubredoxin (Rd) (Mayhew and Peel, 1966 and Atherton 3E_El., 1969 and Mayhew and Massey, 1969) has been isolated from P. elsdenii and will serve the same function as ferridoxin. The electrons of reduced ferridoxin may produce hydrogen gas (Hase) or they may reduce NAD via a transhydrogenase (TH) (Valentine eE_§l., 1962). As shown in this research, NADH can then denote electrons to ETF, from which they are transferred to acrylyl CoA via ACD. Crotonyl CoA, and presumably other unsaturated CoA's produced during fatty acid synthesis, are also acceptors of electrons carried from ETF by ACD. 158 .moxoc ac moumOHmGH mum muospoum Hmcflw cam mamfluoume mcfluumum .ommcmmoupmcmm ¢OU Hhom mow Qua mam .cflououmo>maw mcwuuommcmuuucouuooam How mam .mmmcmmoupacmcmuu “Ow ma .ommcmmoupmc mom omm: .cflx0©o>maw How cam ~cflxocmucsu Mom cm .cflxomflunom Mow pm .mmmuommcmuu d ogxmcoou How 8400 .Eoumwm oflcmmaouocmmocm ocu now mum .memEoo ommcomoucwcom opm>summ ocu MOM uom .mmmcoooummcmm ofluomaucnvm new mac .mmmaoomu mumuoma How ma ”mum musmflm mcu ca moms moEmc mamncm How mcofluma>muccd .mflmmflm .m 5 Shamans» 53630 mam 5302365 3363:-.mm 353m 159 a 1E . H.383 flEzoaoms m8... tic/Ls :98 Somali E 2... <8 #52 .511. 1522.1 wmiaooa . _ E 0 Zoo :28 <8 #zomoE 1 mZZEE .18? “ET 2 ISL DESSUQ 10-0..m <8 sit/54141118 #53 of 160 The mechanism of the conversion of lactyl CoA to acrylyl CoA is not known with any certainty. It was thought that a-phospholactyl CoA was an intermediate (Schneider, 1969), but recent experiments in this laboratory do not support this conclusion (Tung, 1970). Coenzyme A transferase, which presumably activates lactate prior to its dehydra- tion, has been purified to homogeneity (Tung, 1970). F. C. Relation Between Phosphoryla- tibn and Electron Transport in Lactate Metabolism Whether or not a phosphorylated intermediate E exists in the lactate to acrylyl CoA conversion is critical in predicting whether phosphorylation should occur concomitant with electron transport. The equations in Figure 6 (see Literature Review) when applied to the fermentation data in Table l predict that if an equivalent of ATP is required in the conversion of lactyl CoA to acrylyl CoA, then the net yield of ATP in the lactate fermentation is -10.8 mmoles of ATP per 100 mmoles of lactate fermented. Clearly, the organism cannot grow with an energy deficit. Alternatively, if ATP is not required for acrylyl CoA formation, an ATP yield of 30 mmoles of ATP/100 mmoles lactate fermented is predicted. This value is close enough to the yield of 50 mmoles of ATP per 100 mmoles of lactate calculated from growth experiments (Gotts, 1970) to suggest that the lactate fermentation alone may be sufficient to explain observed 161 growth. If this is true then the need to postulate an alternate source of ATP, such as electron transport mediated phosphorylation can be eliminated. The results obtained in this research do not support the existence of any electron transport phosphorylation by the reconsti- tuted electron transport system, but an additional w coupling factor(s) may be required for ATP synthesis. Using purified LDH, ETF and ACD with the radioactive assay for phosphorylation, it should be possible to test other protein fractions from P. elsdenii for their ability to couple phosphorylation to electron transport. In addition, the fermentation balance should be redeter- mined under the same conditions as were used by Gotts (1970) to determine molar growth yields. D. Control of Electron Tran5port As ETF is the common intermediate in electron transport during lactate metabolism by g. elsdenii, control of its activity could control lactate metabolism and hence, ATP production. The observed inhibition of electron transport by phosphate buffer at limiting ETF concentration supports this contention and suggests that the controlling factor(s) may be compounds containing phosphate. Likewise, the stimulation by inorganic phos- phate and by perphosphate compounds of the ferricyanide assay for LDH suggests possible control at this point in the chain. Because P. elsdenii is an anaerobe and must 162 carry out tightly coupled oxidation-reduction reactions involving organic donors and acceptors, control of LDH activity would control ATP production by controlling the availability of pyruvate, the substrate for the only net thioester bond-producing reaction in the metabolic scheme. With the development of procedures to produce purified electron transport enzymes, the tools are now available 5 for more intensive investigations into the control of lactate metabolism by this organism. Z E. Physical State of the L Electron Transport System In mitochondria electron transport and phosphoryla- tion enzymes are normally membrane bound, but in g. elsdenii the enzymes investigated in this research and all other enzymes shown in Figure 25 are soluble. Upon disruption either by sonication or pressure and subsequent centrifugation the enzyme activities all appear in the supernatant with virtually no activity in the pellet. Furthermore, the interactions between the components exhibit saturation kinetics which are characteristic of soluble enzymes. F. ‘Physiological Relevance of the Reconstituted Electron Transport System To determine the physiological relevance of the electron transport measurements presented in this thesis, calculations were made to compare the rate of product 163 formation by whole cells with the rate of ACD reduction by the reconstituted electron transport system. For each valerate and prOpionate produced by whole cells, a pair of electrons must pass through ETF and ACD. Thus, the specific activity of product formation can be normalized to effective ETF concentration and can be compared to the rate of ACD reduction in the cuvette assay, normalized to the same ETF concentration. ETF was chosen as the normalization factor because it is rate limiting in the cuvette assay. Dividing the minimum specific activity of WE“ combined valerate and prOpionate formation (Schneider, 1969) by the specific activity of ETF in crude extracts (4 units/mg), the minimum specific activity for electron transport in whole cells is 17.1 x 10.6 mmoles per minute per unit of ETF. From the experiment which demonstrated, the dependence of the ETF assay upon ETF concentration, the rate of ACD reduction is 40 A450 per minute per mg of ETF. Dividing this value by the extinction of ACD (12 L x A450/mmole) and by the specific activity of the ETF (88 units/mg) and multiplying the dividend by the volume of the assay (2 x lO-4L) gives a specific activity of ACD reduction of 7.6 x 10.6 mmoles per minute per unit of ETF.3 This value compares quite favorably with the minimum specific activity of 17.1 x 10.6 mmoles per ‘40 A450) (1 mmole) (1 mg ETF) 3 (min.)(mg ETF)X (I2L)(A450)X T88 units ETF) 2 x lO-4L x 164 minute per unit of ETF calculated for whole cells. These results suggest that the experiments and conclusions presented in this thesis are, indeed, physiologically relevant to lactate metabolism by cultures of E. elsdenii. Part II. Properties of Electron Transport Enzymes A. D(-)-Lactic Dehydrogenase f The LDH of g. elsdenii is similar to the LDH isolated from anaerobically grown yeast by Iwatsubo and 1 Curdel (1961). It is specific for Dflactate and 2,Efa- “7"" " "' ' 'qrhw hydroxybutyrate (presumably the 27 isomer) but shows a broad acceptor specificity. Dyes, cytochrome c, oxygen and ETF will all serve as acceptors, but NAD and 3- acetylpyridine DPN will not. It is in acceptor speci- ficity that the LDH from P. elsdenii differs from the enzyme from anaerobic yeast (Cremona and Singer, 1964). The latter enzyme will denote electrons only to cytochrome c. Lack of reactivity toward pyridine nucleotides is characteristic of flav0protein lactic dehydrogenases, as is the FAD coenzyme and a metal cofactor. Divalent zinc reactivates E. elsdenii LDH better than cobalt or manganese, but mroe detailed studies with large quanti- ties of pure enzyme are necessary before the identity of the metal cofactor normally present can be definitely determined. Zinc is known to be the metal cofactor of the aerobic yeast LDH (Cremona and Singer, 1964) and it 165 reactivates the enzyme from anaerobically grown yeast (Iwatsubo and Curdel, 1961). The most unusual pr0perty of the LDH is that substrate binding is negatively cooperative and that the Km for Qflactate is extremely high. If an allosteric effector for LDH is eventually found, it will probably function by reversing this negative COOperativity and r thereby lower the Km for Delactate. As mentioned above, control of LDH activity, possibly by changing its Km, may well constitute the main means of control of lactate metabolism in B. elsdenii. B. Electron-Transferring Flavoprotein ETF was not fully characterized. Further studies should be undertaken to determine which of the two flav0proteins in ETF preparations contains the atypical flavin and to identify this flavin cofactor. Mammalian ETF which has been more extensively purified contains FAD as a cofactor but still exhibits a spectrum similar to that shown by ETF preparations from E. elsdenii. Thus, although ETF from g. elsdenii may contain an unusual flavin cofactor, environmental factors might still cause the unusual spectrum with FAD as the cofactor. C. Acyl CoA Dehydrogenase The flav0protein spectrum exhibited by ACD is atypical relative to other flav0proteins but is typical 166 when compared to green ACD from other sources. The origin of the green color is not fully understood, but is thought to be connected with an unusual acyl CoA bound to the enzyme (Engle, 1970). SUMMARY The electron transport system between Eflactate and c,B-unsaturated acyl CoA is composed of three flavo- proteins, 2(-)-lactic dehydrogenase, electron-transferring flavoprotein and acyl CoA dehydrogenase. Electrons from NADH enter the chain via the electron-transferring flavo- protein. The reconstructed electron transport system will not catalyze the phosphorylation of either AMP or ADP concomitant with electron transport. Kinetic para- meters were determined for the interactions of the com- ponents in the system and kinetic and physical prOperties of the individual enzymes were measured. 167 LITERATURE CITED 168 (In LITERATURE CITED Allison, M. J., Bucklin, J. A., and Robinson, I. M. 1966. Appl. Microbiol. 94:807. Atherton, N. M., Garbett, K., Gillard, R. D., Mason, R., Mayhew, S. G., Peel, J. L., and Stangroom, J. E. 1966. Nature. 212:590. Baldwin, R. L. 1962. Unpublished Ph.D. thesis, Michigan State University. Baldwin, R. L., and Milligan, L. P. 1964. Biochim. Biophys. Acta.‘ 99:421. Baldwin, R. L., Wood, w. A., and Emery, R. s. 1965. Biochim. Biophys. Acta. 91:202. Barker, H. A. 1937. Arch. Mikrobiol. 9:415. Baseman, J. B., and Cox, C. D. 1969. J. Bacteriol. 91:992. Bauchop, T., and Elsden, S. R. 1960. J. Gen. Microbiol. 23:457. Baum, R. H., and Dolin, M. I. 1963. J. Biol. Chem. 238:P.C. 4109. Beechey, R. B., Holloway, C. T., Knight, I. G., and Robertson, A. M. 1966. Biochem, Biophys, Res. Comm. 23:75. Beinert, H. 1962. In Methods in Enzymology: 5 ed., p. 5. S. P. Colowick and N. P. Kaplan. Academic Press, Inc., New York. Beinert H. 1963. In The Enzymes, 7 ed., p. 447. P. D. Boyer, H. Lardy, and K. Myrback. Academic Press, Inc., New York. Beinert, H. 1963a. In The Enzymes, 7 ed., p. 467. P. D. Boyer, H. Lardy, and K. Myrback. Academic Press, Inc., New York. Beinert, H., and Frisell, W. R. 1962. J. Biol. Chem. 237:2988. 169 170 Bornstein, B. T., and Barker, H. A. 1948. J. Biol. Chem. 172:659. Brown, E. B., Jr., and Stadtman, E. R. 1958. Unpublished observation cited in Stadtman, 1966. Bryant, M. P., and Robinson, I. M. 1961. Appl. Microbiol. 9:91. Conn, H. J. (ed.). 1957. Manual of Microbiological Methods, p. 16. McGraw-Hill Book Co., Inc., New York. Crane, F. L., and Beinert, H. 1954. J. Am. Chem. Soc. 19:4491. Crane, F. L., and Beinert, H. 1956. J. Biol. Chem. 218:717. Crane, F. L., Mii, S., Hauge, J. C., Green, D. E., and Beinert, H. 1956. J. Biol. Chem. 218:701. "5': - Cremona, T., and Singer, T. P. 1964. J. Biol. Chem. 239:1466. Cronin, J. R., Frisell, W. R., and Mackenzie, C. G. 1967. In Methods in Enzymology, 99, p. 302. Ed. R. W. Estabrook andTM. E. PuIIman. Academic Press, Inc., New York. Curdel, A. 1966. Biochem. Biophys. Res. Comm. 99:357. Davis, B. J. 1964. Annals N.Y. Acad. Sci. 121:404. Deibel, R. H., and Kvetkas, M. J. 1964. J. Bacteriol. 88:858. DeMoss, R. D., Bard, R. S., and Gunsalus, I. C. 1951. J. Bacteriol. 99:499. Dolin, M. I. 1955. Arch. Biochem. Biophys. 99:415. Elsden, S. R., and Lewsi, D. 1953. Biochem. J. 99:183. Elsden, S. R., Volcani, B. E., Gilchrist, F. M. C., and Lewis, D. 1956. J. Bacteriol. 19:681. Engles, P. 1970. Private communication. Ernster, L., Danielson, L., Ljunggren, M. 1962. Biochim. Biophys. Acta.‘ 99:189. 171 and Mackenzie, C. G. 1962. Frisell, W. R., Cronin, J. R., :2975. J. Biol. Chem. 237 Frisell. W. R., Cronin, J. R., and Mackenzie, C. G. 1966. In Flavins and Flavoproteins, p. 367. Ed. E. C. Slater, Elsevier, Amsterdam. ' Gallin, J. I., and VanDemark, P. J. 1964. Biochem. BiOphys. Res. Comm. 91:630. Gaston, L. W., and Stadtman, E. R. 1963. J. Bacteriol. 85:356. Gelbard, A., and Goldman, D. S. 1961. Arch. Biochem. Biophys. 99:228. Gotts, J. W. 1970. Private communication. Green, D. B., Mii, S., Mahler, H. R., and Bock, R. M. 1954. J. Biol. Chem. 206:1. " Gregolin, C., and D'Alberton, A. 1964. Biochem. Biophys. Res. Comm. 99:103. Gregolin, C., Scalella, P., and D'Alberton, A. 1964. Nature. 204:1302. Gregolin, C., and Singer, T. P. 1962. Nature. 1 3:659. Gregolin, C., and Singer, T. P. 1963. Biochim. Bi0phys. Acta. 91:201. Gunsalus, I. C., and Shuster, C. W. 1961. In The Bacteria, 2. Ed. I. C. Gunsalus and R. Y. Stanier. Academic Press, Inc., New York. Gutierrez, J., Davis, R. E., Lindahl, I. L., and Warwick, E. J. 1956. Appl. Microbiol. 1:16. Hardman, J. K., and Stadtman, T. C. 1960. J. Bacteriol. 79:544. Hardman, J. K., and Stadtman, T. C. 1963. J. Biol. Chem. 238:2081. Hardman, J. K., and Stadtman, T. C. 1963a. J. Biol. Chem. 238:2088. Hardman, J. K., and Stadtman, T. C. 1963b. J. Bacteriol. 99:1326. 172 Hatefi, Y., Haavik, A. G., Fowler, L. R., and Griffiths, D. E. 1962. J. Biol. Chem. 237:2661. Hauge, J. G. 1956. J. Am. Chem. Soc. 99:5266. Hauge, J. G., Crane, F. L., and Beinert, H. 1956. J. Biol. Chem. 219:727. . Heytler, P. B. 1963. Biochemistry. 9:357. Hoskins, D. D. 1966. J. Biol. Chem. 241:4472. Hoskins, D. D., and Bjur, R. A. 1965. J. Biol. Chem. 240:2201. Hoskins, D. D., and Mackenzie, C. G. 1961. J. Biol. Chem. 236:177. Iwatsubo, G., and Curdel, A. 1961. Biochem. Biophys. Res. Comm. 9:385. "wc.‘=" '.' - . '..-'-«'. 'l Jacobs, N. 0., and Vandemark, P. J. 1960. J. Bacteriol. -74:532. ' Kilgour, G. L., Felton, S. P., and Huennekens, F. M. 1957. J. Am. Chem. Soc. 12:2254. Koshland, D. E., Jr., Nemthy, G., and Filmer, D. 1966. Biochemistry. 9:365. Ladd, J. N., and Walker, D. J. 1959. Biochem. J. Z9;364. Ladd, J. N., and Walker, D. J. 1965. Annals N.Y. Acad. Sci. 119:1038. Lee, H. A., Jr., and Ables, R. H. 1963. J. Biol. Chem. 238:2367. Levitzki, A., and Koshland, D. E., Jr. 1969. Proc. Nat. Acad. Sci. (USA). 99:1121. Lewis, D., and Elsden, S. R. 1955. Biochem. J. 60:683. Mayhew, S. G., Foast, G. P., and Massey, V. 1969. J. Biol. Chem. 244:803. Mayhew, S. G., and Massey, V. 1969. J. Biol. Chem. 244:794. Mayhew, S. G., and Peel, J. L. 1966. Biochem. J. 00:8OP. Mell, G. P., and Huennekens, F. M. 1960. Fed. Proc. 99:411. 173 Mickelson, M. H. 1969. J. Bacteriol. 100:895. Mizushima, S., and Kitahara, K. 1962. J. Gen. Appl. Microbiol. 99:214. Molinari, R., and Lara, F. J. S. 1960. Biochem. J. 19357. Monod, J. 1942. Recherches Sur Las Croissance Des Cultures Bacterienne. Herman, Paris. Nisman, B. 1954. Bacteriol. Rev. 99:16. Nygaare, A. P. 1961. Acta Chim. Scand. 99:1627. Painter, A. A., and Hunter, F. E., Jr. 1970. Science. 170:552. Pascal, M. C. 1966. Bull. Soc. Fr. Physiol. Veg. 99:123. Peel, J. L. 1958. Biochem. J. 70:7P. Peel, J. L. 1960. Biochem. J. 74:525. Randerrath, K., and Randerrath, E. 1967. In Methods in Enzymology, 9, p. 323. Ed. C. H. W. Hirs. Academic Press, Inc., New York. Roy, B. L. 1964. Nature. 201:80. Schneider, D. L. 1969. Unpublished Ph.D. thesis, Michigan State University. Seubert, W., and Lynn, F. 1953. J. Am. Chem. Soc. 99:2787. Shuster, C. W., and Gunsalus, I. C. 1958. Fed. Proc. 99:310. Smalley, A. J., Jahrling, P., and VanDemark, P. J. 1968. J. Bacteriol. 99:1595. Snoswell, A. M. 1966. In Methods in Enzymology, 9, p. 321. Ed. W. A. Wood. Academic Press, Inc., New York. Stachiewicz, B., Labeyrie, F., Curdell, A., and Sloninski, P. P. 1961. Biochim. Biophys. Acta. 99:45. Stadtman, E. R. 1957. In Methods in Enzymology, 9, p. 932. Ed. S. P. Colowick and N. 0. Kaplan. Academic Press, Inc., New York. Stadtman, E. R. 1966. In Current Aspects of Bioenergetics. Ed. N. 0. Kaplan and—E. P. Kennedy. Academic Press, Inc., New York. 174 Stadtman, T. C. 1962. Acta Biochim. Biophys. 99:36. Stadtman, T. C. 1965. In Non-Heme Iron Proteins, p. 439. Ed. A. San Pietro. Antioch Press, Yellow Springs, Ohio. Stadtman, T. C. 1966. Acta Biochim. BiOphys. 113:9. Stadtman, T. C., and Elliott, P. 1956. J. Am. Chem. Soc. 99:2020. Stadtman, T. C., Elliott, P., Tiemann, L. 1958. J. Biol. Chem. 231:961. Stern, J. R. 1955. In Methods in Enzymology, 9, p. 559. Ed. S. P. ColowiEE and N. O. Kaplan. Academic Press, Inc., New York. Steyn-Parve, E. P. and Beinert, H. 1958. J. Biol. Chem. 233:853. Symons, R. H., and Burgoyne, L. A. 1967. In Methods in Enzymology, 9, p. 314. Ed. W. A. Wood. Academic Press, Inc., New York. Thauer, R. K., Jungermann, K., Henninger, H., Henning, W., and Decker, K. 1968. J. Biochem.. 4:173. Tung, K. K. 1970. Private communication. Valentine, R. C., Jackson, R. L., and Wolfe, R. S. 1962. Biochem. Biophys. Res. Comm. 9:453. Warburg, 0., and Christian, W. 1942. Biochim. Z. 310:384. Wittenberger, C. L., and Haaf, A. S. 1966. Biochim. BiOphys. Acta. 99:393. m4!)l)(l)(ll)(|1(1“WWW“WINNIE 3 1293 03082 0355