'M'^i^'. iiii m^0. ^^;; i^-^-- 1 MM ] Marine Biological Laboratory Library [[ B Woods Hole, Mass. ^ [0 c: I I I Presented by [ Dr. Philip Person Feb. 26, 1962 HABMATIN ENZYMES l.U.B. Symposium Series Volume 19 INTERNATIONAL UNION OF BIOCHEMISTRY SYMPOSIUM SERIES Vol. 1 . The Origin of Life on the Earth— k. I. Oparin et al. (Editors) Vol. 2. Enzyme Chemistry: Proceedings of the International Symposium in Tokyo- Kyoto PROCEEDINGS OF THE FOURTH INTERNATIONAL CONGRESS ON BIOCHEMISTRY VIENNA, 1958 (I) Carbohydrate Chemistry of Substances of Biological Interest (II) Biochemistry of Wood (III) Biochemistry of the Central Nervous System (IV) Biochemistry of Steroids (V) Biochemistry of Antibiotics (VI) Biochemistry of Morphogenesis (VII) Biochemistry of Viruses (VIII) Proteins (IX) Physical Chemistry of High Polymers of Biological Interest (X) Blood Clotting Factors (XI) Vitamin Metabolism (XII) Biochemistry of Insects (XIII) Colloquia (XIV) Transactions of the Plenary Sessions (XV) Biochemistry Vol. 18. Biochemistry of Lipids — G. Popjak (Editor) Vol. 19. Haematin Enzymes (Parts I and 2)— J. E. Falk, R. Lemberg and R. K. Morton Vol. 20. Report of the Commission on Enzymes, 1961 (I.U.B.) proceedings of the FIFTH INTERNATIONAL CONGRESS ON BIOCHEMISTRY MOSCOW, 1961 {Provisional titles) (I) Biological Structure and Function at the Molecular Level (II) Functional Biochemistry of Cell Structures (III) Evolutionary Biochemistry (IV) Molecular Basis of Enzyme Action and Prohibition (V) Intracellular Respiration: Phosphorylating and Non-Phosphorylating Systems (VI) Mechanism of Photosynthesis (VII) Biosynthesis of Lipids (VIII) Biochemical Principles of the Food Industry (IX) Transactions of the Plenary Sessions (X) Abstracts of Papers and Indexes to the Volumes of the Proceedings Vol. 3. Vol. 4. Vol. 5. Vol. 6. Vol. 7. Vol. 8. Vol. 9. Vol. 10. Vol. 11. Vol. 12. Vol. 13. Vol. 14. Vol. 15. Vol. 16. Vol. 17. Vol. 21. Vol. 22. Vol. 23. Vol. 24. Vol. 25. Vol. 26. Vol. ,27. Vol. ,28. Vol. ,29. Vol. ,30. HAEMATIN ENZYMES A SYMPOSIUM OF THE INTERNATIONAL UNION OF BIOCHEMISTRY ORGANIZED BY THE AUSTRALIAN ACADEMY OF SCIENCE CANBERRA 1959 ^^^j J. E. O'Hagan 173 Discussion Native globin . . . . . . . . .190 The Linkage of Iron and Protein in Haemoglobin . . . .190 Early Stages in the Metabolism of Iron by J. B. Neilands 194 The Enzymic Incorporation of Iron into Protoporphyrin by R. A. Neve 207 Discussion The Formation of Metal-Porphyrin Complexes . . . . .211 Co-ordination of Divalent Metal Ions with Porphyrin Derivatives Related to Cytochrome c by J. B. Neilands 211 Metal Incorporation in Model Systems . . . . . .214 On the Enzymic Incorporation of Iron . . . . . .215 Biosynthesis and Metabolism of Cytochrome c by D. L. Drabkin 216 The Location of Cytochromes in Escherichia coli by A. TissiERES ......... 218 Discussion The Origin of the Respiratory Granules of Bacteria .... 223 On the Cytochromes of Anaerobically Cultured Yeast by Paulette Chaix Discussion The Lactate Dehydrogenase of Yeast ...... 233 ' Components of the Respiratory Chain in Yeast Mitochondria . . 233 On the 'Haemoglobin'' Absorption Band o I Yeast .... 233 viii CONTENTS PAGE Irreversible Inhibition of Catalase by the 3-Amino-l:2:4-Triazole Group of Inhibitors in the Presence of Catalase Donors by E. Margoliash and A. Schejter 236 Catalase Oxidation Mechanisms by M. E. WiNFiELD 245 Discussion Oxidation States of Haemoproteins ...... 252 Peroxide Compounds of Catalase and Peroxidase .... 254 The Nature of Catalase-Peroxide Complex I by B. Chance 254 Studies on Problems of Cytochrome c Oxidase Assay by LuciLE Smith and Helen Conrad 260 Discussion Assay of Cytochrome c Oxidase ....... 275 Inhibition of Cytochrome c Oxidase by Cytochrome c . . . . 275 Interaction of Cytochrome c with Other Compounds . . . . 276 The Effect of Cations on the Reactivity of Cytochrome c in Heart Muscle Preparations by R. W. Estabrook 276 Composition of Cytochrome c Oxidase by W. W. Wainio 281 Discussion Function of Copper in Cytochrome Oxidase Preparations . . .301 Cytochrome Oxidases of Pseudomonas aeruginosa and Ox-Heart Muscle and Their Related Respiratory Components by T. HoRio, I. Sekuzu, T. Higashi and K. Okunuki . . . 302 DisciLssion Properties and Nomenclature of Cytochromes di and z^ . . .311 The Prosthetic Groups of Pseudomonas Cytochrome Oxidase by T. Horio and M. D. Kamen 314 The Reaction of Cytochrome c Oxidase with Oxygen . . . .316 The Oxygen-Reducing Equivalents of Cytochromes a and 03 B. Chance .......... 316 CONTENTS IX PAGE The Isolation, Purification and Properties of Haemin a by D. B. MoRELL, J. Barrett, P. Clezy and R. Lemberg . . 320 Discussion Model Systems for Cytochrome Oxidase . . . . . .330 Absorption Spectra of Ferro- and Ferri-Compounds of Haem a by R. Lemberg ......... 330 Cryptohaem a ........ . 333 Mitochrome in Relation to Cryptohaem a .... . 334 Cytochrome Oxidase Components by M. Morrison and E. Stotz 335 The Structure of Porphyrin a, Cryptoporphyrin a and Chlorin flg by R. Lemberg, P. Clezy and J. Barrett 344 Discussion The Structure of Haem & and Haem di2 ...... 358 The Structure of Porphyrin a by M. Morrison ......... 358 The Properties of Haem a^ and Cytochrome Oj by J. Barrett and R. J. P. Williams 360 Extractability of Ferro- and Ferricytochrome c .... 361 A Haemopeptide from a Tryptic Hydrolysate of Rhodospirilluin rubrum Cytochrome c by S. Paleus and H. Tuppy 363 Electrometric and other Studies on Cytochromes of the C-Group by R. W. Henderson andW. A. Rawlinson . . . .370 Discussion Protein Configuration and Linkage to the Prosthetic Group in Cytochrome c 382 Studies of the Haemochrome-forming Groups in Cytochrome c by E. Margoliash ........ 382 The Amino Acid Sequence in Horse Heart Cytochrome c by E. MargoUash and R. Hill 383 < Comments on the Structure of Cytochrome c . . . . . 384 Structure and General Properties of Cytochrome c . . . .385 Comparative Properties of Cytochrome c from Yeast and Heart Muscle by J. McD. Armstrong, J. H. Coates and R. K. Morton Properties of Native Cytochrome c . . . . Reactivity of Native Cytochrome c itt Oxidative Phosphorylation Structure of Bacterial Cytochromes of c-Type . Structure and Redox Potentials of Cytochrome c 385 388 389 389 390 The Electron Transfer from Cytochromes to Terminal Electron Acceptors in Nitrate Respiration and Sulphate Respiration by F. Egami, M. Ishimoto and S. Taniguchi .... 392 Cytochrome Cg by J. PosTGATE 407 Discussion Nature and Properties of Cytochrome Ca . . . . .414 Functional Aspects of Cytochrome Cz . . . . . .415 Evolutionary Aspects of the Sulphate-reducing Bacteria and of Cytochrome Cj 416 The Atypical Haemoprotein of Purple Photosynthetic Bacteria /?7 M. D. Kamen a/2^ R. G. Bartsch 419 Discussion The Functions of Cytochrome b, and of Cytochrome c*^' (Halotolerant Coccus) . . . . . .432 Nomenclature of CO-binding Pigments ...... 432 Cytochrome o by B. Chance 433 On the Oxidase Function of RHP . . . . . .435 Spectrophotometric Studies of Cytochronics Cooled in Liquid Nitrogen by R. W. EsTABROOK ........ 436 Discussion 'Trapped' Steady-states by B. Chance .......... 457 Low-teinpcrature Absorption Spectra of Cytochromes in Relation to Structure ......... 458 CONTENTS XI PAOB Studies on Microsomal Cytochromes and Related Substances by C. F. Strittmatter 461 Discuss ion On the Nature of Cytoplasmic Pigments of Liver Cells by B. Chance 473 Possible Functions of the Cytochromes of the Endoplasmic Reticulum of Animal Cells 476 The Significance of E^ Values of Cytochromes in Relation to Cellular Function ......... 477 The Cytochromes of Plant Tissues by W. D. Bonner, Jr 479 Discussion Cytochromes c^ and b^ of Particulate Components of Plants by R. K. Morton 498 The Cytochromes of Roots ....... 499 The Chemical and Enzymic Properties of Cytochrome b^ of Bakers' Yeast by R. K. Morton, J. McD. Armstrong and C. A. Appleby . 501 Conditions for the Autoxidation of Flavocytochrome b^ by E. BoERi and M. Rippa 524 Kinetic Studies on the Action of Yeast Lactate Dehydrogenase by H. Hasegawa and Y. Ogura 534 Various Forms of Yeast Lactate Dehydrogenase by A. P. Nygaard 544 Studies on Bakers' Yeast Lactate Dehydrogenase by T. Horio, J. Yamashita, T. Yamanaka, M. Nozaki and K. Okunuki 552 Discussion The Problem of Cytochrome hi . . . . . . .558 Properties of Intact and Modified Cytochrome b2 . . . .560 Nature of Bakers' Yeast Lactate Dehydrogenase by T. Horio 560 Nomenclature of Cytochrome bj and Derived Proteins . . . 562 The Substrate Specificity of Cytochrome bz . . • • • 563 On the Cytochrome b Cotnponents in the Respiratory Chain in Yeast . 564 The Kinetics of Reactions Catalysed by Cytochrome hi ■ • ■ 565 The Kinetics of Reduction of Cytochrome b^ by B. Chance 565 The Absorption Spectrum in Relation to the Structure of Cytochrome bj 565 The Contribution of the Prosthetic Groups to the Absorption Spectrum of Cytochrome bt by R. K. Morton 567 The Oxidation-reduction Changes in the Reaction of Lactate with Cytochrome by E. Boeri and E. Cutolo 568 The Function and Bonding of the Flavin Group of Cytochrome hi . .569 The Bonding between the Flavin Group and Apoprotein of Cytochrome bi by J. McD. Armstrong, J. H. Coates and R. K. Morton . .569 Possible Free Radical Formation in Flavoproteins by G. D. Ludwig 572 Autoxidation of Cytochrome hi ....... 573 The Role of Cytochrome h in the Respiratory Chain by E. C. Slatcr and J. P. Colpa-Boonstra .... 575 Discussion The Cross-over Theorem and Sites of Oxidative Phosphorylation . . 592 The Oxidation-reduction Potential of Cytochrome b . . . . 593 On the Redox Potential of Cytochrome h, the Kinetics of Reduction of Cytochrome b and the Existence of Slater's Factor . . . 593 The Influence of Cyanide on the Reactivity of Cytochrome h . . 596 Energy Transfer and Conservation in the Respiratory Chain by B. CiiANcn .... 597 Discussion Mechanism of Oxidative Phosphorylation ..... 622 CONTENTS Xlll PAOB The Significance of Respiratory Chain Oxidations in Relation to MetaboHc Pathways in the Cell by F. Dickens 625 Discussion On ''Additional DPN' of Incubated Mitochondria .... 636 Possible Structure of Complexes of DPN or of DPNH involved in Oxidative Phosphorylation ........ 637 Author Index 641 Subject Index ......... 653 PARTICIPANTS Dr. C. a. Appleby Mr. J. McD. Armstrong Mr. J. Barrett Dr. N. K. Boardman Professor E. Boeri* Dr. W. D. Bonner Mme. Professor P. Chaix Professor B. Chance Dr. p. S. Clezy Dr. E. Cutolo Professor F. Dickens Professor D. L. Drabkin Professor F. P. Dwyer t * Died 1960. Biochemistry Section, Division of Plant Industry, C.S.I.R.O., Canberra, Australia. Department of Agricultural Chemistry, Waite Agricultural Research Institute, University of Adelaide, Adelaide, Austraha. Institute of Medical Research, Royal North Shore Hospital, Sydney, Australia. Biochemistry Section, Division of Plant Industry, C.S.I.R.O., Canberra, Australia. Institute of Human Physiology, University of Ferrara, Ferrara, Italy. Johnson Research Foundation, University of Pennsylvania, Pliiladelphia, U.S.A. Laboratory of Biological Chemistry, Univer- sity of Paris, Paris, France. Johnson Foundation for Medical Physics, University of Pennsylvania, Philadelphia, U.S.A. Institute of Medical Research, Royal North Shore Hospital, Sydney, Australia. Italian Serum Research Institute, Naples, Italy. Courtauld Institute of Biochemistry, Middle- sex Hospital Medical School, University of London, London, England. Department of Biochemistry, Graduate School of Medicine, University of Pennsylvania, Philadelphia, U.S.A. The John Curtin School of Medical Research, Australian National University, Canberra, Australia. XVI Professor F. Egami Dr. R. W. Estabrook Dr. J. E. Falk Professor P. George Dr. R. W. Henderson Dr. T. Horio Professor M. D. Kamen Professor K. Kaziro Mr. J. W. Legge Dr. R. Lemberg, Prof. a. D. (Heidelberg) Mr. W. H. Lockwood Dr. G. D. Ludwig Dr. E. Margoliash Dr. D. B. Morhll Dr. M. Morrison PARTiaPANTS Department of Biophysics and Biochemistry, Faculty of Science, University of Tokyo, Tokyo, Japan. Johnson Foundation for Medical Physics, University of Pennsylvania, Philadelphia, U.S.A. Biochemistry Section, Division of Plant Industry, C.S.I.R.O., Canberra, Australia. John Harrison Laboratory of Chemistry, University of Pennsylvania, Pliiladelphia, U.S.A. Department of Biochemistry, University of Melbourne, Melbourne, Australia. Graduate Department of Biochemistry, Brandeis University, Walthara, U.S.A. Graduate Department of Biochemistry, Brandeis University, Waltham, U.S.A. Biochemical Laboratory, Nippon Medical School, Tokyo, Japan. Department of Biochemistry, University of Melbourne, Melbourne, Australia. Institute of Medical Research, Royal North Shore Hospital, Sydney, Australia. Institute of Medical Research, Royal North Shore Hospital, Sydney, AustraUa. Hospital of the University of Pennsylvania, Philadelphia, U.S.A. Laboratory for the Study of Hereditary and Metabolic Disorders, College of Medicine, University of Utah, Salt Lake City, U.S.A. Institute of Medical Research, Royal North Shore Hospital, Sydney, Australia. Department of Biochemistry, School of Medicine and Dentistry, University of Rochester, Rochester, U.S.A. PARIICIPANTS Professor R. K. Morton Dr. F. J. Moss Professor J. B. Neilands Dr. R. a. Neve Dr. a. p. Nygaard Dr. Y. Ogura Dr. J. E. O'Hagan Dr. L. E. Orgel Dr. S. Paleus Dr. D. D. Perrin Dr. J. N. Phjllips Dr. J. Postgate Assoc. Professor W. A. Rawlinson Professor E. C. Slater Dr. L. Smith Department of Agricultural Chemistry, Waite Agricultural Research Institute, University of Adelaide, Adelaide, Australia. School of Biological Sciences, University of New South Wales, Sydney, Australia. Department of Biochemistry, University of California, Berkeley, U.S.A. Department of Biochemistry, University of California, Berkeley, U.S.A. Nutrition Institute, Blindern, Oslo, Norway. Department of Biophysics and Biochemistry, Faculty of Science, University of Tokyo, Tokyo, Japan. Red Cross Blood Transfusion Service, Brisbane, Australia. University Chemical Laboratory, University of Cambridge, Cambridge, England. Biochemistry Department, Nobel Institute of Medicine, Stockholm, Sweden. Department of Medical Chemistry, John Curtin School of Medical Research, Australian National University, Canberra, Australia. Biochemistry Section, Division of Plant Industry, C.S.I.R.O., Canberra, Australia. Microbiological Research Establishment, Porton, Wiltshire, England. Department of Biochemistry, University of Melbourne, Melbourne, Australia. Laboratory of Physiological Chemistry, University of Amsterdam, Amsterdam, The Netherlands. Department of Biochemistry, Dartmouth Medical School, Hanover, New Hampshire, U.S.A. PARTICIPANTS Dr. C. F. Strittmatter Dr. a. Tissieres Dr. p. Trudinger Professor S. F. Velick Professor W. W. Wainio Professor J. H. Wang Dr. R. J. P. Williams Dr. M. E. Winfield Department of Biological Chemistry, Harvard Medical School, Boston, U.S.A. The Biological Laboratories, Harvard Univer- sity, Cambridge, U.S.A. Biochemistry Section, Division of Plant Industry, C.S.I. R.O., Canberra, AustraUa. School of Medicine, Washington University, St. Louis, Missouri, U.S.A. Bureau of Biological Research and Depart- ment of Physiology and Biochemistry, Rutgers, The State University, New Brunswick, New Jersey, U.S.A. Department of Chemistry, Yale University, New Haven, U.S.A. Inorganic Chemistry Laboratory, and Wad- ham College, Oxford, England. Division of Physical Chemistry, Indus- trial Chemical Laboratories, C.S.I.R.O., Melbourne, Australia. A HAEMOPEPTIDE FROM A TRYPTIC HYDROLYSATE OF RHO DO SPIRILLUM RUBRUM CYTOCHROME C* By S. Paleus and H. Tuppy Mcdicimka Nobelinstitutet, Biokemiska Avdelningen, Stockholm, and Zweites chemisches Institut der Universitdt, Wien For the past few years the structure of cytochrome c has been studied with the aim of relating the catalytic activity of this haemoprotein to certain characteristic features of its chemical architecture. Much attention has been given to that portion of the protein moiety which adjoins the prosthetic group and which is supposed to contribute to a catalytically active site. In previous investigations (Tuppy and Bodo, 1954; Tuppy and Paleus, 1955) the sequence of twelve amino-acid residues, including two haem-bound cysteine residues, has been elucidated and found to be identical in the cytochromes of three different mammalian species (beef, horse and pig). When, however, the homologous amino-acid sequences were also determined in cytochrome c from fish (salmon), bird (chicken) (Tuppy and Paleus, 1955), insect (silk- worm) (Tuppy, 1957) and from a fungus (yeast) (Tuppy and Dus, 1958), they appeared to be dissimilar though resembling each other more or less closely. The present communication will extend this comparative structural investigation to the c-type cytochrome produced by the photo- synthetic bacterium Rhodospirillum rubrum, first described and subsequently studied in more detail by Kamen and his co-workers (Vernon and Kamen, 1954; Kamen and Vernon, 1955; Bartsch and Kamen, 1958). There is a twofold interest in comparative structural work of this kind. In the first place, one may suppose that only those features which turn up invariably in cytochromes c from different sources are likely to be essential to the specific catalytic function, whereas structural differences will indicate points not directly concerned with catalytic activity. Secondly, the species similari- ties and dissimilarities as revealed by comparative investigations of amino- acid sequences may give a clue to the genetic relation between the organisms from which the proteins studied have been obtained. * Reprinted from Acta Chemica Scandinavica, 13, 641, 1959. 363 364 S. Paleus and H. Tuppy MATERIAL AND METHODS Preparation of Rhodospirillum Rubrum Cytochrome c Lyophilized bacterial cells were extracted with H2SO4 at pH 4. From the extract, the cytochrome c was obtamed by fractionation with ammonium sulphate (KeiUn and Hartree, 1937), dialysis and lyophihzation. A sample thus prepared was kindly put at our disposal by Professor M. D. Kamen. It was further purified by chromatography on a column of CM-W cellulose (1-6 X 10 cm) prepared according to Peterson and Sober (1956) and buffered with 0-015 M ammonium acetate of pH 6-5. The preparation of cytochrome c (470 mg) was dissolved in 2 ml of this buffer, reduced by the addition of solid Na2S204 and transferred to the column. A dark brown, fast movmg band separated from the red cytochrome band which moved slowly, if at all. When the buffer of pH 6-5 was replaced by 0-05 m ammonium acetate of pH 7-5, the red material started moving down the column and divided into three bands. The main fraction (R = 0-4, E270/E551 = 0-96) yielded 71-5 mg of dry, salt-free cytochrome c which was used for all subsequent experiments. Preparation of the Haemopeptide and Removal of the Haem After hydrolysis of the bacterial cytochrome c with trypsin, the haemo- peptide formed could be precipitated from the digest with ammonium sulphate and purified by adsorption on talc and by column partition chro- matography as described previously for other haemopeptides (Tuppy and Bodo, 1954; Tuppy and Paleus, 1955). The haemin part of the haemo- peptide was split from the peptide moiety using silver sulphate in acetic acid solution accordmg to the method of Paul (1949; 1950). Performic acid was used to oxidize the thiol groups of cysteine residues to sulphonic acid residues. The haemin-free oxidized peptide obtained was essentially homogeneous as determined by electrophoresis on paper in a formic acid-acetic acid buffer of pH 2 (Werner and Westphal, 1955). Enzymic Hydrolyses The crystalline chymotrypsin used was purchased from the Worthington Biochemical Corporation. The subtilisin was a gift from Ing. M. Ottesen, Carlsberg Laboratory, Copenhagen. Digestions were carried out at pH 7-5 to 7-8 for 20 hr at 25°C (with chymotrypsin) or for 24 hr at 37°C (with subtilisin), the ratio between enzyme and substrate (w/w) being about 1: 15. End Group Determinations For the characterization of N-terminal amino-acid residues the DNP method (Sanger, 1945; see also Sanger and Thompson, 1953) was employed. C-terminal residues were determined using the hydrazinolysis procedure of Akabori, Ohno and Narita (1952). Tryptic Hydrolysate o/Rhodospirillum Rubrum Cytochrome c 365 RESULTS In the hydrolysate of the haemin-free oxidized polypeptide (prepared from the haemopeptide as described in the foregoing section) ten different ami no- acids were identified by paper chromatography — alanine, aspartic acid, cysteic acid, glutamic acid, glycine, histidine, leucine, lysine, phenylalanine and threonine. The lysine present in the polypeptide may be expected to be the C-terminal residue of the amino-acid sequence, since the polypeptide has been formed by the action of trypsin which is known to split bonds involving the carboxyl groups of lysine and arginine residues. The N- terminal residue was shown to be cysteic acid. After dinitrophenylation of the polypeptide and subsequent hydrolysis, however, the hydrolysate was found to contain, besides DNP-cysteic acid, free cysteic acid as well. This finding is accounted for by the presence in the polypeptide of two cysteic acid residues, one of which is terminal and the other is not. Chymotryptic hydrolysis of the polypeptide (Table 1) gave rise to three main split products which were separated from each other by high-voltage electrophoresis at pH 2 and by paper chromatography in butanol-acetic acid : (I) CyS03H-[Leu, Ala, CySOgH, His], (II) Thr-Phe, and (III) Asp-Glu- [Gly, Ala, Asp]-Lys. Split product (I) with N-terminal cysteic acid, must be derived from the N-terminal side of the polypeptide chain and the lysine- containing peptide (III) from the C-terminal side. Thr-Phe, then, must be located between these two. After dinitrophenylation of peptide III and partial hydrolysis (with cone. HCl at 37°C, for 8 days) two ether-soluble products were separated and characterized, DNP-Asp and DNP-Asp-Glu, which demonstrates that glutamic acid must adjoin the N-terminal aspartic acid residue. In the subtilisin hydrolysate of the polypeptide (Table 1) five products were present in relatively high concentration : CySOgli-Leu-Ala, CySOgH- His, Thr-Phe, [Asp, Glu, Gly, Ala], and Asp(NH2)-Lys. Among the split products obtained in small amounts there were two tripeptides containing liistidine: Ala-[CyS03H, His] and CyS03H-[His, Thr]. Other minor components of the hydrolysate were shown to be [Gly, Ala], [Asp, Glu, Gly], [Thr, Phe, Asp, Glu, Gly] and free alanine. From the findings presented above it may be concluded that the amino acid sequence in the haemin-free oxidized polypeptide is 1 2 3 4 5 6 7 8 9 10 11 12 13 CySOsH— Leu— Ala— CySOsH— His— Thr— Phe— Asp— Glu— Gly— Ala— AspCNHg)— Lys The results obtained leave it undecided whether residues 8 and 9 are free aspartic acid and free glutamic acid or whether these amino acids are in the amide form, as asparagine and glutamine residues. The two cysteic acid residues contained in the sequence have arisen from and taken the place of two haem-bound cysteine residues present in the 2 a I- 2 X o z a o < X Z w 5 a >, >-. 5 1 i J 1 1 1 .. 1 i G- X ^ < z-< 1 pf -^ 1 «j < « 1 < < _>. >-, _>> 5^ o _>^ a a T 3 3 3 _3" a ^ a G. cL < 1 u 1 < < CL, a, j:: - x: Q- IS H Xi H 'x 1 X 1 !2 X 1 X X o d C/3 X S ^ ^ S' S es Cj i < ^ < 3' k 1 J 1 1 ffi 1 o" 0" 0" C/3 00 on >^ >-. >~, u u U 12 1 2 3 1 73 T3 c 1 u, .0 l!^ 1 a. ^^^ CL, lo ._2 i 4 cu "S- i '^ ^ ■■5 ^ .0 _c i 1 f '3 2 h^ z s 1 S ■§ ^ § i' •o .S;5 JO TJ Si a c: 'S5 00 .g a a 1 1 1 r. "Y ii O- U-2 z u ^ (Si -< i I y o s ! I on g-3 IV Y I I 368 S. Paleus and H. Tuppy original Rhodospirillwn rubrum cytochrome c and in the haemopeptide prepared from it by tryptic hydrolysis. The specificity of trypsin already mentioned suggests that the amino-acid residue preceding the above sequence must be either lysine or arginine. The cleavage by chymotrypsin of a bond between histidine and threonine (residues 5 and 6) may appear unexpected in view of the general belief that chymotrypsin will split bonds involving the carboxyl groups of the aromatic residues phenylalanine and tyrosine. The finding is in agreement with recent observations of Davis (1956), who, however, has reported that bonds in- volving the carboxyl groups of histidine residues are somewhat susceptible to chymotryptic hydrolysis. The comparison of the amino-acid sequences that have been shown to occur in the vicinity of the haem moiety in different cytochromes demonstrates that they have various structural features in common (Table 2). The presence of two cysteine residues linked to side chains 2 and 4 of the porphyrin of the prosthetic group, first indicated by Theorell (1939) for ox cytochrome c, has been confirmed and found to be characteristic of all c-type cytochromes. The distance between the two cysteine residues along the polypeptide chain is always the same, two other amino acids being situated in between. In- variably a histidine residue has been found to adjoin one of the haem-bound cysteine residues. Stereochemical considerations (Ehrenberg and Theorell, 1955) have indicated that the chain of amino acids in which the two haem- bound cysteine residues and the histidine residue are included is likely to be in the form of an a-helix. This coiling will bring the histidine residue into a steric position favourable for covalent attachment of an imidazole nitrogen to the iron of the haem disc. There are still other structural features encountered in all cytochrome c samples examined so far, such as, for example, the presence of a basic residue (lysine or arginine) close to cysteine and of a threonine residue close to histidine. The significance of these similarities is at present not known. It is remarkable that in so many structural respects the cytochrome c of Rhodospirillwn ruhrum appears to resemble the c-type cytochromes of other organisms, in spite of an overall amino-acid composition, oxidation-reduction potential, and a specificity of action quite different from those of vertebrate and yeast cytochromes (Kamen and Takeda, 1956). REFERENCES Akabori, S., Ohno, K. & Narita, K. (1952). Bull. chem. Soc, Japan 25, 214. Bartsch, R. G. & Kamen, M. D. (1958). J. biol. Chem. 230, 41. Davis, N. C. (1956). /. biol. Chem. 223, 935. Ehrenberg, A. & Theorell, H. (1955). Acta chem. Scand. 9, 1193. Kamen, M. D. & Takeda, Y. (1956). Bioclnm. biophys. Acta 21, 518. Kamen, M. D. & Vernon, L. P. (1955). Bioclnm. biophys. Acta 17, 10. Keilin, D. & Hartree, E. F. (1937). Proc. ray. Soc. 8122, 298. Paul, K. G. (1949). Acta chem. Scand. 3, 1178. Tryptic Hydrolysate t»/RhodospiriIluni Rubrum Cytochrome c 369 Paul, K. G. (1950). Acta chem. Scand. 4, 239. Peterson, E. A. & Sober, H. A. (1956). /. Amer. chem. Soc. 78, 751. Sanger, F. (1945). Biochem. J. 39, 507. Sanger, F. & Thompson, E. O. P. (1953). Biochem. J. 53, 353. Theorell, H. (1939). Enzymologia 6, 88. TUPPY, H. Z. (1957). Naturforsch. 12b, 784. TUPPY, H. & BODO, G. (1954). Mh. Chem. 85, 1024, 1182. TuppY, H. & Dus, K. (1958). Mh. Chem. 89, 407. TuppY, H. & Paleus, S. (1955). Acta chem. Scand. 9, 353. Vernon, L. P. & Kamen, M. D. (1954). /. biol. Chem. 211, 643. Werner, G. & Westphal, O. (1955). Angew Chem. 67, 251. ELECTROMETRIC AND OTHER STUDIES ON CYTOCHROMES OF THE C-GROUP By R. W. Henderson and W. A. Rawlinson Department of Bioehemistry, University of Melbourne There is a very wide distribution in tissues, of pigments with a haemochrome- type spectrum showing absorption peaks in the visible region close to 550 and 520 m/n. A number of haemochromes with different haem prosthetic groups are known, however, to show a similar spectrum (see for example, Lemberg and Legge, 1949; Morton, 1958) and from time to time it has been pointed out that many of these compounds are as yet only spectral pheno- mena. With the isolation of bacterial cytochromes (see for example, Kamen Table 1. Range of £"„' values found in cytochrome c group pigments (see also Morton, 1958) Common symbol Source Eo' (V) RHP-pigment Plants (chloroplasts) Rhodospirillum rubrum Mammalian heart-muscle Micrococcus denitrificans Rhodospirillum rubrum Desulphovibrio desulphur icons -I-0-365 +0-338 +0-255 +0-250 -0008 -0-205 Davenport & Hill (1952) Kamen & Vernon (1955) Rodkey & Ball (1950) Henderson & Rawlinson (1956a) Kamen & Vernon (1955) Bartsch & Kamen (1958) Postgate (1956) Table 2. Eq values obtained after modification of mammalian heart-muscle cytochrome c Treatment Fraction separated on a resin column from TCA-extracted preps. 'Resin-purified' prep, treated with 2-5% TCAfor 18hr, 25°C Proteinase-digested Henderson & Rawlinson (1956b) Henderson & Rawlinson (1956b) Minakami, Titani & Ishikura (1957) cf. Table 1, resin-purified cytochrome c, Eq = + 0-255 V. 370 Electroinelric and Other Studies of Cytochrome 371 and Vernon, 1955) and the modification of mammalian heart muscle cyto- chrome c by treatments such as proteolytic digestion (Tsou, 1951a) many varied properties are apparent in pigments having the same or closely similar absorption spectra. This variation is observed in the important property of oxidation-reduction potential {Eq) even where the haem prosthetic groups are known to be of the mammalian heart cytochrome c (M. heart c) type (Table 1). It is parti- cularly interesting that a difference in E'q at least as large as that existing between the naturally occurring 'c-type' pigments may be obtained by modification of M. heart c without any significant spectral change (Table 2). OXIDATION-REDUCTION POTENTIAL OF YEAST CYTOCHROME C The early controversy regarding the £■„ value of yeast cytochrome c centred around the values of +0-26 V (Coolidge, 1932) and +0-12 V (Green, 1934) (see also Table 3). It was generally considered to have been resolved in Table 3. Eq values of various yeast cytochrome c samples Extraction method Eo (V) pH Reference (i) Ammonium sulphate extract (ii) Cells plasmolysed at boil then treated with SO2 (iii) Whole cell suspension SO2 method of Keilin (1930) Plasmolysed (boiled) cell suspension Washed yeast cell suspension SO2 method of Keilin (1930) then resin (XE-64) treatment +0-26 +0-123 +0-12 to +0-16 +011 to +015 +0-26 to +0-27 7 7 7 6-8 6,7 Coolidge (1932) Green (1934) deToeuf(1937) Baumberger (1939) Minakami (1955) favour of the former by what is now obviously the doubtful procedure of comparing the values already obtained from yeast samples with those from several other sources, especially mammalian heart muscle. Furthermore, Green's criticisms of Coolidge's methods seem quite valid. The problem does not appear to have been attacked again using the isolated pigment until Minakami (1955) reported some potential levels for several ratios of oxidized/ reduced cytochrome c of yeast which were similar to those for M. heart c. It still seemed desirable to carry out a complete electrometric titration of yeast cytochrome c extracted under mild conditions and avoiding the use of Na2S204 as titrant (see, for example, Paul, 1951). This was done using a sample (kindly provided by Prof. R. K. Morton) which had been extracted from dried baker's yeast with NaCl and chromatographed on a resin exchange 372 R. W. Henderson and W. A. Rawlinson column (Armstrong, Coates and Morton, 1958, and this volume, p. 385). The value of +0-28 V with « = 1 at pH 6-4 and 25°C was obtained (Fig. 1 ; see also Henderson and Rawlinson, 1958). r- ~ \ 0-320 \ \ \ \ \ V 0-260 0-240 0220 0-200 \ \ k f^ \ \ 20 40 60 80 100 Reduction, % Fig. 1. Oxidation-reduction titration at pH 6-4 of yeast cytochrome c. Solid line is theoretical curve for Eq = 0-282 V, « = 1. COMBINATION OF CYTOCHROME C FRACTIONS WITH COPPER Inhibition of Copper-catalysed Oxidation of Ascorbic Acid One of the most striking effects of proteolytic digestion of M. heart c is the induction of a high ascorbic acid oxidase activity which parallels a loss in activity in the usual biological oxidase systems (Tsou, 1951b). Margoliash (1954b) observed a similar effect with a fraction separated on a resin column. When examining the ascorbic acid oxidase activity of a number of M. heart c fractions we observed that in the case of the '0-34% Fe' or 'low-iron' content samples, lower oxygen-uptake values were obtained than for the ascorbatc blank. A similar reduction in oxygen-uptake of the blank was obtained by addition of ethylenediaminetetraacetic acid (EDTA) (Fig. 2). Electioiuelric and Other Studies of Cytochrome c 373 The copper catalysis of ascorbate oxidation is well known (see, for example, Barron, de Meio and Klemperer, 1935) and up to about 10^^ g atom/ml the rate of oxidation is directly proportional to the concentration of copper. It seemed hkely that the '0-34 % Fe' content samples were inhibiting an oxidation of ascorbate catalysed by traces of copper present as an impurity in the reagents. A series was therefore run in which there was a constant added 0.5 10 Cytochrome c cone, Fig. 2. Influence of cytochrome c fractions on ascorbate oxidation. (Determina- tions using Warburg manometers, 007 m sodium phosphate buffer, pH 7-3 and 0-02 M sodium ascorbate, at 37°C.) I. With added copper (I /tig J flask). Curve A: Preparations: (i) before resin column treatment, and (ii) colourless protein fraction separated electrophoretically. Curve B: Preparations: (iii) after resin column treatment, and (iv) after electro- phoresis, and (v) after treatment with TCA. Curve C: (vi) cytochrome fraction separated on resin column. II. No copper added. Fractions (i) to (v) inclusive and also 10~^ m EDTA lowered the oxygen uptake to the bottom X; fraction (vi) catalysed the oxygen-uptake to the top X. concentration of copper and varying amounts of the different cytochrome c fractions (Fig. 2). Three distinct types of curves were obtained. The '0-34% Fe' cytochrome c gave a value the same as in the absence of added copper until the point corresponding to approximately 4 cytochrome c molecules/ copper atom, after which there was a rapid rise in uptake inversely pro- portional to the cytochrome concentration. The '0-43% Fe' cytochrome c inhibited the ascorbate oxidation to a much lesser extent than the above material. This indicated that combination was in the main occurring between the copper and a colourless protein fraction which could be removed electro- phoretically (Theorell and Akeson, 1941) or chromatographically (Paleus 374 R. W. Henderson and W. A. Rawlinson and Neilands, 1950; Margoliash, 1954a, b). Presumably the same material is removed when the iron content is increased by the alkaline salt fractionation of Keilin and Hartree (1945) or the boiling CHCI3 treatment of Tsou (1951a). This colourless fraction was accordingly separated in the Tiselius-type electrophoresis apparatus and found to combine with copper producing a curve of type A, Fig. 2. Zinc Precipitation of Myoglobin The question arises as to whether the colourless protein fraction is present in vivo. Margoliash (1954a) considered it to be globin present as an artefact — I r /L V \\ J /' ^v \\ /I v>^ / J 0/ / / ^ ^»«— . ^^"i^^^ 1 B/ 1 L 1 1 1 1_ ,. 1 640 620 600 580 560 540 520 500 480 Wavelength, m/^ Fig. 3. Comparison of characteristic absorption spectra of oxidized and reduced cytochrome c samples from ox-heart muscle. Curve A : salt-extracted and Zn- treatcd. Curve B: the former after resin treatment. Curve O: the oxidized forms of both samples. of preparation due to splitting of myoglobin under the conditions of the extraction with trichloroacetic acid (TCA). In the case of neutral extraction procedures some myoglobin was present. In order to examine further this question, cytochrome c was extracted from heart muscle with saline solution. The resulting solution contained the pigments haemoglobin (Hb) and myoglobin (Mb) in addition to cytochrome c. It had been shown by one of us (Rawlinson, 1938) that Hb and globin may be quantitatively precipitated by Zn++ at slightly alkaline pH. Myoglobin was accordingly tested under similar conditions and found to behave in the same manner. Haemoglobin and myoglobin along with much colourless protein material were therefore removed from the saline extract of cytochrome c by this method. Up to this point, the conditions had been such that no splitting of the prosthetic Electrometric and Other Studies of Cytochrome c 375 groups from Hb or Mb would be expected. The cytochrome c was then concentrated with TCA as in the final stage of the Keilin and Hartree procedure. This material had an iron content of 0-32% and a trace only of zinc was present. Prior to TCA-treatment no absorption at 630-635 mfi due to metmyoglobin or methaemoglobin could be detected and subsequently the absorption spectra of the oxidized forms before and after resin chroma- tography were identical (Fig. 3). No fore-running non-cytochrome fraction containing protohaem could be detected on resin column chromatography as is the case with TCA-extracted samples. However, increased absorption, particularly on the longer wave-length slope of the a-band is apparent in the reduced form of the material prior to resin treatment (Fig. 3) and the band ratios differ considerably (Table 4). Inhibition of the copper-catalysed oxidation of ascorbic acid was similar to that of the usual '0-34% Fe' content preparations, indicating combination with copper. A sample which had been treated with copper under the conditions described in Fig. 2, was given a repetition of the final TCA precipitation of the Keilin and Hartree pro- cedure. It was found that this material would then again take up a similar quantity of copper. It appeared therefore, as in fact would be expected, that the copper originally combined was removed under the acid conditions. Table 4. Comparison of optical-density ratios of several ox-heart cytochrome c samples Ratio of Ratio of Ratio of Ratio of 590 mil 550 m/« 550 m//. 520 m^u Extraction method %Fe (oxidized) (reduced) (reduced) (reduced) 590 m/t 520 m^t. 535 m// 535 m/* (reduced) (reduced) (reduced) (reduced) (a) Saline (Zn-treated) 0-32 1-95 1-55 2-53 1-63 (b) TCA (Keilin and Hartree 0-30 1-93 1-74 3-38 1-94 method) (c) (a) After single resin — 313 1-79 3-92 2-20 column, pH 7 (Margo- liash, 1954b) (d) (b) After double resin 0-44 5-78 1-76 3-86 2-19 column, pH 7 and 9-6 (Margoliash, 1954b) Combination of Copper with Globin Myoglobin was prepared from horse-heart by a combination of the methods of Theorell (1932) and Lawrie (1951) and globin from this by the procedure of Theorell and Akeson (1955). Globin was added in varying concentration to constant copper and ascorbate concentration as for the cytochrome c 376 R. W. Henderson and W. A. Rawlinson fractions (Fig. 2) and the O2 uptake determined (Fig. 4). Great avidity for copper was shown; combination, however, was in the ratio of 2 g atom of Cu/mole globin. ^ 150 Cu blonk (l«5/flosk)J [ 1 1 Ascortxjte blank \- "" 1 V ° u u - 0-5 1-0 1-5 2-0 2-5 3-0 3-5 moles xlO^giobin/flosK Fig. 4. Inhibition by globin (from horse-heart myoglobin) of the copper- catalysed oxidation of ascorbate. (Experimental conditions: Warburg mano- meters; 007 M sodium phosphate buffer, pH7-3; 0-02 m sodium ascorbate, and 1 fig of copper/flask at 37°C.) Eq of Cytochrome c in the Presence of Copper Low-iron content samples in general show a slightly higher Eq (about 15 mV) than the higher-iron content samples (Henderson and Rawlinson, 1956a). This was attributed, however, to the presence of modified cytochrome c. We have now examined the Eq of a '0-34% Fe' content sample of M. heart c before and after addition of copper in shght excess of its binding capacity and find the value to remain unaltered (£"0 = + 0"26 V, « = 1). The sample containing copper showed a somewhat increased rate of autoxi- dation. This electrometric result is compatible with an aggregation where there is no significant interaction one with another between the groups oxidized/reduced (Shack and Clark, 1947), although of course it neither proves nor disproves a state of aggregation. Crystallization of Cytochrome cfrom a 'Low-iron'' Content Preparation Since the first crystallization of cytochrome c from King penguin muscle by Bodo (1955) there have been several reports, notably from the school of Okunuki, of crystallization of the pigment from a number of different tissues (see Morton, 1958). Much work has been carried out on cytochrome (■ from mammalian heart muscle, especially from ox-heart. To the best of our knowledge cytochrome c from this source had been crystallized only after resin-column treatment. Fig. 5. Cytochrome c crystals after 48 lir, phase contrast, x800. Eleclrometrk and Other Studies of Cytochrome c 377 Recently, however, we have been successful in crystallizing the pigment from a '0-31 % Fe' content sample obtained simply by the usual KeiUn and Hartree procedure. The crystallization method of Hagihara, Morikawa, Tagawa and Okunuki (1958) was followed except that an addition of copper was made in the binding ratio of 1 g atom Cu/4 moles cytochrome c (see Henderson and Rawlinson, 1960). Small crystals formed rapidly (1-2 hr) and within 48 hr many rosettes were present along with crystals (Fig. 5) having an appearance very similar to those depicted by Hagihara et al. (1956) and described by them as leaflets. The crystals were in the reduced form. The biological activity and other properties of this preparation have not yet been investigated. Possible Linkage of Cytochrome c in vivo The above experiments show that a colourless protem fraction is associated with M. cytochrome c extracted both with and without the use of acid, and in the latter case when myoglobin and haemoglobin would appear to have been completely removed. This fraction has an anodic mobility at pH 7-4 and from this fact it is to be expected that some affinity exists between this material and M. cytochrome c with its high iso-electric point (pH 10-5). There is some evidence that the fraction is globin but from the above experiments it would seem that there is still room for doubt. The view has been put forward previously that this material is associated with cytochrome c in vivo (see Lemberg and Legge, 1949). This view was based largely on the behaviour of '0-34% Fe' cytochrome c with regard to such properties as heat stability and difficulty of separation except by electro- phoresis. If it is in fact present in vivo it may be that the strong metal- binding capacity shown above is the means of bringing about the particular spatial or other arrangement — even of electron transport — which it seems is necessary (see Slater, 1958) for full endogenous activity. ^Modified'' Cytochrome c After separation of the main fraction of cytochrome c on a resin column (Fraction I, pH 7 according to Margoliash, 1954b), there is left a band of pigment at the top of the column. This material (Fraction II) may be eluted with 0-5 N NH4OH. Margoliash (1954b) found it to have an increased ascorbic acid oxidase activity and a lowered biological activity. Some of the colourless protein fraction is, however, eluted at the same time by the NH4OH. After further chromatography to remove as much as possible of the latter, a relatively high ascorbic acid oxidase activity resulted (curve C, Fig. 2). ^ The Eq value of this fraction has already been mentioned (Table 2) to be some 55 mV above that of the main fraction. It also seems very likely that the state of aggregation is altered, as it is seen from Fig. 6 that the value of 378 R. W. Henderson and W. A. Rawlinson n which is between 1 and 2 at the commencement of titration, approaches 2 as the reduction continues, and after the point of 50 % reduction is quite close to 2. These changes would appear to provide an explanation of the altered biological activity of this fraction. • \ • \ \ \« \ < ^ \ \ V •. \ \ \ \ *- __ __ __ -< 3 \ \ \ \ 20 40 60 80 100 Reduction, % Fig. 6. Oxidation-reduction titration at pH 6-4 of 'modified' ox-heart cyto- chrome c (Fraction II according to Margoliash, 1954b). Observed points •; theoretical curves are for E^ = -fO-310 V; broken line, « = 1 ; solid line, « = 2. The oxidation-reduction potential of 'unmodified' fraction {Eq = +0-255 V) is shown thus O. Margoliash (1954a, b) obtained evidence that in the case of horse-heart cytochrome c, TCA at pH 4-5 was responsible for the formation of Fraction II above. We subjected resin-purified samples from ox-heart to treatment with TCA at approx. pH 1, (2-5% TCA, 18 hr, 25°C) and obtained a similar E'q value to that of Fraction II. Again there was evidence of aggregation although in this case it was not as pronounced as in Fraction II above; the curve was in between n = 1 and n = 2 after 50% reduction. It is of con- siderable interest that the ascorbic acid oxidase activity remained virtually unaltered (Fig. 2). It did not behave on the resin column like Fraction II as it was eluted slowly and diffusely by 0-25 m ammonium acetate at pH 7. Electrometric and Other Studies of Cytochrome c 379 From one such TCA-treated cytochrome c sample a small residue was ob- tained after dialysis against 0-5% NaCl which dissolved in 0-02 n NH4OH. This material had an increased ascorbic acid oxidase activity and a somewhat altered absorption spectrum (Fig. 7). This result was variable, however, and generally no residue was obtained. There was thus evidence for the production of more than one fraction by this treatment, which is in agreement with the €20 580 560 540 Wavelength, m/x Fig. 7. Comparison of characteristic absorption spectra of reduced cytochrome c samples from ox-heart muscle. Curve ^4: resin-treated sample; curve fl: fraction soluble in 0-5% NaCl after 2-5% TCA, for 18 hr, 25°C; curve C: residue after TCA treatment, soluble in 0-02 N NH4OH. observations of Margohash, Frohwirt and Wiener (1959) who have reported a separation of Fraction II into several other fractions with varying properties. Although TCA under various conditions can bring about changes in cytochrome c such as increased susceptibility to proteinase digestion, change in chromatographic behaviour and altered E'q, it seems that with mild treat- ment its biological activity is unaltered (see, for example, Keilin and Hartree, 1955). In fact the modified material — as judged by its chromatographic behaviour — may show unchanged properties in biological systems (see, for example, Hagihara et al., 1958a; Yamanaka et al, 1959). In connexion with the often-quoted autoxidation and combination with CO as criteria of modification, it is noteworthy that Minakami et al. (1958), after acetylation of the e-lysine residues, obtained a product inactive in the succinic oxidase system and autoxidizable but which showed neither ascorbic 380 R. W. Henderson and W. A. Rawlinson acid oxidase activity nor combination with CO. Furthermore, guanidation of these residues (Takahashi et ah, 1958) left the pigment practically unaltered in relation to these properties. These findings along with those of Nozaki et al. (1957, 1958), who showed dependence of resistance to proteolytic attack on the state of oxidation of the pigment, seem to be, as postulated by these workers, an indication of the involvement of the secondary protein structure in the oxidation-reduction reactions of cytochrome c. Eo CHANGE WITHIN THE CYTOCHROME C GROUP An explanation for the variation in E'q values quoted in Tables 1 and 2 may be found in terms of the following general equation derived by Clark et al. (1940). This relates the Eq of a metal co-ordinate complex (of the type with which we are dealing here) to change in degree of association of the ligand with the central metal atom. An outline only of the derivation is given below, and reference should be made to the original for full details of assumptions. From the basic equation— 0, + ne=pR,, (1) where 0„ and R^ are the oxidant and reductant respectively in the fixed states of aggregation designated by n and m, the usual electrode equation may be derived : Co-ordination is then assumed to occur with a nitrogenous base B such that there is no change in the respective states of aggregation ; then : 0, + qB= 0,8, (3) R^ + iB = R^B, (4) from which the respective equihbrium constants Kq and A'^ may be obtained. Under the conditions/? = m = n = 1 and at 30°C, the following equation may be obtained : £, = £, -f 0-06 log f + 006 log ^ -f 0-06 log p^^j||^ (5) where S, = niOJ + n{OM and S^ = m(RJ + m(R,M It can be seen from equation (5) that, other things being equal, E^ is dependent upon the values Kq and Kj^, i.e. the Eq value is dependent upon these two constants. Electromclric and Other Studies of Cytochrome c 381 In the cytochrome c group, the hgands to the 5th and 6th co-ordinate positions are attached to the apo-protein. It is conceivable that owing to differences between the natural apo-protein structures or to modifications which bring about perhaps some unfolding of these structures, there is variation in the strain and in consequence of the stability of the ligand bondings. In such cases the resultant alterations in values of Kq and Kj^ would be reflected in the E'q values. A cknowledgements We wish to thank Dr J. O'Hagan for details of myoglobin separation and Mr E. Matthaei for the photograph (Fig. 5). We are indebted to the Australian National Health and Medical Research Council for a grant towards the cost of the investigation. REFERENCES Armstrong, J. McD., Coates, J. H. «&. Morton, R. K. (1958). Aust. J. Sci. 21, 119. Barron, E. S. G., de Meio, R. H. & Klemperer, F. (1935). /. biol. Cheni. 112, 625. Bartsch, R. G. & Kamen, M. D. (1958). /. biol. Chem. 230, 41. Baumberger, J. p. (1939). Cold Spring Harbor Symposia Quant. Biol. 7, 74. BoDO, G. (1955). Nature, Lond. 176, 829. Clark, W. M., Taylor, J. F., Da vies, T. H. & Vestling, C. S. (1940). J. biol. Chem. 135, 543. Coolidge, T. B. (1932). /. biol. Chem. 98, 755. Davenport, H. E. & Hill, R. (1952). Proc. roy. Soc. B139, 327. Green, D. E. (1934). Proc. roy. Soc. B1I4, 423. Hagihara, B., Morikawa, I., Sekuzu, I., Horio, T. & Okunuki, K. (1956). Nature, Lond. 178, 629. Hagihara, B., Sekuzu, I., Tagawa, K., Yoneda, M. «fe Okunuki, K. (1958). Nature, Lond. 181, 1588. Hagihara, B., Morikawa, I., Tagawa, K. & Okunuki, K. (1958). Biochem. Prep. 6, 1. Henderson, R. W. & Rawlinson, W. A. (1956a). Biochem. J. 62, 21. Henderson, R. W. & Rawlinson, W. A. (1956b). Nature, Lond. Ill, 1180. Henderson, R. W. & Rawlinson, W. A. (1958). Aust. J. Sci. 21, 116. Kamen, M. D. & Vernon, L. P. (1955). Biochim. biophys. Acta 17, 10. Keilin, D. (1930). Proc. roy. Soc. B106, 418. Keilin, D. & Hartree, E. F. (1945). Biochem. J., 39, 289. Keilin, D. & Hartree, E. F. (1955). Nature, Lond. 176, 200. Lawrie, R. a. (1951). Nature, Lond. 167, 802. Lemberg, R. & Legge, J. W. (1949). Hematin Compounds and Bile Pigments. Interscience Publishers Inc., New York. Margoliash, E. (1954a). Biochem. J. 56, 529. Margoliash, E. (1954b). Biochem. J. 56, 535. Margoliash, E., Frohwirt, N. & Wiener, E. (1959). Biochem. J. 71, 559. MiNAKAMi, S. (1955). /. Biochem. Tokyo Al, 749. Minakami, S., Titani, K. & Ishikura, H. (1957). /. Biochem. Tokyo 44, 537. Minakami, S., Titani, K. & Ishikura, H. (1958). /. Biochem. Tokyo 45, 341. Morton, R. K. (1958). Rev. pure appl. Chem. 8, 161. Nozaki, M., Yamanaka, T., Horio, T. & Okunuki, K. (1957). J. Biochem. Tokyo 44, 453. NozAKi, M., Mizushima, H., Horio, T. & Okunuki, K. (1958). /. Biochem. Tokyo 45, 815. Palieus, ?. & Neilands, J. B. (1950). Acta chem. Scand. 4, 1024. Paul, K. G. (1951). The Enzymes, Vol. 2, Pt. 1. Ed. Sumner, J. B. & Myrback, K. Academic Press Inc., New York. 382 Discussion PosTGATE, J. R. (1956). J. gen. Microbiol. 14, 545. Rawlinson, W. a. (1938). Aust. J. exp. Biol. med. Sci. 16, 303. RODKEY, F. L. & Ball, E. G. (1950). J. biol. Cliein. 182, 17. Shack, J. & Clark, W. M. (1947). /. biol. Chem. 171, 143. Slater, E. C. (1958). Advanc. Enzymol. 20, 147. Takahashi, K., Titani, K,, Furuno, K., Ishikura, H. & Minakami, S. (1958). J. Biochem. Tokyo 45, 375. Theorell, H. (1932). Biochem. Z. 252, 1. Theorell, H. & Akeson, a. (1941). J. Amer. chem. Soc. 63, 1804. Theorell, H. &. Akeson, A. (1955). Ann. Acad. Sclent. Fennicae All, 60, 309. ToEUF, G. DE. (1937). J. chem. Phys. 34, 740. Tsou, C. L. (1951a), Biochem. J. 49, 362. Tsou, C. L. (1951b). Biochem. J. 49, 367. Yamanaka, T., Mizushima, H., Nozaki, M., Horio, T. & Okunukj, K. (1959). J. Biochem. Tokyo 46, 121. DISCUSSION Protein Conjiguiation and Linkage to the Prosthetic Group in Cytochrome c Studies of the Haeraochrorae-forming Groups in Cytochrome c By E. Margoliash (Utah) Margoliash: For all the typical haemochrome properties there is no sharp breaking point between native cytochrome c on the one hand and the pepsin digested cytochrome c 'core', which can be considered as the most degraded form of the cytochrome c haemo- chrome, on the other. Quite the contrary, by isolating by column chromatography a series of cytochromes c showing increasing degrees of denaturation it could be shown that all the haemochrome properties vary gradually with the degree of denaturation. These fractions of denatured cytochrome c were defined by their chromatographic properties on Amberlite IRC-50, the more protein being de- natured the larger the number of e-NHz groups of lysine available to the resin and for reaction with a reagent such as l-fluoro-2-nitro-4-benzene sulphonate. Native cytochrome c was at one end of the scale with maximal enzymic activity in the succinic oxidase and cytochrome oxidase systems, a very low rate of auto-oxidation, a small per cent combination with CO at near neutral pH values, and a very wide pH-range of stability of the reduced spectrum. With increasing degrees of denaturation the enzymic activities became lower, the rate of auto-oxidation and the per cent combination with CO increased ; the pH-range of stability of the reduced spectrum became gradually smaller moving toward the alkaline side. With respect to all these properties the cytochrome c 'core' is in every way identical to the most denatured form of cytochrome c obtained, even though it contains only 1 1 amino acids out of the full complement of 96-98 present in denatured cytochrome c. The cytochrome c 'core' shows on reduction the typical cytochrome c haemo- chrome spectrum, but in contradistinction to native cytochrome c, the reduced spectrum of which is completely unaffected by pH over a very wide range from pH 2-3 to pH 13, the 'core' acts like a normal chemical haemochrome. Measuring at the maximum of the a band, one finds that the spectrum of the 'core' reaches the full extinction of native cytochrome c only at about pH 11 after passing through 3 sets of inflexions with mid-points at 5-4, 7-6 and 9-5. The addition of a large excess of bases containing a- or e-NH, groups did not effect the maximal extinctions obtained at pH 11, whereas the addition of bases containing only an imidazole group decreased the height of the a-band to 5-8 % below that of native cytochrome c. This was in- terpreted as indicating that the haemochrome-forming groups in cytochrome c are not both imidazoles. We therefore investigated the possibility that the e-NHg group of the lysine and the imidazole group of the histidine, both amino acids being the residues adjacent to the cysteines involved in the thio-ethcr bonds holding the haem to the protein, were the actual haemochrome-forming groups in cytochrome c. Electrometric and Other Studies of Cytochrome c 383 Ehreoberg and Theorell {Acta chem. Scand. 9, 1193, 1955) had shown that if the peptide chain of the 'core' were made into an a-helix cither the imidazole or the c-NHo group could be made to coordinate with the central hacm iron atom, but not both at the same time. However, building a suitable model it could readily be shown that if the intervening chain of 4 amino acids were extended, these two groups could indeed be made to occupy coordination positions 5 and 6 of the iron atom, above and below the plane of the haem. A study of the ultracentrifugal sedimentation characteristics of the 'core' indicated that although the 'core' was probably a small polymer at intermediate pH values, at those alkaline pH values at which its spectrum maximally approached that of native cytochrome c the 'core' was a monomer and the haemochrome must necessarily have been intramolecular, rather than intermolecular as could have been the case for a polymer. These studies (Biochem. J. 71, 559, 1959) have led us to conclude that: (a) the haemochrome-forming groups in cytochrome c are probably the e-amino group of the lysine in position 3 and the imidazole group of the histidine in position 8 of the amino-acid sequence of the pepsin-digested cytochrome c 'core' (see Paleus' paper); (b) the properties particular to the haemochrome of native cytochrome c, as contrasted with those of a normal chemical haemochrome, appear to be determined by the effect of the entire protein in its native configuration on the iron- ligand bonds, rather than by the particular chemical groups involved in the haemo- chrome of cytochrome c. The Amino-acid Sequence in Horse-heart Cytochrome c By E. Margoliash and R. Hill (Utah) Margoll«lSh: I should like to report on two lines of work dealing with the amino-acid sequence of horse-heart cytochrome c. (i) Leucine aininopeptidase digestion of cytochrome c. Using a 20 % molar ratio of a highly purified hog kidney leucine aminopeptidase to cytochrome c, the liberated amino acids were separated by dialysis and the remaining partly digested cytochrome c isolated by column chromatography on Amberlite IRC-50. Either directly or after suitable acid hydrolysis both fractions were analysed on an automatic amino-acid analyser. It was shown that about 30 amino acids are liberated after 20 hr of digestion from the N-terminal sequence of cytochrome c, and that the liberated amino acids as well as those remaining in the partly-digested protein add up rather accurately to the composition of the original cytochrome c. These 30 residues contain the single tryptophane, one of the two arginines, one of the three histidines, one of the three valines and one of the four prolines in the entire protein. In addition two serines, which have previously not been found in horse-heart cytochrome c were also liberated. Since practically the entire haem was found in the remaining partly digested cytochrome c, the haem must be located at least 30 amino acids away from the N-terminal end of the peptide chain. (ii) Amino-acid sequence of a chymotrypsin-digested cytochrome c 'core\ Using a partial digestion with chymotrypsin it was possible to isolate a chymotryptic 'core' of horse-heart cytochrome c, by column chromatography on Amberlite IRC-50 followed by high voltage paper electrophoresis. This peptide contained 42 amino acids. N-terminal and C-terminal residues were determined by reaction with fluoro- dinitrobenzene and carboxypeptidase digestion, respectively. The chymotryptic 'core' was digested with trypsin; the peptides formed were isolated by various combinations of paper electrophoresis and paper chromatography and their composition deter- mined using an automatic amino-acid analyser. A partial sequence thus established for the chymotryptic 'core' showed some unusual features. Towards the C-terminal end there was a remarkable concentra- tion of aromatic and long-chain aliphatic amino acids, enclosing a single glutamic 384 Discussion acid, and in the centre of the sequence a concentration of basic residues (3 or 4 lysines and one histidine). Between these two regions was located a single proline. Mammalian heart cytochrome c has no known covalent bonds holding the protein together, it is readily denatured by acid and alkali and although all of its acidic and basic residues are titratable, only 8-9 of the e-NH, groups of its 18 lysines will react rapidly, in the native protein, with a water soluble reagent such as l-fluoro-2-nitro-4- benzene sulphonate. I should therefore like to propose the hypothesis that concen- trations of long chain aliphatic and aromatic amino acids in specific positions provide the required hydrophobic microenvironment to stabilize particular acid-base ionic bonds which serve to keep the protein in its native folded configuration. What was observed in the chymotryptic 'core' would presumably be one such region. Another consequence of this work is that the haem is attached at least 30 residues away from the N-terminal end and at least 30 residues away from the C-terminal end, thus putting the haem to probably within 15 residues of the centre of the peptide chain. Comments on the Structure of Cytochrome c Paleus: As to the spectrophotometric curve shown (by Margoliash) of the peptic horse haemopeptide, and the effect of addition of histidine to the solution, I want to refer to the results of Tuppy and Bodo (M/j. Chem. 85, 1024, 1182, 1954), as well as those of Ehrenberg and Theorell {Acta chem. Scand. 9, 1193, 1955). These authors added histidine (in the case of Tuppy and Bodo also a-benzoyl-histidine) to a solution of the tryptic horse haemopeptide of cytochrome c (Tuppy and Bodo) and to the peptic beef haemopeptide (Ehrenberg and Theorell) at pH 7-3 and 8-9 respectively. They found that the extinction at 550 m// was raised to that found in native cytochrome c. Margoliash, however, observed this phenomenon first at pH 11. As to the haem-Unkage, I want to refer to the results of Tuppy and Bodo, who were able to show how the autoxidizable haemopeptide lost most of its autoxidizability on adding histidine at the same time as the extinction at 550 m^ of native cytochrome c was reached. Ehrenberg and Theorell showed spectrophotometrically the inability of ferro- and ferric protoporphyrin at pH 7 to form any compounds with lysine or glycyl-glycine. What does Margoliash now think is the N-terminal group of his cytochrome c preparation? Is it histidine as he stated in 1955? Did the N-terminal group show up at the degradation of cytochrome c with leucine aminopeptidase, the experiment which he has shown us today? However, Minakimi, Titani and Ishikura (/. Biochem. Tokyo, 45, 341, 1958) could not degrade cytochrome c with leucine aminopeptidase. Their results do not correspond to yours. As to your opinion of the electrostatic holding together of the cytochrome c molecule it may be rather important. I also want here to mention that one can abolish the capacity of the enzyme for electron-transfer by, e.g. acetylation of the e-NHo-lysine group, but not by guanidation, as shown by Minakami and co-workers. Margoliash: I do not think that the results of Tuppy and Bodo (who found that adding a large excess of a-benzoylhistidine to the tryptic core of cytochrome c resulted in a considerable increase in the reduced a-band extinction) conflict with our own results. If I remember correctly, Tuppy and Bodo obtained an increase of the a-band extinction to about 80-90% of that of native cytochrome c at somewhat alkaline pH. This is very similar to our own results with imidazole and the pepsin digested cytochrome c 'core', and does not influence the fact that no externally added bases are required to obtain a spectrum virtually identical to that of native cytochrome c, at pH 11, with this 'core'. You asked whether we managed to establish an N-terminal amino acid for cytochrome c with leucine aminopeptidase. No, this has not been done. Leucine aminopeptidase docs not produce an appreciable digestion at too low a concentration. On the other hand, at concentrations that will work, the rate of digestion is so rapid that it is not easy to establish by a kinetic study which is the first amino acid liberated. There is of course the added difliculty that if the first peptide bond to be split is broken slowly as compared to the next few bonds along the chain, all these amino Electronwtric and Other Studies of Cytochrome c 385 acids will be liberated practically simultaneously, making it impossible to decide which is the N-terminal. Williams: I believe that the observations of Margoliash can also be interpreted in terms of Theorell's original suggestion that the cytochrome was a di-imidazole complex. Rather than go into detail here it is proposed to submit a full analysis of these experi- ments elsewhere (see also, this volume, p. 141). Structure and General Properties of Cytochrome c Comparative Properties of Cytochrome c from Yeast and Heart Muscle By J. McD. Armstrong, J. H. Coates and R. K. Morton (Adelaide) Morton: This paper relates particularly to the contribution by Henderson and Rawlinson (this volume, p. 369) and describes studies briefly reported elsewhere (see Armstrong, Coates and Morton, 1958; Morton, 1958). Air-dried baker's yeast (as used for isolation of crystalline cytochrome b.^; see Appleby and Morton, 1954, 1959) was extracted with m NaCl for about 6 hr at room temperature (about 22°C) or for 30 min at 55-60°C. The supernatant obtained on centrifuging was dQuted with three volumes of water and adjusted to pH 5-7. The cytochrome c in the extract was absorbed batchwise onto resin (IRC-50(Na+)), and after washing the resin, the cytochrome c was eluted either with 0-8 m NaCl in 0-1 m sodium phosphate pH 7-8, or 40% saturated ammonium sulphate pH 8-5-90. The cytochrome c was purified by chromatography on a column of the same resin until the ratio of E549 tsxfij^^n ^l<^ was constant (1-05-1 -12) for the eluted cytochrome. Electrophoresis by the moving boundary method was carried out in buffers of ionic strength of 0-2, at 11 pH values from pH4-18 to pH 1007, At all pH values only one protein component was detected, although some asymmetry of the boundary was apparent. Only microheterogeneity could be detected by reversal of polarity after prolonged electrophoresis at pH 9-93, which is close to the iso-electric point of the cytochrome. Figure 1 shows the pH-mobility plot for the yeast cytochrome c in comparison with the plot obtained by Theorell and Akeson (1941) and by Tint and Reiss (1950) for ox-heart cytochrome c. The iso-electric point for yeast cytochrome c is pH 9-85 ± 0-05 under our conditions, as compared with pH 10-6 for the ox-heart cytochrome c. Sedimentation at 60,000 rev/min in a Spinco analytical ultracentrifuge showed only a single component. Preliminary studies using the Archibald approach-to-equilibrium procedure indicate a weight/mole of approximately 15,000 g, in reasonable agreement with the minimum weight/mole of 14,700 g estimated from the iron content of 0-38 %. The Eq value at pH 6-4 was determined as +0-282 V by Henderson and Rawlinson (see this volume, p. 371), as compared with +0-255 V for ox-heart cytochrome c under similar conditions. Table 1 is a summary of the comparative properties of yeast and heart-muscle cytochrome c (see also Morton, 1958). It is apparent that the two proteins are not identical. The lower iso-electric point for yeast cytochrome c is consistent with the lower lysine content (15-16 and 18 residues/mole in yeast and heart-muscle cytochrome c respectively) reported by Nunnikhoven (1958). It is noteworthy that only one protein component was detected in these studies. Nunnikhoven (1958) extracted cytochrome c from yeast by Keilin's (1930) procedure, concentrated the extract on resin, and fractionated with ammonium sulphate after acidification with trichloroacetic acid (TCA). By electrophoresis at pH 8-3, two haemoprotein components were detected (containing 0-36 and 0-42% of iron for the faster and slower component respectively). Tsou (1956) obtained yeast cytochrome c by autolysis of dried yeast in m NaCl, concentrated it on a synthetic zeolite, eluted with 72 % saturated ammonium sulphate, and precipitated the cytochrome with TCA. Two haemoprotein components were detected, capable of separation by electrophoresis at pH 6-3. The major component had an iron content of 0-43 %. 386 Discussion However, Minakami (1955, 1956), who also avoided use of TCA, obtained only a single electrophoretic component in his preparation of yeast cytochrome c, which had an iron content of 0-37%, in close agreement with the results obtained in our work. In addition to the differences indicated in Table 1, we have observed that solutions of yeast cytochrome c are rather unstable as compared with heart-muscle cytochrome c. Prolonged dialysis against distilled water at 0-2°C may lead to autoxidation of yeast ferrocytochrome c, and the oxidized form is no longer reduced by cytochrome b^ and lactate. Similar results have been obtained with solutions of yeast ferrocytochrome c Effect of pH on mobility of c ytochrorre c 8 - • o Yeosf cytochrome C Beef cytochrome c Theorell and Akeson Tint ondReiss _ \ « b A \; b - \ \ o \ ^ - V \ ^.» - o V* • '^»^ - o ^ o^^ A - \ \ - • — 1 — , pH Fig. 1. pH/mobility plot for yeast ferricytochrome c and for heart-muscle ferricytochrome c (results from Theorell and Akeson, 1941). in the presence of ammonium sulphate. Denaturation of the cytochrome c may occur on oxidation of the reduced form with excess ferricyanide, and also on precipitation from solutions of low ionic strength with acetone at — 10°C. The prosthetic group was completely split from native yeast ferrocytochrome c very rapidly by treatment with silver sulphate and 2n acetic acid at room temperature (about 22"C), whereas the reaction with heart-muscle cytochrome c is slow under these conditions, and complete splitting requires higher temperature conditions as described by Paul (1951). However, the product obtained from the yeast cytochrome c appeared to be similar to those obtained by Paul from heart-muscle cytochrome c. As shown in Table 1, the absorption bands of yeast ferrocytochrome c at room temperature are displaced slightly towards the blue as compared with the heart-muscle protein, and these differences are even more pronounced at — 190°C (Estabrook, 1956). It is of interest that there is considerable similarity between the absorption spectra at — 190°C of yeast ferrocytochrome c and of the TCA-modified heart-muscle ferro- cytochrome c (Estabrook, 1956; and this publication, p. 444). Yeast ferricytochrome c obtained by Nozaki and co-workers (1957, 1958) was more readily digested by bacterial proteinase than heart-muscle ferricytochrome c. If the haemoprotein used «> a, 549-3 /J, 520-0 y. 414-7 6.314-0 U.V., 275 (This paper) 8 o£ |3 P < lis 388 Discussion in these studies was undenatured, the results suggest that native yeast cytochrome c has a more 'open' folding of the peptide chains than has native heart-muscle cyto- chrome c. The major reason for this study was to provide well-characterized, native cytochrome c from yeast for use with crystalline yeast lactic dehydrogenase (cytochrome bo). Both the yeast and heart-muscle cytochrome c were found to be equally reactive with the crystalline yeast enzyme. Both yeast and heart-muscle cytochrome c also react with heart-muscle cytochrome c oxidase preparations, as also found by others (Li and Tsou, 1956; Minakami, 1955, 1956). The results thus show that both yeast and heart-muscle cytochrome c are closely similar enzymically but have differences in physical and chemical properties which are rather greater than the differences as yet found for cytochrome c preparations from different animal sources. Although the prosthetic groups of yeast and heart-muscle cytochrome c would appear to be similar, there are undoubtedly considerable differ- ences in the secondary protein structure of the two cytochromes. It would appear that the reactivity of the prosthetic group of cytochrome c with other compounds (including enzymes such as cytochrome Zjj) is relatively independent of the folding of the peptide chains of the protein. REFERENCES Appleby, C. A. & Morton, R. K. (1954). Nature, Lond. 173, 749. Appleby, C. A. & Morton, R. K. (1959). Biochem. J. 71, 492. Armstrong, J. McD., Coates, J. H. & Morton, R. K. (1958). Abstr. Annual General Meeting, Aust. Biochem. Soc., Adelaide, August, 1958; Aust. J. Sci. 21, 119. Ehrenberg, a. (1957). Acta cheni. Scand. 11, 1257. Estabrook, R. (1956). J. biol. Chem. 223, 781. Keilin, D. (1930). Proc. roy. Soc. B106, 418. Leaf, G., Gilles, N. E. & Pirie, R. (1958). Biochem. J. 69, 605. Ll, W. C. & Tsou, C. L. (1956). Acta physiologica Sinica, 20, 50. Loftfield, R. B. & BoNNiCHSEN, R., quoted by Boeri, E. & Tosi, L. (1954). Arch. Biochem. Biophys. 52, 83. Minakami, S. (1955). /. Biochem. Tokyo 42, 749. Minakami, S., Ishikura, H. & Staako, K. (1956). /. Biochem. Tokyo 43, 575. Morton, R. K. (1958). Rev. pure appl. Chem. 8, 161. NozAKi, M., Yamanaka, T., HoRio, T. & Okunuki, K. (1957). /. Biochem. Tokyo 44, 453. NozAKi, M,, MizusHiMA, H., HoRio, T. & Okunuki, K. (1958). /. Biochem. Tokyo 45, 815. Nunnikhoven, R. (1958). Biochim. biophys. Acta 28, 108. Paul, K. G. (1951). Acta chem. Scand. 5, 389. Paul, K. G. (1951). The Enzymes, Vol. 2, Pt. 2, p. 375. Ed. Sumner, J. B. & Myrback, K. Academic Press Inc., New York. Theorell, H. & Akeson, a. (1941). /. Amer. chem. Soc. 63, 1804. Tint, H. & Reiss, W. (1950). /. biol. Chem. 182, 385. Properties of Native Cytochrome c HoRio: I would like to make a comment on the properties of cytochrome c. As found by Margoliash, native cytochrome c can be separated easily from modified cytochrome c by resin chromatography. Of the modified cytochrome c fractions, some fractions show the same absorption spectra as native cytochrome c. One can recognize easily which cytochrome c is native, by comparing various extraction- and purification-procedures, by comparison of the properties (ascorbate- oxidizing activity, redox-potential, etc.) and by comparing the properties of each sample after treatments such as lowering and raising pH, or heating. In the oxidized state, native cytochrome c is rapidly digested by a bacterial proteo- lytic enzyme. The modified cytochrome c is much more rapidly digested. This fact Electromctric and Other Studies of Cytochrome c 389 ndicates that the modified cytochrome c has a more unfolded protein-structure than native cytochrome c. With the same enzyme, native cytochrome c crystallized from baker's yeast is much more rapidly digested than those from ox- and horse-heart. Using native cytochrome c, the reduced form is much more resistant to digestion by the proteolytic enzyme than the oxidized form. In fact, if the rapid autoxidation of the reduced cytochrome c, (which is catalyzed by proteolytic products obtained by digestion of oxidized cytochrome c) is prevented by using anaerobic conditions, the reduced form is not attacked by the proteolytic enzyme at all. On the other hand, if one uses crystalline enzymes, for example, bacterial amylase, yeast alcohol dehydrogenase, animal triose-phosphate dehydrogenase, animal lactate dehydrogenase, the enzymes in their enzymically-active state are not digested by the proteolytic enzyme, while the denatured enzymes (regardless of whether the denatura- tion is reversible or irreversible), are rapidly digested. Kunitz and co-workers have reported that crystalline yeast hexokinase is digested by trypsin and that the enzyme becomes more resistant against the digestion in the presence of the substrates than in the absence (glucose > fructose > mannose). These considerations indicate that the reduced form of cytochrome c has a more rigidly folded protein-configuration than the oxidized form. As mentioned by Smith in connexion with estimation of cytochrome c oxidase activity (p. 260), using cytochrome c modified in its monomer state, one may not find serious differences in enzymic reduction and oxidation of native and modified cyto- chrome c. But generally speaking, enzyme proteins lose their enzymic activity at the same time as they lose the folded (coiled) protein-configuration. This concept does not fit with the enzymic reduction and oxidation of modified cytochrome c. From this point of view, I am just wondering what function is coupled with the folded con- figuration of the native cytochrome c, and with the alteration in the folded protein- configuration which may result from the oxidation-reduction reaction of cytochrome c. Reactivity of Native Cytochrome c in Oxidative Phosphorylation Morrison: As Henderson and Rawlinson (p. 369) have pointed out, the modified heart- muscle cytochrome c (fraction 2) obtained by the chromatographic method of Margo- liash does not appear to have markedly decreased biological activity as judged by its ability to transport electrons. This is not true when the role of cytochrome c in oxidative phosphorylation is assayed (Morrison, HoUocher and Stotz, Arch. Biochem. Biophys. 92, 33S, 1961. When this modified cytochrome c fraction was added to tightly-coupled rat liver mitochrondria, there was no increase in this rate of phosphorylation with added cytochrome c. This is in marked contrast with the results obtained with the so-called 'native' fraction of cytochrome c (fraction 1). These results point to the direct participation of the cytochrome c molecule in the oxidative phosphorylative mechanism. Structure of Bacterial Cytochromes of c- type Postgate: My question is addressed to Margoliash. Several of the bacterial cytochromes of the c-type are neutral or acid (see Morton, Rev. pure appl. Chem. 8, 161, 1958). If electrostatic links between polar groups in the protein are responsible for the very great thermal stability of this group of proteins, would not this thermostability be pH- dependent? If so, would not a study of the pH-dependence of the thermostability of a basic and a neutral cytochrome c show differences bearing on the nature of the linkage in folding of the peptide chain ? Margoliash: My idea of hydrophobically shielded electrostatic bonds as the bonds conferring on cytochrome c its native folded configuration and relative resistance to some types of denaturation procedures, such as heating, was advanced for the common mammalian types of cytochrome c which are strongly basic proteins. The thermo- stability of such cytochromes c is certainly pH-dependent. 390 Discussion There is, however, no particular reason why similar bonds could not occur in neutral or acid cytochromes c, since all that is required is a concentration of hydrophobic amino-acid side chains around the region where such an electrostatic bond may be formed. The effect of such an 'umbrella' would be to change the dielectric constant at the point where the electrostatic bond is to be formed, so as to render it more stable than it would be in a normal water environment. Without the specific hydrophobic microenvironment an electrostatic bond would be quite unstable even at neutral pH in an aqueous medium. Structure and Redox Potentials of Cytochrome c The effect of denaturation on E^ values for cytochrome c Margoliash: To obtain what we have called the trichloroacetic acid modified or denatured cytochrome c it is necessary to have the cytochrome c in solution during the denatura- tion procedure, and not precipitated. This is done by carefully controlling the pH. Under such conditions, or by using ethanol as a denaturing agent, one obtains a product that can be separated into fractions of increasing degree of denaturation by column chromatography on Amberlite IRC-50. These fractions have the full com- plement of amino acids present in the original native protein. The more the degree of denaturation, the lower the E^ value. The actual values obtained range from that of native cytochrome c at -fO-255 V to not far from 0-0 V. The ligands to iron in cytochromes of the c-type Perrin: The difference of 0-54 V in Eq between cytochrome c and cytochrome c peptide indicates that in the latter the ferric form is stabilized by a factor of 10^ relative to the ferrous form, compared with cytochrome c. This represents a preferential stabiliza- tion of the ferric cytochrome c peptide by a AG of about 13 kcal. This difference is much greater than is found in the iron porphyrin haemochromes where the ligands are neutral molecules and I should like to suggest that in the peptide the sixth link to the metal is through a carboxyl group (e.g. of glutamic acid) or some other anion because anions form much more stable complexes with ferric than with ferrous iron. The very large difference between cytochromes c^ and C3 would then indicate that in the former the sixth position ligand is neutral (e.g. amino-N or imidazole-N) while in the latter it is carboxyl. Falk and I (p. 66) have indicated why we think that the absorption spectra of such complexes are insensitive to the particular ligands other than the porphyrin groups. Margoliash: I wish to reply to Perrin's suggestion that the lowering of the Eq value of cytochrome c on denaturation may be due to an exchange of a — COO~ group for one of the normal haemochrome-forming ligands in the native protein, during denaturation as well as during proteolytic digestion to form the 'core'. I think that the spectrophotometric evidence rules out such a possibility. Whether a ligand is acidic or basic, thespectrophotometrically-observed pkfor the dissociation of a particular ligand would be at a pH below that of the normal pAT value for the ligand in solution, since as pointed out previously by Perrin, one must take into con- sideration the constants involved in the reaction of the ligand with the haem iron. In the ferro- form of the cytochrome c 'core', the lowest apparent pA" value we have observed spectrophotometrically, using the visible region of the spectrum, was at pH values between 5 and 6. This clearly rules out the participation of a carboxyl as a liaemochrome-forming group in the reduced form of the 'core'. With reduced denatured cytochrome c the results will naturally vary with the degree of denaturation of the fraction studied, all the way from an apparent pA: at pA:2-3 for the native protein, to a behaviour essentially identical to the 'core' for the most denatured material, again showing the lowest pA" at pH values of 5-6. Since as dis- cussed previously, the ZTq' value will decrease with an increasing degree of denaturation and one would expect that the introduction of a carboxyl group would result in a decrease in Eq, Perrin's suggestion cannot be reconciled with the spectrophotometric Electrometric and Other Studies of Cytochrome c 391 results, at least for those derived forms of the protein we have studied, which show pAT values apparently shifting towards the alkaline side as denaturation proceeds. I would, however, like to stress that our results with the E^' values for denatured cytochrome c fractions are compatible with the theory that it is not the particular ligands involved that establish the properties peculiar to the cytochrome c haemo- chrome as contrasted with those of an ordinary chemical haemochrome, but rather the native configuration of the protein. This holds not only for the haemochrome properties I discussed following Paleus' paper, but also for the oxidation-reduction potentials. Influence of dilhionite on redox potential of cytochrome c Henderson: According to Paul {The Enzymes, Academic Press Inc., New York, 1951) dithionite when used to titrate electrometrically cytochrome c gives the low value of +0-12 V instead of +0-26 V obtained with other reductants. This may be the explanation for the similar value of Green {Proc. roy. Soc. B114, 423, 1934) for yeast cytochrome c. This would presumably be due to some compound formation with the cytochrome c. Has Estabrook observed any difference in the low-temperature spectra between cyto- chrome c reduced with dithionite and the same material reduced by other means, as they might show such compound formation ? Estabrook: We have used mainly enzymic reduction where possible to try and detect any difference in spectra and find that there is no obvious difference between the spectral properties at low temperature of samples of purified haemoproteins reduced enzymically or by dithionite treatment. In the cases of the haemopeptide of cyto- chrome c or the alkaline haemochromes there was, of course, only dithionite reduction employed. THE ELECTRON TRANSFER FROM CYTOCHROMES TO TERMINAL ELECTRON ACCEPTORS IN NITRATE RESPIRATION AND SULPHATE RESPIRATION By F. Egami, M. Ishimoto and S. Taniguchi Department of Biophysics and Biochemistry, Faculty of Science, University of Tokyo, Tokyo, and Department of Chemistry, Faculty of Science, Nagoya University, Nagoya Until 1949 the cytochrome system was considered to participate only in the electron transfer to oxygen. The participation of a cytochrome in anaerobic oxido-reduction was first suggested by Sato and Egami (1949) in the case of nitrate reduction in E. coli. Recently a cytochrome was found even in a strict anaerobe, Postgate (1954) and Ishimoto, Koyama and Nagai (1954a, b) independently demonstrating the participation of a cytochrome in the reduction of sulphur compounds in Desulphovibrio desiilphuricans. Nitrate reduction with participation of a cytoclirome system has been called by us 'nitrate respiration', because its physiological meaning is not nitrogen assimilation, but energy-yielding through oxido-reduction similar to aerobic respiration (Sato, 1950; Taniguchi, Sato and Egami, 1956). The electron transport mechanism functioning in such processes also appears to resemble that of aerobic respiration. It has been shown that the electron- transferring system in nitrate respiration in E. coli resides on particulate structures containing cytochrome b-^ as in the case of the analogous system involved in aerobic respiration. On the other hand, enzyme systems of different nature in soluble form without participation of cytochromes have been shown to be responsible for the reduction of nitrate in 'nitrate assimi- lation' (Nason, 1956). Similarly sulphate reduction in Desulphovibrio desulphur leans, being an energy-yielding mechanism with participation of cytochrome C3, may be called 'sulphate respiration'. Our studies on the mechanisms of enzymic electron transfer in nitrate respiration and sulphate respiration until 1957 were summarized by Tani- guchi, Asano, lida, Kono, Ohmachi and Egami (1958) and Ishimoto, Kondo, Kamayama, Yagi and Shiraki (1958) respectively at the International Symposium on Enzyme Chemistry held in Tokyo in 1957. However, the 392 Electron Transfer in Nitrate and Sulphate Respiration 393 electron transfer from the cytochromes to corresponding terminal electron acceptors had been scarcely investigated on the enzyme level till the symposium. The present paper describes the recent results obtained in Egami's labora- tories by Taniguchi, Ohmachi and Itagaki and by Ishimoto and Fujimoto on the terminal electron transfer in nitrate respiration and sulphate respiration respectively. NITRATE RESPIRATION Cytochrome b^ and Nitrate Respiration A highly active nitrate-reducing system of nitrate respiration type was found in E. coli grown in anaerobiosis with nitrate. As shown in Table 1, anaerobic cells had very little activity in the aerobic reduction of nitrate. A remarkably opposed tendency could be seen in cells grown aerobically in the same medium with nitrate. They exhibited aerobic reduction much more than anaerobic reduction (Table 1).* Table 1. Nitrate-reducing activity NaR of E. coli cells* Reaction mixture contains 100 /(moles of KNO3, 1 //mole of formate, 240 //moles of phosphate buffer. pH 71, and E. coli cells in total volume of 3-0 ml. Temperature: 30°C. Conditions of nitrate reduction in vitro Nitrate-reducing activity of cells under various growth conditions (in //moles of NO3- reduced/mg N/hr) Aerobic Anaerobic Aerobict (^ NH, - . . .) Anaerobic^ (-* NOj") 30-50 10-20 0-5 200-500 t In Warburg vessels under vigorous shaking. J In Thunberg tubes under anaerobic conditions. The addition of nitrate in the anaerobic culture stimulates not only the bacterial growth but also the cytochrome b^ content of the cells to the extent of 2 to 3 times higher (almost to the same level as that of aerobic cells) as compared with that of cells grown in anaerobiosis without nitrate. * Abbreviations The following abbreviations have been used in this paper: nitrate reductase, NaR cytochrome, cyt.; formate dehydrogenase, FDH; /)-chloromercuribenzoate, PCMB reduced methyl- or benzylviologen, MVH or BVH; phenazinemethosulphate, PMS methylene blue, Mb; flavin adenine dinucleotide, FAD; flavin mononucleotide, FMN adenosine-5'-phosphosulphate, APS; 3'-phosphoadenosine-5'-phosphosulphate, PAPS adenosine triphosphate, ATP; 2-heptyl-4-hydroxyquinoline-N-oxide, HOQNO; diphos- phopyridine nucleotide, DPN; reduced di- or tri-phosphopyridine nucleotide, DPNH or TPNH; 2-amino-2-hydroxymethylpropane-l :3-diol, Tris. I) 394 F. EgAMT, M. ISHIMOTO AND S. TaNIGUCHI The anaerobic cells cultured anaerobically with nitrate and containing the active mechanism of nitrate respiration were used throughout this study. The nitrate reductase activity was found in the particulate fraction (sedimented at 2,000-20,000^) with cytochrome b^ and formate dehydro- genase after alumina or sonic destruction of cells (Table 2). This particulate Table 2. Distribution of nitrate reductase, formate dehydrogenase AND PROTOHAEM IN CELLULAR FRACTIONS OF E. CoU 40 g of wet cell paste was ground with alumina and subjected to the foUowing centrifugal fractionation. All determinations at 30°C. Fraction Total N NaR activity* FDH activityt Protohaem (mg) Total Specific Total Specific Total m/ig/mgN Original 36-5 4250 117 1880 51 1860 51 (2,000^ extracts) Large particle 10-2 2360 235 3400 340 1230 121 (2,000-20,000^) Small particle 2-9 224 77 50 160 300 103 (20,000-110,000^?') Soluble fraction 26-5 235 9 450 17 (110,000-) * The donor for NaR was MVH. One unit of NaR is defined as an amount of the enzyme which produces 1 /-hydroxybutyrate and glucose, but also hydrogenase (production of H., from reduced methyl- viologen, MVH), and of reduced tri- and di-phosphopyridine nucleotide (TPNH and DPNH) oxidase activity. Contrary to the very high activity of +0.3r 432 [Reduced] formate " Coxidized] 10 min.- preincubation [Reduced]-[oxidized] formaie+KN03 (IOMmoles,solid) after 1 minute ^0.l -0.1'- 412 Fig. 1 . Difference spectra of particulate preparation. Tlie solution in Thunberg tube-type cuvette contained large particles (0-6 mg N/ml), phosphate buffer 50/imoles, pH 7-1 and formate 50/imoles in 3-0 ml under anaerobic conditions at 30°C. formate dehydrogenase and oxidation of formate by nitrate, aerobic oxida- tion of formate was found to be weak. Thus, the particle seems to be highly characterized by the well-developed formate dehydrogenase — cytochrome b^ — nitrate reductase system, i.e. the enzyme system of nitrate respkation type. Nitrate reduction by the particulate nitrate reductase with various electron donors is summarized in Table 3. Cytochrome Zj^ in the particulate fraction was rapidly reduced with DPNH or even more rapidly with formate under anaerobic conditions. TPNH was quite inactive as an electron donor. Cytochrome b^ thus reduced was rapidly reoxidized with nitrate (Fig. 1). In this case nitrate reduction was strongly inhibited by 2-heptyl-4-hydroxy- quinoline-N-oxide (HOQNO), a specific inliibitor for cytochrome b^. On the contrary, nitrate reduction by reduced dyes such as MVH was not inhibited by HOQNO (Table 4). 396 F. Egami, M. Ishimoto and S. Taniguchi Table 4. Inhibition and activation of the PARTICULATE NITRATE REDUCTASE ACTIVITY BY SEVERAL REAGENTS Inhibition (%) Concentration (M) Reagent With formate With MVH Amytal 2 X 10-3 75 -20* Menadione 5 X 10-" -40* Dicoumarol 3 X 10-5 70 25 Dicoumarol + menadione 3 X 10-5 5 X 10-« 70 - HOQNO 10-5 85 PCMB 5 X 10-* 60 20 PMS 3-3 X 10-" -160* 5 * Activation. Nitrate reduction by the particulate system with formate as an electron donor was strongly inhibited not only by CN- and N3-, but also by dicou- marol, amytal and /7-chloromercuribenzoate (PCMB). This finding suggests the participation of flavoprotein, dicoumarol-sensitive factor and SH group besides heavy metals in the electron transfer sequence. On the other hand it was activated by vitamin K3 and phenazine methosulphate (PMS) (Table 4). From these findmgs the following electron transfer sequence for the particulate system can be proposed : Formatel DPNH I -^ (FAD) -> Cyt. bi Nitrate Reductase (NaR) ^ NO3 t MVH Phosphorylation Coupled with the Electron Transfer The decrease of inorganic phosphate was observed coupled with nitrate reduction with formate as an electron donor by the particulate fraction prepared with 2-amino-2-hydroxymethylpropane-l:3-diol (Tris) buffer (0-02 M, pH 7-2). Adenosine triphosphate (ATP), glucose, yeast hexokinase, Mg++ and F~ were added in the reaction mixture besides inorganic phosphate. Glucose was scarcely utilized as an electron donor. It was found that the particulate fraction itself contained ATP-ase which was activated by Mg++ and inhibited by F-. The decrease of inorganic phosphate during the nitrate reduction indicates the occurrence of an anaerobic phosphorylation coupled with the electron transfer from formate to nitrate through cytochrome h^. These experiments are still in a preliminary stage and the fate of phosphate incorporated remains to be elucidated. Electron Transfer in Nitrate and Sulphate Respiration 397 Solubilization and Purification of Particulate Nitrate Reductase The greater part of nitrate reductase, estimated with MVH as an electron donor, of the particulate fraction was solubiHzed, when heat-treated at 60°C for 5 min and kept at pH 8-3, at 4°C for 20 hr. Formate dehydrogenase and cytochrome b^ were not solubilized by the procedure. Nitrate reductase thus solubilized was purified to about 1 ,000 times the activity of cell paste (yield 10-20%) by adsorption onto and elution from calcium phosphate gel, ammonium sulphate fractionation, and ultracentrifugation (1 10,000 g, 3-4 hr). The purified nitrate reductase with yellowish-brown colour is homogeneous in ultracentrifugal and electrophoretic analysis and has a specific activity as high as about 200,000 /^moles NOa" formed/mgN/hr (Table 5). Table 5. Solubilization and purification of nitrate reductase NaR Specific Step Fraction N (mg) activity (units X 10"*) activity (units/mg N X 10-3) 1 Frozen cell paste 270 g wet wt. 018 2 Large particle (2,000-20,000^) 2500 135 0-54 3 Supernatant after heat treatment and incubation in cold 580 100 1-72 4 Calcium phosphate gel eluate 227 90 3-95 5 (NH4)2S04— 1st ppt. (38-53% satu- ration) 55-2 80 14-5 6 (NH4)2S04— 2nd ppt. (3040% satu- ration) 11-2 50 56-2 7 1st pellet of ultracentrifuge (1 10,000^, 3-4 hr) 3-04 40 132 8 3rd pellet of ultracentrifuge 1-50 28 186 It has no specific absorption peak except that of protein (275-280 m/i) and the absorbance decreases gradually over the entire near-ultra-violet and visible region with increasing wavelength. The difference spectrum, (oxidized — reduced enzyme) showed a broad peak around 445-450 m// which instantly disappeared on the addition of nitrate, with simultaneous production of nitrite (Fig. 2). The purified enzyme preparation contains about 40 atoms of Fe per molecule, on the basis of its weight of approx. 10^ g/mole, as estimated from measurement of sedimentation constant, 8%^ ^ = 25-0 s, and diffusion coefficient, 1)20, w = 2-27 X 10~^ cm^ sec^^, and assuming its partial specific volume (Foo) as 0-75 ml/g. By emission spectrographic analysis using the cathode layer method, the existence of Mo (1 atom/molecule) besides Fe was 398 F. Egami, M. Ishtmoto and S. Taniguchi observed (Table 6). Analysis of flavin in the homogeneous preparation, as recorded in Table 7, led to the conclusion that the flavin content in the enzyme is close to the limit of error of determination. So the flavin content is, if any, only a trace and the enzyme may be regarded as a metalloprotein O.D. 0.2 Oxidized] -[Reduced] NaBH4 + KN03(IOpmoles) 400 500 600 rn^L Fig. 2. Absorption spectrum of purified nitrate reductase. The solution contained NaR preparation (specific activity: 15 x 10* units/mg N) in final concentration of 1-9 mg protein/ml in phosphate buffer 0-05 m, pH 7-1 in Thunberg tube-type cuvette at 30°C. but not as a metalloflavoprotein. In our previous preUminary communica- tion (Taniguchi and Itagaki, 1959), it was considered to be a metalloflavo- protein. Purified solubiUzed nitrate reductase does not reduce nitrate w^ith formate or DPNH, but reduces nitrate with MVH and, less actively, with reduced flavin adenine dinucleotide (FAD), riboflavin phosphate (FMN) and free riboflavin (Table 8). CN" and Ng^ strongly inhibit the reaction (Table 9). CO (1 atni in the dark) has no effect. Nitrate reductase thus obtained may be regarded as a solubiUzed form of the terminal oxidoreductase in the particulate fraction and so is an anaerobic variant of terminal oxidases (Fig. 3). HOQNO Formate (FAD)— t-cyf b, Incubation in cold after 5 min heating of eCC.pH 8.3. -«-NaR t FADH?, MVH no; phenosofranine - Hg Fiu. 3. Particulate nitrate-reducing system. Table 6. Metal content of purified nitrate reductase Experiment Specific activity (units/mg N X 10 ") Duration of dialysis in hours* Number of atoms/NaR molecule on the basis of its molecular weight (lO**) Mo Fe 1 2 16 18 20 15 50 l-2-0-8t Vl-O-lt 45 43 36 * Dialysis was against deionized water. i Internal standard method with Pd and V. X Visual method. Table 7. Content of flavin in nitrate reductase Experiment Specific activity of NaR (units/mg N X 10-^) Content of riboflavin in figjmg dry wt Number of flavin molecules/NaR molecule 1 2* 3 4 5 22 22* 18 15 20 2-8 00 0-8 0-2 0-009 0000 0002 00005 002t * The sample used was the same preparation as in Experiment 1 but with preliminary digestion by crystalline chymotrypsin (8% of weight of NaR) for 3 hr at pH 8-15, 37°C. t This value was obtained by D-amino acid oxidase assay. Table 8. Comparison of electron donors for purified nitrate reductase activity Electron donors Vmax K,„ for the donor (reduced compound in each case) (relative value) (M X 10^) Methylviologen 100* cyt . c^ < > NH2OH DyestulTs such as benzyl- or methylviologen will replace cytochrome c^ in such reactions. Reactions of this kind may be expected to occur in any system involving low-potential electron transport enzymes; the possibly 'artificial' character of certain accepted steps in the reduction of nitrate was pointed out by Senez and Pichinoty (1958). 412 J. POSTGATE Reduction of Sulphite Oxidation of c^ by sulphite can be observed spectroscopically in intact bacteria, but the enzymological evidence for participation of cytochrome Cg in sulphite reduction is inconclusive. Some enzyme preparations show small stimulations of sulphite reduction when extra cytochrome Cg is added. Recent work by the author has shown that the sulphite reductase system is complex, as indicated by the following results. Cells of DesuJphovibrio desuJphuricans (Hildenborough) grown in continu- ous culture were disrupted by treatment with liquid Ng (Moses, 1955) and treated with deoxyribonuclease (to keep the preparation liquid). Centri- fugation yielded a red particulate fraction having most of the sulphite reductase activity (together with much bound cytochrome Cg) and a super- natant fraction without which the sulphite reductase activity of the particulate preparation was negligible unless benzylviologen were added. The particles then showed 1-5 to 3% of the sulphite reductase activity of the original cells; the liquid fraction had only slight reductase activity though hydrogenase alone was plentiful. The liquid 'co-sulphite reductase' could be further fractionated into two heat-labile proteins (one soluble in 36% ammonium sulphate, the other not) and a dialysable factor that could be recovered by freeze drying the dialysate. Addition of cytochrome Cg to the complete preparation (particles and soluble fraction) did not influence the rate of reduction of sulphite at all (there is probably sufficient cytochrome Cg bound with the particles) even if cytochrome Cg had previously been removed from the soluble fraction by ion-exchange. Complete enzyme preparations were stimulated further by benzylviologen but not by diphosphopyridine nucleo- tide, triphosphopyridine nucleotide, coenzyme A, a-lipoic acid, hypoxanthine, methyl-l:4-naphthoquinone, pyridoxal phosphate, adenosine-3'-phosphate, pyruvate, succinate, cysteine, glutathione, a concentrate of desulphoviridin, or by salts of the following metals : Mo, V, Mn, Co, Ni, Mg, Zn or Ca. Modest stimulation was observed with adenosine triphosphate (ATP) and acetyl phosphate and a slight effect occurred with ferrous ions ; these materials did not, however, replace the dialysable component referred to earlier. Some quantitative data obtained with particulate sulphite reductase preparations are given in Table 3. These observations show that the sulphite reductase system involves several components : 1. hydrogenase 2. thermolabile protein A 3. thermolabile protein B 4. cytochrome Cg 5. a dialysable co-factor 6. a sulphite reductase. Cytochrome Cg 413 Table 3. Activity of sulphite reductase preparations Experiments were conducted in conventional Warburg manometers under hydrogen gas at ZTC in 0-5°^KHjPO4(pH 6-9) with CdClj in centre well. Negative (?h2 values for Na.jSOj recorded (in //I H3 absorbed/mg N/hr) from five experiments. Preparations Hydrogen absorbed (/ — CH— CH3 + CO2 + RSH SR Are these catalase-free organisms sensitive to oxygen, and is anything known of their pyruvate metabolism ? Postgate: The organism is totally deficient in catalase; its metabolism is stopped by oxygen but it is not killed even after several hours in air. Its pyruvate metabolism is phosphoclastic, yielding acetyl phosphate, CO2 and either free hydrogen or ethanol. Falk : The question of the synthesis of protohaem by some anaerobes is a very interesting one. In animal tissues, oxygen is required \n the biosynthetic pathway for the formation of succinyl-coenzyme A, and again for the conversion of two propionic acid groups to vinyl groups : all the remaining steps can be performed anaerobically. It has been found recently (Falk et al.. Nature, Land. 184, 1217, 1959) that for protoporphyrin synthesis in chicken erythrocytes in vitro there is a sharp optimal oxygen tension at about 0-07 atmospheres ; oxygen tensions higher than this are increasingly inhibitory and at the 0-2 atm. of sea-level air the inhibition is already some 25 %. We have found that this inhibition by oxygen operates somewhere after porphobilinogen has been formed, and before uroporphyrin appears, and we are currently looking for the sensitive reaction. We feel that this inhibition is directly related to the adaptation of haemopoiesis to oxygen tension (e.g. altitude) and to such adaptations as those found in Daphnia and other Crustacea and invertebrates, though other factors also are no doubt involved. It appears possible that this control is one of the factors in adaptive cytochrome synthesis in micro-organisms also; in fact it was the finding of Moss {Aust. J. exp. Biol, med. Sci. 34, 395, 1956) that the synthesis of cytochromes by Aerobactor aerogenes shows an optimum against oxygen tension which led us to these studies. To return to the less common but definite anaerobic formation of protohaem by micro-organisms, the requirements, as different from animal tissues, are that the formation of succinyl coenzyme A and of vinyl-groups must occur anaerobically. As far as the latter is concerned, it is easy to postulate a dehydrogenase with a terminal acceptor other than oxygen. In a way, we have an analogy for this, though in a reductive, not an oxidative, dehydrogenase-type reaction, in the enzymic, non photo-catalytic conversion of protochlorophyll to chlorophyll in certain plants. Kaziro: In connexion with the problem now under discussion, I would like to mention that there are some bacteria, not strictly anaerobic, which produce hydrogen peroxide as a metabolite. Since they have no catalase, the hydrogen peroxide accumulates in the medium as they proliferate. It might be of some interest, as suggested now by Lemberg, to refer to a peculiar disturbance observed in the cases of congenital deficiency of catalase in man. I have 418 Discussion had the opportunity of finding some cases of these deficiencies and to investigate them. In most of these cases a peculiar malignant inflammation over the oral region was observed which started from the root part of the teeth and resulted in severe pro- gressive gangrenous ulcers. Often, the ulcer progressed to the jaw bone. The dis- turbance has been designated as Takahara's disease after the name of the first dis- coverer. The pathogenesis has so far been explained as the result of bacterial infections and the tissue disturbance is caused by a toxic action of hydrogen peroxide produced continuously by the infecting bacteria. These grow well at the gingiva, especially where gingiva and teeth root touch. In some cases, the infection was reported to have started from fissures in the tonsil. Among bacteria which are found in the infected gingiva are strains of Lactobacillus, Streptococcus haemolyticus, Diplococcus pneumoniae and others. They are known to produce hydrogen peroxide and have no catalase. The above-mentioned abnormality has long been reported in Japan and so far in Japan only. It was found for the first time by Takahara in 1946 at the polyclinic of the Okayama University. Until now, we have had 35 cases of the abnormality originating from 16 families which have been described by various physicians in different places in Japan (see Nishimura at al.. Science 130, 330, 1959). THE ATYPICAL HAEMOPROTEIN OF PURPLE PHOTOSYNTHETIC BACTERIA By M. D. Kamen* and R. G. Bartsch* Graduate Department of Biochemistry Brandeis University, Waltham, Massachusetts INTRODUCTION All PHOTOSYNTHETIC systciiis contain large amounts of haemoproteins bound to photoactive subcellular aggregates (Kamen, 1957). These haematin compounds are in many ways quite similar to those of the familiar respiratory systems and can be included in the categories known to biologists since the classic work of Keilin. However, there appears to be a novel type of haemo- protein which can be extracted in soluble form from all the common species of purple photosynthetic bacteria (Vernon and Kamen, 1954; Kamen and Vernon, 1955; Newton and Kamen, 1956; Bartsch and Kamen, 1958) and which cannot be placed in any single recognized class of haemoproteins, e.g. cytochromes, haemoglobins, catalases, etc. On the occasion of the present symposium, it is appropriate to summarize briefly our knowledge about this atypical haemoprotein. GENERAL PROPERTIES Distribution of the Atypical Haemoprotein {'RHP') The first specimen to be noted and isolated was found in trichloroacetic acid extracts prepared from cell suspensions of the facultative photohetero- troph, Rhodospirillum rubrum (Vernon and Kamen, 1954). At first, the compound was thought, on spectroscopic grounds, to be a "pseudohaemo- globin'. In further studies, the revelation of its anomalous character created uncertainty mirrored in a series of progressively less definite names which culminated finally in the present noncommittal 'Rhodospirillum haem protein' , abbreviated to 'RHP'. We now customarily use 'RHP' to refer to all examples of the atypical haemoprotein, regardless of the species from which it may be obtained. RHP has been found in representative species of all the presently- recognized genera of the purple non-sulphur bacteria (Athiorhodaceae) and * Publication No. 61 in the series 'Publications of the Graduate Department of Biochemistry, Brandeis University.' 419 420 M. D, Kamen and R. G. Bartsch in the only readily available species (strain D of the genus Chroinatiwn) which is representative of the purple sulphur bacteria (Thiorhodaceae) (Kamen and Vernon, 1955). Preliminary attempts to detect RHP in other photosynthetic systems, such as green plants, algae and the green sulphur bacteria, have been unsuccessful to the present. If further studies, now in progress, should confirm that the occurrence of this unique haemoprotein is restricted to the purple bacteria, this result would occasion no great surprise. The same condition exists with regard to the chlorophyll and carotenoid components of the purple bacteria, relative to those of other photosynthetic tissues (Van Niel, 1944; Goodwin, 1955; Goodwin and Land, 1956^; Goodwin, 1956; Goodwin, Land and Sissins, 1956; Goodwin and Land, 1956ft). It is possible that a haematin compound similar to RHP is present in a halotolerant nitrate-reducing coccus studied by Taniguchi, Asano, lida, Kono, Ohmachi and Egami (1958). A soluble fraction obtained from cell suspensions of this micro-organism contains an enzyme active as a hydroxyl- amine reductase, which exhibits absorption spectra and chemical properties closely similar to those of RHP (see discussion in a later section). The fact that RHP occurs in the purple photosynthetic bacteria, and possibly in the nitrate-reducing bacterium mentioned above, points to the desirability of searching for RHP among bacterial species which are chemo- synthetic analogues of the photosynthetic bacteria, e.g. sulphate-reducers, sulphur oxidizers, ammonia and hydrogen oxidizers, nitrate-reducers, etc. Many studies on the absorption properties of bacterial suspensions using the elegant methods of dynamic spectrophotometry developed by Chance (1954), Duysens (1954) and others have revealed the existence of spectro- scopic entities among a variety of bacterial species (Bacillus subtilis, Proteus vulgaris, Achromobacter fischeri. Staphylococcus albus, etc.) which appear to resemble RHP closely (Smith, 1954). These pigments appear to function as terminal oxidases, and the term, 'cytochrome o\ has been suggested to denote them as a class (Castor and Chance, 1959). In the particular case of R. rubrum, isolated RHP and the cytochrome o moiety are so much alike spectroscopically, there is strong temptation to equate them and to consider RHP as a prototype for the cytochrome o class. We shall return to this point later. Physical Properties RHP has been isolated in varying degrees of purity from all the purple photosynthetic bacteria, but in only two cases has it been obtained sufficiently pure and in such amounts as to permit extensive investigation of its physical and chemical properties. These are the original RHP of R. rubrum and that of the strict photoanaerobe, Chromatium (strain D). Salient physical properties of these two preparations are listed in Table 1. The references Haemoprotein of Purple Photosynthetic Bacteria 421 cited in this table contain detailed descriptions of methods used. It will be seen, not only in connection with these particular properties but with others discussed later, that RHP's vary somewhat in detailed physical nature. However, the variation is not more than that encountered in other estabhshed categories of haematin compounds, such as cytochromes of the c type, b type, haemoglobins, etc. Table 1. Physical properties of rhp* (Cf. Newton and Kamen. 1956; Bartsch and Kamen, 1958) Property R. rubrum Chromatium Isoelectric point (pH) V(cmVg) ^t (pH 5-0, ionic strength = Mol. wt. £o' (pH 7-0) = 0-l)t 4-43 2-7 X 10-13 8-65 X 10-7 0-731 + 6-8 26,000 (±2,000) -0-008 V§ 5-5 ± 01 3-2 X 10-13 7-5 X 10-7 0-706 + 5-4 36,000 (±3,000) -0005 V§ * Values in this table corrected in proof; see Horio and Kamen, 1960. t C.G.S. units. % Electrophoretic mobilities at other pH and ionic strengths are omitted. See original references cited above. § Electrometric titration shows that oxidation-reduction of RHP is reversible and involves a one-electron change. Chemical Properties The chemical behaviour of RHP has been studied mainly using the R. rubrum preparation, but recent studies (Bartsch and Kamen, 1960) have shown that results obtained with this preparation are duplicated in every detail using the Chromatium RHP. We shall summarize here the chemical properties most relevant for the present discussion. RHP is soluble in water. It appears to have minimal solubility below pH 5, as is expected from its isoelectric point. In general, it salts out with ammonium sulphate at concentrations which differ slightly from those effective with the corresponding cytochrome c protein obtained from the same source. However, RHP shows solubility behaviour so much like cytochrome c that it is advantageous to devise extraction procedures which minimize contamination with cytochrome c before commencing extensive fractionation of RHP (Bartsch and Kamen, 1958). RHP resembles mammalian cytochrome c in heat stability and resistance to denaturation by acid or base. However, at strongly alkaline pH (greater than 1 1), RHP undergoes a reversible denaturation in which it assumes the spectrochemical form of a haemochrome hke cytochrome c, with char- acteristic triply-banded spectra in oxidized and reduced forms. The original 422 M. D. Kamen and R. G. Bartsch myoglobin-like spectra are obtained when the pH is brought back to neutrahty. Unhke cytochrome c, RHP is rapidly auto-oxidizable. It can be oxidized and reduced reversibly using the reagents commonly employed for this purpose in chemical manipulations of haematin compounds. The alkaline haemochrome form of RHP can also be oxidized and reduced reversibly. However, in the presence of a detergent such as lauryl sulphate, repeated cycles of oxidation and reduction at pH > 11-8 result in progressive loss of haem. This result is not observed at pH 7. The irreversible oxidative degradation in alkah is reminiscent of the behaviour of haemoglobin in the presence of detergents at physiological pH. If RHP, while still partially denatured in dilute alkali and in the presence of lauryl sulphate, is incubated with 4-methylimidazole, it forms a typical haemochrome. Denaturation with strong alkali is required before ligands such as cyanide, pyridine, fluoride and azide will bind the central iron atom of RHP. The prosthetic group cannot be split by cold acid-acetone, as in myoglobin, haemoglobin, or cytochrome b; reductive cleavage with sodium amalgam, strong acid and heat, or fission with silver salts and acid (Paul, 1951) are required as in cytochrome c. The products obtained by the sodium amalgam procedure (Davenport, 1952) are identical with those derived from cytochrome c. The pyridine and cyanide haemochromes derived after denaturation in strong alkali are identical spectroscopically with those of cytochrome c (Vernon and Kamen, 1954). Hence, it is likely that the prosthetic group of RHP is identical in type and nature of binding with that of the haem in cytochrome c, that is, the haem group is bonded by saturated (thioether) linkages between the vinyl side chains of the haem moiety and cysteine residues of the protein. RHP is most peculiar in its response to reagents which normally form reversible addition products with haem. Thus, it does not react at neutral pH with HgS, O2, NO, NaNg, NaCN and 4-methylimidazole, either in the reduced or oxidized form. In this respect, it resembles cytochrome c or cytochrome b. However, it will bind carbon monoxide reversibly in the reduced form to give a typical haemochrome, which is extremely light- sensitive. If the pH is kept below 4, the carbon monoxide complex is very light-stable. The chemistry of RHP can be understood on the basis that it is a variant of cytochrome c, capable of reversible denaturation in varying degree, as evidenced by its response to various ligands. The particular variation intro- duced in RHP may be assumed to be a coiling of the peptide chains, so that one or both of the usual basic co-ordination groups of the protein cannot bind to the extraplanar valences of the central iron atom. Falk and Nyholm (1957) have presented correlations between electro- chemical potential, spectra, mode of binding and substituents on the Haemoprotcin of Purple Photosynthetic Bacteria 423 /^-carbons of the pyrrole moieties of porphyrins and metalloporphyrins. These support the suggestion (Bartsch and Kamen, 1958) that the protein presents an electronegative group, such as carboxyl, to the central iron atom of the haem (see also, Williams, this Symposium, p. 41 ) and that it is this functional grouping (possibly as part of a long chain) which prevents approach of a basic co-ordinating group. It will be of great interest to attempt isolation of a haem peptide from RHP and if successful, to ascertain its structure. Immunochemical Properties All of the haemoproteins, including the RHP type, derived from either R. rubrum or Chromatium, can function as antigens, evoking characteristic antibody synthesis in rabbit sera. The antibody against any one haemo- protcin, when tested by the Ouchterlony technique, by precipitation, by complement fixation, etc., does not appear to interact with any other haemo- protein, whether of a different type from the same bacterium or of the same type from a different bacterium. For example, the anti-RHP of Chromatium will not interact with the R. rubrum RHP, nor with cytochrome c from either Chromatium or R. rubrum. Thus, there are strain differences in RHP from different sources, just as there are in the haemoglobins and myoglobins. The antigenic properties of RHP preparations afford a means for applying immunochemical techniques as aids in structure determination, and also as a basis for monitoring denaturation during extraction. Spectroscopic Properties In Figs. 1 to 5, characteristic features of the absorption spectra under various conditions are exhibited. The main properties may be summarized as follows. 1. Spectra of the oxidized and reduced forms (Figs. 1-3) resemble those for a myoglobin pigment, except for a general sliift of all absorption maxima toward shorter wavelengths than those found for myoglobin or peroxidase. Tliis shift accords with the apparent mesohaem character of the prosthetic group of RHP. 2. There is a characteristic haematin band at 630-640 m^ which dis- appears on reduction and reappears on oxidation. This band is used for preliminary monitoring of extracts for the presence of RHP. 3. The Soret band of the reduced form is complex, showing a persistent shoulder some 5-10 m/^ on the long wavelength side of the maximum absorp- tion at 423-426 m^. There appears to be a more pronounced shoulder in the R. rubrum compound (Fig. 1) than in the Chromatium compound (Fig. 3). 4. Carbon monoxide forms a complex with reduced RHP, showing a typical haemochrome spectrum (Fig. 5). As compared with unreacted reduced RHP, the Soret peak of the complex RHP sharpens, intensifies, and 1.2 _ 1 1 1 1 1 0.4- I.I - "'''"/A 0.3- 1.0 - \ / '' \ °'^~ >_ V_^/ \ \ t 09 " " N^ .V^-'^s °'~ ^08 ~ 500 550 600 650 Sot - f\ - H o 06 - 05 / 1/ - 04 ./ \ - 0.3 - ^\ - 0.2 - \ \ - 0.1 - \ \CZ^:^'<^ - 1 1 1 1 400 450 500 550 600 650 WAVELENGTH {mfi) Fig. 1 . Spectrum of oxidized (broken line) and reduced (solid line) R. rubrum RHP at pH 7. For the main curves, 0-66 x 10"^ //moles/ml, and for the inset curves 2-7 X 10-2 /j) Fig. 3. Spectrum of oxidized and reduced Chiomatium RHP at pH 6-^ spectroscope, but are hardly discernible in the Gary traces, particularly those of the R. rubrum compound. The wavelength maxima shown in Fig. 3 are those determined visually, using the spectroscope. The absence of a j3-band in the usual position (approx. 520-525 m^) is another criterion employed in monitoring RHP. 6. The haemochrome spectrum of RHP found in alkaline medium (Fig. 4) exhibits absorption maxima located at the same positions as those of corresponding cytochrome c at physiological pH, but relative intensities of the absorption peaks differ markedly (Bartsch and Kamen, 1958). In Table 2 we have listed values for a number of extinction coefficients, both for absolute and difference absorption spectra. The preparations measured all had purities in excess of 95%, based on spectroscopic data. They were homogeneous by criteria based on sedimentation and diffusion studies. 400 450 500 550 600 650 WAVELENGTH (mM) Fig. 4. Spectrum of oxidized (broken line) and reduced (solid line) R. rubnim RHP at pH 11-8 (phosphate buffer, 01 m). Relative concentrations for main curves and inset curves as in Fig. 1. (Reproduced by permission of the Journal of Biological Chemistry.) OA loz A B _ z So.2 - - 2 0.1 o -0.1 -0.2 - r^^. - -1 1 - 450 500 550 600 WAVELENGTH (m;i) Fig. 5. Difference spectrum of R. rubnim RHP-CO compound minus reduced RHP, at pH 4-75. In A, Q-ll X IQ-^ //mole/ml and in B, 1-1 x IQ-^ /j) 7\{m}i) Fig. 2. Photodissociation difference spectra comparing cytochrome o with cytochromes a^ and 03 (left) and with RHP-CO (replotted from Kamen and Bartsch, this volume, p. 426). higher extinction coefficient of RHP-CO may strengthen the possibility that this haem is a polymer, perhaps a dimer. (This has recently been found true for both R. rtibrum and Chroinatium RHP (Kamen and Horio, personal communication).) A comparison of the difference spectra of the two compounds is afforded by Fig. 2. On the left, cytochrome o is compared with and clearly distingushed from cytochromes Oi and a-i. On the right, cytochrome o is compared with RHP-CO and a considerable similarity is observed. The speed with whicli cytochrome o is oxidized can at present be determined in the rapid flow apparatus by following the a band of cytochrome of type c (Fig. 3), a method that indicates only the minimum value of the reaction velocity. The reaction is slow for M. pyogenes var. alhus, but is somewhat faster for A. suhoxydans (right). A comparison of such kinetics with those of RHP in R. nibriiin is not possible Haemopiolein of Purple Photosynthetic Bacteria 435 because of the optical artifacts caused by the flow of the spirillum in the observation tube. A similar study with Chromatiuin gives satisfactory results, but the oxidation of the cytochromes is very slow. 3,2>jM Oxygen 554-540m>i log I0/I-O.OOI5 l6jjM 5 10 Time after flow stops (sec) 50 Time after flow stops (sec.) Fig. 3. Rapid flow studies of the speed of oxidation of cytochrome of type c by cytochrome o in M. pyogenes var. albus (left) and A. suboxydans (right). On the Oxidase Function of RHP Smith : In regard to the oxidase function of the RHP of Rhodospirilluin rubrum, as dis- cussed by Kamen, we have not been able to find any respiratory activity in isolated chromotophores of these organisms, although other fractions of the broken-cell suspensions can respire. Also we cannot find any evidence for the formation of the CO-compound of reduced RHP in anaerobic suspensions of R. rubrum that have been grown in the dark with aeration, although they respire very rapidly. Finally, an oxidase function for RHP is very puzzling in Chromatium, since it is an obligate anaerobe. Kamen: We have noted that RHP could hardly function as an oxidase in Chromatium. In fact we regard it as likely that the protein in R. rubrum and other facultative bacteria may function as an oxidase whereas in Cluomatium it may retain only an electron-transferring property. Another uncertainty relates to the fact that cyanide inhibits R. rubrum respiration whereas there is no spectroscopic evidence of combination between RHP and cyanide. However, this also appears to be the case with Pseudomonas oxidase (as reported by Horio et al., this volume, p. 302) where cyanide inhibits without showing visible alterations in the pigment spectrum. As for the experiments on R. rubrum grown in the dark, it would be a matter of great difficulty to interpret results of such an approach. We know there are profound changes in haematin enzyme content with such culture conditions and that not only RHP, but also catalase, peroxidase and the various cytochromes fluctuate greatly. At present, there are no compelling reasons for discarding the oxidase hypothesis. Addendum (in proof). Horio, working in Chance's laboratory with Dr. C. P. S. Taylor, has now found (September, 1960) that the CO-action spectrum for oxidase activity in dark-grown aerobic R. rubrum is identical with the CO-reduced RHP spectrum. SPECTROPHOTOMETRIC STUDIES OF CYTOCHROMES COOLED IN LIQUID NITROGEN By R. W. ESTABROOK Johnson Foundation for Medical Physics, University of Pennsylvania, Philadelphia 4 INTRODUCTION One of the primary problems associated with the studies of biological oxidations is the recognition and identification of those components partici- pating in the electron transport chain during the oxidation of substrates by oxygen. The classical spectroscopic studies of Keilin (1925) showed the presence of absorption bands characteristic for specific intracellular haemo- proteins, which he called cytochromes. These studies, coupled with the considerable numbers of enzymic tests carried out by Keilin and his collabora- tors, have served as the foundation for our present understanding of cellular respiration. The fact that the various haemoproteins have, in their reduced form, specific absorption bands in the visible spectra has served as the means of identification, thus permitting an investigation into the mode of interaction of cytochromes. An excellent summary of the spectral properties of cyto- chromes has recently been published by Morton (1958). One of the more recent developments in the spectroscopic identification of the cytochromes are the observations, first pubhshed by Keilin and Hartree in 1949, describing the remarkable effect of lowering the temperature on the sharpening of the associated absorption bands of reduced cytochromes — thus permitting the resolution of overlapping absorption bands. This effect coupled with an intensification resulting from the devitrification of the glycerol media employed by Keilin and Hartree (1949) permitted them to recognize and characterize cytochrome c^ (Keilin and Hartree, 1955) having in vitro properties which suggest its function as a member of the respiratory chain, thus confirming the earlier contention of Yakushiji and Okunuki (1940) concerning the presence of such a pigment as cytochrome c^. In addition, Keilin and Hartree (1949) recognized the splitting of some of the absorption bands of reduced cytochrome c giving rise to fine structure or what they termed satellite bands. The use of low temperature spectroscopy is not new. Many varied types of pigments such as porphyrins, nucleotides, chlorophyll and carotenoids have been investigated at low temperature and the appearance of additional 436 Cytochromes Cooled in Liquid Nitrogen 437 band structure has frequently been observed. The organic solvent systems (such as the ether-pentane-alcohol (EPA) mixture) generally employed by chemists in the spectroscopic studies of these pigments are inadequate, however, for work with proteins. Thus the introduction by Keilin and Hartree (1949) of a glycerol-water-mixture as a medium and the resultant five to tenfold intensification of absorption bands obtained with such a mixture, has served as the basis for the identification of pigments in such low amounts that they would not otherwise be discernible. We at the Johnson Foundation, recognizing the importance of Keilin's and Hartree's (1949) work on the low temperature effect but cognizant of the restriction imposed by simply describing spectra observed in a low dispersion spectroscope, have adapted a wavelength scanning recording spectrophotometer in order to plot automatically the spectra of haemo- proteins at low temperature. These studies have confirmed in every way the excellent reports of Keilin and Hartree. Indeed, we can honestly say that we have not been able to add much to the basic knowledge brought forth by them other than to extend the range of material studied. The present paper will deal with only two facets of the studies carried out at the Johnson Foundation; the first is principally concerned with the satelhte band structure of soluble haemoproteins, and the second with the complexities observed in the spectra obtained with particulate materials. METHODS Figure 1 presents a schematic description of the instrument employed. This is a more modern version of the basic instrument which has been designed and developed by Chance and his colleagues (Chance, 1951; Yang and Legallis, 1954). A Bausch and Lomb diffraction grating monochromator serves as the source of monochromatic light; the monochromatic light from the diffraction grating is divided into two equal parts by an oscillating mirror and then passes through the cuvettes in the sample holder to an end-on photomultiplier. As with all spectrophotometric studies of turbid material the geometry of the sample to the light detector is most important, i.e. the cuvettes must be placed as close to the face of the photomultiplier as possible so that the greatest solid angle of transmitted and scattered light will be included. For use at low temperatures (Estabrook, 1956), an unsilvered Dewar flask is placed in the position normally localized for the sample. By filling this Dewar with liquid nitrogen and using an appropriate sample holder, one is able to cool samples and thus record the spectra of haemo- proteins at the temperature of liquid nitrogen. Figure 2 shows in greater detail the arrangement of the sample holder and Dewar flask. Two cuvettes are employed, one being essentially a reference, the otlier the sample to be investigated (Note 1). The two light beams pass alternately through the two cuvettes and the difference in light absorption, 438 R. W. ESTABROOK ©Tungsten or high pressure Hg arc lamp Fig. 1. Schematic description of the wavelength scanning recording spectro- photometer developed by Chance (1951). split light beam from 30 c p s oscillating mirror adjustable (sliding) cuvette holder liquid air Fig. 2. Detail of the cuvettes and holder used for low temperature spectrophotometry. Cytochromes Cooled in Liquid Nitrogen 439 after appropriate amplification, is then recorded automatically as a function of wavelength. The sample holder is arranged so that the samples may be raised out of the Dewar, put into liquid nitrogen, or aligned in the light path from the monochromator. A classic demonstration of the performance of a spectrophotometer is the spectrum which one can obtain with a didymium nitrate solution. In Fig. 3 350 400 450 500 550 600 Wavelength Cm/j) Fig. 3. Spectra of didymium nitrate. A 10% solution of didymium nitrate was mixed with an equal volume of glycerol and the spectrum was recorded at room temperature (A-A) using cuvettes with a 10 mm light path. A sample of the didymium nitrate-glycerol mixture was placed in a cuvette of 1 mm light path and cooled in liquid nitrogen and the spectrum recorded (B-B). are plotted two spectra as obtained using the instrument developed by Chance. The spectrum marked AA is that of didymium nitrate at room temperature — a rather complex spectrum. When the sample is cooled in liquid nitrogen and then warmed to devitrify the glycerol mixture and then recooled in liquid nitrogen, one obtains the spectrum BB (Note 2). There is a tenfold difference in the optical depth of the cuvettes employed in the two experiments: that is, the upper curve, BB, should be ten times larger than the curve A A. Of interest here, other than seeing the appearance of additional absorption bands at liquid nitrogen temperature, is the resolution of absorp- tion bands which one can obtain with the instrument we are using. For 440 R. W. ESTABROOK example, at 522 niju we are able to resolve absorption bands which are approximately 10 A apart. Another point of interest is the fact that one obtains as great an intensification of the associated absorption bands in the visible region around 570-580 m/u as can be obtained at 350 mju or 445 m/^. That is, there is a fifteen to twenty fold intensification of all absorption bands throughout the spectrum. CYTOCHROMES OF THE c-TYPE Cytochrome c of Heart Muscle Figure 4 shows the type of spectrum one obtains (Estabrook, 1956) with cytochrome c purified from horse heart (Note 2). The sample is reduced. 400 430 460 490 520 550 580 Fig. 4. A comparison of the spectral properties of heart muscle cytochrome f at room temperature and at — 190"C. Purified cytochrome c was diluted in 01 m phosphate buffer, pH 7-4, a few crystals of Na^S204 were added, and then an equal volume of glycerol. The reference cuvette contained a mixture of an equal volume of glycerol and 0- 1 m phosphate buffer. The sample and reference cells were cooled in liquid nitrogen, warmed to induce devitrification of the glycerol mixtures, and recooled in liquid nitrogen "(Condition II). Optical path of the cuvettes was 1 mm. in this case, by sodium dithionite although identical results are obtained if the cytochrome c is reduced enzymically. The spectrum of the same sample of reduced cytochrome c, as obtained at liquid nitrogen temperature, has been included. One of the first differences observed is that the a-band of reduced cytochrome c, which at room temperature looks relatively sym- metrical, splits into three absorption bands (Note 3) which we have termed <^ai' <^"a2' ^aa- The jS-band shows an even more complex structure and by precise spectral analysis one sees nine absorption bands. In addition one sees an intensification of the light absorption by these a- and /5-bands of five to six fold. In contrast, the Soret band does not appear as greatly intensified. In addition one sees no clear splitting of the Soret band, indeed all that is Cvtochroines Cooled in Lufuul Nitrogen 441 observed is the appearance of a shoulder at about 430 m/^ indicative of another possible absorption band. A point of interest is the appearance of what have now been termed (Estabrook and Sacktor, 1958) beta prime (/5') bands, a triple band structure at about 470 mfi. Although of low extinction ABSORBANCY 490 520 550 580 Fig. 5. The effect of glycerol on the appearance of the c^2 absorption band of reduced cytochrome c. Horse heart muscle cytochrome c was diluted in 0-2 m phosphate buffer, pH 7-4 and reduced by the addition of Na2So04. Various volumes of a mixture of 75 % glycerol, 25 % 0-2 m phosphate buffer, pH 7-4 (v/v) were added to portions of the reduced cytochrome c solution to give varying concentrations of glycerol. Curve A represents no glycerol, curve B was with 14% glycerol, and curve C contained 71 % glycerol. Optical depth of cuvettes, 3 mm. these three bands appear to be present in all the purified cytochrome haemo- proteins that have been studied so far. Of interest is the nature of the splitting of the a- and /5-bands observed in samples cooled in liquid nitrogen. One immediately asks does this imply heterogeneity of the sample, thus indicating the presence of more than one haemoprotein? If the split of the a- and /3-bands as observed is not due to contamination by more than one haemoprotein, what is the significance of this split. How may the fine band structure, as observed when samples are cooled in liquid nitrogen, be modified and what factors influence these bands ? It may also be asked whether the changes seen in the spectra of modified samples of cytochrome c can be related to the biological activity of the haemdproteins, to the protein structure itself, or to the spatial arrangement of the constituent parts of the molecule ? G 442 R. W. ESTABROOK Effect of Glycerol When samples of reduced cytochrome c are diluted in various concentra- tions of glycerol and then cooled in liquid nitrogen, it is observed that glycerol is required in order to resolve clearly the c^^ and c^o absorption bands. This is illustrated in Fig. 5 where three representative spectra are presented. When the concentration of the glycerol is less than 5 % (v/v) the absorption band attributed to c^2 appears merely as a shoulder on c^^. In addition, some of the /?-bands as well as the Cg^g-band are less distinct. This failure to obtain maximal resolution in the absence of glycerol may be due to the type of crystal struc- ture formed when the sample is cooled to — 190''C, or it may represent a hitherto unrecognized reaction of reduced cytochrome c with glycerol forming a derivative with distinctive absorption bands discernible only at hquid nitrogen temperatures. In the subsequent discussion most of the measure- ments have been carried out in the presence of about 50% (v/v) of glycerol in order to measure the maximal effect of low temperature on the absorption bands of reduced haemoproteins. Fig. 6. A comparison of the spectral properties of reduced cytochrome c of heart muscle, the alkaline haemochrome of cytochrome c, and the peptide core of cytochrome c. Curve A — Samples of peptide core of cytochrome c were dissolved in 01 M phosphate buffer of pH 7-4 and a few crystals of Na^S.^Oj added. This was then mixed with an equal volume of glycerol. Curve B — Samples of horse heart cytochrome c were diluted with 1 n NaOH, permitted to incubate lOmin at room temperature, reduced with Na2S204, and then mixed with an equal volume of glycerol. Curve C — Samples of heart muscle cytochrome c were treated as described in Fig. 4. Optical depth of cuvette was 1 mm. Cytochromes Cooled in Liquid Nitrogen 443 Modified Cytochrome c of Heart Muscle The extreme in modification of a- and /5-band splitting is observed when one determines the spectrum of the peptide core of cytochrome c (Ehrenberg and Theorell, 1955). This is shown in Fig. 6 (Note 4). Included with the spectrum of digested and reduced cytochrome c is a spectrum of alkali- modified cytochrome c as well as that of biologically-active heart muscle TCA Treuied Cytochrome C 1 I I I I I I I I ' I I 490 520 550 Wavelenglh (mp) Fig. 7. Spectra of trichloroacetic acid-modified cytochrome c. Samples of horse heart cytochrome c, obtained from Dr. Margoliash, were treated as described in Fig. 4. Curve A represents fraction la while Curve B represents fraction He as obtained from the ion exchange resin IRC 50 during the purification of cytochrome c. cytochrome c. One sees with the peptide core a single a-band, broad and symmetrical, with indications of a Ca^g-band at 536 mfx. The presence of a complex /9-band is apparent but the many small bands generally observed with the undigested cytochrome c are not detectable. Alkali-treated cyto- chrome c has also lost much of the fine band structure. Of interest is the 1-5 m/^ difference in absorption band maxima between the two modified samples. The spectrum of the peptide core of cytochrome c resembles most closely the alkaline haemochrome of reduced cytochrome q (see below). Figure 7 shows the extremes of a series of samples which we were fortunate to obtain from Dr. Margoliash. These represent the first (A) and the last (B) of a series of six fractions of cytochrome c obtained from IRC 50 resin (Margoliash, 1954a) during the purification of horse heart cytochrome c prepared by the conventional Keilin and Hartree (1945) method. Margoliash 444 R. W. ESTABROOK (1954b) has termed these type I and type II cytochrome c. One sees that the first material to be eluted (Type la) has very well defined c^^- and c^^a-bands. As subsequent samples are eluted, the degree of resolution of these two bands gradually diminishes until we see in the last sample to be eluted from the column (Type lie) that there is merely a small shoulder indicative of the presence of r„2- Margoliash, Frohwirt and Wiener (1959) have characterized this last fraction as a trichloroacetic acid (TCA)-denatured cytochrome c, and have investigated in detail the chemical and enzymic activities of these fractions, showing that the TCA-modified cytochrome c has lost a large part of its enzymic activity. Thus another means of modifying the split of the absorption bands observed at low temperature is by acid denaturation of the cytochrome c. Cytochrome cfrom Other Sources One further means of obtaining a difference in the fine band structure of cytochrome c is to prepare the pigment from other sources. We have 1 ' -190 ' 1 520 X mjj) 1 ' 550 580 Fig. 8. A comparison of the spectral properties of reduced cytochrome c prepared from heart muscle and from yeast. Samples of cytochrome c diluted in 0-1 M phosphate buffer, pH 7-4 were reduced with Na2S204 and diluted with an equal volume of glycerol. Curve A represents the spectrum obtained with yeast cytochrome c while Curve B is that obtained using heart muscle cytochrome c. Optical depth equals 1 mm. Condition II. purified the pigments and investigated the spectra of cytochrome c from wheat germ, flight muscle sarcosomes of the house fly, rat Hvers, and yeast. All of the samples assayed, except that from yeast (Note 5), siiowed essentially Cytocliroines Cooled in Liquid Nitrogen 445 the same type of fine band structure of the a- and /^-bands as heart muscle cytochrome c, with some modifications in the location of the maxima. Figure 8 presents a comparison of the spectra obtained when samples of heart muscle cytochrome c and yeast cytochrome c are analysed. The spec- trum of the yeast cytochrome c looks similar to that obtained with TCA- denatured or with alkali-denatured heart muscle cytochrome c. It should be noted, however, that the locations of the absorption bands of the denatured material and of the yeast cytochrome c are different. Other Types of Cytochrome c Some other interesting variations on pigments of the type of cytochrome c are shown by results obtained with cytochrome c^ of Azotobacter vinelandii. 43 O 460 490 52 O Wavelength , mp 550 580 Fig. 9. A comparison of the spectral properties of reduced cytochrome c prepared from heart muscle and cytochrome C4 of Azotobacter vinelandii. Curve A represents the spectrum obtained with cytochrome C4 while curve B is that obtained with heart muscle cytochrome c. Optical depth equals 1 mm. Condition II. One sees in Fig. 9 the spectrum of reduced cytochrome q (curve A) compared to that of the heart muscle cytochrome c. Of interest is the fact that even though the fine band structure of the a- and /9-bands are apparently identical in both samples, the location of the maxima of the absorption bands of reduced cytochrome q from bacteria are shifted about 1-5-2 mp, to the red. It was difficult to convince ourselves of the vahdity of this observation, but repeated experiments have proven that this is not an artifact of the machine or the method of carrying out the experiments, but truly a reproducible difference between the two pigments. The multiplicity of Soret bands ob- served with cytochrome c^ prepared from Azotobacter is in contrast to the 446 R. W. ESTABROOK results shown for cytochrome c from heart muscle where very slight changes were seen when the sample was cooled in hquid nitrogen. The bacterial cytochrome q shows much greater detail in the Soret region, at least five absorption bands being apparent. Another series of r-type pigments which have been investigated are the pigments cytochrome q of heart muscle and Cg of Azotobacter. These are 460 490 520 550 Wovelength (rryj) Fig. 10. A comparison of the spectra of cytochrome c^ and c^. Curve A represents the spectrum of cytochrome Cj obtained as a difference spectrum of cytochrome c^ reduced by ascorbic acid minus the oxidized pigments of a heart muscle DPNH-cytochrome c reductase. Curve B is the spectrum of cytochrome C5 of Azotobacter vinelandii obtained with Na2S204 as reducing agent. Condition II. shown in Fig. 10. The spectrum of heart muscle cytochrome c is added as a dotted curve to orient one in locating wavelengths. One of tiic first observa- tions is that cytochromes (v, and c\ do not show a distinct splitting of the alpha absorption bands. There is however, a pronounced splitting of the ft bands. Although these two pigments have been derived from entirely different organisms, i.e. heart muscle and Azotobacter vinelandii, they have identical spectral characteristics in every way thus far measurable. It should be mentioned, however, that there are differences in biological activity as well as chemical properties between the two pigments. Cytochrome c^ functions (Keilin and Hartree, 1955; Estabrook, 1958) between cytochrome Z? Cytochionics Cooled in Liquid Nitrogen AAl and c in mammalian tissues and has an oxidation-reduction potential of about +0-145 V, while cytochrome c^ of Azotobacter has an oxidation- reduction potential of about +0-300 V (Tissieres and Burris, 1956; Tissieres, 1956). In addition, cytochrome c^ is enzymatically active in particles of Azotobacter, whereas it is not active in the heart muscle system. CYTOCHROMES OF THE 6-TYPE Cytochrome h^ A pigment which we have been most fortunate to have available (Note 7) is cytochrome b^ from the halotolerant bacteria. As shown in Fig, 11, the A (m;i) Fig. 1 1 . Low temperature spectra of purified cytochrome A,i from halotolerant bacteria. The upper curve represents the spectrum observed when purified cytochrome b^ is diluted in 0-1 M phosphate buffer, pH 7-4 and reduced with Na.jSoOi. The sample was then mixed with an equal volume of glycerol and the spectrum recorded. The spectrum of the alkaline haemochrome of cytochrome b^ was obtained in a similar manner except that the sample was diluted in 1 n NaOH. Optical depth equals 1 mm. Condition II. a and /7 absorption bands of reduced cytochrome b^ show the most remarkable splittiiig of any pigment yet observed. The two a absorption bands which are observed at low temperature have their maxima about 6 nyx apart. Another 448 R. W. ESTABROOK anomalous property of this pigment is the very high extinction of the /i bands. In addition it must be mentioned that cytochrome Z)4 is really not a ^-type cytochrome but has proven to be a c-type cytochrome. This is evidenced by the location at 549 mw of the alpha band of its reduced alkaline haemo- chrome. The impure samples of cytochrome b^ first reported by Egami, Itahashi, Sato and Mori (1953) have since been shown to be contaminated by another haemoprotein with a typical b-typQ absorption spectrum. Cytochromes b^, bg and bg Spectral studies have been carried out on cytochrome b^ of anaerobic yeast, cytochrome b^ (yeast lactic dehydrogenase) and cytochrome b^ of Fig. WAVELENGTH ( m;j ) 12. The low temperature absorption spectra of cytochromes b^ and h.^. Solid line: cytochrome b^ in intact cells of anaerobically grown yeast, stationary phase of growth, reduced with NaaS.jOi. The dashed line represents cytochrome bo purified from baker's yeast, reduced with Na2S20i. Optical depth was 1 mm. Condition II. liver microsomes. All three pigments show essentially the same type of absorption spectrum at low temperature (Chance, Klingenberg and Boeri, 1956; Lindenmayer and Estabrook, 1958). As shown in Fig. 12 one observes a split of the a-band of reduced cytochromes /;, and /^a; it should be noted that in contradistinction to the fine band structure observed with the a-band of reduced cytochrome c of heart muscle, the band of lower extinction Cytochroines Cooled in Liquid Nitrogen 449 resides on the red side (long wavelength side) of the major band in these b- type pigments, whereas it is on the blue side of the spectrum with cytochrome c. Also of interest is the fact that the major absorption band of cytochromes /?!, b^ and b^, that is the band of highest extinction, has its absorption maxi- mum at about 553 m/*, approximately the same as that seen for cytochrome c\ (Fig. 10). The alkaline haemochromes, however, differ between cytochromes 1^ 400 500 550 Fig. 13. The pigments of rat liver microsomes. Microsomes were diluted in 0-1 M phosphate buffer, pH 7-4 and then a few crystals of Na2So04 were added. The sample was then diluted with an equal volume of glycerol. The resulting spectrum (solid line curve) shows the double a-band of cytochrome b^. A similar sample reduced by NajSoOi was treated by bubbling with CO to give the dashed line curve. Optical depth was 1 mm. Condition II. of Z)2 and b^ type and the cytochrome c\ type, the latter having the maximum of the reduced alkaline haemochrome at 549 mfx. Of principal interest is the type of spectrum one obtains with cytochrome 65 of liver microsomes. The spectra of the reduced, and CO binding pigments of the rat liver microsomes are presented in Fig. 13. In addition to the split of the a and /i bands one sees that the addition of CO causes the appearance of a large absorption band at about 448 m/^, confirming the previous observa- tion of Khngenberg (1958). The split of the a- and /5-bands does not appear to be influenced by the presence of CO. Of interest are two contradictory experiments concerning the interpre- tation of the split of the a-band of cytochrome ^5 into two distinct absorption bands. In an attempt to resolve the question of whether this represents the contribution of two haematin components (cf. Chaix, Petit, Monier and Zajdela (1958)) or one, the spectral properties of cytochrome b^ purified by 450 R. W. ESTABROOK Dr. Garfinkel (Garfinkel, 1957) from two different sources were investigated. As shown in Fig. 14 there is a difference in the magnitude of the b^^^ absorp- tion band when comparing the spectrum of the two samples. These samples of cytochrome b^ were reduced enzymatically by a specific DPNH-cytochrome Dr. Garfinkel. The variation observed would CYTOCHROME B= 490 520 550 580 Wavelength (rriAi) Fig. 14. A comparison of the spectral properties of cytochrome 6,,, purified from two different sources. Samples of purified cytochrome Zjj were diluted in 01 m phosphate buffer, pH 7-4 and a purified reduced diphosphopyridine nucleotide (DPNH)-cytochrome h-^ reductase and DPNH were added. The sample was then diluted with an equal volume of glycerol and the spectrum recorded. Curve A represents cytochrome b^ purified from rabbit liver microsomes while curve B is that obtained using pig liver as the source of microsomes. The dotted curve represents the spectrum of reduced heart muscle cytochrome c. Optical depth was 1 mm. Condition II. indicate that the splitting of the a-band seen with cytochrome b^ may be due to the presence of more than one pigment, the haemoprotein contributing to the b-^i-hand having been partially removed during the course of purification. If this hypothesis is correct, i.e. that more than one haemoprotein con- tributes to the spectrum of cytochrome b^, then a preferential reduction of one or the other pigment might be expected when the system is titrated with DPNH. As shown in Fig. 15 this is not the case. Addition of small amounts of DPNH and then rapid cooling of the sample (see below) shows that both b^^i and br^^.y appear together. Tliis indicates cither that the two bands observed are associated with a single pigment or that two pigments are being reduced simultaneously to the same extent when a substrate such as DPNH is added. Cytochromes Cooled in Liquid Nitrogen 451 A point worthy of mention is the possible confusion which may arise because of the similarity of absorption band maxima of reduced cytochrome q and b^^2- This situation reaffirms the contention that spectral studies alone may be misleading and must be coupled with enzymic studies employing ABSORBANCY INCREMENT = 0.10 I I I I I I I I ' I I ' I 490 520 550 580 610 WAVELENGTH (m>j) Fig. 15. Spectra showing the titration of cytochrome b^ of rat liver microsomes by DPNH. Samples of rat liver microsomes were suspended in 0- 1 m phosphate buffer, pH 7-4. One portion was removed and placed in the reference cuvette. To the remainder were added small amounts of DPNH. The sample was then cooled by plunging the cuvettes into liquid nitrogen and the difference spectrum recorded. Curve A represents the spectrum obtained when DPNH sufficient to reduce 27% of the cytochrome be, was added, curve B was the spectrum for 54% reduction, curve C for 81 % reduction while curve D was the spectrum obtained with a large excess of DPNH. Optical depth of the cuvette was 3 mm. various specific inhibitors or substrates in order to distinguish clearly one pigment from another. PARTICLE-BOUND PIGMENTS As is apparent from the above, low temperature studies are at present purely descriptive and of principal importance for identification of pigments. The technique has therefore been extended to a study of the respiratory pigment complement of particle-bound haemoproteins from a variety of tissues. Among the first samples to be investigated were the particulate heart muscle preparations; these studies v/ere carried out in collaboration with Dr. Mackler (Estabrook and Mackler, 1957). As shown in Fig. 16 one 452 R. W. ESTABROOK obtains a resolution of the a absorption bands associated with the cyto- chromes h, c and c\ of heart muscle. Indeed, this method is the most sensitive for determining the presence of cytochrome c^ and distinguishing cytochrome c from cytochrome c^. An extension of this type of study is the spectrum shown in Fig. 17 of the respiratory pigments of sarcosomes isolated from the flight muscles of the Fig. 16. The low temperature spectra of the reduced haemoproteins of a Keilin and Hartree heart muscle preparation. Samples of heart muscle preparation were diluted with 01 m phosphate buffer, pH 7-4, appropriate reagents added, and then mixed with an equal volume of glycerol. Curve A represents the spectrum of the oxidized pigments while curve B shows those pigments reduced by succinate in the presence of cyanide and curve C represents those reduced by NaoSjOi. Optical depth was 1 mm. Condition II. fly (Estabrook and Sacktor, 1958). These studies, carried out with Dr. Sacktor, show the apparent absence of cytochrome q in this type of material. It is known from other studies, however, that although cytochrome q appears to be absent, a pigment functioning similarly to cytochrome q but of entirely different absorption characteristics is present. Another type of material studied was the cytochrome complement of wheat germ mitochondria as presented in Fig. 18. This work was carried out with Dr. Stern of the University of Toronto. Again one sees variation in the type of cytochrome spectra obtained. The spectral properties of the cyto- chromes of plants will be described in greater detail by Dr. Bonner at this symposium. Thus it appears that every new source that we have tried shows a variability in cytochrome pattern. One must admit that it becomes most Cytochromes Cooled in Liquid Nitrogen 453 Fig. 17. The low temperature spectra of the reduced pigments of flight muscle sarcosomes. Curve A represents the spectrum obtained when the pigments are reduced enzymically. The sample cuvette contained a mixture of 0-4 ml of sarco- some preparation (6-8 mg protein), 002 ml of 0-5 M a-glycerol phosphate, and 0-4 ml glycerol. Curve B represents the spectrum of those pigments reduced by sodium dithionite. The sample cuvette contained a mixture of 0-4 ml of sarcosome pre- paration (4-5 mg protein), a few crystals of sodium dithionite, and 0-4 ml glycerol. The reference cuvette contained a mixture of equal volumes of glycerol and buffer. Optical depth of cuvettes was 1 mm ; half effective band width, 0-7 m^a; Condition II. Temperature, -190°. WAVELENGTH (m^) Fig. 18. The low tem- perature spectra of re- duced haemoproteins of wheat germ mitochon- dria. Samples of mito- chondria were treated with various reagents, mixed with an equal amount of glycerol, and cooled in liquid nitrogen. Curve A represents those pigments appearing on the addition of Na2S204 ; curve B is the pigments reduced by succinate in the presence of cyanide; curve C represents the absorption spectrum ob- tained when no reducing agent was added. Optical depth equals 1 mm. Con- dition II. 430 460 490 Wavelength ( ' 1 ' 520 ' 1 ' 550 ' i ' 580 ' 1 610 454 R. W. ESTABROOK intriguing to unravel the nature of the respiratory pigments present in various sources, and indeed certainly makes the hypothesis of a unified cytochrome complement untenable. At the present time the use of low temperature spectroscopy appears to be the most promising approach to unravel the complexity of variable pigment content in different types of materials. Trapped Steady States Recently Chance and Spencer (1959), and more recently Chance and Bonner (unpublished), have succeeded in recording the difference spectrum T . AO.D.= 0.005 I — I — \ — \ \ — I — I — \ — I 500 550 Fig. 19. The low temperature spectrum of the trapped steady states of rat liver mitochondria. Samples of mitochondria, diluted in isotonic buffer, were treated with substrate (state 4). One portion was treated with ADP to oxidize the pigments to state 3. The samples were then rapidly cooled in liquid nitrogen and the resulting difference in light absorption plotted as a function of wavelength. of pigments in various steady states of oxidation. This is shown in Fig. 19, where the difference between the extent of cytochrome reduction in states 4 and 3 of phosphorylating rat liver mitochondria is presented. A clear resolution of the crossover point between cytochromes b and c^ is demon- strated. The combined increase in resolution of absorption bands coupled with a means of rapidly trapping a steady state holds great promise in solving in part the action of the respiratory pigments during oxidative phosphory- lation. DISCUSSION The cooling of samples of reduced haemoproteins, in particular the cyto- chromes, causes a sharpening of the associated absorption bands and a splitting of some of these bands. The observed narrowing of the bands is caused by restricting the energy loss from the excited vibrational states of the absorbing molecule to the solvent molecules. The lowering of tempera- ture therefore accomplishes in part a change comparable to the sharpening of absorption bands seen when going from the liquid to the vapour state with organic molecules such as benzene derivatives. The origin of the splitting of the absorption bands is the question of the Cytochromes Cooled in Luinid Nitrogen 455 greatest interest. Since at present interpretation of the absorption bands of haemoproteins is sketchy, any discussion of factors influencing these bands must be largely general, by analogy with simpler systems. The appearance of an absorption band such as the c\^-b?ind of heart muscle cytochrome c probably represents the resolution of a vibrational component of the c^^- band which may be supposed to be a single electronic transition. This and other vibrational bands are not apparent at higher temperatures because, as the bands broaden with the increased temperature, they overlap and obscure all fine structure. The question of how various factors influence the resolution, i.e. the narrowing, of an absorption band such as c^^ can only be surmised. Different solvents or solutes that can interact slightly with the haemoprotein molecule may affect the rate of energy loss from the vibrational levels, and hence narrow or broaden the absorption lines corresponding to excitation of the system to these levels. The viscosity of the medium is an important factor. Changes in solvent-bonding may affect the lifetime of the vibrational states. Adsorption of the molecule on crystals of the solvent or ice surfaces (cracks) may, by distorting the molecule, change the degree to which both weak electronic transitions and vibrational transitions are forbidden. The observation that a loss of c\<^ accompanies the loss in enzymic activity resulting from a modification of the protein, implies that these bands are profoundly influenced by the associated protein structure surrounding the haemin molecule. The modification of the protein introduces a perturbation resulting in the change of planarity or symmetry of the ring. These changes may be so small, however, that they could not be detected by chemical analysis. Of interest is the shift in absorption maxima of the a- and /)-bands observed with cytochrome c,^ with the retention of an associated fine band structure closely resembling that seen with heart muscle cytochrome c. The somewhat lower electronic energy levels observed in this instance, yet the appearance of vibrational transitions the same distance apart, may indicate a small change in a substitutent location resulting in the change of an electronic energy level. This might be accomplished by a change in the arrangement of the associated side chains on the porphyrin ring. As our understanding of the factors influencing the spectral characteristics of less complicated molecules such as porphyrins and haemins increases, and our knowledge of the chemistry of haemoproteins such as cytochrome c increases, we should be able to explain better the changes observed in spectral properties under a variety of conditions. In addition studies which are planned at liquid helium temperatures, giving better resolution of the associ- ated absorption bands, may establish more precise data for accurate spectral analysis. The technique as described, therefore, is best applied to resolve absorption bands of a mixture of pigments in studies of cytochrome inter- action in various particulate materials. 456 R, W, ESTABROOK SUMMARY 1. The method of recording spectra of samples of haemoproteins cooled to the temperature of liquid nitrogen has been described. 2. The application of this method of study of the fine band structure of heart muscle cytochrome c, and the modification of the fine band structure by alteration of the protein, has been described. 3. The spectral properties of reduced cytochromes c, q, c^, c^ as well as b^, h^, b^ and b^ are presented. 4. The usefulness of the technique for resolving the complexity of the cytochrome complement is illustrated by studies on heart muscle particles, insect flight muscle sarcosomes, and wheat germ mitochondria. 5. Speculations on the interpretation of modifications of the fine band structure of cytochrome c are discussed. A ckno wledgemen ts The author is indebted to Dr. Britton Chance for his advice and guidance during the course of these experimental studies. The generosity of many other investigators in supplying samples and materials for study has made many of the experiments reported possible. The constructive criticism of the manuscript by Dr. Walter Bonner is gratefully acknowledged and the assistance of Mr. Patrick Taylor in the spectral analysis is greatly appreciated. This research was supported by a Senior Research Fellowship, SF-206-C, of the U.S. Public Health Service. 1. Recently Dr. Walter Bonner had modified the cuvette holder permitting a more precise and convenient method for recording difference spectra and for assuring the maintenance of the temperature of the sample at — 190°C. 2. These spectra have been termed 'apparent absolute absorption spectra' to distinguish them from 'difference spectra'. The former describes spectra obtained with turbid samples when an induced turbidity, such as obtained with glycerol water mixtures, is used as a reference. 3. The nomenclature introduced to describe these spectra identifies by subscripts the various absorption bands discernible at low temperature — numbering from longer to shorter wavelengths and retaining the conventional distinction of a, /3 and y bands and the general classification of a-, b- or c-type of cytochrome. 4. The sample of the peptide core of cytochrome c was obtained from Dr. Anders Ehrenberg of the Nobel Institute, Stockholm, Sweden. 5. It is reassuring to report that Dr. Hagihara has examined his samples of crystaUine cytochromes c from a variety of sources and confirms the observation that the yeast cytochrome c differs from cytochrome c obtained from mammalian sources in the manner described here. Cytocliromes Cooled in Liquid Nitrogen 457 6. The samples of cytochrome c^ and c^ were prepared by Dr. R. Burris and N. Neumann of the University of Wisconsin. 7. Cytochrome Z?4 was prepared by Dr. Ryo Sato of Osaka University, Osaka, Japan. REFERENCES Chaix, p.. Petit, J., Monier, R. & Zajdela, F. (1958). Bull. Soc. Cliim. biol. 40, 215. Chance, B. (1951). Rev. Sci. lustrum. 22, 634. Chance, B., Klingenberg, M. & Boeri, E. (1956). Fed. Proc. 15, 231. Chance, B. & Spencer, E. L. (1959). Disc. Faraday Soc. 27, 200. Egami, p., Itahashi, M., Sato, R. & Mori, T. (1953). /. Biochem. Tokyo 40, 527. Ehrenberg, a. & Theorell, H. (1955). Nature, Lond. 176, 158. ESTABROOK, R. W. (1956). /. biol. Chem. 223, 781. EsTABROOK, R. W. (1958). J. biol. Chem. 230, 735. EsTABROOK, R. W. & Mackler, B. (1957). J. biol. Chem. IIA, 637. EsTABROOK, R. W. & Sacktor, B. (1958). Arch. Biochem. Biophys. 76, 532. Garfinkel, D. (1957). Arch. Biochem. Biophys. 71, 100. Keilin, D. (1925). Proc. roy. Soc. B. 98, 312. Keilin, D. & Hartree, E. F. (1945). Biochem. J. 39, 289. Keilin, D. & Hartree, E. F. (1949). Nature, Loud. 164, 254. Keilin, D. & Hartree, E. F. (1955). Nature, Lond. 176, 200. Klingenberg, M. (1958). Arch. Biochem. Biophys. 75, 376. Lindenmayer, a. & ESTABROOK, R. W. (1958). Arch. Biochem. Biophys. 78, 66. Margoliash, E. (1954a). Biochem. J. 56, 529. Margoliash, E. (1954b). Biochem. J. 56, 535. Margoliash, E., Frohwirt, N. & Wiener, E. (1959). Biochem. J. 71, 559. Morton, R. K. (1958). Revs, pure appl. Chem. 8, 161. TissiERES, A. (1956). Biochem. J. 64, 582. TissiERES, A. & BuRRis, R. H. (1956). Biochim. biophys. Acta 20, 436. Yakushiji, E. & Okunuki, K. (1940). Proc. Imp. Acad. (Tokyo) 16, 299. Yang, C. C. & Legallais, V. (1954). Rev. Sci. lustrum. 25, 801. DISCUSSION 'Trapped' Steady States By B. Chance (Philadelphia) Chance: Various experiments indicate that there is only a small change in the steady state of cytochromes c, b and a in the transition from room temperature to liquid nitrogen temperatures (77°K). Explanations for this are described elsewhere (Chance and Spencer, Disc. Faraday Soc. 27, 200, 1959), and centre about the very similar temperature coefficient of the inter-cytochrome reaction velocities. This method has important applications in the identification of functional cytochromes and in locali- zation of inhibitory interaction sites in the respiratory chain. Methods The method is described in Chance and Spencer (Disc. Faraday Soc. 27, 200, 1959), and very briefly consists of modifications of Keilin's and Estabrook's procedures suggested by Bonner and Yocum, namely the omission of glycerol. Modified cuvettes for 1 and 3 mm paths are used. Instead of absolute spectra, difference spectra are used. The steady-state oxidized reference material is washed and starved where possible. It is strongly oxygenated to ensure an adequate supply of oxygen in the H 458 Discussion filling and freezing interval. The steady-state reduced material is similarly oxygenated and is treated with substrate just prior to pipetting into the cuvette. The cuvette may be pre-chilled so that freezing occurs very nearly on contact with the cooled surfaces. The difference spectrum is then plotted in the split-beam recorder (see Chance, Methods in Eiizymol. 4, 273, 1957). Identification of Respiratory Enzymes It was pointed out in the discussion of Chaix's paper that the distinction between respiratory and accessory pigments (for definition see Chance, this volume, p. 476), could be simply made by observation of their steady-state reduction. Trapping of this steady state at liquid nitrogen temperatures has the great advantage of com- bining the precise identification of cytochrome type afforded by the sharpening and splitting of the bands at low temperatures with the test for respiratory function. An example of this technique is aff"orded by Fig. 3 of Chance (this volume, p. 607) which illustrates a frozen steady state for succinate-treated ox heart mitochondria. In addition to the reduction of cytochromes a and c to values characteristic of active electron transfer, we find the steady-state reduction of cytochrome c^ to be readily observed for the first time in this material and, in addition, can calculate its extent of reduction as a percentage of the fully reduced material. This value is very close to that of cytochrome c. Inhibitory Interaction Sites The cross-over theorem (see Chance, this volume, p. 607) affords a unique method for the localization of inhibitory interaction sites in the respiratory chain. The changes in steady state of cytochromes b and c may be small and difficult to m.casure, and in any case, cytochrome q cannot be discerned at room temperatures. Tiie application of trapped steady states to the study of cross-over behaviour is illustrated by several figures given in the oral presentation, one of which is included in the text of this volume (Estabrook (Fig. 19), this volume, p. 454). We can demon- strate the state 2-A and 3^ transitions in the presence and absence of azide. The participation of cytochrome c^ in these transitions is clearly shown and a cytochrome b-Ci crossover is demonstrated (see Chance, this volume, p. 607). The technique has also been applied to cross-over phenomena in intact cells and the increased reduction of cytochrome b and c characteristic of the inhibited phase of glucose metabolism can readily be observed, although these changes are too small to allow a resolution of cytochrome q. One of the greatest limitations of the low-temperature technique, namely that it affords only identification and gives no indication of function, appears to be overcome with the trapped steady-state method. It is probable that most of the functional and identification test for cytochromes will be done at low temperatures in the future. Low Temperature Absorption Spectra of Cytochromes in Relation to Structure Possible Differences in the Prosthetic Groups ofc-type Cytochromes Morton: With reference to the differences in the low temperature absorption spectra noted by Estabrook, it seems possible that there may be small differences in the prosthetic groups of cytochrome c from various sources. Not only does yeast cyto- chrome c differ from heart-muscle cytochrome c in absorption spectra both at room temperature and at — 190°C, but the prosthetic group of native yeast cytochrome c is more readily split by treatment with silver sulphate in acetic acid, as previously noted (Armstrong, Coates and Morton, this publication, p. 386). In one sample of aggregated yeast cytochrome c, Armstrong in my laboratory observed that the prosthetic group was split by treatment of the cytochrome with 2 n acetic acid only. Unfortunately, treatment with acetone-HCl was not investigated. Falk, Appleby and Porra {Soc. exp. Biol. Sympos. 13, 73, 1959) have reported the isolation from effective Cytochromes Cooled in Liquid Nitrogen 459 soy bean root nodules, by simple extraction with ethyl acetate-acetic acid, of a haem similar to haem c. It appears unlikely that the prosthetic group of yeast cytochrome c differs from that of heart-muscle cytochrome c, yet these observations suggest that there may be only one thio-ether link to the polypeptide chain in yeast cytochrome c, or possibly true ether bonds instead of thio-ether bonds. J. Keilin {Biochem. J. 64, 663, 1956) has pointed out a somewhat related position in respect of cytochrome It. Is there any information as to the influence of the protein structure itself on the splitting of the prosthetic group of cytochrome c by Paul's silver sulphate method? Margouash: We have repeatedly observed that Paul's silver sulphate method for the splitting off of the prosthetic group of cytochrome c works very readily with native horse-heart cytochrome c and that the haem is entirely liberated within about one hour at 60'C. However, when the peptic 'core' of cytochrome c, containing only an 1 1 -amino acid peptide is used, the reaction proceeds much more slowly and takes from 18-24 hr to go to completion. This is contrary to what may be expected, since if anything, the presence of the entire intact protein should interfere with and not facilitate an attack on a bond presumably holding the haem in a crevice. Lemberg: While it appears to me unlikely that yeast cytochrome c contains mesohaem (it could not then combine by thioether linkages with the protein) I should like to make a plea to chemists interested in cytochromes c to investigate the compound obtained by the silver sulphate method splitting in acetic acid more carefully. It should e.g. be possible to differentiate between haematoporphyrin, vinyl-a-hydroxyethyl deutero- porphyrin and ethyl-a-hydroxyethyl deuteroporphyrin. Unfortunately we have obtained some evidence that the silver sulphate method may yield artifacts and requires further study to exclude secondary alterations. Origin of tiie Fine Structure in the Absorption Spectra of Cytochromes at — 190°C Orgel: According to Piatt's {Radiation Biology 3, 101, 1956) theory of porphyrin spectra the lowest excited state of a metal porphyrin with four-fold symmetry is doubly degenerate and it is the splitting of this degeneracy which produces the more com- plicated spectra of the free porphyrins since the latter have only two-fold symmetry. I wonder if the splitting which Estabrook observes at low temperatures is a similar but smaller splitting of the degeneracy of the excited state caused : 1 . by the different side chains ; 2. by the unsymmetrical environment of the protein. One would anticipate that such splittings would be very small and might only be revealed when the absorption bands become very narrow as they do at sufficiently low temperatures. It should be noted that splittings due to different types of asymmetry might either add up or cancel out depending on the exact geometry of the situation. As suggested by Margoliash, the asymmetry might be envisaged as due to different groups bound to co-ordination positions 5 and 6 of the iron, above and below the plane of the haem. Estabrook: These possibilities certainly exist. One difficulty in interpretation which must be considered is that one does not see a split of the absorption bands of haemo- globin, myoglobin, catalase, or cytochrome a at the temperature of liquid nitrogen. The inability to see a split with these compounds may be resolved when we complete adapting the instrument for spectra at the temperature of liquid helium. Williams: The resolution of the fine structure of the a- and /?-bands of cytochromes on cooling to low temperature is to be contrasted with the absence of resolution in the Soret band region. If we accept that the resolution is due to coupling of vibrational with electronic transitions then there are at least two explanations. The first is that given by Orgel in which the vibrational structure is due to a breakdown of the four-fold symmetry by substituents ; the other arises from the different natures of the electronic transitions involved and would occur for a totally symmetrical porphyrin. The position can be understood by reference to the spectrum of benzene which has a long wave-length band with vibrational structure and a shorter wave-length band without 460 Discussion vibrational structure. Moffitt (/. chem. Phys. 22, 320, 1954) has shown how the differ- ence in coupling between two electronic states and the vibrational states can arise through the change in symmetry property of the ground as compared with the two excited states. Until it is known that symmetrical porphyrins do not show vibrational structure of the a- and /?-bands it will not be safe to look for explanations in terms of the ground state symmetry of the ring. STUDIES ON MICROSOMAL CYTOCHROMES AND RELATED SUBSTANCES By C. F. Strittmatter Department of Biologieal Chemistry, Harvard Medical School, Boston, Massachusetts In this paper I shall attempt to review the development of our knowledge concerning the microsomal cytochrome of mammahan liver and related compounds, as this story provides a case history of the problems in identi- fication, characterization, nomenclature and elucidation of biological significance of cytochromes which is one of the major concerns of this Symposium. IDENTIFICATION, LOCALIZATION AND CHARACTERIZATION The ''Liver-type^ Cytochrome Spectrum The pioneer spectroscopic studies of MacMunn (1884, 1886) established that haemochrome-like compounds are widely distributed in tissues of vertebrate and invertebrate animals, and that the haemochrome complement of different animal cells differs both with type of tissue and with the species. In examinations of what he termed the 'histohaematin' spectra of tissues, MacMunn noted that heart and skeletal muscle from a variety of species showed, particularly in the presence of a reducing agent, a characteristic absorption spectrum which he labelled 'myohaematin'. This myohaematin spectrum was the four-banded spectrum of what is now commonly regarded as the typical 'muscle-type' cytochrome system, first intensively studied by Keilin (1925) with the strong a-bands of cytochromes a (and a-^ at about 605 m//, of cytochrome b at about 562-566 m/i and of cytochrome c (and c^) at about 550-553 m^ respectively, and the fused weaker /5-bands of cyto- chromes at 520-530 vl\[x. MacMunn noted that the 'histohaematin' spectra of many other tissues were similar to that of muscle, but that spectra of some tissues, most notably liver and the analogous green gland of certain in- vertebrate groups, and the adrenal medulla, were distinctly different. Char- acteristically, these 'liver-type' spectra showed a feeble band in the 605 m/^ region, and the bands of cytochromes b and c appeared to be replaced by a single broad band, generally with no distinct absorption maximum, which apprbximately spanned the region normally bounded by the oc-bands of cytochromes b and c. 461 462 C. F. Strittmatter Although the 'liver-type' cytochrome spectrum was subsequently observed, with some variations, in homogenates or particulate preparations from several mammalian tissues, its significance remained uncertain. The centre of the broad band, whose full development required the presence of a fairly strong reducing agent, such as dithionite, fell in the range of 555-560 m/f, thus coinciding approximately with the absorption maximum of the microbial component known as cytochrome b^ (Keilin, 1934) or b' (Fujita and Kodama, 1934). Keilin and Hartree (1940) consequently referred to the broad band of the 'liver-type' spectrum as the bi band, and Yoshikawa (1951) ascribed the band to a haemochrome-like cytochrome b', but the component or components responsible were not more definitely established. The early microbial studies (Fujita and Kodama, 1934) had indicated that at least some b^-\[ke bands resulted from a fusion of the a-bands of cytochromes b and c, and the studies by Keilin and Hartree (1949) of low temperature spectra suggested that b^ bands in some cases reflected presence of cytochromes b and c and a third haematin component absorbing between b and c, possibly cytochrome e, now known as cytochrome fj. Slater (1949) also pointed out that in some instances absorption between the bands of cytochromes b and c may reflect presence of contaminants such as denatured protein-proto- haemochrome, with an a-band at about 558 m/i. Extracts of liver prepara- tions with organic solvents also showed a haemochrome spectrum with maximum at 560 m/t which was considered to be possibly referable to cytochrome b (Kun, 1951). Localization of a Distinct Microsomal Cytochrome The nature of the intrinsic cytochrome components responsible for the Miver-type' haemochrome spectrum was established in the course of studies in our laboratory on the intracellular distribution of cytochrome components and oxidative enzymes in rat liver (Strittmatter and Ball, 1952, 1954). Spec- troscopic and spectrophotometric examination of washed cellular particles isolated from homogenates of perfused rat liver by differential centrifugation indicated that a complete 'muscle-type' cytochrome system, with absorption maxima at 605, 562, 551 and 520-530 m^i in the reduced state, was localized in the mitochondria, although the cytochrome components were present in lower concentration than in muscle. The localization of these cytochrome components in the mitochondria was correlated with the concentration in this particulate fraction of cytochrome oxidase and other oxidative enzyme activities involving terminal electron transport to oxygen. The microsome fraction, which showed only low capacities for terminal respiration, possessed no spectroscopically discernible amounts of the typical cytochrome com- ponents of the 'muscle-type' cytochrome system, but contained instead as its major intrinsic pigment a cytochrome component with a-band at 557 m// which was not observable in the mitochondria. The concentration of the Studies on Miciosonial Cylocfironies 463 microsomal cytochrome was of the order of 6 x 10^° molcs/mg protein in rat Hver microsomes, accounting for nearly 50% of the total haem found in these particles, and no other haemochrome component was discernible provided that the liver had been perfused, or the isolated microsomes adequately washed to reduce to a minimum such contaminants as adsorbed haemoglobin or denatured protein haemochromes. The microsomal cyto- chrome was a firmly bound, intrinsic component and could be removed from its particulate binding only by enzymic digestion or by treatment with lipid solvents or bile salts, whereas a single washing sufficed to remove readily adsorbed contaminants such as haemoglobin (Strittmatter and Ball, 1952; Strittmatter, 1952 and unpublished work). As it was specifically concentrated in the microsome fraction, the microsomal cytochrome was tentatively labelled 'cytochrome /;/', pending its more definitive character- ization (Strittmatter, 1952; Strittmatter and Ball, 1954). The haemochrome absorption spectrum of liver was therefore quantitatively interpretable as due to masking of the rather weak 'muscle-type' cytochrome spectrum of the mitochondria by the relatively intense absorption by a greater concentration of the microsomal cytochrome. Properties of Particulate Microsomal Cytochrome The properties and possible biological function of the microsomal cyto- chrome were first examined in isolated rat liver microsomes, with the cyto- chrome still in particulate form and presumably still in essentially physiologic state (Strittmatter and Ball, 1952, 1954; Strittmatter, 1952). In the oxidized state, the visible range of the spectrum showed only a slight broad hump with maxima at about 530 and 565 m/n and an intense Soret band at 414 m/(. When reduced by addition of dithionite to a suspension of microsomes, the typical haemochrome spectrum of the cytochrome appeared, with a sharp a-band at 557 m/n, a ^-band at 527 m^ and the Soret band shifted to 423 m/Li. Neither carbon monoxide nor cyanide (10~- m) at neutral pH altered the oxidized or reduced spectra appreciably. The prosthetic group of the microsomal cytoclii-ome was identified as iron protoporphyrin from spectro- scopic studies on the haem extracted from the microsomes and from the spectrum of the pyridine haemochrome formed with intact microsomes. The apparent standard potential of the microsomal cytochrome in its parti- culate complex was roughly estimated to be about —0-12 V at pH 7 from reduction titrations in the presence of oxidation-reduction indicators. As a microsomal component, the cytochrome was reduced rapidly and essentially completely by reduced di- and tri-phosphopyridine nucleotides (DPNH and TPNH), suggesting that it may serve as acceptor of electrons from reduced pyridine nucleotide-cytochrome reductases concentrated in the rhicrosomes. The cytochrome was also readily reduced by cysteine and partially reduced by ascorbate, but was not appreciably reduced by succinate 464 C. F. Strittmatter or /^-phenylcne diamine. Following reduction by DPNH or dithionite, the cytochrome of microsomal suspensions could be reoxidized by presence of air, or, more rapidly, by added cytochrome c or ferricyanide ; the microsomal cytochrome might therefore provide a link in a DPNH-cytochrome c reduc- tase system. As the reoxidation by air was not inhibited by cyanide (10"- m) or azide, it appeared that the microsomal cytochrome was not reoxidized via cytochrome oxidase, but might be autoxidizable. The sum of the proper- ties of the microsomal cytochrome suggested that it should be classified in the Z>-group of cytochromes. Studies with Isolated Microsomal Cytochromes The nature and relations of the microsomal cytochrome were contirmed and extended in studies with isolated microsomal cytochrome by Strittmatter and Velick (1956a, b) and by Vehck and Strittmatter (1956), who first obtained the cytochrome in highly purified state from rabbit liver microsomes, using a procedure involving separation from microsomes by treatment with pancre- atic lipase and purification by ammonium sulphate fractionation. The purified cytochrome, with a molecular weight of about 17,000, possessed as prosthetic group one iron protoporphyrin unit per molecule and contained no appreciable non-haemin iron or flavin (Strittmatter and Velick, 1956a). In its absorption spectra the isolated cytochrome agreed well with the par- ticulate form, showing the Soret band at 413 m^, the oxidized state and peaks at 423, 526 and 556 m/^ after reduction with cysteine or other reducing agents. With the purified substance it was possible also to obtain accurate values for the absorbancy indices of the absorption peaks and to observe new broad absorption bands at 355-370 m^ in the oxidized state and at 320-340 m/^ in the reduced state. Oxidation-reduction titrations and equilibrium measurements indicated that the isolated cytochrome behaved as a single, univalent electron acceptor and donor, and from the oxidation-reduction equilibria, the standard potential was calculated to be -f 0-02 V at pH 7 (Velick and Strittmatter, 1956). This potential value is more than 0-1 V higher than the values esti- mated for the cytochrome in particulate suspension from either rat or rabbit liver (Yoshikawa, 1951 ; Strittmatter and Ball, 1952), but the significance of the difference is uncertain at present. The significance of observed apparent standard potentials, particularly those for particle-bound components, are necessarily subject to uncertainty. If it is a real difference, the shift in the apparent standard potential of such a reversibly reduced substance on libera- tion from its particulate complex might reflect a differential interaction of the oxidized and reduced cytochrome with some components of the microsomal complex. This would be analogous to the sliift of apparent standard potential of diphosphopyridine nucleotide (DPN) in the alcohol dehydro- genase reaction which occurs in the presence of stoichiometric amounts of Studies on Microsomal Cytochromes 465 the enzyme, as a result of differential complex formation of the oxidized and reduced form of the nucleotide with the enzyme protein (Theorell and Bonnichsen, 1951; Hayes and Velick, 1954). Alternatively the cytochrome may have been altered during its isolation, perhaps in the incompletely defined enzymic liberation from its particulate complex, although the retention of specific enzymic activities and other properties provides no evidence of such alteration. The isolated microsomal cytochrome was readily reduced by such reducing agents as dithionite and high concentrations of cysteine. Unhke the particu- late form, the purified cytochrome was not directly reduced by DPNH or TPNH, and reduction by DPNH or TPNH required the presence of cyto- chrome reductases, which have been obtained in soluble form from liver microsomes (Strittmatter and Velick, 1956a, b). A flavin adenine dinucleotide (FAD)-linked reductase specific for both DPNH and microsomal cytochrome, but essentially inactive with TPNH as electron donor or cytochrome c as acceptor, was highly purified from calf liver microsomes (Strittmatter and Velick, 1957), and a reductase of similar specificities has been obtained from pig liver particles (Mahler, Raw, Molinari and doAmaral, 1958). On the other hand, the microsomal cytochrome was not reduced by typical DPNH- cytochrome c reductase preparations from heart muscle (Strittmatter and Vehck, 1956b). The reduced form of isolated microsomal cytochrome was very rapidly oxidized by cytochrome c, with a direct cytochrome-to-cyto- chrome electron transfer, by ferricyanide, and by various dyes of appropriate potential; it was also oxidized, less rapidly, by oxygen (Strittmatter and Vehck, 1956a). The highly active DPNH-microsomal cytochrome reductase and the microsomal cytochrome therefore form an efficient microsomal system for electron transport from DPNH to cytochrome c and this pathway, which is antimycin A-insensitive, can account for at least one-half of the total DPNH-cytochrome c reductase capacity observed with isolated micro- somes and added cytochrome c (Strittmatter and Vehck, 1956b). In addition to the isolation from rabbit liver microsomes, the microsomal cytochrome has been purified from liver microsomes of pig (Garfinkel, 1957; Krisch and Staudinger, 1958), rat (Strittmatter, unpublished work) and calf (Strittmatter and Velick, 1957), by somewhat similar procedures, and, possibly, from pig liver particles (Raw, Mohnari, doAmaral and Mahler, 1958). In so far as the preparations have been compared, the purified cyto- chromes from the various sources appear to be essentially the same protein, with properties referable to a particular type of group b cytochrome. In current usage, the term 'cytochrome b-^ is widely used in referring to the microsomal cytochrome of mammalian liver, and to a number of apparently similar pigments in other biological materials. 466 C. F. Strittmatter DISTRIBUTION OF MICROSOMAL CYTOCHROME AND SIMILAR HAEMOCHROMES There have been observations in many biological materials and derived preparations of haemochrome-like materials variously described as identical with or similar to microsomal cytochrome, cytochrome m or cytochrome b^,, but in many cases the identity of the haemochrome-like materials and their possible relationsliip to the well-characterized microsomal cytochrome are uncertain by reason of insufficiently rigorous or specific tests of identity. Many of these substances are defined solely or largely on the basis of the absorption spectra of an unpurified preparation, particularly by the appear- ance of an absorption band in the 555-560 m/t region under fairly strong reducing conditions, or by the demonstration of protoporphyrin haemochrome with such preparations. In many instances these observations appear to represent non-specific absorption, since protohaem may be combined with various non-specific or denatured proteins and with other possible nitro- genous components or contaminants of tissue preparations to yield haemo- chromes with reduced absorption spectra almost identical with that of the microsomal cytochrome, and adsorbed haemoglobin or denatured protein haemochromes are common contaminants of tissue preparations unless care is exercised to exclude or remove them (cf. Slater, 1949; Stritt- matter and Ball, 1952; Strittmatter, 1952; Klingenberg, 1958). Moreover, as the early studies of 'Z>i-like' bands already indicated, an apparently homo- geneous absorption band may result from a group of absorbing substances (not necessarily including one absorbing maximally at the observed absorp- tion maximum) and caution must be exercised even in interpretation of the absorption spectrum of an optically clear extract or 'solution', particularly one which shows uncharacteristic asymmetries (cf. Fujita and Kodama, 1934; Slater, 1949; Strittmatter, 1952; Strittmatter and Ball, 1952, 1954; Hulsman, Elliott and Rudney, 1958). On the other hand, of course, minor differences of properties between different preparations or different purified substances need not exclude the essential identity of the active substances concerned. Distribution in Mammalian Tissues Haemochrome materials with an absorption maximum in the region 555-560 m/t have now been reported in preparations from a number of mammalian tissues in addition to fiver, including adrenal medulla (MacMunn, 1886; Huszak, 1942; Ryan and Engel, 1957; Spiro and Ball, 1958), mam- mary gland (Bailie and Morton, 1955, 1958), intestinal mucosa (Bailie and Morton, 1955), ovary (MacMunn, 1886; Yoshikawa, 1951), pancreas (Palade and Siekevitz, 1956), and kidney (Yoshikawa, 1951), but the identity of the absorbing materials in many cases is not clearly estabhshed. As to intracellular distribution, studies on localization of well-defined microsomal Studies on Microsomal Cytochromes 467 cytoclirome-likc material in cell tVactions indicate that it is specifically concentrated and firmly bound in the microsome fraction, suggesting that in the intact cell the cytochrome is an intrinsic component of the endoplasmic reticulum or related membranous structures which give rise to the microsome fraction by the usual fractionation procedures. The possible presence of microsomal cytochromc-like substances in other cell compartments is not excluded, but direct and compelling evidence is lacking. There is indirect evidence (Spiro and Ball, 1958; Bailie and Morton, 1955, 1958; Spiro, 1959) that the cytochrome may appear in certain cytoplasmic 'granules', which in at least some cases appear to be derived from the endoplasmic reticulum, being formed by accumulation of newly synthesized material within the vacuoles of the endoplasmic reticulum (cf. Schneider and Hogeboom, 1956; Siekevitz and Palade, 1958). In addition, Raw and Mahler (1959) reported that pig liver mitochondrial preparations contained cytochrome b^-\ike material in appreciable concentrations, although the unambiguous designa- tion of this material as an intrinsic mitochondrial cytochrome b^ was not possible. Previous studies with rat liver had indicated that the mitochondrial fraction contains either no discernible cytochrome b^ or only a small amount attributable to microsomal contamination (Strittmatter, 1952; Strittmatter and Ball, 1952, 1954; Chance and Williams, 1955). Bailie and Morton (1955) have reported that bovine mammary gland microsomes in the presence of dithionite show a strong 557 m/* band of cytochrome b^ but no other detectable haemochrome bands, while the mitochondria of this tissue showed a 'muscle-type' cytochrome spectrum with no evidence of the microsomal cytochrome. The predominant haemochrome-like band of the adrenal medulla was variously reported in early studies to be at 559 m/n by Huszak (1942) or 561 m/ii by Tsou (1951) and was considered by them to represent cytochrome b. Spiro and Ball (1958, 1961) have recently reported that the medullary haemochrome material responsible for this absorption band is apparently concentrated in a 'microsomal fraction' and in the 'epinephrine-rich granules', which probably represent distinct entities and are partially separable from a 'mitochondrial fraction' in which cytochromes b, c, a and a-^ appear to be concentrated. The apparent localization, absorption spectrum and ready reducibility of the predominant medullary haemochrome material by DPNH, cysteine and ascorbate suggest a possible relation to the liver microsomal cytochrome, but the problem awaits a more rigorous characterization and establishment of the unitary nature of the medullary pigment. The presence of a microsomal cytochrome-like substance in the adrenal is indicated anew by the recent isolation of a 'cytochrome b-^' from a 'microsomal fraction' of whole pig adrenals by Krisch and Staudinger (1958). The properties of this preparation in general closely resemble those of the isolated liver microsomal cytochrome, although the question of possible inhomogeneity posed by 468 C. F, Strittmatter incomplete reduction of the substance with DPNH and by apparent shght spectral asymmetry is not resolved as yet. Observations in Invertebrates While there have been many observations in invertebrate tissues of haemo- chromes absorbing maximally in the range 555-560 m/^, the most exphcit demonstration of a microsomal cytochrome-hke component in invertebrates has come from studies of cytochrome components in the Cecropia silkworm. Many tissues of the larval form, most notably the midgut wall, showed on reduction a broad indivisible absorption band over the range 551-562 m//, with maximum at about the centre, which was ascribed to a cytochrome x by Sanborn and Williams (1950). In further studies with larval midgut homogenates, Pappenheimer and Williams (1954) concluded from the absorption spectra and such properties as cyanide-insensitive autox- idizability that this Cecropia pigment was a Z?-type cytochrome and called the material 'cytochrome b^. They considered it to be closely related to microsomal cytochrome, but also tentatively identified the pigment with cytochrome e (now termed q). The insect cytochrome b^ has not been purified and its precise relationship with the microsomal cytochrome is not established, but the analogy between the two pigments has recently been extended by observations on intracellular distribution: in both the larval midgut mucosa and the wing epithelium of the developing adult Cecropia, the cytochrome b^ was localized in the microsome fraction while cytochromes b, c, a and a^ were concentrated in the mitochondria (Shappirio and Williams, 1957). Plants An apparently analogous haemoprotein in certain plant tissues is cyto- chrome ^3, first described by Hill and Scarisbrick (1951). In studies of a variety of non-photosynthetic plant tissues, Martin and Morton (1955, 1957) subsequently localized cytochrome b^ in microsomal fractions. Similar to the mammalian microsomal cytochrome, cytochrome />;} shows absorption bands in the reduced state at about 558-560 m/n, 525-529 m/n and 425 m/u, is apparently autoxidizable and does not combine with carbon monoxide. Micro-organ isms Other than having the general characteristics of a Z?-type cytochrome, cytochrome b^ of various bacteria shows no particular relationship to the microsomal cytochrome, but preparations of cytochrome bj from yeast resemble microsomal cytochrome closely in certain haemoprotein properties, including the position of the haemochrome absorption bands, the reduction of the haem moiety by specific enzyme-linked flavin, and its rapid reoxidation by cytochrome c or ferricyanide. However, the nature and definition of the Studies on Microsomal Cytochromes 469 cytochrome h^ molecule is in question, since, as crystallized by Appleby and Morton (1954), the molecule contained one unit each of riboflavin phosphate and haem and possessed lactate dehydrogenase activity, while a crystalline preparation of Yamashita et al. (1957) was devoid of intrinsic flavin or lactate dehydrogenase activity. POSSIBLE BIOLOGICAL SIGNIFICANCE The firm association of the microsomal cytochrome in significant and rather constant amounts with specific metabolically-active intracellular structures in the liver and possibly in other tissues bespeaks for it specific biological roles, probably catalytic in nature, but it is not yet possible to define with assurance the particular biological significance of this cytochrome. However, it might be profitable to review again in the light of present in- formation the several possible roles which we considered in earlier studies to merit special consideration (Strittmatter, 1952; Strittmatter and Ball, 1952, 1954). Storage Form or Precursor The microsomal cytochrome might represent a storage form of haemin or a precursor in the formation of other cytochromes or haemin components, but there is no direct evidence that it serves in such relatively inert capacities. Such a role was early suggested for cytochrome b^ in certain micro-organisms (Keilin, 1933), and was apparently supported by observations that the typical 'muscle-type' cytochrome spectrum of aerobically grown resting yeast cultures is altered during anaerobic growth or in a phase of exponential aerobic growth, with replacement of the cytochrome b and c bands by a b^ band at about 555 m// (Ephrussi and Slonimski, 1950; Chaix and Heyman- Blachet, 1957). However, low temperature studies suggest that the Z>i band represents a fusion of bands (Chaix and Heyman-Blachet, 1957). It is likely that essentially all intracellular haem is maintained in bound forms, since addition of low concentrations of free haematin to cell-free preparations markedly inhibits oxidative activities (Keilin and Hartree, 1947). Protohaem might therefore be stored in bound forms with nitrogenous components, most probably proteins, to form haemochromes that possess absorption spectra very similar to that of the microsomal cytochrome, and in the absence of more specific identification such bound forms might in some preparations be mistaken for the cytochrome. Active Role in Electron Transfer The specific and highly active oxidation-reduction properties of the micro- somal cytochrome suggest for it a more active role in biological oxidation- reduction reactions. Liver microsomes possess a high capacity for oxidation of DPNH and TPNH; indeed, the DPNH-cytochrome c reductase activity 470 C. F. Strittmatter of liver is concentrated most markedly in this fraction (Hogeboom, 1949; Strittmatter and Ball, 1954). As noted earlier, the microsomal cytochrome and its reductases comprise an efficient antimycin A-insensitive system that could account for a quantitatively important flow of electrons from DPNH (or TPNH) or other donors. However, the question arises as to whether this system functions biologically, and if so, what substances serve as electron acceptors physiologically. Terminal Electron Transport The microsomal cytochrome might serve as a link in terminal electron transport to oxygen either directly or via mediators. The autoxidizability of microsomal cytochrome may suggest a role as a terminal oxidase, but the low rate of autoxidation militates against any major quantitative signihcance of such a role. The turnover number of purified rabbit liver microsomal cytochrome with oxygen as electron acceptor is of the order of 1/min, in contrast with the very rapid rate of turnover with cytochrome c as acceptor (Strittmatter, unpublished work). A hmiting value for the rate of autoxidation physiologically is suggested by the report that the turnover numbers for aerobic reoxidation of microsomal cytochrome in intact liver microsomes in the presence and absence of CN~ are 1-2 and 3-6/min respectively (Chance and Williams, 1954). Thus the DPNH oxidase activity of liver microsomes is very low compared to the DPNH cytochrome c reductase capacity of these particles or the DPNH oxidase activity of the mitochondria (Strittmatter and Ball, 1954; Hogeboom, 1949; Hogeboom, Claude and Hotchkiss, 1946). Nevertheless, a terminal oxidase role has been postulated for cytochrome b^ in systems with low rates of oxygen consump- tion, such as the cyanide-insensitive respiration of insects (Pappenheimer and Williams, 1954) and the ascorbate-dependent, cyanide-insensitive DPNH oxidase activity of a preparation from adrenal 'microsomes' (Kersten, Kersten and Staudinger, 1958). Alternatively, reduced microsomal cytochrome is known to be oxidized rapidly in vitro by a variety of reversibly reducible substances such as cyto- chrome c, ferric salts and a number of dyes, and analogous electron acceptors in vivo might serve as mediators of terminal respiration, with electrons being transported further to oxygen either non-enzymically, if the mediator is autoxidizable, or enzymically. Model oxidase systems may be constructed by addition of mediator systems to microsomal cytochrome, but there is no direct evidence for such a system in vivo. Thus, while addition of cytochrome c and isolated mitochondrial preparations to microsomal cytochrome permits rapid electron transport from the microsomal cytochrome to oxygen in vitro (Strittmatter, unpublished work), it is not known whether significant terminal electron transport from microsomal cytochrome via cytochrome c and mitochondrial cytochrome oxidase could occur in the intact cells. In Studies on Micrusoinal Cytocliroincs All particular, the availability of soluble cytochrome c and the accessibility of extramitochondrial electron donors such as soluble cytochrome c to the terminal respiratory chain of intact mitochondria in vivo are questionable (cf. Chance and Williams, 1955; Wainio and Cooperstein, 1956; Slater, 1958). If a microsomal cytochrome-like substance were localized within the mitochondrial structure, it might more readily be involved significantly in terminal respiration. While, therefore, microsomal cytochrome might play a role in terminal respiration of cells with low rates of oxygen consumption and may contribute to the usually minor cyanide-insensitive respiration, there is as yet no com- pelling evidence that this substance is responsible for a significant portion of terminal respiration in actively respiring cells. ''Reductive'' Synthetic Reactions Another possibility which merits consideration is that microsomal cyto- chrome primarily funnels electrons available from DPNH (or TPNH) to acceptors in synthetic processes that are reductive in character, such as the reductive synthesis of fatty acids or steroids, or which involve elements of both reduction and oxidation, such as hydroxylation reactions (requiring both DPNH (or TPNH) and molecular oxygen). The rather low oxidation- reduction potential of microsomal cytochrome would make it a suitable mediator for transfer of electrons in such processes requiring reductive capacity. The localization of the cytochrome in microsomes would be in keeping with such a role, as the microsomes, or the endoplasmic reticulum from which they are derived, are recognized as the site of or are required for a variety of synthetic processes, including protein biosynthesis (Littlefield, Keller, Gross and Zamecnik, 1955), cholesterol synthesis in liver (Bucher and McGarrahan, 1953), and a variety of hydroxylation reactions required for formation and metabolism of steroids, various aromatic compounds, etc., in the liver or adrenal (Ryan and Engel, 1957; Mason, 1957; Kersten, Leonhauser and Staudinger, 1958). It may therefore be significant that microsomal cytochrome is present in large concentration in the rather abundant endoplasmic reticulum of liver, an organ with a relatively reductive internal environment and a high capacity for many synthetic reactions, and that other tissues in which microsomal cytochrome or apparently similar pigments are reportedly present are also sites of specialized synthetic pro- cesses or of transport phenomena which may involve elements of synthesis. The possible participation of microsomal cytochrome in hydroxylation of steroids by adrenal microsome preparations has recently received considerable attention (see above). A model system of chelated iron salts, ascorbic acid and oxygen can produce hydroxylation reactions (Udenfriend, Clark, Axelrod and Brodie, 1954), and it has been suggested that the haemin of microsomal cytochrome might similarly provide the site at which electrons from DPNH 472 C. F. Strittmatter or TPNH and molecular oxygen might be combined to produce the grouping required for biological hydroxylation reactions. However, there is at present no direct evidence for participation of microsomal cytochrome in hydroxy- lation or other synthetic processes. Clearly, the question of the biological functions of microsomal cytochrome and other apparently related cytochrome ^5-like substances is still unanswered, but the lively current interest in this problem gives hope that further insight into the biological significance of these compounds may be obtained in the near future. REFERENCES Appleby, C. A. & Morton, R. K. (1954). Nature, Loncl. 173, 749. Bailie, M. & Morton, R. K. (1955). Nature, Loud. 176, 111. Bailie, M. & Morton, R. K. (1958). Biochem. J. 69, 35. BucHER, N. L. R. & McGarrahan, K. (1953). /. biol. Chem. Ill, 1. Chaix, p. & Heyman-Blachet, T. (1957). Biochim. biophys. Acta 26, 214. Chance, B. & Williams, C. R. (1954). /. biol. Chem. 209, 945. Chance, B. & Williams, G. R. (1955). /. biol. Chem. Ill, 395. Ephrussi, B. & Slonimski, p. P. (1950). Biochim. biophys. Acta 6, 256. FujiTA, A. & KoDAMA, T. (1934). Biochem. Z. 113, 186. Garfinkel, D. (1957). Arch. Biochem. Biophys. 71, 111. Hayes, J. E. Jr. & Velick, S. F. (1954). /. biol. Chem. 207, 225. Hill, R. & Scarisbrick, R. (1951). New Phytol. 50, 98. HoGEBOOM, G. H., Claude, A., Hotchkiss, R. D. (1946). /. biol. Chem. 165, 615. HoGEBOOM, G. H. (1949). /. biol. Chem. Ill, 847. Hulsman, W. C, Elliott, W. B. & Rudney, H. (1958). Biochim. biophys. Acta 11, 663. HuszAK, I. (1942). Biochem. Z. 312, 330. Keilin, D. (1925). Proc. roy. Soc. 5 98, 312. Keilin, D. (1933). Ergebn. Enzymforsch. 2, 239. Keilin, D. (1934). Nature, Lond. 133, 290. Keilin, D. & Hartree, E. F. (1940). Proc. roy. Soc. B 129, 277. Keilin, D. & Hartree, E. F. (1947). Biochem. J. 41, 503. Keilin, D. & Hartree, E. F. (1949). Nature, Lond. 164, 254. Kersten, H., Leonhauser, S. & Staudinger, H. (1958). Biochim. biophys. Acta 29, 350. Kersten, H., Kersten, W. & Staudinger, H. (1958). Biochim. biophys. Acta 11, 598. Klingenberg, M. (1958). Arch. Biochem. Biophys. 75, 376. Krisch, K. & Staudinger, H. (1958). Biochem. Z. 331, 37. Kun, E. (1951). Proc. Soc. exp. Biol. N.Y. 11, 441. Littlefield, J. W., Keller, E. B., Gross, J. «& Zamecnik, P. C. (1955). J. biol. Chem. Ill, 11. MacMunn, C. a. (1884). /. Physiol. 5, Proc. xxiv. MacMunn, C. a. (1886). Phil. Trans. Ill, 267. Mahler, H. R., Raw, I., Molinari, R. & doAmaral, D. F. (1958). /. biol. Chem. 233, 230. Martin, E. M. & Morton, R. K. (1955). Nature, Lond. 176, 113. Martin, E. M. & Morton, R. K. (1957). Biochem. J. 65, 404. Mason, H. S. (1957). Advanc. Enzymol. 19, 79. Palade, G. E. & SiEKEVTTZ, P. (1956). J. Biochem. Biophys. Cytol. 2, 671, Pappenheimer, a. M. & Williams, C. M. (1954). /. biol. Chem. 209, 915. Raw, I., Molinari, R., doAmaral, D. F. & Mahler, H. R. (1958). /. biol. Chem. 233, 225. Raw, I. & Mahler, H. R. (1959). /. biol. Chem. 134, 1867. Ryan, K. & Engel, L. L. (1957). /. biol. Chem. 225, 103. Sanborn, R. C. & Williams, C. M. (1950). /. gen. Physiol. 33, 579. Schneider, W. C. & Hogeboom, G. H. (1956). Annu. Rev. Biochem. 25, 201. Studies on Microsomal Cytochromes 473 SiEKEViTZ, P. & Palade, G. E. (1958). /. Biochem. Biophys. Cytol. 4, 203. Shappirio, D. G. & Williams, C. M. (1957). Proc. roy. Soc. B. 147, 218. Slater, E. C. (1958). Advanc. Enzvmol. 20, 160. Spiro, M. J. & Ball, E. G. (1958)"^. Fed. Proc. 17, 314. Spiro, M. J. & Ball, E. G. (1961). /. biol. Chem. 236, 225. Slater, E. C. (1949). Biochem. J. 45, 1. Strittmatter, C. F. (1952). Ph.D. Thesis, Harvard University. Strittmatter, C. F. & Ball, E. G. (1952). Proc. Nat. Acad. Sci. Wash. 38, 19. Strittmatter, C. F. & Ball, E. G. (1954). /. cell. comp. Physiol. 43, 57. Strittmatter, P. & Velick, S. F. (1956a). /. biol. Chem. Ill, 253. Strittmatter, P. & Velick, S. F. (1956b). /. biol. Chem. Ill, 273. Strittmatter, P. & Velick, S. F. (1957). /. biol. Chem. 228, 785. Theorell, H. & Bonnichsen, R. K. (1951). Acta chem. Scand. 5, 1105, Tsou, C. L. (1951). Biocfiem. J. 49, 658. Udenfriend, S., Clark, C. T., Axelrod, J. & Brodie, B. B. (1954). /. biol. Chem. 201, 731. Velick, S. F. & Strittmatter, P. (1956). /. biol. Cliem. Ill, 265. Wainio, W. W. & Cooperstein, S. J. (1956). Advanc. Enzymol. 17, 331. Yamashita, J., HiGASHi, T., Yamanaka, T., Nozaki, M., Mizushima, H., Matsubara, H., HoRio, T. & Okunuki, K. (1957). Nature, Loud. 179, 959. Yoshikawa, H. (1951). /. Biochem. Tokyo 38, 1. DISCUSSION On the Nature of Cytoplasmic Pigments of Liver Cells By B. Chance (Philadelphia) Chance: There has been much speculation on the possible function of the cytoplasmic pigment, cytochrome b^, in liver cells, and the desirability of direct experimental data is clearly indicated. We have attempted to determine whether cytoplasmic cytochrome 65 is oxidized and reduced simultaneously with the mitochondrial cytochromes. ; 0.04 b 440 mg rat iver (wef weight) 540 550 560 570 Fig. 1 . Steady-state oxidized minus reduced difference spectrum for a rat liver slice. (Expt. 708a). Liver ^//'ce5— Studies on liver slices can be carried out by a simple modification of the double-beam spectrophotometer (Chance, in Methods in Enzymology, 4, p. 273. Academic Press, Inc., New York, 1957). Thin slices of perfused liver are floated in Ringer's solution in either an oxygen or a nitrogen atmosphere. Figure 1 records the aeiiobic-anaerobic difference spectra of such material. There is a peak attributable to cytochromes c and c^ at 552 m/t and a shoulder caused by cytochrome b absorption at — 562 m/<. There is not, however, any indication that an appreciable portion of J 474 Discussion cytochrome b^ (557 m//) changes its oxidation-reduction level in synchronism with the cytochromes of the respiratory chain. Other, much earlier work clearly shows what we now term the cytochrome 65 component to be reduced in liver cells. MacMunn (Phil. Trans. Roy. Soc. London 177, 267, 1886) and Keilin and Hartree (Proc. roy. Soc. B 129, 277, 1940) record reduced bands of this component in liver. Thus, cytochrome b„ is accessible to reducing substances, presumably cytoplasmic reduced diphosphopyridine nucleotide (DPNH), but not to the mitochondrial oxidase, in accord with studies of isolated mitochondria (Chance and Williams, /. biol. Chem. 217, 395, 1955). Isolated liver cells — If, instead of the tissue, we use a liver cell suspension (work in collaboration with W. D. Rutter), we find in the low temperature 'absolute' spectra of Fig. 2, the absorption bands of reduced cytochromes c, q, b and o + Og (dashed Normal Liver Cells Absolute" Low temperature Spectra 500 550 600 Fig. 2. 'Absolute' low temperature spectra of succinate-reduced (dashed trace) and dithionite-reduced (solid trace) liver cell suspensions. (Expt. 885a). trace) but no b^ band. If dithionite is added (solid trace), the absorption of the b components increases considerably and the bands of cytochrome b^ can be identified. Thus it appears that some if not all of the 65 component is oxidized in the liver cell suspension and can be reduced only in the presence of dithionite, in contrast with the results on solid tissues. It is possible that cytoplasmic reducing substances, presumably DPNH, have been oxidized or removed in the preparation procedure. A further examination of the cytoplasmic pigments of a liver cell suspension at low temperatures is aff"orded by Fig. 3 in which the oxidized minus succinate-reduced spectrum of the respiratory pigments (solid trace) is compared to that corresponding to the difference between succinate-reduced and DPNH-reduced cells (dashed trace). In the latter, the spectrum of the b components appears, showing the a^-band of cytochrome b-^. (The slight difference in the maxima of Figs. 2 and 3 can be attributed to differences between an absolute (Fig. 2) and a difference (Fig. 3) spectrum.) The room-temperature difference spectrum for the same conditions (Fig. 4) does not show the bands described by Williams and myself (Chance and Williams, /. biol. Chem. Ill, 395, 1955) at 426 and 557 mji, but instead shows an apparently different 'cytoplasmic pigment' with absorption bands at 423-5 and 555 m/t. The shoulder on the long wavelength side of the oc-band suggests the presence of some cytochrome 65. Microspectrophotometry — To localize the pigments in question more accurately, we have constructed a microspcctrophotometer capable of recording spectra of respiratory and accessory pigments in situ (Chance, Perry, Akerman and Theorell, Rev. Sci. Instrum. 30, 735, 1959). Figure 5 shows a spectrum of an area 1-5 /t in dia- meter in a single cell teased from a portion of liver tissue after centrifugation of the tissue and rendered anaerobic due to its own respiration (Theorell and Chance, Exptl. Cell Research 20,43, 1960). Traces 1 represent 'absolute' spectra of various portions of the cell containing the mitochondrial granules; traces 2 represent portions of the Studies on Microsomal Cytochromes 475 P 1 1-*— Reduced - Reduced + DPNH %' I b (Azide + Succinate In botti) Oxidized -Reduced (Azide ■^ Succinate) — I — r 500 T 1 1 1 1 1 1 1 1 1 1 550 600 A{m>i) Fig. 3. Oxidized minus succinate-reduced (solid trace) and succinate-reduced minus DPNH-reduced low temperature spectra for liver cell suspensions. (Expt. 976-1.) 423.5 nyj 400 T A0.D.=a005 ± 450 500 ;\(rnM) T— T — T — I — r 550 T 1 1 1 1 600 Fig, 4. Room temperature difference spectrum for succinate and DPNH reduced liver cell suspension. (Expt. 979.) 476 Discussion endoplasm where microsomes containing cytochrome A-, would be expected. The absorption peaks correspond to those of reduced cytochromes a^, b, and c in the mitochondria and to 65 or the 'cytoplasmic pigment' in the microsomes. The peak positions are appropriate to the reduced forms of these substances in an 'absolute' rather than a difference spectrum. These data for the single cell teased directly from the tissue indicate a considerable concentration of the reduced 'cytoplasmic pigment' without the addition of DPNH or dithionite, as was necessary for recording absorbency changes of the cells in suspension (Figs. 2 and 3). This result agrees with those of MacMunn and Keilin for liver tissue but clearly distinguishes the location of the cytochromes O3, b, c and b^ spectra. Observations of the cytoplasmic pigment in the transition from aerobiosis to anaerobiosis in cell suspensions support the observations of Fig. 1 : no change is Fig. 5. Microspectrophotometric recording of the spectra of a single, teased and centrifuged, anaerobic liver cell. B, base line; 1, mitochondria; 2, endo- plasm. (BT-4.) Reprinted from Exptl. Cell Research 20, 43, 1960. -X observed. However, significant increases of reduction are observed as a function of time after beginning the experiment, which might well be due to variations in the level of cytoplasmic DPNH. Nomenclature of accessory and cytoplasmic pigments — These studies and those reported by Chaix (p. 225) and by Bonner (p. 479) lead us to emphasize the use of the term accessory pigment (Chance and Williams, Advanc. Enzymol. 17, 65, 1955) in contrast to respiratory pigment. An acceptable preliminary test for the latter is the existence of a steady state of the pigment in respiratory activity as observed at room temperatures or as 'trapped' at liquid nitrogen temperatures. Another test is based on observations of the increasing reduction of the pigment that closely follows the time- course of the termination of respiration as oxygen is exhausted in an aerobic-anaerobic transition. An accessory pigment may not change its oxidation-reduction level at all during this transition or may change so slowly that its kinetics violate one of the two theorems mentioned elsewhere in this symposium (see p. 607). A critical test is afforded by a comparison of the kinetics of its oxidation and reduction with the rate of electron transfer through the system. There should be a close relationship for the respiratory pigment and a remote one for the accessory pigment. This does not, however, eliminate the participation of the accessory pigment from slow phases of biological oxidation. If the localization of the accessory pigment is known, then the term may be appropriately modified, as in this case, to cytoplasmic pigment; if the pigment is localized elsewhere, such terms as endochrome, mitochrome, nucleochrome, etc., might ultimately become more useful and more sharply define the general term cytochrome. Possible Functions of the Cytochromes of the Enchplosniic Reticulum of Animal Cells Dickens: My remarks about tlie paper by Stritlmatter (p. 461) concern the possible significance of the reduced di- and triphosphopyridine nucleotide (DPNH and TPNH) cytochrome reductase activity of microsomes. Whereas in the isolated microsome fraction of liver and other tissues cytochrome 65 readily interacts with DPNH and Studies on Mkrosonml Cytochroincs All TPNH, and the reduced isolated cytochrome is known to be capable of being oxidized efficiently by cytochrome c, during the extraction procedure the reactivity towards DPNH and TPNH is lost. It would be of great interest to know precisely on what the TPNH reductase activity depends, for instance, whether a specific pyridine nucleotide reductase such as that found for DPNH is involved, and the nature of its terminal acceptor. Is it possible that, in the intact cell, the fact that the mitochondria lie in the folds of the endoplasmic reticulum, which may therefore be in very close proximity to mitochondria, could facilitate microsomal-mitochondrial interchange? However, Chance's preceding remarks seem to make such an interaction improbable, since there is little change in the oxidation-reduction level of cytochrome b^ in relation to changes in that of other cytochromes. On the other hand the microsomes are the seat of an important series of hydroxyla- tion, aromatization and ring-closure reactions which specificially require TPNH, and might therefore represent a main route of TPNH oxidation in microsomes. My colleagues, G. Clock and P. McLean, found for washed liver microsomes a ratio of DPN+/DPNH of 5/1, suggesting that DPNH is probably capable of oxidation in micro- somes, and since DPNH does not appear to be utilized there in the same pathways as those which utilize TPNH, cytochrome b^ plus its specific cytochrome reductase system would be suitable for DPNH oxidation provided some effective terminal acceptor can be found. There remains, of course, the possibility raised by Chance's comment, that alteration of the haemoprotein occurs during the extraction procedure used. Strittmatter : The possibility of microsomal-mitochondrial interchange to provide a mechanism and possible function for the large pyridine nucleotide oxidizing capacity of the microsomes is indeed attractive, but, as noted in my paper (p. 470), it finds no direct support in the meagre data available. The observation of Chance that micro- somal cytochrome in perfused liver slices or cell suspensions is not reduced under mild reducing conditions, in contrast to the mitochondrial cytochrome components, also militates against but does not rule out microsomal-mitochondrial interchange or a respiratory role for cytochrome b^; reduction of cytochrome 65 requires a relatively strong reducing agent, such as DPNH, which may be present in the intact cell of solid liver but, as noted by Chance, may have been removed by his procedures for preparing slices or cell suspensions. The Significance of E'q Values of Cytochromes in Relation to Cellular Function Estabrook: Strittmatter and Velick (/. biol. Cheni. 221, 265, 1956) have reported on the oxidation-reduction potential of cytochrome 65. I wonder about the change in poten- tial as determined with the particle bound haemaprotein (—0-120 V) and the purified haemoprotein (+0-02 V), and the significance of this change in any consideration of oxidation-reduction potential and the relevance to a thermodynamic consideration of such potentials in energy transfer. Morton: There is a similar type of discrepancy, but in the opposite direction, in the reported E'q values for soluble cytochrome b of heart-muscle (—005 V; Sekuzu and Okunuki, /. Biochem. Tokyo 43, 109, 1956) and for cytochrome b in particulate preparations (+0077 V; Colpa-Boonstra and Holton, Biochem. J. 11, iv, 1959). It would appear that the reactivity of a cytochrome is altered by association with lipo- protein material of cytoplasmic particles. Wainio: To demonstrate that the redox potential of an enzyme need not be different when soluble and insoluble, we need only to consider the values obtamed for cytochrome c oxidase by Ball (+0-29 V) who employed the heart muscle mitochondrial fragment, and by Wainio (+0-285 V), who employed a deoxycholate-solubilized partially- purified preparation of the oxidase. Strittmatter: As was noted in my paper (p. 464), the significance of this apparent change in, potential on isolation of the particle-bound cytochrome cannot be assessed with assurance at present. The change, if real, could reflect either the inevitable changes in specific linkages, structural relations and physico-chemical milieu on separation 478 Discussion of the cytochrome from its particulate site, or other structural changes resulting from the isolation procedures. The possiblity of either type of change must be considered in assessing the properties of any particulate component removed from its physiological environment, and, on the other hand, the uncertain significance of potential deter- minations on a particulate component in situ must also be kept in mind. Henderson: There is doubt as to the validity of £'„ values determined on particulate systems in so far as they apply to some molecular species on or within the particle (see also Paul, The Enzymes, Academic Press, 1951). This is shown e.g. by the results of Baumberger (Cold Spring Harbor Sympos. Quant. Biol. 7, 195, 1939) where the degree to which yeast cells were washed and aged caused a variation from -HO- 12 to +0-31 V in the potential of the medium at which the bands of reduced cytochrome c appeared. Apart from any other factors it would seem rather unlikely that the com- plete thermodynamic reversibility necessary for correct E'q evaluation would exist in a complex particulate system and this would be especially the case with the whole cell. Postgate: I think that the absurdity of discussing redox potentials of materials bound at specific sites in cells is illustrated by considering the effect of pH on this quantity. If a bacterium of volume 1 fi^ has an internal pH of 7, a little arithmetic shows that there are only about 60 hydrogen ions to be expected there. The meaning of pH and hence Eh, at specific sites in the cell is even less meaningful. THE CYTOCHROMES OF PLANT TISSUES By W. D. Bonner, jr. Johnson Research Foundation, University of Pennsylvania, Philadelphia 4, Pennsylvania INTRODUCTION Knowledge concerning the cytochrome components and their function in plant tissues has not developed as rapidly as has been the case in animal tissues. This slow development is not entirely due to the fact that, until recently, it was not always considered respectable to work on plant tissues. Visual observation of cytochromes in plant tissues is difficult because of the presence of plastid pigments and the low haem concentration. Plant physio- logists themselves, because of their preoccupations with the copper-protein 'oxidases' have been partially responsible for the lack of knowledge of plant cytochromes. The development of techniques for the extraction and concentration of cytoplasmic inclusions, coupled to the development of rapid and sensitive spectrophotometers, has resulted in considerable progress toward under- standing the nature of some of the plant cytochromes. In spite of this progress our present understanding of plant cytochrome components is inadequate and our knowledge concerning the path of electron transport is negligible. Plant cells possess a rigid cell wall, an immediate barrier to rapid and effective homogenization. This difficulty can be partly overcome by using dark-grown material, the absence of hght preventing the large-scale pro- duction of lignin. The cytoplasm represents only a modest fraction of the total cell volume, the major portion of a mature plant cell being vacuole. Frequently the vacuolar contents are very acid and on rupture of the cell this acid can produce a considerable and undesirable drop in the pH of the homogenate. This high acidity that is frequently encountered presents an obstacle to the extraction and fractionation of the cytoplasmic constituents of the cell. A completely suitable isotonic extracting medium with low salt content and the necessary buffering capacity remains to be found. The isolation of the 'mitochondrial fraction' from plant tissues has been more or less stereotyped since the methodology was introduced in 1951 (Millerd, Bonner, Axelrod and Bandurski, 1951; Axelrod, 1955). This method involves homogenization of the tissue in isotonic sucrose together with a rather high concentration of phosphate butfer. In spite of the fact that active, phosphorylating particulate preparations have been reported from 479 480 W. D. Bonner, jr. many laboratories and with a wide variety of plant tissues, there have been no reports that such particulate preparations exhibited control of respiration by adenosine diphosphate (ADP) concentration. It may be, of course, that in plant tissues there is a more complex control of the respiratory process than that we now understand in animal systems (Chance and Williams, 1956; Chance, 1959). Many plant tissues exhibit a respiration which is insensitive to the usual ligands, cyanide, azide, and carbon monoxide, a situation analogous to instances recorded in some animal and insect tissues. There are three known types of cyanide-insensitive respiration, viz : (a) The so-called 'ground respiration', easily demonstrated through the prolonged washing of many tissues, e.g. carrot. 'Ground respiration' appears after a marked decrease in respiratory rate during washing. (b) The development of cyanide-insensitive respiration in 'aged' tissue slices ; the development of the cyanide-insensitive respiration is accom- panied by a marked increase in respiratory rate. (c) Tissues in which the rate of respiration is close to maximum and which normally exhibit cyanide insensitivity. These different situations under which cyanide-insensitive respiration appears in plant tissues add com- plexities over and above those already enumerated to the under- standing of the sequence and function of the cytochromes found in plant tissues. Cytochromes were first observed in plant tissues by Keilin (1925) who described the reduced spectrum in a wide variety of plant species. Keilin observed that untreated plant tissues showed a two-banded spectrum, a wide diffuse band at 556 m// and a weaker band at 524 m/x and suggested that this spectrum was due to "modified cytochrome" or cytochrome b'. The nature and significance of cytochrome b' remains clouded in obscurity. Following dithionite reduction of the tissue Keihn found that the two-banded spectrum was replaced by a spectrum showing components a, b and c. Curiously, the region of the spectrum between components b and c appeared shaded and frequently was observed as one wide absorption band. For fourteen years following Keilin's initial observations the study of cytochromes in plant tissues remained a depressed area. During this time interval, Yakushiji (1935) published on the cytochromes of germinated soy beans in which he described absorption maxima at 640, 603, 560 and 550 mju. The 640 m/< component was erroneously described as cytochrome «2- Mann (1938; see also Hill and Hartree, 1953), made an important contribution. He showed that, in general, the greatest amount of haemoprotein, estimated as pyridine haemochrome, was found in the most actively respiring part of the plant tissue, particularly in the non-vacuolated meristematic region. The Cytochiomes of Plant Tissues 481 The year 1939 was a vintage year in the study of plant cytochromes because of the work of three separate groups in Japan, England and America. Okunuki (1939), continuing investigations initiated in 1937, pubUshed three papers devoted to pollen derived from ten different kinds of plants. The cytochrome spectrum of pollen is intense, showing absorption maxima at 605, 561 and 550 m/^. Okunuki did not comment on the difference in maxima of the pollen b component (561 mfi) and that of yeast and muscle (564 m^i). Okunuki also studied the effect of carbon monoxide on the respiratory rates of the pollen under investigation and in most instances found a photoreversible inhibition. Remarkably, there were some instances where carbon monoxide had little effect on the respiratory rate, an observa- tion Okunuki interpreted as being due to species variation in carbon monoxide affinity. Okunuki, in 1939, was successful in preparing and concentrating cell-free pollen extracts which contained a number of dehydrogenases and which were capable of catalysing the rapid oxidation of succinate to fumarate. It is unfortunate that these remarkable contributions of Okunuki have largely been passed over. Okunuki was primarily interested in the oxidative characteristics of particulate cell-free preparations from pollen. During the same time and using very similar techniques. Hill and Bhagvat (1939) showed that particulate cell-free preparations could be obtained from a variety of flowering plants. Such preparations contained the bulk of the plant cytochromes, observed at that time as components a, b and c, and these cytochromes were observed to undergo reversible oxidation and reduction in plant preparations. Hill and Bhagvat (1939) also observed that their preparations oxidized succinate, and that this oxidation was promoted by the addition of soluble cytochrome c and was found to be inhibited by cyanide and azide. The first brief report of Hill and Bhagvat (1939) was elaborated in a later paper (Baghvat and Hill, 1951) in which they stated that components a, b and c could be observed in every tissue studied by them. In each case the cytochromes were associated with cytochrome oxidase and succinic dehydrogenase activities. Bhagvat and Hill (1951) concluded : 'the cytochrome system in plants behaves in the same way as that of yeast and animal tissues, thereby showing the presence of a respiratory mechanism identical with that characteristic of animal tissues'. This very sweeping generalization needs some modification in the light of our present knowledge, much of which has been subsequently contributed by Hill and his colleagues. Marsh and Goddard (1939) used a different approach for studying the cytochromes of plant tissues. Their approach was to investigate the effects of cyanide, azide, and carbon monoxide on respiratory rates of carrot slices and leaves. By determining the ratio of affinity of the enzymes which mediate the respiration of carrot tissues and young carrot leaves for oxygen and 482 W. D. Bonner, jr. carbon monoxide, these workers deduced that the respiration of these tissues was mediated by cytochrome oxidase. However, mature carrot leaves did not show any appreciable inhibition with azide, cyanide, or carbon monoxide. From these widely varying approaches to plant respiration substantial evidence was available already in 1939, showing that cytochrome oxidase does mediate the reaction of plant tissues with oxygen and that plants possess a full complement of other cytochromes. In spite of this fact, progress in understanding the cytochromes of plant tissues has continued to be slow for the reasons outlined above. THE CYTOCHROME COMPONENTS THAT HAVE BEEN DESCRIBED AND NAMED IN PLANT TISSUES As mentioned above, both Okunuki (1939) and Hill and Bhagvat (1939) and Bhagvat and Hill (1951) described components a, b and c, and prior to this, Keilin had described 'modified cytochrome' or cytochrome b' as well as components a, b and c. Hill and Scarisbrick (1951) described a b type cytochrome present in the leaves of a wide variety of higher plants and also extracted from the leaves of Vicia faba. The soluble component, called cytochrome b^, was auto-oxidizable and did not combine with carbon monoxide or cyanide. The presence or absence of cytochrome b^ has neither been confirmed nor denied, although Martin and Morton (1955, 1957) have described a similar component in the microsomal fractions derived from chlorophyll-free portions of silver beet petioles and from wheat roots. The absorption maxima given by these two groups of investigators are sufficiently different to warrant the view that Martin and Morton were not dealing with the same component described and extracted by Hill and Scarisbrick (see Martin and Morton, 1957). While studying the cytochrome components in the plastids of etiolated barley leaves, Davenport (1952) observed a b cytochrome which was later observed in Chlorella and named cytochrome b^ by Hill (1954). There is an additional /)-type cytochrome component which has been described in the tissues of Arum macidolum and called 'cytochrome b^"" (Bendall and Hill, 1956). The term 'cytochrome 6,' has also been used by Bendall (1958), by Hackett and Haas (1958) and by Chance and Hackett (1959). Evidence that cytochrome h.^ differs from previously described b cytochromes rests on the visual spectroscopic determination of oxidation- reduction potentials in tissue preparations. This determination is based on the absorption characteristics of preparations that are in equilibrium with various solutions which differ in their oxidation-reduction potentials. The cytochrome A-type components which have been described in plant tissues are summarized in Table 1. All observations point to the fact that the cytochrome c of higher plants is very similar to that of other organisms. Goddard (1944) was the first to The Cytochromes of Plant Tissues 483 extract cytochrome c from plant tissues. The cytochrome c, extracted from wheat embryos appeared to be identical with the cytochrome c extracted from horse heart. More recently, Bonner and Smith (1961) and, independently, Estabrook (this volume, p. 444) have shown that cytochrome c from wheat germ has a low temperature absorption spectrum which is identical with that of horse heart cytochrome c. Hagihara, Tagawa, Morikawa, Shin and Okunuki (1959) have shown that the crystalline form of wheat germ cyto- chrome c differs from that of cytochrome c from other sources. Table 1. Cytochromes b that have been described in plant tissues Room temperature -190 C a-max Component oc-max /^-max (m//) (m/t) Reference b b. b, 564 559 560 562 560 530 529 (525) 529 560 556-5 558-5 Keilin (1925) Estabrook (1956) Hill and Scarisbrick (1951) Martin and Morton (1955, 1957) HUl (1954) Bendall and Hill (1956) Chance and Hackett (1959) It has not been established definitely that plant tissues have a component identical with cytochrome q of animal tissues. Martin and Morton (1957) have used the term cytochrome c\ to describe an absorption band observed in mitochondria of wheat roots and of silver-beet petioles. Cytochrome /, which is in many respects rather similar to cytochrome c, was first described and extracted by Hill and Scarisbrick (1951). The extraction procedure was improved by Davenport and Hill (1952). Cytochrome/appears to be located in the chloroplasts. Bendall and Hill (1956) have described a cytochrome which is similar to /but not identical with it and which they have hesitated to name. The a-components of plant tissues appear to be cytochromes a and a^. Although the evidence is meagre, the few absorption spectra that are available (see Chance and Hackett, 1959; Hackett and Haas, 1958; Yocum and Hackett, 1957; Yocum and Bonner, unpubUshed work; Hartree, 1957; Martin and Morton, 1957 and Bonner and Smith, 1961) makes it probable that the cytochrome a -\- a^ band is identical with that found m heart muscle and other sources. There have been many demonstrations of photo- reversible carbon monoxide inhibition of a variety of plant tissues. As yet there are no photochemical absorption spectra of carbon monoxide-inhibited plant tissue respiration. 484 W. D. Bonner, jr. From the foregoing it may be readily appreciated that in the tissues of higher plants the complexities of the cytochrome components are not found in the c and a components but in the number and diversity of the cytochromes b. It is the complexity of the b components of plant tissues that gives to them the characteristic visual absorption pattern on reduction, a broad band in the green portion of the spectrum, a band very reminiscent of the absorption spectrum one can observe in well perfused liver slices. The complexity of the number of b components in plant tissues immediately raises the question as to their reahty. Is it possible that there are as many as five b components ? Since, on the basis of visual spectroscopy, there have been described since 1951, five different b components, is there a possibility that the use of more refined techniques will reveal even more ? It is of the utmost importance to answer these questions and also to understand the role that each of the cyto- chromes b plays in electron transport to oxygen and the role, if any, of those components not directly involved in the main transport chain. It is possible that some cytochromes are involved in photosynthesis and in the reactions of photomorphogenesis. All these questions are closely coupled to the structure and function of the various inclusions found in the cytoplasm of plant cells. Therefore, it is of great importance to know the location of each of these different cytochromes b, should they turn out to be real. We can begin to answer some of the above questions and try to point a way to the efforts which will have to be invested in an attempt to answer the others. EXPERIMENTAL During the past months we have carried out a systematic investigation of the cytochromes of plant tissues with two primary objectives, viz., (a) what are the cytochrome b components of plant tissues and where are they located in the cell? {b) are there species variation in the cytochrome composition of the higher plant tissues ? This investigation has been carried out using intact tissue as well as various cell-free preparations derived from them. The cytochrome components have been dehneated through the use of the divided beam spectrophotometer developed by Chance (1957), and adapted for low temperature work by Estabrook (1956). A full discussion of the methodology is given by Estabrook (this volume, p. 436). The reaction cell, illustrated in Fig. 1, was made according to the modification of Yocum and Bonner (unpublished work) who have shown that the glycerol and devitrifi- cation procedure of Keihn and Hartree (1949) need not be adhered to. Good spectra can be obtained by freezing in aqueous solution provided a con- centrated suspension of particulate matter is used. This procedure lends itself well to the study of frozen steady states. It should be emphasized that the absorption maxima obtained at low temperature in aqueous solution are shifted to slightly longer wave-lengths when compared to those obtained in 50 % glycerol at the same temperature. The Cytochromes of Plant Tissues 485 Since it was impossible to determine the osmotic concentrations of each tissue used in this investigation, one standard procedure for isolation of cytoplasmic inclusions was used; this procedure is outlined in Fig. 2. In cases where the tissues were known to be fairly acid, 0-05 m potassium bicarbonate or 0-01 m Tris (2-amino-2-hydroxymethylpropane-l :3-diol) buffer, pH 7-8 were used in place of the phosphate. A more detailed description of the isolation of plant particles is given by Bonner and Smith (1961). TEMP ABSORPTION CELL Fig. 1 . The absorption cell used for low temperature studies. During the optical measurements, the foot remains in the liquid nitrogen, thus insuring that the temperature of the absorption cell and its contents remains close to — 190°C. The use of low temperature spectrophotometry is essential for the investi- gation of plant cytochromes for two reasons : (a) the relatively low concen- tration of cytochrome components in plant tissues and in cell-free prepara- tions derived from them, a situation where absorption intensification at low temperature is almost mandatory ; and (b) the multiplicity of the cytochrome b components of close absorption maxima requires the sharpening and definition that is obtained at low temperatures. In this investigation, four types of difference spectra were used to delineate the cytochromes in particulate preparations. In each case, the spectra were representative of the difference in absorption between an aerated preparation and a preparation treated in one of the following ways : 1 . Substrate reduction to different steady states. 2. Reduction with reduced diphosphopyridine nucleotide (DPNH) in the presence of cyanide. 486 W. D. Bonner, jr. 3. Reduction with DPNH in the presence of n-heptyl-8-hydroxy-quinoUne- N-oxide (HOQNO). 4. Reduction with sodium dithionite. It will be seen below that these four methods of treatment coupled to the low temperature absorption spectrophotometry represents a powerful approach to the study of plant cytochromes. Homogenate 2000 g X 10 min Supernafant Discard 10,000 g x 20 min Resuspend in 0-5 M iO.OOO g X 20 min mannitol/phospiiate and tesediment / "mitochondria" ... . /\ ^ ... Discard 30,000 g x I hour Sediment / \ Supernatant "Microsomes Discard Fig. 2. Preparation and fractionation of plant tissue homogenates. 200 g of plant tissue was ground in 200 ml of 0-5 m mannitol, 001 m phosphate (pH 7-3) and 0-1% of cysteine or glutathione and 10"^ m versene. The material was either hand-ground in mortar and pestle with sand, or disintegrated in a Waring blender run at a low speed. The ground material was squeezed through muslin, giving the homogenate. In the following, 'component 558' is used to indicate the cytochrome having an a-absorption band at — 190°C in phosphate buffer at 558 m//, and other cytochromes are indicated by a corresponding notation. As shown by Okunuki (1939), pollen is a material relatively rich in cyto- chromes. Dithionite-reduced pollens derived from a variety of sources show three well defined absorption bands when studied with a microspectroscope. The low temperature difterence spectrum, resulting from the difference in absorption between a well aerated corn pollen suspension and a dithionite- reduced suspension is shown in Fig. 3. It can be seen that this spectrum shows, AO DfO 01 CORN POLLEN SUSPENSION ' I ' ' I ' M ' ' I ' 520 550 580 610 ;)(nn>.) Fig. 3. Difference spectrum (at — 190°C) representing the difference in absorption between oxidized and dithionite-reduced com pollen suspensions. 655 558 520 1^ 550 I I I I I ■ 580 OX-HOONO/DPNH L^L- T OX-CN/DPNH 520 I ' ' I ' • I ' 550 580 610 A (nip) Fig. 4. Difference spectra (at — 190^C) obtained from a corn pollen homogenate. The spectra represent the difference in absorption between oxidized homogenates and homogenates reduced with dithionite (left), reduced with DPNH in the presence of HOQNO (upper right) and DPNH reduced in the presence of cyanide (lower right). Note the difference in scale between the spectrum on the left and those on the right. Fresh pollen was supplied by Mr. D. B. Walden of Cornell University, Ithaca, N.Y. 488 W. D. Bonner, jr. in addition to the {a + (h) band at 598 m//, and the c band at 549 m/i, two well defined peaks at 558 and 554 m/x. These latter two components are seen as one absorption band when observed with the microspectroscope at room temperature. In order to delineate the cytochromes of corn pollen more clearly, a homogenate, prepared in 01 m sodium phosphate buffer was studied. The ■CAULIFIQWER. A0D.=0.01 Ox.-SjO^ Ox.-DPNH/CN" Ox.-DPNH/HOQNO ■I I I I I I I I I I I I I 520 550 580 610 Wovelenglh (m;j) Fig. 5. Difference spectra (at -190°C) ob- tained from cauliflower mitochondria. Curve A, oxidized minus dithionite; Curve B, oxidized minus DPNH and cyanide; Curve C, oxidized minus DPNH and HOQNO. Fig. 6. Diff"erence spectra (at -IWC) obtained from etiolated black valentine bean mitochondria. The spectra repre- sent the difference in absorption between an oxidized mitochondrial suspension and suspensions reduced with A, dithio- nite; B, cyanide and DPNH; C.DPNH and HOQNO. low temperature spectra resulting from this study are shown in Fig. 4. Inspection of this figure shows that the dithionite reduced homogenate gives essentially the same spectrum as does the whole pollen suspension. A remarkable effect is observed on reduction of the homogenate by DPNH in the presence of cyanide, under wliich conditions the ^b" components disappear almost entirely and only cytochromes c and {a + ^3) remain reduced. Such behaviour is very different from yeast and muscle preparations, but has been demonstrated in two aroids Arum maculatuni (Bendall and Hill, 1956) and Syinplocarpus foctidus (Chance and Hackett, 1959; Yocum and Bonner, unpublished work). The Cytochromes of Plant Tissues 489 Reduction of the corn pollen homogcnate in the presence of HOQNO also produces a remarkable spectrum. Under such conditions in the Keilin and Hartree muscle preparations or in yeast suspensions, only cytochrome b Anfimycin A / OPNH Fig. 7. Difference spectra (at — IWC) obtained from etiolated mung bean hy- pocotyls. Tlie spectra represent the dif- ference in absorption between oxidized mitochondrial suspensions and suspen- sions treated as indicated in each curve. Fig. 8. Difference spectra (at — 190°C) ob- tained from cabbage mitochondria. The spectra represent the difference in absorption between oxidized mitochondrial suspensions and suspensions treated as indicated on each curve. (at 560 m/Li) remains reduced while the (a + ^3) and c components become more oxidized. In the com pollen homogenate three well defined peaks appear in the presence of HOQNO, these peaks have their maxima at 555, 558 and 564 m/n. Other illustrations of this behaviour toward HOQNO or antimycin A, will be given below. All plant tissues that have been examined so far characteristically show these same three absorption peaks following treatment with HOQNO or antimycin A. K 490 W. D. Bonner, jr. Figures 5 to 9 show low temperature difference spectra of 'mitodiondrial' suspensions derived from various sources. These sources include cauliflower buds, etiolated black valentine bean and mung bean hypocotyls, cabbage and SKUNK CABBAGE Abffjrbaiicy Ox- DPNH/CN" Ox -DPNH/HOONO I M M I I I I 490 520 550 580 Wovelength (mp) Fig. 9. Diflference spectra (at — 190°C) obtained from skunk cabbage spadix mitochondria. The spectra represent the difference in absorption between aerobic mitochondrial suspensions and suspensions treated as indicated on each curve. skunk cabbage flowers. All of these preparations show the same character- istic behaviour with HOQNO or Antimycin A and to reduction by DPNH in the presence of cyanide. Reference to the figures showing the low temperature spectra of the various plant preparations shows that there is much more dithionite-reducible cytochrome 'Z)' than can be found following substrate reduction, even with anaerobiosis. This behaviour of the 7/ components is also characteristic of all plant tissues that have been investigated ; so far any reasonable explana- tion of this phenomenon has evaded us. An indication of the magnitude of The Cytocluomcs of Plant Tissues 49i this dilTerencc is shown by Ihc following figures: substrate reduction of mung bean particles; a + a.^, 93%; component 558, 43-6% and component 555, 40-8%. Microsomes The microsomal fraction, when reduced with dithionite, characteristically shows an asymmetric band of maximum 559 m/^ (Fig. 10) or in some instances 558.6 MUNG BEAN "MICROSOMES" 580 580 ,1(n 520 550 X (nn>i) Fig. 11. The difference spec- trum (at -190°C) obtained by measuring the difference in absorption between oxidized mung bean microsomes and a suspension of microsomes re- duced with DPNH. Fig. 10. Difference spectra (at — 190^^C) obtained from the mung bean hypocotyl microsomal frac- tion. The curves were obtained from the difference in absorption between aerated microsomal sus- pensions and suspensions reduced with dithionite (solid line) and DPNH (dashed line). a two-peak spectrum of maxima 559 and 553 mfi. Reduction of this fraction with DPNH gives a two banded spectrum at 558 and 552 mfi (Fig. 1 1). The amount of 552 component varies among different plant species. Chloroplasts Cytochromes /and b^, the two haemoproteins that have been described as chloroplast components, are always found in the reduced state; these two components are not easily oxidized chemically (Davenport and Hill, 1954). For this reason, difference spectra were difficult to carry out on isolated chloroplasts. Before entering into a study of the low temperature spectra of isolated chloroplasts it was thought advisable to obtain the low temperature spectrum of cytochrome /, the one chloroplast component that has been 491 W. D. Bonner, jr. purified. The extraction and purification of cytochrome /, from parsley leaves, was carried out according to Davenport and Hill (1952). The low temperature absolute absorption spectrum of partially purified cytochrome /is shown in Fig. 12. With a sharp maximum at 552 m^ cyto- chrome /can be easily distinguished, at —190°, from the other cytochrome components that have been described here. No attempt has been made yet Z\0. D. = 0.02 i CYTOCHROME f I I I I I I I 520 550 Fig. 12. The absolute absorption spectrum (at cytochrome /preparation. 190°C)ofa to investigate the absorption properties of the chloroplast b component, nor of whole chloroplast suspensions. Roots Particulate preparations were obtained from roots with difficulty and in small yields. Particulate preparations from barley roots showed maxima at 596, 558 and 551 m/n. The spectrum for a particulate preparation derived from bean roots is shown in Fig. 13. It may be seen that reduction with DPNH shows bands corresponding to cytochrome components (a + a^), b and c ; after reduction with dithionite, a very large amount of dithionite- reducible cytochrome appears. Cyanide Insensitive Respiration Only brief mention will be made here concerning cyanide-insensitive respiration; for a detailed discussion see Bendall and Hill (1956), Chance and Hackett (1959) and Bonner and Smith (1961). Those tissues which exhibit, characteristically, cyanide-insensitive respiration have been examined in detail for their cytochrome components. 'Mitochondrial fractions', prepared from cyanide-insensitive tissues have been found to contain exactly The Cvtochromes of Plant Tissues 493 the same cytochrome components as found in cyanide-sensitive plant tissues. A series of spectra of a particulate preparation from skunk cabbage is shown in Fig. 9. Such particulate preparations are hardly affected in their rate of oxidation of succinate by sodium azide in concentrations as high as 0-01 m. It may be seen that the cytochrome spectrum does not differ in any way from Fig. 13. Difference spectra (at — 190°C) obtained from a black valentine bean root particulate preparation. The spectra represent the difference in absorption between aerated particulated suspensions and suspensions reduced with dithionite (solid line) and DPNH (dashed line). that of cauliflower mitochondria (Fig. 5) which are inhibited fully by rela- tively small concentrations of sodium azide. DISCUSSION In the preceding section a large number of absorption spectra have been presented and now one may well ask what positive contributions to our understanding of plant cytochromes have been made here. How far can one go in delineating the cytochromes of plant tissues? Which specific com- ponents are involved in electron transport to oxygen and in the photo- synthetic apparatus? What are the immediate problems that now face us? These are important questions and most of the answers are not immediately apparent. We now have available considerable lore relating to the cytochrome components of plant tissues and this paper adds to this. However, the results presented in this study of plant cytochromes show quite clearly that, in spite of the complexity of the problem, there is an undercurrent of unity. It has been shown here that plant tissues appear to contain some of the same cytochrome components and these components are unique to plants. In 494 W. D. Bonner, jr. short, higher plant tissues possess a common unity of cytochromes components. It becomes increasingly clear that there is no unity of cytochrome com- ponents throughout the biological world ; such a unity exists only within each major biological group (cf. Estabrook, this volume, p. 454). At the present time there is insufficient evidence to assign a localization or a specific name to many of the plant cytochromes. However, one can consider what we know about cytochromes involved in electron transport to oxygen in the cyanide-sensitive portion of the respiratory chain of plant mitochondria. It would appear that the reaction with oxygen is mediated by cytochrome oxidase and, on the basis of spectral evidence and inhibitor studies, cyto- chromes a and a^ in plants are very nearly identical to those of animal tissues. This conclusion is by no means a new or surprising one. The substrate of cytochrome oxidase, cytochrome c, is likewise present in plant tissues and in good concentration. Again, the demonstration of cyto- chrome c in plant tissues is by no means new. It is interesting that the plant cytochrome c, as delineated most clearly by its reduction upon addition of substrate to mitochondria in the presence of cyanide or azide, has an a-band absorption maximum (— 190°C) which is consistently 2 m/^ toward the red as compared to animal preparations and pure animal or wheat germ cytochrome c solutions, all measured under the same conditions. Such a modest optical variation is not at all surprising. The important thing is that the plant c responds very similarly to its animal counterpart in steady state studies. A component corresponding to cytochrome q, as defined for animal tissues, has not been found in any plant tissues so far examined. On the basis of the above discussion we formulate the electron transport system in cyanide-sensitive plant mitochondria as follows : DPNH->? Substrate -> -* c^- a-^ a^-^O.^ Succinate -> ? It is the question marks that are difficult to discuss at our present state of knowledge. The question marks, of course, refer to the baffling nature of the cytochrome b components. This problem is further comphcated by the fact that possible mitochondrial contamination by plastid fragments, and consequently plastid cytochromes, has not been eliminated. The spectral data that have been presented in this study show no cyto- chrome component that can be clearly designated as 'cytochrome b\ Keilin and Hartree heart-muscle preparations and yeast suspensions show on substrate reduction in the presence of Antimycin A or HOQNO, one sharp absorption band with a maximum (at — 190°C) of 560 m/f. The plant preparations do not reveal a 560 m/^ component when similarly treated. Chance and Hackett (1959) also found no evidence for cytochrome b in their study of skunk cabbage mitochondria (on reduction by substrate). The Cytochromes of Plant Tissues 495 In spite of the apparent lack of 'cytochrome f in plant tissues we are still presented with the problem of finding a home for the cytochromes b that have been specifically named in plant tissues, viz., cytochromes ^3, b^ and b^. The name 'cytochrome b^ has been given to the microsomal pigment (Martin and Morton, 1953, 1957) wliile cytochrome b^ has been localized in plastids (Hill, 1954). Of the named Z> cytochromes in plants we are left with cytochrome b-, and yet one has to account for the three banded spectrum that char- acteristically appear on reduction of plant mitochondria in the presence of HOQNO and Antimycin A. It is felt that any attempt to name specifically any of the components b that have been presented in this paper would be premature. There are two characteristics of the /?-type cytochromes of plant tissues that differ markedly from the cytochrome b component of heart-muscle preparations. Steady state substrate reduction studies of substrate reduction in the presence of cyanide both show that plant b cytochromes possess a high degree of auto-oxidizability. The plant tissues and preparations also show an unusually high amount of dithionite-reducible b. An important question concerning this dithionite- reducible material is whether it is the same as one of the substrate-reducible components already described, or whether it represents a separate, non- enzymic component. Chance and Hackett (1959) found that the dithionite- reducible component differed in its light absorption characteristics from the substrate-reducible component. Not only were the y- and a-bands at different positions but the ratio of their intensities also was different, a situation analogous to mammalian tissue preparations (Chance, 1958). The answer to the nature of the dithionite-reducible material is not available at the present time nor is it known if this material represents one or more than one component. Mention has been made concerning the cytochrome components known to function in electron transport to oxygen in cyanide-sensitive plant mito- chondria. "While this is not the place for a detailed discussion of cyanide- insensitive respiration, this topic should be considered briefly. Bendall and Hill (1956) have suggested that cytochrome b-, could act in electron transport to oxygen in the cyanide-insensitive particles extracted from the spadix of Arum macula turn. Chance and Hackett (1959), from their study of skunk cabbage, concluded that the time was still premature to consider alternate pathways of electron transport. On the other hand, Bonner and Yocum (unpublished) feel that their evidence with the skunk cabbage does support the alternate pathway hypothesis; although they do agree with Chance and Hackett (1959) that really definite support for an alternate pathway is still lacking. The evidence presented in this paper has shown the cytochrome components of cyanide-insensitive and cyanide-sensitive tissues to be exactly comparable. 496 W. D. Bonner, jr. In short, there are no cytochrome components unique to cyanide-insensitive tissues, a fact that requires a further reinterpretation of the alternate pathway hypothesis. Such a reinterpretation has been made by Bonner and Smith (1961) who propose that all tissues possess the capacity for exhibiting cyanide- insensitive respiration. Such a hypothesis presumes two pathways of electron transport to oxygen, perhaps as illustrated in Fig. 14. Some possible consequences of such a proposal are discussed by Bonner and Smith (1961). However, considerable effort is needed to provide proof CONTROL SUBSTRATE 'Q CYAMDE "sensitive * 2 C_YAJ^LDE_ _ . ^ insensitive 2 Fig. 14. Hypothetical pathways of electron transport in plant tissues. that plant tissues can, or cannot, transport electrons to oxygen through a cytochrome system working along with, or alternately to, the c, a, a^ system. Although there has been considerable progress in the last few years in information concerning the cytochromes of plants, there remain enormous gaps in our knowledge. SUMMARY 1. A survey of the cytochromes in various plant tissues has been carried out utiUzing the remarkable sensitivity and resolution of the low temperature divided-beam spectrophotometer. 2. Previous observations, from other investigators showing that plants contain cytochromes a -f a^ and c have been confirmed and extended. 3. Plant tissues do not appear to contain cytochrome q, as defined in animal tissues, nor cytochrome b. 4. The cytochromes 'f of plants are unusually auto-oxidizablc as shown by substrate steady-state reduction of the pigments of mitochondria in the presence and absence of cyanide. 5. Plant mitochondria, when reduced with DPNH in the presence of HOQNO or antimycin-A, show three well-defined absorption maxima at 555, 558 and 564 m//. 6. The microsomal fraction, derived from various plant tissues, show, on DPNH reduction, a two-banded spectrum of maxima 552 and 558 m//. Reduction of this same fraction with dithionite results in an asymmetric band of maximum 559 m/^. The Cytochromes of Plant Tissues 497 7. The cytochromes of roots are exactly comparable to those of other plant parts. 8. The low temperature absorption spectrum of purified cytochrome /has been obtained. 9. It is concluded that it is premature to designate names to the cytochrome b components described here. Acknowledgements The author expresses warm appreciation to Dr. Britton Chance for his enthusiasm and advice during the course of this study. He also thanks Dr. R. W. Estabrook for reading the manuscript, and Mr. N. Meadow and Miss S. Yao for help in the preparation of cytochrome /. This work was supported by a grant from the National Science Foundation. REFERENCES AxELROD, B. (1955). Methods in Enzymology, Vol. 1, p. 19. Ed. Colowick, S. P. & Kaplan, N. O. Academic Press, Inc., New York. Bhagvat, K. & Hill, R. (1951). New Pliytol. 50, 112. Bendall, D. S. (1958). Bioc/iem. J. 70, 381. Bendall, D. S. & Hill, R. (1956). New Pliytol. 54, 206. Bonner, W. D. Jr. & Smith, S. (1961). Plant Physiol. To he published. Chance, B. (1957). Methods in Enzymology, Vol. 4, p. 273. Ed. Colowick, S. P. & Kaplan, N. O. Academic Press, Inc., New York. Chance, B. (1958). /. iyiol. Chem. 233, 1223. Chance, B. (1959). Ciba Foundation Symposium on Cell Metabolism. Ed. Wolstenholme, J. & O'Connor, C. London: J. & A. Churchill Ltd. Chance, B. & Hackett, D. P. (1959). Plant Physiol. 34, 33. Chance, B. & Williams, G. R. (1956). Advanc. Enzymol. 17, 65. Davenport, H. E. (1952). Nature, Lond. 170, 1112. Davenport, H. E. & Hill, R. (1952). Proc. roy. Soc. B, 139, 327. Estabrook, R. W. (1956). J. biol. Chem. 223, 781. Goddard, D. R. (1944). Amer. J. Botany, 31, 270. Hackett, D. P. & Haas, D. W. (1958). Plant Physiol. 33, 27. Hagihara. B., Tagawa, K., Morikawa, L, Shin, M. & Okunuki, K. (1959). J. Biochem. Tokyo 46, 321. Hartree, E. F. (1957). Advanc. Enzymol. 18, 1. Hill, R. (1954), Nature, Lond. 174, 501. Hill, R. & Bhagvat, K. (1939). Nature, Lond. 143, 726. Hill, R. & Hartree, E. F. (1953). Ann. Rev. Plant Physiol. 4, 115. Hill, R. & Scarisbrick, R. (1951). New Phytol. 50, 98. Keilin, D. (1925). Proc. roy. Soc. B, 98, 312. Keilin, D. & Hartree, E. F. (1949). Nature, Lond. 164, 254. Mann, T. (1938). Archiwum Towarzystwa Naukowego we Lwowie, 9, 421. Marsh, P. B. & Goddard, D. R. (1939). Amer. J. Botany, 26, 724. Martin, E. M. & Morton, R. K. (1955). Nature, Lond. 176, 113. Martin, E. M. & Morton, R. K. (1957). Biochem. J. 65, 404. MiLLERD, A., Bonner, J., Axelrod, B. & Bandurski, R. S. (1951). Proc. natl. Acad. Sci. Wash. 37, 855. Okunuki, K. (1939). Acta Phytochim. (Tokyo) 11, 27 and 65. Yakushiji, E. (1935). Acta Phytochim. {Tokyo) 8, 325. YocuM, C. S. & Hackett, D. P. (1957). Plant Physiol. 32, 86. 498 Discussion DISCUSSION Cytochromes c^ and bg of Particulate Components of Plants By R. K. Morton (Adelaide) Morton: In 1953-1955 in collaboration with E. M. Martin, I investigated some of the haemoproteins of plants. From silver-beet petioles and from wheat roots, mitochondria and microsomal fractions were prepared, the origin and nature of which were investigated by electron microscopy (Hodge, Martin and Morton, /. Biochem. Cytol. 3, 61, 1957). We showed that the microsomal fraction largely originated from two types of lipoprotein membranes which we designated as 'endo- plasmic reticulum' and 'golgi zones' respectively. Incidentally, this was the first identification of such structures in higher plants. When plant mitochondria were heated (to destroy other haemoprotein components), and reduced with ascorbate, cytochrome c (a-band, approximately 550 m/< at room temperature) was detected. However, when reduced with DPNH, and to some extent, with succinate, and especially with dithionite, a band appeared close to 554 mfi (at room temperature), which could be readily resolved by lovz-dispersion microspectro- scopy from the cytochrome c band at 550 m// and from the cytochrome b band at 560-564 m/t. The band at 554 m// at first suggested to us contamination with cyto- chrome /(a-band, 555 m/{ at room temperature). However, when it was detected even more strongly in mitochondria from plant roots, it was recognized as a component not previously described in plants and called cytochrome Cj of plant mitochondria (Martin and Morton, Biochem. J. 65, 404, 1957; see also Morton, Rev. pure appl. Chem. 8, 161, 1958). This nomenclature was, at that time (1958), consistent with the report of cytochrome c^ as a haemoprotein of a-band at approximately 554 mn (at room temperature) in liver and muscle mitochondria and in other animal tissues. At that time, no function had been allocated to cytochrome c^; and, indeed, we would be on very dangerous ground today if we attempted to classify cytochromes on a functional basis. Moreover, it appeared undesirable to introduce a new alphabetical designation for every component found in each new tissue examined. Lundegardh (Biochim. biophys. Acta 11, 355, 1958) confirmed the presence of this component from spectro- photometric studies of plant roots, and showed that its oxidation-reduction state changed during aerobiosis and anaerobiosis. Moreover, in our studies of the eff"ect of inhibitors such as antimycin A (Martin and Morton, Biochem. J. 64, 221, 1956; Biochem. J. 65, 404, 1957) and later of HOQNO and amytal (Wiskich, Morton and Robertson, Aiist. J. Sci. 13, 109, 1960) on plant respiration, we observed that cyto- chrome Ci was partly reduced in the presence of these respiratory inhibitors. We therefore tentatively included this cytochrome as a component of the respiratory chain of plant mitochondria (see also Wiskich, Morton and Robertson, loc. cit.). This, of course, needs verification. Chance's trapped steady-state procedure (this volume, p. 457) will undoubtedly be valuable for clarifying this question. In the microsomal fraction, at room temperature only a single oxidizable-reducible haemoprotein was detected, with an a-band, when reduced, at 559 m// (at room temperature). This new component we first called 'cytochrome b^' (Martin and Morton, Nature, Lond. 176, 113, 1955) (not wishing to add another cytochrome name to the literature) and later 'cytochrome b^ (M & M)' (Martin and Morton, 1957, loc. cit.) to distinguish this pigment from the component described by Hill and Scarisbrick (New Phytol. 50,98, 1951). Cytochrome b^ of wheat-root microsomes was reduced very rapidly by DPNH and more slowly by TPNH ; no other substrate was found. The ferrocytochrome is very rapidly oxidized in air -so rapidly that, by analogy with rates experienced with cytochrome b-, in microsomes, it appeared possible that a specific cytochrome 63 oxidase was present (Martin and Morton, 1955, loc. cit.). In this connexion, we isolated from silver-beet petiole a very small amount of an oxygen-combining haemoprotein which appeared to have some analogy with the haemoglobins of yeast and fungi described by Keilin and Tissieres (see Martin and Morton, 1957, loc. cit.) The The Cytochromes of Plant Tissues 499 oxidation of ferrocytochrome b^ is partly cyanide resistant, suggesting a possible in- volvement of this cytochrome in the cyanide-resistant respiration of certain plant cells (Martin and Morton, 1955, loc. cit.). Lundegardh (1958, loc. cit.) has also detected cytochrome b^ in plant cells, and Crane {Plant Physiol. 32, 619, 1957) has observed cytochrome b^ in plant microsomes. We considered that cytochrome b^ of plant microsomes is closely analogous to cytochrome b^ of animal microsomes (cf. Martin and Morton, 1955, loc. cit.; Bailie and Morton, Nature, Land. 176, 111, 1955; Morton, 1958, loc. cit.) The observations reported here by Bonner have all been made at — 190°C and it is therefore difficult to relate the various absorption bands detected by him to the components observed by other workers, and especially to the components which Martin and I designated as cytochrome Cj and cytochrome b (in mitochondria) and cytochrome b^ (in microsomes). This difficulty arises because of the variable shifts and variable splitting of the a-absorption bands of the different ferrocytochromes when cooled to — 190°C (see Estabrook, this volume, p. 436) and because Bonner's results give difference spectra (reduced minus oxidized component) whereas our results give absorption bands of the reduced component. The broadening and in- tensification of the absorption at about 560 mjj, (at room temperature) on treatment of plant mitochondria with dithionite after addition of substrate under anaerobic conditions is readily observed. We attributed this to modification, probably involving denaturation, of the labile cytochrome b of plant mitochondria by dithionite. In our experience, dithionite readily produces denaturation of labile haemoproteins such as yeast cytochrome c and yeast cytochrome b^ even when the system is adequately bufTered against changes of pH. Finally, in agreement with Slater, we have suspected that CN" may combine with cytochrome b of plant mitochondria, thus accounting for the loss of absorption near 560 mn in the presence of this ligand and substrates. Bonner: As regards the microsomal component I think that this is a single pigment which, like cytochrome 65 shows (at — 190°C) a^ and aj bands. It differs from cytochrome 65, as shown in Fig. 10 in my paper, in that, on reduction with DPNH with cyanide present, it no longer retains the double peak but shows a single a-band. The component with an a-band at 552 m^t (at — 190°C) in plants is not reduced by ascorbate. Since cytochrome Cj of animal tissues is reduced by ascorbate, it does not appear to be identical with cytochrome c^ Estabrook : A further point to be considered in assigning the term cytochrome Cj to the pigment of plant mitochondria, in addition to its inability to be reduced by ascorbic acid, is the fact that an absorption band characteristic of this pigment appears in the presence of antimycin A. Thus applying the functional definition descriptive of cytochrome c^ of mammalian systems (cf. Estabrook and Sacktor, Arch. Biochem. Biophys. 76, 509, 1958) one must conclude that this absorption band is not associated with a pigment analogous to cytochrome c^ of mammalian systems. The Cytochromes of Roots Chance: In the writing of a review on cytochromes in roots with Lucille Smith (Smith and Chance, Ann. Rev. Plant Physiol. 9, 49, 1958), some aspects of Lundegardh's experi- mental data on cytochromes b kinetics appeared to be inconsistent with those of other components of the respiratory chain. For example, alternating oxygen and nitrogen in the wheat root bundle causes the absorption bands of cytochromes a, c and b to appear in less than a minute after the aerated medium is replaced by oxygen-free medium. Thereafter large and slow absorbency changes occur in the region of cytochrome b. Similarly in his important experiments on activation of anion respiration, there is a rapid change of cytochrome 03 and a slower change of cytochrome b which is maximal in about 10 min. The latter change is in fairly good agreement with the time course of the chloride absorption. In the course of the work on plant cytochromes, it has occurred to Bonner and me that the large amounts of accessory b pigments (see Chance, this volume, p. 476) found in other parts of the plant might also occur in the roots and thereby afford an 500 Discussion explanation for the discrepancies of cytochromes a^, a, c and the large cytochrome b band of Lundegardh. Figure 13 of Bonner's paper (p. 493) shows that treatment of bean roots with CN" and DPNH clearly shows (in a low-temperature difference spectrum) the bands of the cytochrome O3, a, c, b system in the usual proportions (corresponding roughly to one each of the components). The oxidized minus dithionite spectrum, however, clearly shows a great excess of accessory '6' pigments, that completely obscure the b and c bands of the respiratory components. This result for bean root suggests that the large amounts of accessory '6' pigments may be a general phenomenon and provide an explanation for Lundegardh's data. In this case we reach the interesting conclusion that the accessory pigment (of 6-type) is best correlated with the anion transport, thereby providing a haemoprotein of the type required by Davies' theory on this topic. THE CHEMICAL AND ENZYMIC PROPERTIES OF CYTOCHROME b., OF BAKERS' YEAST By R. K. Morton,* J. McD. Armstrong* and C. A. AppLEBYf Department of Agricultural Chemistry, The Waite Agricultural Research Institute, University of Adelaide, and Division of Plant Industry, C.S.I. R.O., Canberra Dixon and colleagues (Bach, Dixon and Keilin, 1942; Bach, Dixon and Zerfas, 1946) first described cytochrome ^2 as a haemoprotein associated with highly purified preparations of lactate dehydrogenase of bakers' yeast. Bach et al. (1946) concluded that 'cytochrome b.^ forms an essential part of the enzyme system either as the dehydrogenase itself or as an essential inter- mediate carrier between lactate and methylene blue.' Subsequently, Morton and co-workers (Appleby and Morton, 1954, 1959a, b, 1960; Armstrong and Morton, 1959) continued studies of cytochrome b,^, to determine the relation- ship between the haemoprotein and lactate dehydrogenase activity of bakers' yeast. The cytochrome b^ was isolated as a crystalline deoxyribonucleo- protein which contained equimolecular proportions of protohaem and riboflavin phosphate (Appleby and Morton, 1954, 1959a, b, 1960). Appleby and Morton (1954, 1959b) showed that the flavin group was essential for enzymic activity. Thus cytochrome b^ was identified as a flavohaemoprotein, a unique type of haematin enzyme with dehydrogenase activity. Boeri and colleagues (Boeri, Cutolo, Luzzati and Tosi, 1955; Boeri and Tosi, 1956), and Nygaard (1958) have confirmed the stoichiometric occurrence of flavin and haem groups in this enzyme. The isolation procedure (Appleby and Morton, 1954, 1959a; Morton, 1955a, b) involves few treatments and depends on the use of organic solvents (n-butanol and acetone) rather than on ammonium sulphate fractionation and adsorption and elution procedures (Bach et al., 1946; Boeri et al., 1955; Yamanaka, Horio and Okunuki, 1958). Lipid is extracted first from finely-ground, air-dried bakers' yeast by treatment with n-butanol (Morton, 1950, 1955a, b). The cytochrome b.;, and cytochrome c are then extracted from the yeast with lactate solution; the solution is fractionated with acetone at about -5°C, and the cytochrome bo, is crystallized free of cytochrome c by dialysis to low ionic strength against dilute lactate at about pH 6-8 (Appleby * University of Adelaide. t C.S.I.R.O. 501 502 R. K. Morton, J. McD. Armstrong and C. A. Appleby and Morton, 1959a). A yield of 10-20 rag of crystals/kg of dried yeast is consistently obtained and, with only minor modifications, the procedure as described in detail by Appleby and Morton (1959a) has been successfully used for isolation of crystalline cytochrome b.^ from a number of different strains of commercial bakers' yeast obtained in Australia and from other countries. Boeri (personal communication) has recently reported isolation of crystalline cytochrome b^ from Italian yeast by this procedure. Because of the ready dissociation of flavin from this cytochrome (see later), it is essential to observe the precautions indicated by Appleby and Morton (1959a) in order to avoid modification of the cytochrome during isolation. oxygen Ferricyanide methylene blue 2:6-dichlorophenol indophenol Riboflavin >- protohoem >- cytochrome Q phosphate oxygen c ytochrome bj Fig. 1. Scheme for interaction of substrate and acceptors with cytochrome 62 (Morton, 1955a; Appleby and Morton, 1954, 1955). Crystalline cytochrome b^ catalyses the reduction by lactate of ferricyanide, cytochrome r, methylene blue, 2 : 6-dichlorophenol indophenol and oxygen (Appleby and Morton, 1954, 1959b; Armstong and Morton, 1959). Based on detailed chemical and preliminary kinetic studies, it was proposed (Morton, 1955; Appleby and Morton, 1954, 1955) that various hydrogen (or electron) acceptors interact with cytochrome b^ as shown in Fig. 1. This scheme postulates that reduction of cytochrome c and of certain dyes involves an intramolecular hydrogen (or electron) transfer between the flavin and haem groups attached to the one protein molecule, whereas in other systems such as the heart-muscle succinate-cytochrome c reductase, transfer takes place between the flavin and haem groups of different proteins, succinate dehydro- genase (flavoprotein) and cytochrome b (haemoprotein) respectively (Morton, 1955). The chemical and kinetic evidence presented here further establishes that cytochrome b^ is a flavohaemoprotein and shows that reduction of cytochrome c, ferricyanide and dyes involves different mechanisms. CHEMICAL PROPERTIES OF CRYSTALLINE CYTOCHROME b. AND OF DERIVED PROTEINS Prosthetic Groups Maximum lactate dehydrogenase activity (approx. 7750 /tmoles of ferri- cyanide or 5870 //moles of cytochrome c reduced/hr/mg of enzyme at 20°C) Li E ^ 8-? « o II 1: < li If oo o . TfdNc5N^rsvbrrr4vbvb66.^r5irnfN6-^'* -^ V Alanine Amide Arginine Aspartic acid i-Cystine Glutamic acid Glycine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Proline Serine Threonine Tryptophan Tyrosine Valine c ^ o ii; ll E o 8^ II- < •^E E 1 1? r-i o "^ ^ (N ri 2 is frequently accompanied by the appearance of flavin fluorescence (Appleby and Morton, 1954; Boeri et al., 1955); this is prevented, and activity is protected by exclusion of oxygen, by 506 R. K. Morton, J. McD. Armstrong and C. A. Appleby the presence of ethylenediaminetetra-acetate (EDTA) and by substrate (lactate) (Morton, 1955a; Appleby and Morton, 1959b). Under some conditions autoxidation leads to formation of HoOa, since catalase partially, but not completely, protects against inactivation (Table 2). The formation of H2O2 under these conditions may indicate a slow reaction of oxygen with the flavin group (see Fig. 1), or, more hkely, autoxidation of some dissociated flavin since free riboflavin phosphate may act as an acceptor with cytochrome bo. Oxygen also reacts slowly with the haem group (Fig. 1 and vide infra). Protection by EDTA is due to chelation of metals, such as copper, which Table 2. Protection of enzymic activity of crystalline cytochrome b^ BY catalase Crystalline cytochrome b., (40 /(g/ml) in m sodium lactate at pH 68 was held aerobically with and without added crystalline catalase (about 100 / pH 9) conditions, under which conditions activity with ferri- cyanide and with cytochrome c is greatly decreased (Appleby, 1957; Appleby and Morton, 1959b). The enzyme is considerably protected by lactate, and, at the same time, there is marked reduction in appearance of flavin fluorescence. The ready appearance of flavin fluorescence in the oxidized enzyme, and protection by lactate and EDTA, suggest that the dissociation constant of the flavin group of oxidized cytochrome 02 is very much greater than in the reduced cyto- chrome. The dissociation constant of the oxidized form of the 'old yellow enzyme' is over 100 times greater than that of the reduced form (Nygaard and Theorefl, 1955; Vesthng, 1955). If the flavin group is bound by thiol groups as proposed, oxidation of thiol groups after dissociation of the flavin would prevent recombination of the flavin and haemoprotein. The dissociation of flavin from cytochrome b^ is therefore strictly reversible only under conditions which prevent oxidation of the thiol groups of the apoproteins. The substrate probably protects enzymic activity by maintaining the FMN group in the reduced state and thus lowering the dissociation from the apoprotein. The Haem-ProteUi Link As In other cytochromes, the haem group is strongly bound by the apo- protein, but may be split by treatment with acid-acetone (Armstrong and 508 R. K. Morton, J. McD. Armstrong and C. A. Appleby Morton, 1959). The positions and extinction coefficients of the a, /i and y- bands of reduced cytochrome bo (Fig. 2 and Table 4) are very close to those of pyridine protohaemochrome and suggest that positions 5 and 6 of the iron atom are co-ordinated with strongly basic groups. Since Ehrenberg and Theorell (1955) found that the a-NHa group of glycylglycine and the e-NHg *fr.V - 1 /I "B c 1 '"-I ••'V "i" 300 400 Wavelength, m/f Fig. 2. Absorption spectra of cytochrome b.^. Crystalline cytochrome b.. (de- oxyribonucleoprotein) at pH 6-8: (A) reduced with lactate; (J5) oxidized; (C) oxidized riboflavin phosphate at a concentration equivalent to that of the flavin in cytochrome bo. group of lysine do not readily co-ordinate with the iron of either ferri- or ferro-protoporphyrin, it seems probable that at least one of the groups involved in co-ordination is an imidazole nitrogen of a histidine residue. There are 6 such residues/mole of haem (Table 1). The absorption spectra of lactate-reduced cytochrome ^2 ^t pH 4-7, 6-5 and 100 are shown in Fig. 3. At pH 10-0, the solution was strongly fluor- escent, indicating extensive dissociation of the riboflavin phosphate, but there was no detectable shift in the positions of the a-, /i- or y-bands of the haemo- chrome. However, as compared with the spectrum at pH 6-5, the spectrum at pH 4-7 and 100 shows a decrease in the ratio of fj.^banci/^a-band- The increased absorption in the region 440-500 m/i at pH 10-0 as compared with Properties of Cytochrome by 509 that at pH 6-5 probably reflects the dissociation of the riboflavin phosphate group. Table 4. Principal absorption bands of crystalline cytochrome b^ AND OF ITS derivatives Wavelengths (A) are m//, extinction values (.E) are given for mM concentration of protein-bound haem ; some are calculated values, and may be subject to slight revision. Values are given for neutral pH at room temperature unless otherwise stated. A. Bands in reduced compounds Compound a. /J (?) )' d (?) Crystalline* cytochrome 6, (A 556-5 528 460 423 330 265 (£■ 38-8 18-6 232 52-2 199 Nucleotide-free- cytochrome (i 557 528 460 423 330 268 b., 36 25 224 135 Haemoprotein' derivative el 557 528 460 422 IE 38 22 6 200 Transient intermediate* A 567 533 (lactate complex) Nucleotide-free cytochrome I 557-5 (aj) 532-5 (Jij) ftjCat -190°C)^ 552-5 (a,) 525-0 (/?2) B. Bands in oxidized compounds Crystalline* cytochrome fc> (i 560* 530* 460t 413 359 275 10 14 25 137 47-6 198 Nucleotide-free- cytochrome a 560» 530» 460t 412 359 b2 IE Haemoprotein* derivative a 563* 530* 460t 411 359 278 IE ^ 18 140 30 106 * Broad band. t Shoulder only. 1 Crystalline cytochrome b^ of Appleby and Morton (1954, 1959a) (flavohaemoprotein containing deoxyribosepolynucleotide). 2 Nucleotide-free cytochrome b., (Boeri et al., 1955; Armstrong and Morton, 1959) (flavohaemoprotein). 3 Enzymically-inactive haemo protein derivative of cytochrome Ag (Yamashita et al., 1957, 1958). The extinction values given for this compound are relative only and may be subject to revision. * Transient intermediate ('lactate'-cytochrome 63 complex of Appleby and Morton (1959b), see text). s Nucleotide-free cytochrome b., (Boeri el al., 1955) at temperature of liquid air (Linden- mayer and Estabrook, 1958). Table 4 compares the extinction values and positions of the principle absorption bands of reduced crystalline cytochrome ^2' of nucleotide-free cytochrome ^2' and of the flavin-free haemoprotein derived from cytochrome b^. In the fully reduced enzyme, the flavin appears to have Uttle influence on the contribution of the haem group to the absorption spectrum (see also Fig. 2). The high extinction values of the principal absorption bands of cytochrome Zjg ^s compared with those of other cytochromes of group B (see Morton, 1958, Table 5) therefore does not seem to be due to interaction between the flavin and haem prosthetic groups. Moreover, if there were direct interaction between the resonant systems of the fully reduced proto- haem and flavin groups, the positions of the absorption bands of cytochrome 510 R. K. Morton, J, McD. Armstrong and C. A. Appleby Z?2 might be expected to shift as compared with those of pyridine proto- haemochrome. Table 4 shows that this is not so. However, when a freshly-prepared pellet of crystalline cytochrome bo is washed and dissolved in 0-5 m NaCl, containing 0-1 niM EDTA at pH 6-8 and at 0°, and then gently shaken in air, the intense absorption bands of the 250 300 350 400 450 500 550 600 Wovelength, m/^ Fig. 3. Lactate-reduced crystalline (deoxyribonucleoprotein) material at pH 6-5 {A), 4-7 {B) and 100 (C). Curve D is of lactate-reduced nucleotide-free flavohaemoprotein at pH 6-8. reduced enzyme at 556-5 and 527 m/^ slowly fade and are replaced by some what weaker bands at 567 and 533 m,a (Appleby, 1957 ; Appleby and Morton, 1959b). These bands fade completely on further oxidation. Addition of further lactate, or of NaaSgOj, causes immediate re-appearance of the bands of the reduced enzyme, at 556-5 and 527 m/^. The oxidized cytochrome b^ has a typical ferrihaemochrome-type spectrum, with a weak, broad band between 530 and 560 m/t (Fig. 2). Appleby and Morton (1959b) therefore considered that the bands at 567 and 533 m/« were those of the enzyme at an intermediate state of oxidation. If there was formed a stabilized (lavohaemoprotein-lactate complex, in which the resonance pathway is lengthened by interaction of the 77-electron systems of the flavin and haem prosthetic groups, it would be expected that the absorption bands would be at longer wavelengths as compared with those of the fully-reduced flavo- haemoprotein. Our interpretation of the transient appearance of absorption bands at 567 and 533 m/i is that they are indeed the bands of an intermediate flavohaemo- protein-lactate complex, and that they are due to the influence of lactate and Properties of Cytoelirome bo 511 flavin on the resonance system of the haem group. The bands at 567 and 533 mn are unhkely to be due to a semi-quinone form of the flavin group. Such semi-quinone intermediates, observed in some flavoproteins, have a rather weak band at about 560 m// but show no band near 533 m// (see Beinert, 1957) whereas a well defmed two-banded spectrum is observed with the transient form of cytochrome ho. Tliis also suggests that the haem, as well as the flavin group, is functional in the lactate dehydrogenase activity of cytochrome b^. Kinetic studies, as discussed below, provide evidence con- sistent with this interpretation. The Deoxyribose Polynucleotide Conipoiient of Crystalline Cytochrome b.^ and the Nucleotide-Protein Link The crystalline cytochrome b^, of Appleby and Morton (1954, 1959a) has a prominent absorption band at 265 m/* (see Table 4 and Figs. 2 and 3). This is due to a deoxyribose polynucleotide, which is not dialysable, and which is about 6 % of the dry weight of the material (Appleby, 1957 ; Morton, 1955a, 1958; Appleby and Morton, 1960). It contains about 15 nucleotide residues/mole of haem. Re-crystallized cytochrome b^ contains no ribose or adenine + thymine . uracil (Table 1). The molecular ratio of : : — m the poly- guanme + cytosme nucleotide is 2-60 and in yeast deoxyribonucleic acid is 1-79 (Appleby and Morton, 1960; Montague and Morton, 1960). The polynucleotide is a unique and integral component occurring in constant amount in different batches of crystalhne cytochrome b.^.* More recently, Nygaard (1958) has found dialysable and non-dialysable ribonucleotides associated with non-crystalline cytochrome b^. This type of material remains in the mother liquor when cytochrome b^ is crystallized (Appleby and Morton, 1960). The deoxyribose polynucleotide is not essential for lactate dehydrogenase activity of cytochrome b^. It may be split from the enzyme by digestion with deoxyribonuclease ; by precipitation of the protein with ammonium sulphate ; and by electrophoresis at a suitable pH (Appleby and Morton, 1960). The function of this component is not known. Uptake of ^sp-labelled inorganic phosphate during lactate oxidation is not catalysed by cytochrome /?2 (Appleby and Morton, 1960). ENZYMIC PROPERTIES OF CRYSTALLINE CYTOCHROME b^ Substrate Specificity Armstrong and Morton (1959) have shown that crystalline cytochrome b<2_ catalyses reduction of ferricyanide by a number of a-hydroxy-monocarboxylic * Added in proof. The polynucleotide has been obtained as homogeneous material of molecular weight approx. 10,000 (Montague and Morton, 1960). 512 R. K. Morton, J. McD. Armstrong and C. A. Appleby acids (Table 5). Of the compounds tested, lactate is clearly the most effective substrate. Table 5 also shows that a number of other hydroxy acids strongly inhibit oxidation of lactate; D-malate is particularly inhibitory. Neither reduced di- or tri-phosphopyridine nucleotide (Yamashita et al., 1957; Yamanaka et al, 1958) nor malate (Yamanaka et «/., 1958; Yamashita et al, 1958) are substrates of the crystalline enzyme. The results show that there is a high degree of specificity at the substrate binding site. Table 5. Substrate specificity of crystalline cytochrome 6, Relative activities were determined spectrophotometrically by the rate of reduction of potassium ferricyanide at pH 80 in 033 m sodium pyrophospliaie-HCl buffer containing 10^^ m EDTA. Relative inhibition was determined under generally similar conditions with DL-sodium lactate (01 m) as substrate. Where appropriate the DL-compounds were used in all cases, unless otherwise specified, at OT m concentration. Results of Armstrong and Morton (1959). Relative rate (%) Relative Compound of reduction inhibition (%) of potassium of oxidation ferricyanide of lactate Glycollate 3 21 Lactate 100 Malate 95 D-malate* 86 L-malate^ 53 Glycerate 40 L(+)-tartrate» 5 TPNH Mandelate — 72 Hydroxy-malonate 8 a-hydroxy-glutarate' 7 a-hydroxy-n-butyrate 30 45 a-hydroxy-iso-butyrate 32 a-hydroxy-n-caproate 18 65 a-hydroxy-iso-caproate 17 51 a-hydroxy-iso-valerate 76 1 D-malate (commercially prepared material), 001 m con- centration. 2 L-malate (commercially prepared material), 005 M con- centration. ^ L(+)-tartrate and L(+)-a-hydroxy-glutarate, 0-05 M con- centration. Reactivity of Crystalline Cytochrome h^ with Oxygen, Ferricyanide, Cyto- chrome c and Dyes Crystalline cytochrome />2 containing very small amounts of lactate is slowly oxidized aerobically. At pH 7-4 and at 0°C in the presence of 0-5 m NaCl, 0-1 mM EDTA, and 001 m Tris (2-amino-2-hydroxymethylpropane- 1 : 3-diol)-HCl buffer, the rate of oxidation was equivalent to the oxidation Properties of Cytochrome b.. 513 of 80 moles of lactate/hr/mole of haem (Appleby and Morton, 1959b). With prolonged exposure to air, small amounts of HoOo may be formed {vide supra). With 0-5 M lactate and niM EDTA, cytochrome b^ is only slowly oxidized by air. Under these conditions, the oxygen probably reacts slowly with the haem group rather than with the flavin group, so that HgOa is not formed (see Fig. 1). However, by contrast, Yamashita et al. (1958) have found that the flavin- free haemoprotein derivative of cytochrome b^ 'has a remarkable aut- oxidizability'. Other typical cytochromes of group B, such as cytochrome b^, also are fairly rapidly autoxidizable (Strittmatter and Velick, 1956; Morton, 1958). If the crystalline haemoprotein derived from cytochrome b2 by the procedure of Yamashita et al. (1957) is not seriously denatured, it would appear that the presence of the flavin group modifies the reactivity of oxygen with the haem group of intact cytochrome b^. This is further evidence in support of the close juxtaposition of the two prosthetic groups on the one apoprotein in the intact cytochrome b.^. As compared with the rates with oxygen, the cytochrome /72-catalysed reduction of ferricyanide, cytochrome c and dyes is very much greater. The optimum pH values for reduction of these acceptors vary somewhat accord- ing to the purity of the enzyme (Appleby and Morton, 1959b). For crystafline cytochrome b2, the optimum pH values are: ferricyanide, 8-0; cytochrome c, 7-0; methylene blue, 6-8. The relative rates of reduction of ferricyanide, heart-muscle cytochrome c and methylene blue are substantially the same for crystalline cytoclii-ome 62 as for an extract of dried yeast (Table 6). This shows that the reactivity of Table 6. Comparative lactate dehydrogenase actfvities of a crude extract OF bakers' yeast and of crystalline cytochrome 60 WITH different acceptors For the crude extract, dried yeast was extracted with 015 m NaCl at 38'C for 1 hr and then centrifuged at 10* X g for 1 hr at 5^C. The clear supernatant, containing 2-6 mg of protein N/ml, was used. Activities were determined under comparable conditions: these are sub-optimal in respect of cyto- chrome c concentration and pH (see Appleby and Morton, 1959a). Activities are expressed as units/ mg Cumoles of acceptor rcduced/hr/mg of protein at 18-20°C). Results of Appleby and Morton (1959a). Concen- tration (niM) pH Lactate dehydrogenase activities Hydrogen acceptor Yeast extract Crystalline cytochrome b.^ Units/mg Relative activity Units/mg Relative activity Ferricyanide Cytochrome c Methylene blue 0-5 002 008 8-0-8-4 60-7-0 5-2 36 6-6 2-9 100 18 8 7750 1730 1030 100 22 13 514 R. K. Morton, J. McD. Armstrong and C. A. Appleby the enzyme with these acceptors is not changed during the purification pro- cedure, and the enzyme is unhkely to have been modified during isolation. However, prolonged storage of cytoclirome b^, under aerobic conditions, even in the presence of EDTA, causes partial or complete dissociation of the flavin group and a corresponding decline in activity with ferricyanide (Apple- by, 1957; Appleby and Morton, 1959b). After this type of modification, some of the haem of the preparation is only slowly reducible by lactate (Table 7). Table 7. Effect of aerobic storage on the enzymic ACTIVITY OF cytochrome b.. Crystalline cytochrome li< was suspended in water and held aerobically at CC. Portions of the suspension were diluted into 0-5 M sodium lactate containing 10~^ m EDTA at pH 6-8 and at 0°C at the times indicated. Activity with ferricyanide was measured immediately. The position and extinction of the Soret band was measured after 10 min at 0°. (From Appleby and Morton. 1959b.) Storage time Relative activity vk/ith ferricyanide (7o) Position of Soret band (Amax in m/0 Relative amount of total haem reduced by lactate (7o) 10 min 18 hr 100 55 21 424 423-5 419 100 87 43 The considerable autoxidizability of the flavin-free derivative of cyto- chrome ^2) and the slow reduction of inactive cytochrome b^ by active enzyme (Appleby and Morton, 1954; Yamashita et al., 1958), are complicating features in dealing with mixtures of inactive and active cytochrome b.^. Stepwise modification of crystalline cytochrome bo, by (a) removal of the polynucleotide, and (b) subsequent dissociation of the flavin, is being carried out and the physicochemical, chemical and enzymic properties of the derivatives of the crystalline enzyme are being studied. However, intact crystalline cytochrome b^ has been used throughout the kinetic studies described below. KINETICS OF LACTATE OXIDATION CATALYSED BY CYTOCHROME b. Differences between the mechanisms of reduction of ferricyanide and of cytochrome c by cytochrome bo appeared likely when it was found (Boeri et al., 1956; Appleby, 1957; Appleby and Morton, 1959b) that reduction of ferricyanide was apparently of zero order, and of cytochrome c of first order. Further kinetic studies by Armstrong and Morton (1959) were carried out with a double-beam recording spectrophotometer (Optica CF 4). Extinctions (E) were measured at 420 m/^ (ferricyanide), 550 m^ (cytochrome c) or 600 m/j, Properties of Cytoclironie b._. 515 (2:6-dichlorophenoIindophenol, DCIP), with 003 M sodium pyrophosphate - HCl buffer at pH 8-0, 10^ m EDTA, 0-1 m DL-litliium lactate, and crystalline cytochrome b^, (1 to 2 x 10~^ m). With ferricyanide, plots of E versus time (T) were non-hnear, except for short time periods (up to 2 min) at high ferricyanide concentrations (> 1-4 mM). The results conform approximately with those expected for a first-order reaction. Essentially similar behaviour was observed with heart-muscle cytochrome c as acceptor. With these acceptors, the first-order rate constant {k) was estimated and different K^^ values for the acceptors were estimated from plots of 1/A:S against 1/S. With DCIP, however, the reaction was substantially of zero order over the range 0-22-0022 mM dye. Table 8 summarizes the apparent kinetic para- meters for the various acceptors. In Table 8, the values given are called 'apparent kinetic parameters', since the concentration of lactate was not varied, and there is a possibility that the meaning of K,^ and V given in Table 8 do not have the usual significance accorded to them. The true kinetic parameters for the reaction with heart- muscle cytochrome c, as calculated from the studies reported below (mechan- ism ///) are given in Table 9. These values are not so very difi'erent from the apparent K^ and apparent V as given in Table 8. Table 8. Values of apparent Km and apparent V* for reaction of CYTOCHROME b.^ WITH VARIOUS HYDROGEN ACCEPTORS AT pH S'O AND 30°C Results of Armstrong and Morton (1959); for experimental conditions, see text. Acceptor Concentration range for acceptor (mM) V (/2 with different acceptors has been obtained from kinetic studies in which the concentration of lactate (S) was varied at several different concentrations of hydrogen acceptor (A), in order to differentiate between several possible enzymic mechanisms. Oxidations catalysed by flavin-linked enzymes may conform to any one of at least three different types of mechanisms. Using a simplified notation (in place of the detailed notation shown in the Appendix (p. 521)), these mechanisms and the corresponding steady state rate equations are as follows. /. Oxidation of a binary complex by acceptor (cf. Chance, 1953). E + S ^ ES ES-f Aox--E-f P + Ared (where Aox and Areci represent the oxidized and reduced forms of the acceptor respectively; other symbols follow usual conventions). Here K„, ' + (1) Both V and K^ are functions of the acceptor concentration [A], and have no theoretical finite limit; —^ is not generally constant (see Appendix). //. Formation of (a) two or more binary complexes or (b) a ternary complex (see Alberty, 1956). (a) or (b) E-I-S^ES E + S^ES ES + A ^ EP + H2A ES + A ^ ESA EP ^ E + P ESA ^ E + P + HoA Properties of Cytochrome bg 517 K^n + 1 + K. Ka^ K, y.= [A] [A][S] [S] 1 + [A] and K,„ = [A] 1 + Ka [A] These mechanisms cannot be distinguished between by steady state kinetics. V^ and K„,^ are the values obtained from Lineweaver-Burk plots at any acceptor concentration [A]^,, and approach limit values of V and K^ as [A]„ approaches infinity. The slope of the Lineweaver-Burk plots, K,„JV^ depends on [A]j.. It will be noted that V^ is a hyperbolic function of [AJ^., while K^^ contains terms which are hyperbolic and inverse functions of [A]^. III. Formation of an intermediate reduced form of enzyme (see Alberty, 1956). E + S ^ ES ES ^ E + P E + Aox -^ E + Ared A / /" ^o. i>. / / ^ ^ / y^O, '/[A], Fig. 4. Types of plots obtained for various mechanisms of flavin-linked dehydro- genases (see text). Plots A and B. Plots of \\v against 1/[S] for varying acceptor concentrations [A]. Curves (oi, a^ and a^ are shown for conditions [A]i < [AJa < [AJs. Plot A is for mechanisms / and //. Plot B is for mechanism ///. Plots C and D. Plots of \\K,,,^ or \\V^ against 1/[AL, where [AL is any acceptor concentration. Plot C is for mechanism /. Plot D is for mechanism ///. 518 R. K. Morton, J. McD. Armstrong and C. A. Appleby Here, + [A] "^ [S] V^ = 1 + K and K,„ = K,. 1 + K [AL [A], In this case, K.^JV^ is constant, and plots of 1/K^ or \jK„^^ against 1/[A]^ are linear, with intercepts on the vertical axis of IjV or 1/A^„j, and on the hori- zontal axis of IjK. From the results presented by Slater (1955), it appears that most flavo- proteins conform to either mechanism / or //. The few exceptions (such as 25 I / / 20 ■ // ^ 15 / / ° 10 ' / / n 25 20 1'^ 10 - 5 - , 1 1 1 1/[S] '/[A], Fig. 5. Lactate-ferricyanide reductase activity of about 2 x 10^* m crystalline cytochrome b^__ in 0-03 M sodium pyrophosphate-HCl buffer at pH 80 and 30°C. Units are as follows: v and K,., Af'joom/f/min ; [S], L(+)-lactate, mmoles/1.; [AJa;, potassium ferricyanide, mmoles/1. Plot I. Curves for 0255 mw ferricyanide ( — O — ) and 0-720 mM ferricyanide ( — □ — ). Compare Fig. 4, Plots A and B. Plot II. Curves for \IK,„^ (— O— ) and l/K, (— □— ) against 1/[A]:,. (The lowest value oillVy. was not considered in fitting the curve, as the rate was clearly inhibited at high ferricyanide concentrations). Compare Fig. 4, Plots C and D. notatin and DPNH-cytochrome c-reductase of heart muscle) may be ex- plained in terms of the relative magnitudes of the rate constants involved in V.^ and K^ so that the plot of K^ against Vy. passes through the origin (see Appendix). Figure 4 shows how the three mechanisms may be differen- tiated. As shown in Fig. 5, results obtained for the lactate-ferricyanide reductase activity of cytochrome h.,^ appear to conform to mechanism / or //, although the latter seems more probable. However, as shown by Fig. 6, the lactate- cytochrome c reductase activity clearly conforms with mechanism ///. From the considerations presented above, we interpret the results in the following way. Properties of Cytochrome b^ 519 The enzyme-lactate complex is one in which the greatest part of the two electrons of the lactate are donated to the flavin moiety. The oxidation of this complex by ferricyanide takes place through a ternary complex. This is consistent with the general behaviour observed for flavoprotein enzymes. The reaction with cytochrome c requires an intermediate reduced form of the enzyme. It is postulated that the protohaem moiety of cytochrome b^ reacts at the same site as the ferricyanide, oxidizing the lactatc-flavin complex, Fig. 6. Lactate-cytochrome c reductase activity about 2 x 10"^ m crystalline cytochrome b^ in 0-03 m sodium pyrophosphate-HCl buflFer at pH 8-0 and 30°C. Heart-muscle cytochrome c purified by resin chromatography was used. Units are as follows: v and K„ AE550 m///min ; [S], L(+)-lactate, mmoles/1.; [A].r, heart-muscle cytochrome c, mmoles/1. Plot I. Curves for 1 5 /m cytochrome c ( — O — ) and 82 /, 539. Appleby, C. A. & Morton, R. K. (1960). Biochem. J. 75, 258. Appleby, C. A., Morton, R. K. & Simmonds, D. H. (I960). Biochem. J. 75, 72. Armstrong, J. McD. & Morton, R. K. (1959). Unpublished results. Armstrong, J. McD., Coates, J. H. & Morton, R. K. (1960). Nature, Loud. 186, 1033. Bach, S. J., Dixon, M. & Keilin, D. (1942). Nature, Loud. 149, 21. Bach, S. J., Dixon, M. & Zerfas, L. G. (1946). Biochem. J. 40, 229. Barron, E. S. G. (1951). Advanc. Enzymol. 11, 201. Beinert, H. (1957). J. biol. Chem. 225, 465. Boeri, E. & CUTOLO, E. (1958). IVth Int. Congr. Biochem., Vienna. Abstracts of Com- munications, No. 5-36, p. 60. Boeri, E., Cutolo, E., Luzzati, M. & Tosi, L. (1955). Arch. Biochem. Biophys. 56, 487. Boeri, E. & Tosi, L. (1956). Arch. Biochem. Biophys. 60, 463. Chance, B. (1953). Techniques of Organic Chemistry. Vol. 8, p. 627. Ed. Friess, S. L. & Weisberger, A. New Yorlc: Interscience Publishers Inc. Chance, B., Klingenberg, M. & Boeri, E. (1956). Fed. Proc. 15, 231. Ehrenberg, a. & Theorell, H. (1955). Acta chem. Scand. 9, 1193. Green, D. E. & Beinert, H. (1955). Annu. Rev. Biochem. 24, 1. HoRio, T., Yamashita, J. & Okunuki, K. (1959). Biochim. biophys. Acta 32, 593. Lindenmayer, a. & EsTABROOK, R. W. (1958). Arch. Biochem. Biophys. 78, 66. Mahler, H. (1956). Advanc. Enzymol. 17, 233. Mahler, H. & Green, D. E. (1954). Science 120, 7. Montague, M. D. & Morton, R. K. (1960). Nature, Land. 187, 916. Morton, R. K. (1950). Nature, Lond. 166, 1092. Morton, R. K. (1955a). Society of Biological Chemists, India: Silver Jubilee Souvenir, p. 177. Morton, R. K. (1955b). In Methods of Enzymology, Vol. 1, p. 25. Ed. Colowick, S. P. and Kaplan, N. O. Academic Press Inc., New York. Morton, R. K. (1958). Rev. pure appl. Chem. 8, 161. Nygaard, a. p. (1958). Biochim. biophys. Acta 30, 450. Nygaard, a. p. (1959). Biochim. biophys. Acta 33, 517. Nygaard, A. P. & Theorell, H. (1955). Acta chem. Scand. 9, 1587. Slater, E. C. (1955). Disc. Faraday Soc. 20, 231. Strittmatter, p. & Velick, S. F. (1956). /. biol. Chem. 221, 253. Vestling, C. S. (1955). Acta chem. Scand. 9, 1600. Yamanaka, T., Horio, T. & Okunuki, K. (1958). /. Biochem. Tokyo 45, 291. YAMASiiiTA, J., Higashi, T., Yamanaka, T., Nozaki, M., Mizushima, H., Matsubara, H., Horio, T. & Okunuki, K. (1957). Nature, Lond. 179, 959. Yamashita, J., Horio, T. & Okunuki, K. (1958). /. Biochem. Tokyo 45, 707. CONDITIONS FOR THE AUTOXIDATION OF FLAVOCYTOCHROME h. By E. BoERi* AND M. RiPPAf Institute of Human Physiology, University of Ferrara, Fetrara INTRODUCTION FLAVOCYTOCHROME b^ is the L^+)-lactate dehydrogenase of aerobic yeast. Several acceptors can be used for the oxidation of lactate by this enzyme ; most of these acceptors are artificial. Ferricytochrome c (which is also present in aerobic yeast) is probably the physiological acceptor. Oxygen is also an acceptor, the flavocytochrome Z?2 having been described as autoxidiz- able by Appleby and Morton (1954), by Boeri and Tosi (1956) and as 'very sHghtly autoxidizable' by Morton (1958). It is the purpose of the present investigation to gain information on how efficient an electron acceptor oxygen can be for flavocytochrome b^,. This information can tell us how likely is a direct by-pass of the usual cytochrome route. Furthermore, a study of the conditions needed for the autoxidation of a cytochrome can give useful general information, as the reasons making certain cytochromes (like cytochrome a^ and b^ readily autoxidizable and others (Uke cytochrome c) only autoxidizable under extreme conditions are still poorly understood. METHODS Extraction of Flavocytochrome h^ Compressed aerobic bakers' yeast was obtained from the factory and dried. Lots of 3 kg of dried yeast were used each time. The yeast was ground, suspended in 3 x its weight 0- 1 m K2HPO4, blended in a Waring blendor, and autolyzed for 45 min at 37°C. After centrifugation, the supernatant was brought to pH5-6 with 0-5 m lactic acid and heated for 15 min at 52°C. After cooling and centrifugation, the supernatant was diluted with an equal amount of distilled water and brought to pH 5-4 with lactic acid. Calcium phosphate gel (45 mg dry weight/ml) was added in a ratio 1 : 6 and the mixture was slowly agitated for 30 min. The centrifuged gel was treated with a solution which was 01 m in K2HPO4, 002 m in sodium lactate and 20% * Professor Enzo Boeri died in October, 1960 (Editors). t Investigator of the Comitato Nazionale per le Ricerche Nucleari, Divisione Biologica Group 3 C. 524 Aiitoxidalion of Flaiocytoc/ironie b., 525 saturated with ammonium sulphate. This solution caused the clulion of the enzyme. The eluted material was treated with solid ammonium sulphate to a final saturation of 90% and kept at 4°C for at least overnight. The precipitate was centrifuged and washed with 60% saturated ammonium sulphate, and finally dialyzed for 3 hr in the cold against 0-05 m sodium lactate under the exclusion of the air. The solution was then treated with cold acetone, which was slowly added under agitation until the cytochrome spectrum disappeared from the supernatant. The excess of acetone was removed by centrifugation and vacuum suction. The precipitate was sus- pended in 0-01 M lactate and dialysed as before. Insoluble material was discarded. The enzyme solution was brought over a column of DEAE- cellulose according to Nygaard (1958) and purified according to the same author. The eluted material was finally adsorbed onto and eluted from calcium phosphate gel as before, precipitated with 70 % saturation of am- monium sulphate and dialysed (as before) to remove the salt. In summary, this procedure contains steps from the purification according to Bach, Dixon and Zerfas (1946), Appleby and Morton (1954, 1959) and Nygaard (1958). The yield is about 0-1 micromole flavocytoclirome bjkg of dried yeast. We were unable to obtain the crystalline preparation described by Appleby and Morton (1954, 1959; but see Note, p. 533). Our best preparation had a content of one flavin monophosphate and one haematin residue for each 140,000 g of protein. The turnover number was 16,000 moles ferricyanide reduced by one mole of enzyme/min in the presence of excess lactate. This corresponds to 6,900 Morton units/mg of protein. Other Techniques Oxygen consumption was measured in a Warburg manometric apparatus at 37°C. L(4-)-Lactate determinations were performed with an enzymic method, by measuring the amount of ferricyanide reduced in the presence of excess flavocytochrome b^. The lactate used was 90% in the l(-|-) form and 10% in the d(— ) form. Pyruvate determinations were performed according to the method of Friedmann and Haugen (1943). Catalase was extracted from horse liver according to Sarkar and Sumner (1955). RESULTS Effect of the Enzyme Concentration Figure 1 shows the effect of the concentration of flavocytochrome b^ on the oxygen consumption for the oxidation of L(+)-lactate as catalysed by this enzyme. The effect is linear. 526 E. BOERI AND M. RiPPA Effect of the Concentration of Lactate L(+)-Lactate is the specific substrate. There was no oxygen consumption without addition of lactate. Equally, no oxygen consumption was found in the absence of the enzyme. It is known that when ferricytochrome c or / o o / 5 10 15 20 /2 m moles Fig. 1. The eflect of the concentration of flavocytochrome b.y on the oxidation of lactate by oxygen. Abscissa: m/mioles of enzyme in the Warburg vessel. Ordinate: //I. of oxygen consumed in 15 min at 37 C, 120 strokes/min. Phosphate buffer, 001 m, pH 6. Lactate concentration 0002 m. Enzyme added at time zero from the side arm. Gas phase: oxygen. ferricyanide is used as the oxidant, the velocity increases with increasing lactate concentration up to a saturation value, these curves allowing cal- culation of the MichaeUs constant. When instead oxygen was the oxidant, a clear inhibiting effect of the substrate was seen, as it appears from Fig. 2. / \ / -Ok. logdactnte) Fig. 2. The effect of the concentration of L(+)-lactate. Abscissa: logarithm of themolarity of lactate. Ordinate: oxygen consumption in 25 min. Enzyme added: 3-75 m/«moles. Total volume: 31 ml. Other conditions as in Fig. 1. Aiitoxidutiun of Flavocytoclnonie b^ 527 Effect of the Oxygen Pressure Figure 3 shows the effect of the oxygen pressure on the reaction rate. It appears that the reaction rate is larger when the oxygen pressure is increased. Asymptote 7 /570 / /■ ,428 // 7 / 7/ / /285 / • 1 / 1/// <> -143 Ify / / / f / / / / y / 1 Ko.5 = 570mmHq Fig, 2 4 6 8 10 l/pxIQ-" 4. Lineweaver and Burk plot of the ex- FiG. 3. The effect of the oxygen pressure on the rate of the reaction. Abscissa: perimental data of Fig. 3. Abscissa: the re- time in min. Ordinate: oxygen con- ciprocal of the oxygen pressure (in mm Hg). sumption. The number near each tracing Ordinate: the reciprocal of the oxygen con- shows the oxygen pressure in mm Hg. sumption (//I. in 15 min). Enzyme added: 17m/ 2CH3COCOO- + 2H2O (2) Table 1. Oxidation of lactate by cytochrome b.^ with oxygen Determination of the stoichiometry of the reaction. Reaction time: 60 min. Enzyme added: 30 m/ the prep- aration has a value of 260, which proves rather low in comparison with other oxidases. Lactate, which is the substrate, inliibits the autoxidation. This means that there is no oxygen consumption without lactate, but the oxygen consumption 532 E. BOERI AND M, RiPPA is blocked by increasing the lactate concentration. As metal ions increase the oxygen consumption, the obvious explanation seems that lactate binds the metal ions and therefore inhibits the autoxidation. Figures 9 and 10 show that this explanation is correct. Among metals tested, iron was found the most active catalyser. Less active were cobalt (C0CI2), manganese (MnClg) and molybdenum (Na2Mo04). Copper (CUSO4) strongly inhibited the autoxidation. The effects of o-phenanthroline were difficult to interpret. It was expected that o-phenanthrohne, as a complexing agent for bivalent iron, would remove the effect of this metal. However, it was found that when o-phenan- throhne and ferrous ions were added, the effect depended on the ratio of the two reagents. With a ratio of 1:1, the catalytic effect on the autoxida- tion is even higher than with ferrous ions alone. It is known that o-phenan- throline is a tridentate chelating agent and binds ferrous ions up to formation of the complex Fe"-(o-phenanthroline)3. When this complex was tested, it was still found to be catalytically active on the autoxidation rate, although less active than ferrous iron alone. Actually, this fact appears less strange when we think that a similar compound, ferricyanide, which also has six valences of iron blocked in a diamagnetic compound, is very accessible to flavocytochrome b^, being the acceptor of choice for testing its activity (Appleby and Morton, 1954). Finally, it must be remembered that high amounts of non-haematin iron accompany the enzyme during its purification, only the last step removing the last traces of it. This seems the reason why the unpurified enzyme appears much more autoxidizable than purified preparations. Our present prepara- tions have only traces of non-haematin iron. Factors which favour the autoxidation (high oxygen and metal and low lactate concentrations) tend to inactivate the enzyme. We have computed from activity analysis that the enzyme has lost about half its activity when it has cycled 750 times with oxygen as the acceptor. The decay is infinitely less when ferricytochrome or ferricyanide are the acceptors. According to Morton (1958), the fact that removal of heavy metals and oxygen protects the enzyme is in favour of a bond between — SH groups of certain of the cysteine residues of the protein and the imino group of the riboflavine phosphate. The present results are in favour of such an explanation. The effect of ionic strength should be compared to the effect shown when other oxidants are used. When ferricyanide is the oxidant, there is no effect due to the ionic strength. When cytochrome c is the oxidant, high ionic strength inhibits the reaction. With oxygen, instead, there is an optimal ionic strength. In conclusion, as noted by Morton (1958), flavocytochrome b^ is a very slightly autoxidizable substance. The complete inhibition of autoxidation Autoxidation of Flavocytochrome b.> 533 exerted by lactate and the effect of metals place flavocytochrome b^ among substances like cysteine and ascorbic acid, which are only autoxidizable in the presence of metals. SUMMARY The oxidation of lactate to pyruvate, catalysed by flavocytochrome h^, was studied using oxygen as the electron acceptor. The effects of the concentra- tion of enzyme, lactate, oxygen, hydrogen ions, and of ionic strength were studied. Under optimal conditions, 27 moles of oxygen were reduced/mole of enzyme/min. One molecule of oxygen was consumed for each two molecules of lactate dehydrogenated to pyruvate. Ferrous and ferric ions actively catalyse the autoxidation of flavocyto- Since preparation of this paper, crystalline flavocytochrome b^ has been obtained according to the procedure described by Appleby and Morton (1954, 1959). A ckno wledgemen ts These investigations were supported by the Rockefeller Foundation and the Comitato Nazionale per le Ricerche Nucleari. We thank Mr. Angelo Trussardi, Director of the Societa Anonima Distil- lerie Italiane, Ferrara, for generous gifts of yeast. REFERENCES Appleby, C. A. & Morton, R. K. (1954). Nature, Lond. 173, 749. Appleby, C. A. & Morton, R. K. (1959). Biochem. J. 71, 492. Bach, S. J., Dlxon, M. & Zerfas, L. G. (1946). Biochem. J. 40, 229. BoERi, E. & Tosi, L. (1956). Arch. Biochem. Biophys. 60, 463. Friedmann, E. & Haugen, G. E. (1943). /. biol. Chem. Ul, 415. Morton, R. K. (1958). Revs, pure appl. Chem. 8, 161. Nygaard, a. (1958). Biochim. biophys. Acta 30, 450. Sarkar, N. K. & Sumner, J. B. (1955). Methods in Enzymology, Vol. 2, p. 777. Ed. Colowick S. P. & Kaplan, N. O. Academic Press Inc., New York, KINETIC STUDIES ON THE ACTION OF YEAST LACTATE DEHYDROGENASE By H. Hasegawa and Y. Ogura Department of Biophysics and Biochemistry, University of Tokyo Since Bernheim (1938) studied the properties of lactate dehydrogenase using a crude cell free extract obtained from bakers' yeast, many investigators have worked on the properties of this enzyme. Using an enzyme preparation of sufficient purity, Dixon and his co-workers (Dixon and Zerfas, 1939; Bach, Dixon and Zerfas, 1942, 1946) observed that cytochrome b^ was always present in the enzyme system. At that time, however, they could not decide whether or not the protohaem present was a prosthetic group of lactate dehydrogenase itself. Five years ago, Appleby and Morton (1954, 1959) succeeded in crystallizing this enzyme and established that the enzyme contains two different prosthetic groups, namely protohaem and flavin mono- nucleotide. The fact that the flavin mononucleotide forms an essential part of the enzyme itself, was also ascertained by Boeri and his co-workers (Boeri, Cutolo, Luzzati and Tosi, 1955; Boeri and Tosi, 1956). Based on their further observations, they inferred that in addition to the two different prosthetic groups mentioned above, eight iron ions bound to the enzyme protein seemed to participate as the essential factors in the enzymic action. The investigation to be reported here was undertaken to obtain further information concerning the mechanism of lactate dehydrogenase. Our main interest was to elucidate in what manner the protohaem and flavin mono- nucleotide behave as electron carriers in the mechanism of the enzymic action. EXPERIMENTAL Bakers' yeast supplied from the Oriental Yeast Co. was used as the source of the enzyme. The procedures of enzyme purification were almost the same as those described by Bach et al. (1946). After the cells were plasmolysed by the addition of ethyl acetate, they were subjected to extraction with phosphate buffer. The purification was performed by selective adsorption with calcium phosphate gel followed by fractional precipitation with ammonium sulphate. The enzyme samples used throughout these experiments were spectrophoto- metrically free from cytochrome c. In the kinetic experiments, the rates of reduction of dyes were measured under anaerobic conditions with a photoelectric colorimeter using Thunberg 534 Actio/! of Yeast Lactate Dehydrogenase 535 tubes, whereas those of cytochrome c and ferricyanide were determined spectrophotometrically. Prior to the experiments, the concentrations of dyes were determined by careful titration with a standardized dithionite solution under anaerobic condition. In most of the experiments racemic lactate was used as the substrate. The concentrations of substrate described in the data were those of L-lactate contained in the reaction mixtures, since it has been confirmed that D-lactate did not show any effect on the enzymic reaction, as mentioned by Boeri et al. (1955). RESULTS AND DISCUSSION Properties of the Enzyme The spectrum of the enzyme, reduced by the addition of lactate, showed maxima at 557, 528 and 423 m//; the Soret maximum of the oxidized form was at 412 m/<. These results were in good agreement with those reported by other authors (Appleby and Morton, 1954, 1959; Boeri et al, 1955). The spectrum of flavin mononucleotide attached to the enzyme protein was not observed by direct spectrophotometry, since it was largely covered by that of cytochrome b^- Using a tonometer-type chamber reported by Nakamura (1958), we could measure the spectrum of the former compound by the following procedures. After the air in the chamber was evacuated and replaced by nitrogen gas, a small amount of lactate (final concentration about 50 /m) was added to the buffered enzyme solution placed in the main cham- ber. The complete reduction of cytochrome bo was checked spectrophoto- metrically, and then the enzyme solution was titrated carefully with an oxygen-free methylene blue solution. When the blue colour of the oxidized form of the dye appeared shghtly, the titration was stopped and the mixture was allowed to stand for several minutes. Figure 1 shows the difference spectrum between the enzyme solution thus treated and the reduced one (in which both the protohaem and flavin mononucleotide had been completely reduced). The spectrum, which showed two maxima at 385 and 445 m/v, seemed to be due to the oxidized form of flavin mononucleotide. Conceivably, the flavin mononucleotide moiety of the enzyme was oxidized by the treat- ment described above, while the protohaem moiety remained in a reduced state. Assuming that the millimolar extinction coefficient of flavin mono- nucleotide attached to the enzyme protein was the same as that of the free molecule, its concentration in the mixture was estimated to be 16-4 /iM from the value of eox — Cred = 9-0 cm~i mM~i. Since the concentration of cyto- chrome Z)2 was found to be 18-6 /^m, it was deduced that the molecular ratio of flavin mononucleotide to protohaem present in one molecule of enzyme is 1 : 1 as reported by Appleby and Morton (1954). The effect of riboflavin upon the enzymic reaction was investigated as 536 H. Hasegawa and Y. Ogura follows. After the enzyme solutions were mixed with various concentrations of riboflavin, the rate of reduction of thionine was measured (by the extinc- tion at 660 m/i) in the presence of an excess of substrate. As may be seen from the results presented in Fig. 2, the inhibition caused by riboflavin increased with time as the reaction proceeded. It is worth mentioning that no inhibitory eff"ects of riboflavin could be observed when the rate of reduction of cytochrome b^ was measured under similar conditions. The data presented in Fig. 2 seem to suggest that the flavin mononucleotide bound to the enzyme 0-2 o 400 450 Wavelength, m// Fig. 1. Difference spectrum of flavin mononucleotide attached to the enzyme protein. protein became mobilized and exchanged with the added riboflavin when the enzyme was put into its function. It was further checked whether or not the flavin mononucleotide attached to the enzyme could be removed on the addition of an excess of riboflavin. After the enzyme solution was mixed with an excess of riboflavin, the mixture was allowed to stand for several hours and then treated with ammonium sulphate solution. The precipitate was washed several times with ammonium sulphate solution. The flavin moiety of the enzyme thus treated was identi- fied as riboflavin by paper chromatography. On adding the substrate to the enzyme thus treated, the protohaem moiety of the enzyme was not reduced ; the reduction could be achieved only by treatment with dithionite. The observation made above indicates that the flavin mononucleotide moiety of the enzyme can be replaced by riboflavin leading to the formation of an inactive enzyme complex, and that the flavin mononucleotide attached to the enzyme acts as the direct acceptor of electrons derived from substrate prior Action of Yeast Lactate Dehydrogenase 537 to the reduction of cytochrome b^ as was first assumed by Appleby and Morton (1954). To elucidate further the situation of electron transfer in the enzymic reaction, spectrophotometric measurement was made of the oxidation- reduction potential of cytochrome b^. An enzyme solution was placed in a 240 sec Fig. 2. Inhibitory effect of riboflavin on the enzymic reaction. Concentrations of riboflavin were 200 //m (curve 1), 100 /2 can be brought about by dyes such as methylene blue or thionine, with oxidation reduction potentials lower than that of cytochrome b^. These data seem to suggest that in the enzymic reaction electron transfer to the dye can occur directly from reduced cytochrome b^, but not from the reduced flavin mononucleotide. In order to interpret all the findings described above, the following set of Action of Yeast Lactate Dehydrogenase Table 2. Further kinetic data for yeast lactate dehydrogenase Values of V/e, K^ and Ai4 obtained by double reciprocal method (see text). Experiments were carried out at 22'C and at pH 7-2. 543 Vie X 10-2 K,„ X 101 ki X 10-« Acceptor (sec-i) (M) (M-i sec-i) Methylene blue 1-5 20 2-7 Thionine 2-7 20 8-6 Toluylene blue 1-5 1-6 3-2 2 :6-dichlorophenolindophenol 20 1-9 4-9 Ferricyanide 1-5 1-5 2-7 Cytochrome c 1-4 1-8 4-3 equations may be proposed as representing the mechanism of lactate dehydro- genase. E + S ^ ES A-,, ES->E' + P A-3 E' ^E" E" + A-^E + AHo (1) (2) (3) (4) where E represents the oxidized form of the enzyme, ES the enzyme-substrate complex, E' the reduced form of the enzyme, in which the flavin mono- nucleotide moiety is reduced, E" another reduced form, in which the proto- haem moiety is reduced, S the substrate, A the hydrogen acceptor, P the product and ki_^ the rate constants. A cknowledgemen t We are indebted to Prof. H. Tamiya for his valuable criticism in this work. This work was supported by a grant from the Ministry of Education. REFERENCES Appleby, C. A. & Morton, R. K. (1954). Nature, Land. 173, 749. Appleby, C. A. & Morton, R. K. (1959). Biochem. J. 71, 492. Bach, S. J., Dixon, M. & Zerfas, L. G. (1942). Nature, Loud. 149, 48. Bach, S. J., Ddcon, M. «& Zerfas, L. G. (1946). Biochem. J. 40, 229. Bernheim, F. (1928). Biochem. J. 11, 1179. BoERi, E., CuTOLO, E., LuzzATi, M. & Tosi, L. (1955). Arch. Biochem. Biophys. 56, 487. BoERi, E. & Tosi, L. (1956). Arch. Biochem. Biophys. 60, 463. Dixon, M. & Zerfas, L. G. (1939). Nature, Lond. 143, 557. LiNEWEAVER, H. & BuRK, D. (1939). J. Amer. chem. Soc. 56, 658. Nakamura, T. (1958). Biochim. biophys. Acta 30, 44. Singer, T. P. & Kearney, E. B. (1957). Methods of Biochemical Analysis, 4, 307. VARIOUS FORMS OF YEAST LACTATE DEHYDROGENASE By A. P. Nygaard Johan Throne Hoist's Institute for Nutrition Research, Blinder n, Oslo Several forms of lactate dehydrogenase have now been isolated from aerobic bakers' yeast (Nygaard, 1958; 1959a, b). The method involved: (1) ex- traction by the stirring of a yeast suspension with micro glass beads, (2) fractionation with acetone, (3) adsorption on calcium phosphate gel, (4) ammonium sulphate fractionation, and (5) separation on an N,N-diethyl- aminoethylcellulose (DEAE-cellulose) column. The different forms differed greatly in their affinity for DEAE-cellulose. In the following, three fractions (lactate dehydrogenase (LDH) I, II and III) will be described; they were eluted in that order from the column. As a reference, the homogeneous and crystalline LDH of Appleby and Morton (1954, 1959) has been used. This enzyme is well defined and well characterized and will be referred to as 'crystalline LDH'. The properties have been described by Appleby and Morton, 1954, 1955, 1959a, b and Appleby, 1957 (see Morton, 1958). COMPOSITION AND SPECTRAL PROPERTIES OF THE ISOLATED ENZYMES Lactate Dehydrogenase I {LDH I) This was eluted with 0-04 m phosphate around pH 6-5. Flavin and haem were eluted as one peak in the ratio 1:2. As in the crystalline enzyme of Appleby and Morton (1954, 1959a, b), the a-band was located at 557 m/z and the Soret band in the reduced and oxidized state was located at 423 m/* and 413 m// respectively. In contrast to the crystalline enzyme, however, the heights of the reduced and oxidized Soret band were about equal. The extinction coefficient was about 100 X 10^ m~i cm^^ compared with 240 X 10' for the crystalline enzyme (Appleby and Morton, 1959a, b). Further- more, all the reduced bands were broader for this fraction. A specific proteolytic attack at the site involved in the binding of the haem may have modified the spectrum. The enzyme gave a normal pyridine protohaemo- chrome spectrum (Nygaard, 1959b). 544 Various Forms of Yeast Lactate Dehydrogenase 545 Lactate Dehydrogenase If (LHD II) This was eluted at pH 5-3 with 0-04 m NaCl. The enzyme was not eluted as a distinct protein peak, as the fraction was contaminated by (partly dialysable) polypeptides and also polynucleotides. Proteolytic and nucleolytic degradation took place during the elution procedure. Flavin and haem were eluted as one distinct peak. £"280/^260 reached a minimum with this peak, thus indicating affinity of the enzyme for the polynucleotides. The purest pre- paration contained one haem/97,000 g of protein, and was thus as pure as the crystalline enzyme. After dialysis, the enzyme still contained 2-3% of polynucleotides. These were all of the ribose type. The crystalline enzyme contains deoxyribonucleotide (Appleby, 1957; see Morton, 1958). Ribo- nuclease and deoxyribonuclease did not affect fraction II, just as they did not inactivate the crystaUine enzyme (Appleby and Morton, 1955, 1960). Between LDH II and III, cytochrome c peroxidase was eluted as a brown zone with one haem/200,000 g of protein. Some preparations of this material underwent inactivation suddenly (in the course of minutes) when held at 0°C. This was frequently accompanied by the formation of a colourless protein precipitate. This precipitate, which contained much of the nucleotides, may be a part of the enzyme protein. One haem/68,000 g of polypeptides remained dissolved. The inactivation was probably due to proteolytic enzymes, which were present even in the best preparations. Crystalline trypsin and chymotrypsin increased the rate of inactivation. Inactivation was associated with the appearance of fluorescence (see Appleby and Morton, 1954). The rate of dissociation of the flavin was increased by sodium chloride in the acidic range (around pH 5), thus sug- gesting that ionic bonds were involved (Theorell and Nygaard, 1954). The dissociation was irreversible. />-Chloromercuribenzoate (10 * m) caused fluorescence to appear immediately. REACTION PROPERTIES Crude Extracts When the acetone precipitate (cf. Nygaard, 1958, 1959a, b) (dissolved in water or in phosphate buffer at pH 7) was incubated at 55-57°C for 5 min. or treated with /7-chloromercuribenzoate (pCMB, 3 x 10^ m), 90% or more of the LDH activity was destroyed. The properties of the remaining enzyme were quite different from those of the original solution. In the first place, the ratio of the reduction of cytochrome c to the reduction of ferricyanide and 2:6-dichlorophenolindophenol was high for the stable enzyme. Secondly, the reaction of the stable enzyme with cytochrome c was much more aftected by changes in the salt concentration. Figure 1 illustrates the reaction properties of a crude preparation which contained 10% of heat-stable enzyme. It is 546 A. P. Nygaard seen from the figure that both the original preparation and the stable enzyme followed zero-order kinetics with cytochrome c as acceptor in the region 10"^ M to 10"* M when the phosphate concentration was very low (ji 0-01). When the buffer concentration was increased to n 0-08, however, the activity CytcxWM — - Fig. 1 . The effect of the concentration of cytochrome c on the rate of oxidation of lactate (dl, 002 m) by a crude preparation of LDH and by the heat- and /7CMB-stable enzyme contained in this preparation; pH 71; 0001 m ethylene- diamine tetra-acetate (EDTA) was added. AAA Original preparation, in phosphate buffer, /< 001 + + + Original preparation in phosphate buffer, /« 008 • • • Heat- and/jCMB-stable enzyme, in phosphate buffer, /t 001 O O O Heat- and/jCMB-stable enzyme, in phosphate buffer, /< 008 Ordinate: arbitrary velocity units; values were calculated from AE at 550 m/<, using £mM = 18-6. of the original enzyme was practically unaltered, whereas the stable enzyme in this buffer followed first-order kinetics with respect to cytochrome c. The ratio of rates of reduction of cytochrome cVferricyanide/2:6-dichlorophenol- indophenol ('relative activities') was 12: 1 : 1 for the stable enzyme compared with 0-8:1:1 for the original preparation. LDH I This reduced cytochrome c, ferricyanide, and indophenol at substantially the same rate. The turnover number of a preparation was 6,000 moles Various Forms of Yeast Lactate Dehydrogenase 547 cytochrome c reduced/mole flavin. The enzyme was very labile and half of the activity decomposed in a few hours at 0°C. LDH II and III From many preparations only minute amounts of LDH I were obtained. In these preparations LDH II constituted the main part of the LDH activity. The amount of LDH III obtained corresponded roughly to the amount of stable enzyme in the crude preparation. LDH II was completely inactivated by incubation at 55-57°C for 3 min (pH 71) and by 10^ M/7CMB (0°C, pH 71). The reduction of cytochrome c by LDH III was unaffected under the same conditions. The reduction of ferricyanide and indophenol, however, was as labile in LDH III as in LDH II. Thus, the relative activities of LDH III could be changed by heat or /?CMB. The relative activities of LDH II could not be altered in this way, and they remained roughly the same also during inactivation due to ageing, freezing, and proteolytic digestion. This is illustrated in Table 1 . Table L Absolute and relattve activities of freshly prepared and of partially inactivated ldh h The activity was determined spectrophotometricaUy at 20°C. The composition of the medium was: phosphate buflfer pH 71, ionic strength 001; versene 0001 m; DL-lactate 0005 m; cytochrome c2 X 10~^M, or ferricyanide to give 10~^M, or 2: 6-dichlorophenolindophenol 5 X 10~*m. Turnover numbers were calculated from Af at 550 m// (cytochrome c, £mM 18-6), 420 m/« (ferri- cyanide, f jpM 1 6), and 600 m/( (2 : 6-dichlorophenolindophenol, £niM 10). Turnover number is defined as the equivalents of acceptor reduced/mole of flavin/min. Turnover numbei (/-type cytochrome' from bakers' yeast. Using the preparation, Okunuki and Yakushiji (1940) proposed a scheme for the terminal electron- transferring system as follows: Substrate -^ Coenzyme ->■ 'Cytochrome f -^ Cytochrome c -- Oxygen t t t Dehydrogenase Diaphorase Cytochrome oxidase system Borei (1945) reported that 'cytochrome f was not extracted according to the method of Yakushiji and Mori. On re-examination, the method, with shght modifications, was found suitable for the extraction of cytochrome c and cytochrome Z)2, but not for the extraction of a typical cytochrome b. After the extraction of both cytochromes, the cytochrome b still remained in the cellular particles. The description by Yakushiji and Mori is therefore regarded by us as the first detection and partial purification of cytochrome ^3- Both cytochromes c and b^ extracted from bakers' yeast have been crystallized (Hagihara, Horio, Yamashita, Nozaki and Okunuki, 1956; Yamashita, Higashi, Yamanaka, Nozaki, Mizushima, Matsubara, Horio and Okunuki, 1957). Appleby and Morton (1954, 1959a, b) succeeded in crystallizing lactate dehydrogenase (Y-LDH) from bakers' yeast. Their crystalline Y-LDH (molecular weight, 80,000) contained riboflavin phosphate, deoxyribo- polynucleotide (Morton, 1955, 1958; Appleby, 1957) and haem iron, but no non-haem iron. Moreover, the enzyme was capable of reducing cyto- chrome c and methylene blue at the expense of lactate in the absence of any coenzyme and showed absorption spectra just like a Zj-type cytochrome. In its lactate-reduced form the absorption peaks of the preparation were at 556 mfi (a-band), 527 m/i (/:/-band) and 423 m/i (y-band), and in its oxidized * Present address: Graduate Department of Biochemistry, Brandeis University, Waltham 54, Mass., U.S.A. t Present address: Department of Agriculture, University of California, Berkeley 4, California, U.S.A. 552 Bakers' Yeast Lactate Dehydrogenase 553 form at 413 m/t (y-band). This enzyme was called cytochrome b,^, by Bach, Dixon and Zerfas (1946) who partially purified the cytochrome simultane- ously with increasing the specific enzymic activity. On the other hand, the cytochrome /?2 crystallized by us (Yamashita et al., 1957) contained one haem iron atom in each 22,000 g protein, but no flavin nor non-haem iron. More- over, it showed no enzymic activities in all tests tried in the presence and absence of flavin (flavinadenine dinucleotide, FAD ; riboflavin phosphate, FMN; and riboflavin, RF). Hereafter in this paper, in order to avoid confusion, the enzymic and non-enzymic cytochromes will be called Y-LDH and cytochrome Z^g, respectively. In addition to the method of Yamashita et al. (1957), by which cytochrome b^ was directly extracted at pH 8-0 from the cells of bakers' yeast previously disrupted by ethyl acetate, it was found that cytochrome b^ could be split from the Y-LDH purified according to the method of Bach et al. (1946) by treating it at high pH. The present paper deals with some studies on relationships between the lactate dehydrogenase moiety and the cytochrome Z^g rnoiety, both of which seemed to be the main moieties of the whole enzyme, Y-LDH. ENZYMIC ACTIVITIES OF YEAST LACTATE DEHYDROGENASE (Y-LDH) The preparation of Y-LDH purified mainly according to the method of Bach et al. (1946) has been found capable of oxidizing various substances other than lactate: a-glycerophosphate (Lehmann, 1938), glycerate (Dickens and Williamson, 1956), malate and reduced triphosphopyridine nucleotide, TPNH (Yamanaka, Horio and Okunuki, 1958), and a-hydroxybutyrate (Yamashita, unpublished). If these substrates are added to the enzyme preparation purified to a specific activity of about 6,000 units (as Q^q at 30°C) Y-LDH immediately exhibits its reduced absorption spectrum to a variable extent, dependent on the substrate. With equal concentrations of the substrates, the extent of reduction is highest with lactate, somewhat lower with a-hydroxybutyrate, and very low with the others. The reduction by malate gradually increases as its concentration rises to the maximum examined (0-5 m). On the other hand, Y-LDH is known to be able to deliver electrons from the substrates to cytochrome c, methylene blue, ferricyanide, phenazine methosulphate, etc. The rates of reduction of these electron acceptors by the substrates vary in proportion to the extent to which Y-LDH itself is reduced by these substrates. Relationship between Y-LDH and Cytochrome b^ During the purification of Y-LDH, the sample is always contaminated with the cytochrome b^ not bound to Y-LDH. If lactate is aerobically added in excess to such an enzyme preparation, all Y-LDH is immediately reduced. 554 T. HoRio, J. Yamashita, T. Yamanaka, M. Nozaki and K. Okunuki but the free cytochrome b.^ is only shghtly reduced because of its high aut- oxidizabiUty (Yamashita, Horio and Okunuki, 1958). Most of the free cytochrome b^ can be removed from tiie enzyme preparation by taking advantage of different affinities of calcium phosphate gel for the two proteins. Reductions of Cytochrome c and Cytochrome bo by Lactate, oi-Hydroxybutyrate and Malate in the Presence of Y-LDH Under anaerobic conditions at pH 6-4, cytochrome c is reduced by Y-LDH 42 times as fast as cytochrome b^. When cytochrome c or cytochrome b^ is used as a final electron acceptor of Y-LDH, reduction of cytochrome c by lactate, a-hydroxybutyrate, and malate shows optimal pH at pH 6-5, 5-5 and 8-0, respectively, while reduction of cytochrome b^ shows optimal pH at about 5, 5, and 8-9, respectively. With lactate and a-hydroxybutyrate, activity-pH curves for the cytochrome c reduction reactions are fairly sharp and roughly symmetrical on both sides of each optimal pH. However, the activity-pH curves for the reduction of cytochrome b^ by lactate and oc- hydroxybutyrate are asymmetrical, showing appreciable activity even at pH 4-0 on the acid side of the pH optimum, but dropping steeply on the alkaline side. In contrast, the activity-pH curve for the reduction of cyto- chrome b^ by malate is nearly constant between pH 8-9, and falls steeply at pH lower than 8. The remarkable differences between the activity-pH curves for the reduction of cytochrome c and cytochrome b^ by the various substrates might be explained by the following pathway: (Y-LDH) Most easily damaged here T Dehydrogenase moiety y^ ^^^ > Cytochrome b^, moiety > Cytochrome c (5) Cytochrome b.^ (not bound) where the box expresses the whole enzyme, Y-LDH, consisting of two parts, the dehydrogenase and the cytochrome b.2, moieties. Reaction (1) expresses a dehydrogenase activity, and reaction (2) an electron transport in the Y-LDH complex. If the externally added cytochrome c is reduced by Y-LDH as fast as is the cytochrome h^ moiety, then reaction (2) proceeds 42 times faster than reaction (4). In this scheme of electron transport, the intimate junction between the dehydrogenase and cytochrome b^ moieties may be broken at the lower and higher pH ranges, and reaction (4) may take the place of reaction (2). Modification of the Enzymic Activity of Y-LDH When Y-LDH is allowed to stand at pH 7-0 and 4-5''C in the absence of the substrates, its enzymic activity gradually decreases. During storage, Bakers'' Yeast Lactate Dehydrogenase 555 the ability to reduce methylene blue, ferricyanide and cytochrome c decreases much faster than does the abihty to reduce phenazine methosulphate. Figure 1 shows a notable difference in the enzymic activities between the original enzyme and the enzyme stored for 72 hr (Horio, Yamasliita and Okunuki, 1959). Such partial modification of Y-LDH can be demonstrated within a much shorter time if the storage is carried out at pH 5-0, at which pH it has been suggested {vide supra) that the electron-transferring pathway 5 10 15 ENZYME CONCENTRATION IN UNITS Fig. 1. Partial modification of bakers' yeast lactate dehydrogenase. The enzymic activities of the original and treated Y-LDH were assayed by two different methods: phenazine methosulphate method (curves, Opu and Mp^), and methylene blue method (curves, Omb and Mmb)- Reduction of methylene blue by lactate was carried out in Thunberg tubes having an oxygen-free nitrogen gas phase at 35°C. The enzymic activities were compared between freshly-prepared Y-LDH (curves, OpM and Omb) and a sample of the Y-LDH stored in the absence of lactate at pH 70, and at 4-5°C, for 72 hr. One unit of Y-LDH (original and treated) was defined as the quantity of the enzyme capable of consuming 20 //I. of oxygen in the presence of phenazine methosulphate and lactate in the first 5 min at 38°C. in Y-LDH might be most easily damaged between the dehydrogenase and cytochrome b^ moiety. If lactate is aerobically added to the modified Y-LDH, the modified enzyme shows the reduced absorption spectrum of cytochrome b.2 only slightly or not at all, even at concentrations at which the original and modified enzymes show the same enzymic activity by the phenazine methosulphate method, and at which the dithionite-reduced absorption spectra of both samples are easily measurable. The modified enzyme has not yet been purified to a state free of cytochrome b^. Protection by Substrates against the Inactivation of Y-LDH by Heat Bach et al. (1946) found that lactate remarkably protects Y-LDH against heat-denaturation, and they adopt this property to their Y-LDH purification procedure. The other substrates can protect against denaturation, although they are inferior to lactate in their protective ability. The protective abilities of the various substrates vary in the same manner as do the rates at which 556 T. HoRio, J. Yamashita, T. Vamanaka, M. Nozaki and K. Okunuki the substrates are oxidized by Y-LDH: lactate, very strong; a-hydroxy- butyrate, a little weaker than lactate; malate, very weak. Throughout the time course of the denaturation of Y-LDH by heat or by lowering the pH in the presence of either lactate of a-hydroxybutyrate, the reactivity of the enzyme toward the two substrates maintains a constant ratio. This is the case even if the activity is assayed with the use of phenazine methosulphate, ferricyanide, methylene blue, or cytochrome c (Yamashita, unpublished). This fact indicates that lactates and a-hydroxybutyrate are dehydrogenated not only by the same enzyme (Y-LDH), but also by the same dehydrogenase moiety (reaction (1)). Since Y-LDH shows much lower affinities for the other substrates as compared with lactate and a-hydroxybutyrate, it has been found difficult to apply tliis method to the dehydrogenation of the other substrates. DISCUSSION Boeri, Cutolo, Luzzati and Tosi (1955) and Boeri and Tosi (1956) have found that their best preparation of Y-LDH contains one FMN residue, one haem group and eight iron atoms not bound to haem for each 230,000 g protein, and they demonstrated the absence of a diphosphopyridine nucleotide-dependence for the dehydrogenase activity. Therefore they proposed the following scheme for electron transport in Y-LDH : Lactate —> FMN — >■ non-haem irons > haem iron — ^ cytochrome c (Y-LDH) The extensive characterization of the crystalline Y-LDH by Appleby and Morton (see Morton, 1955, 1958; Appleby and Morton, 1954, 1959a, b) may require some modifications of the scheme as follows : Lactate —> riboflavin phosphate > haem iron — '-> cytochrome c (Y-LDH) Moreover, the successful crystallization of Y-LDH may support the concept that both dehydrogenation of an organic compound and the transport of the resulting electrons to a cytochrome can be carried out by one enzyme. The linkage between the dehydrogenation of succinate and the reduc- tion of cytochrome h may fit this concept. Singer and co-workers (Singer, Kearney and Bernath, 1956; Singer, Massey and Kearney, 1957) have succeeded in preparing mammalian succinate dehydrogenase in a truly homo- geneous, water-soluble state. Their dehydrogenase contains flavin dinucleo- tide and non-haem iron(s). The purest enzyme can oxidize succinate with the aid of phenazine methosulphate, but not with methylene blue and Bakers' Yeast Lactate Dehydrogenase 557 ferricyanide, while in the crude state the enzyme can reduce all of them. On the other hand, cytochrome b has been purified in an apparently water- soluble state with the aid of detergents and proteinase digestion (Sekuzu and Okunuki, 1956). Attempts to link cytochrome b with the solubilized suc- cinate dehydrogenase have not yet succeeded. Chance (1954) has assumed that the electron transport between the succinate dehydrogenase and cyto- chrome b may be mediated by another flavoprotein, but all attempts at separation of such a flavoprotein have been unsuccessful. Keilin and King (1958) have demonstrated that the water-soluble succinate dehydrogenase purified by the method of Wang, Tsou and Wang (1956) can transfer electrons SUBSTRATES SUBSTRATES ^ ^ /" ^^ DEHYDROGENASE z"*^. DEHYDR0GENAS8?|A Cmcetv fe.,J I MOIETY j\y ^- -^ V^ ^^_ ^CYTOCHROME-C CYTOCHROME-C (A) (B) Fig. 2. Models of bakers' yeast lactate dehydrogenase. Schematic representations of the relationship between bakers' yeast lactate dehydrogenase activity and the reduction of the haem of bakers' yeast lactate dehydrogenase. liberated from succinate to the cytochrome system contained in a mito- chondrial particle preparation which has no succinate dehydrogenase. The similarity in the modification of the enzymic activities between Y-LDH and succinate dehydrogenase might indicate similarities in their electron- transferring systems (compare, however, Morton, 1955). At present, the two models of Y-LDH may be considered to be as shown in Fig. 2, where {A) indicates that Y-LDH is a protein complex which is composed of a cytochrome plus a dehydrogenase containing a bound flavin. In this model the dehydrogenase moiety may be equivalent to the modified Y-LDH or to solubilized succinate dehydrogenase and the cytochrome moiety may be equivalent to crystalhzed cytochrome b^ or to highly purified cytochrome b. Other concepts of an enzymic protein complex may be deduced from studies of phosphorylase a and b (Green and Cori, 1943) and L-amino acid oxidase (Blanchard, Green, Nocito and Ratner, 1944, 1945, 1946). At present, however, there is no evidence to deny the scheme (5), where the dehydrogenase protein itself holds both the 6-type haem and a flavin, except the observation that: Y-LDH and cytochrome b^, show identical a-, /?-, y- and <5-absorption maxima (in wavelength and in ratio). This is in contrast to the behaviour of cytochrome c, for native cytochrome c and cytochrome c digested by proteolytic enzymes are spectrophotometrically different, though the native cytochrome c is not spectrophotometrically distinguished from the cytochrome c modified only in its secondary structure (Nozaki, 558 Discussion Yamanaka, Horio and Okunuki, 1957; Nozaki, Mizushima, Horio and Okunuki, 1958; Mizushima, Nozaki, Horio and Okunuki, 1958; Yamanaka, Mizushima, Nozaki, Horio and Okunuki, 1958). Insofar as cytochrome b^ and cytochrome c are comparable, this behaviour may indicate that the bound cytochrome b^ in the Y-LDH complex and purified cytochrome Z>2 are identical, except perhaps with regard to secondary structure. REFERENCES Appleby, C. A. & Morton, R. K. (1954). Nature, Land. 173, 749. Appleby, C. A. (1957). Ph.D. Thesis: University of Melbourne. Quoted by Morton, R. K. (1958). Rev. pure appl. Cheni. 8, 161. Appleby, C. A. & Morton, R. K. (1959a). Biochem. J. 71, 492. Appleby, C. A. & Morton, R. K. (1959b). Biochem. J. 73, 539. Bach, S. J., Dixon, M. & Zerfas, L. G. (1946). Biochem. J. 40, 229. Blanchard, M., Green, D. E., Nocito, V. & Ratner, S. (1944). J. biol. Chem. 155, 421. Blanchard, M., Green, D. E., Nocito, V. & Ratner, S. (1945). J. biol. Chem. 161, 583. Blanchard, M., Green, D. E., Nocito, V. & Ratner, S. (1946). J. biol. Chem. 163, 173. Boeri, E., Cutolo, E., Luzzati, M. & Tosi, L. (1955). Arch. Biochem. Biophys. 56, 478. Boeri, E. & Tosi, L. (1956). Arch. Biochem. Biophys. 60, 463. Boeri, E. (1945). Ark. Kemi. 20, A, Nr. 8. Chance, B. (1954). The Mechanism of Enzyme Action, p. 389. Ed. McEIroy, W. D. & Glass, B. Johns Hopkins Press, Baltimore. Dickens, F. & Williamson, D. H. (1956). Nature, Lond. 178, 1118. Green, A. A. &. Cori, G. T. (1943). /. biol. Chem. 151, 21. Hagihara, B., Horio, T., Yamashita, J., Nozakj, M. «& Okunuki, K. (1956). Nature, Lond. 178, 629. HoRio, T., Yamashtta, J. & Okunuki, K. (1959). Biochim. biophys. Acta 32, 593. Keilin, D. & King, T. E. (1958). Nature, Lond. 181, 1520. Lehmann, J. (1938). Skand. Arch. Physiol. 80, 237. Mizushima, H., Nozaki, M., Horio, T. & Okunuki, K. (1958). /. Biochem. Tokyo 45, 845. Morton, R. K. (1955). Society of Biological Chemists, India: Silver Jubilee Souvenir, p. 177. Morton, R. K. (1958). Rev. pure appl. Chem. 8, 161. Nozaki, M., Yamanaka, T., Horio, T. & Okunuki, K. (1957). J. Biochem. Tokyo 44, 453. Nozaki, M., Mizushima, H., Horio, T. & Okunuki, K. (1958). J. Biochem. Tokyo 45, 815. Okunuki, K. & Yakushiji, E. (1940). Proc. Imp. Acad. {Japan) 16, 144. Sekuzu, I. & Okunuki, K. (1956). J. Biochem. Tokyo 43, 107. Singer, T. P., Kearney, E. B. & Bernath, P. (1956). /. biol. Chem. 223, 599. Singer, T. P., Massey, V. & Kearney, E. B. (1957). Arch. Biochem. Biophys. 69, 405. Wang, T. Y., Tsou, C. L. & Wang, Y. L. (1956). Scientia Sinica 5, 73. Yakushiji, E. & Mori, T. (1937). Acta Phytochim. Tokyo 10, 113. Yamanaka, T., Horio, T. &. Okunuki, K. (1958). /. Biochem. Tokyo 45, 291. Yamanaka, T., Mizushima, H., Nozaki, M., Horio, T. & Okunuki, K. (1959). /. Biochem. Tokyo 46, 121. Yamashita, J., Higashi, T., Yamanaka, T., Nozaki, M., Mizushima, H., Matsubara, H., Horio, T. & Okunuki, K. (1957). Nature, Lond. 179, 959. Yamashita, J., Horio, T. & Okunuki, K. (1958). /. Biochem. Tokyo 45, 707. DISCUSSION The Problem of Cylochronie h^ Dickens: We now come to the Discussion of these very interesting papers and it has been suggested that, as Chairman, I should summarize briefly the main points which have arisen so far. I am bound to say that I find this a virtually impossible task, but I Cytochrome b, 559 shall try to bring out a few points, which the authors of the papers presented are free to put right if they do not agree. In my view, there are first of all the three groups headed by Morton (Australia), Bocri (Italy) and Nygaard (Norway). The first two of these have no really essential difference in regarding the yeast enzyme as a flavohaemo- protein and, whether they prefer to call it 'cytochrome b^' or 'flavocytochrome 62', these authors are satisfied that they are dealing with a single unit containing proto- haem, flavin and a single enzyme protein, though these compounds can be caused to dissociate under some treatments. There are, however, points of difference between these groups of authors, particularly in regard to the behaviour of the enzyme with different terminal acceptors, and also in regard to substrate specificity. Morton and co-workers find that their well-crystallized cytochrome 60 is almost free from non-haem iron and is only slightly autoxidizable while Boeri and Rippa find that their uncrystallized but highly active material is autoxidizable especially at higher oxygen concentrations, but that this autoxidizability is inhibited by excess lactate, probably by binding o active metal (Fe) ions by substrate. Morton and co-workers find that dissociation of part of the flavin increases autoxidizability of the crystalline material, with a decline in activity towards ferricyanide. The best substrate is L-lactate, though oxidation of some other a-hydroxy acids (but not glycerate) occurs, as reported by Morton and co-workers. Nygaard is evidently dealing with a very different material, or rather three different non-crystalline materials, prepared from Norwegian yeast. By purification including DEAE-cellulose adsorption, he has obtained these fractions with different activities and different heat-stabilities, all able to oxidize lactate, though with differences in reactivity towards different hydrogen acceptors and in optical specificities. Whereas Morton's preparation contained some deoxyribonucleotide, Nygaard's had poly- ribonucleotides, retained even after dialysis. How much these preparations represent breakdown during extraction is not yet clear. The Japanese workers also have made studies of this enzyme. Ogura's kinetic studies establish that the flavin group is involved in the enzymic reaction, supporting the original work of Morton's group. However, Hasegawa and Ogura do not seem to agree entirely with Morton and co-workers on the reaction with ferricyanide. The Japanese work at Osaka demonstrates the alkaline extraction from Japanese yeast of a crystalline haemoprotein having an absorption spectrum resembling the other workers' cytochrome b.^, but devoid of lactate dehydrogenase activity, and having no flavin or non-haem iron. It was also formed by alkaline treatment of a 'yeast lactate dehydrogenase' of the flavohaemoprotein type. Horio and co-workers call this haemoprotein 'cytochrome bo' and suggest that in the intact system a dehydrogenase, containing a bound flavin, is linked to their cytochrome b^, somewhat similarly to the composition of succinate oxidase. The substrate specificity of the partially-purified complex is wider than that observed by Morton and co-workers with their material, and includes glycerate, malate and TPNH. While this enzyme system is of a novel and most interesting type, there is one question that I would like to put to the various contributors : what is the metabolic function of L-lactic dehydrogenase in the intact yeast cell? Lactate is not generally considered to be an important metabolite in yeast, although it is known that added lactate can be freely utilized (Meyerhof, Biochem. Z. 162, 43, 1925). There is obviously a problem here. My second point concerns substrate specificity. This varies in the different pre- parations studied, and the reason for this interests me very much. With the original preparation of Bach, Dixon and Zerfas, we found (Dickens and Williamson, Nature, Loud. 178, 1118, 1956) that at pH 6-7 glycerate was oxidized to hydroxypyruvate at half the rate of lactate ; only the L-isomer was attacked, and a purified yeast lactate dehydrogenase kindly supplied by Boeri gave a similar result. This raises the possi- bility that an a-hydroxyacid other than lactate might be the natural substrate. . Finally I would like to add a correction to a proposed mechanism for enzymic dehydrogenation of lactate and glycerate to pyruvate and hydroxypyruvate (Dickens and Williamson, Biochem. J. 68, 74, 1958). 560 Discussion Because our crystalline lithium salts of these two acids occur as monohydrates, and infra-red spectroscopy by Bellamy and Williams (Biocliem. J. 68, 81, 1958) failed to detect the presence in either solid salt of a keto group or of a molecule of water of cry- stallization, we tentatively suggested a hydration of the keto- to a diol-form as part of the enzymic reaction. While this is still not impossible, recent studies kindly carried out by D. A. Long (of University College, Swansea) on the Raman spectra in aqueous solution do not support the above findings based on the solid substances; in fact they indicate predominantly a keto-form in aqueous solutions of lithium or sodium pyruvate. So far, fluorescence difficulties have prevented similar observations on hydroxypyruvate solutions. Properties of Intact and Modified Cytochrome b^ Nature of Bakers' Yeast Lactate Dehydrogenase By T. HoRio (Osaka) HoRio: I would like to consider again the true physiological function of bakers' yeast lactate dehydrogenase. Of course, the enzyme extracted and purified from bakers' yeast can dehydrogenate lactate and other substrates: namely, a-hydroxybutyric acid, and exceptionally TPNH. When these substrates are added to the enzyme solution, the cytochrome Z)2-moiety bound on the enzyme is reduced. Of the various substrates tested, lactate is most rapidly attacked by this enzyme, so most workers use lactate as its substrate. However, using cytochrome c in its reduced form as an electron donor, the back reaction from pyruvate to lactate has never been demonstrated to take place physiologically, as reported by Boeri and co-workers. Furthermore, any meta- bolic pathway to produce lactate has never been found in bakers' yeast, as far as I know. From this point of view, I am just wondering the reason why bakers' yeast cells have such a lot of lactate dehydrogenase. When purified, bakers' yeast lactate dehydrogenase is quite water-soluble. However, our procedure for its extraction from bakers' yeast cells is very complex and different from that of other workers. In our experiments, pressed bakers' yeast is disrupted by a minimum amount of ethyl acetate, then washed with a great deal of water. The washed cells are extracted with 20 %-saturated ammonium sulphate at a neutral pH. Most of cytochrome c can be extracted in the salt solution by this extraction, but almost none of the enzyme by the procedures up to this step. This debris is suspended again in water, then the enzyme comes out of the debris to a great extent. When the activity of bakers' yeast lactate dehydrogenase purified mainly according to the method of Bach, Dixon and Zerfas (Biochem. J. 40, 229, 1946) is assayed by the phenazine methosulphate-method of Singer and co-workers, and by methylene blue-, ferricyanide- and cytochrome c-methods with the use of lactate as the substrate, the ratio of the phenazine methosulphate activity to the methylene blue, etc., activities varies for diff"erent enzyme preparations. As a rule, we might say that the more native the enzyme is, the lower is the ratio of phenazine methosulphate activity to methylene blue activity (see Fig. 1, Horio, Yamashita, Yamanaka, Nozaki and Okunuki: this volume, p. 555). As we have already pointed out, the modification of yeast lactate dehydrogenase which occurs on various treatments suggests to us the scheme of electron transport which may be broken at different sites (see Horio et al., this volume, p. 554). The intimate junction between the 'dehydrogenase' and 'cytochrome b.," moieties may be broken at the lower and high pH ranges, and the reaction (4) may take the place of reaction (2). It may be thought that the differences of the results presented by Ogura, Nygaard, Boeri, Morton and ourselves are dependent upon these changeable properties of the enzyme itself, as well as the original source of bakers' yeast cells as suggested by Nygaard. To my regret, I have not yet succeeded in crystallizing this enzyme although I have used four kinds of yeast cells made in Japan, and although our extraction method works well with all these yeasts. Therefore, we have been compelled to use a rather crude enzyme preparation for this experiment. Cytochrome bg 561 Figure 1 shows some mutual relationships of cytochrome b.^, flavin, and phenazine methosulphate and methylene blue activities during the course of the modification of the enzyme at a lower pH (Yamashita, Horio and Okunuki, unpublished). The curve 'cytochrome b.,' was made by dithionite-reduction of the a-absorption maximum of cytochrome 60. The curve 'flavin' was measured by fluorescence with the use of supernatant of an ammonium sulphate-saturated sample. Because of the use of a crude enzyme preparation, I think that the amount of liberated flavin does not Fig. 1. Modification of bakers' yeast lactate dehydrogenase. The figure compares the percentage of remaining activity with phenazine methosulphate and with methylene blue after treatment of the enzyme at low pH, with the amount of remaining dithionite-reducible haemoprotein (expressed as 'cytochrome ^3) and the amount of liberated flavin (measured as fluorescence). correctly show the inter-relationship with the enzyme modification. However, it may be assumed that the liberation of flavin has a certain important role in the partial modification of bakers' yeast lactate dehydrogenase, as indicated by Appleby and Morton (.Nature, Lond. 173, 749, 1954; Biochem. J. 73, 539, 1959). Compared with the enzyme, the 'cytochrome />.,' (split from the enzyme) is very stable, and the cytochrome so formed can be easily crystallized. The experimental conditions for direct extraction of cytochrome 63 from bakers' yeast cells, involving as it does extraction at a higher pH (pH 8) and other results, clearly indicate that 'cytochrome b.^ does not exist independently apart from the enzyme, at least in its physiologically native state. In order to investigate the mode of existence of cyto- chrome 62 in the enzyme, the absorption spectra of cytochrome bo and the enzyme in the visible wavelength were compared, and it was found that there is no notable diff"erence above experimental error. If cytochrome c is used as a model case, variously diff^erent kinds of cytochrome c's can be easily prepared : a native one having its coiling structure, a modified one damaged in the coiling structure, and a proteolysed one. The former two show the same absorption spectra which are remarkably difi'erent from the latter. This means that the absorption spectrum of cytochrome c is not seriously changed by damaging the coiling structure, but is remarkably altered by shortening its amino acid composition (see Horio et al., this volume, p. 557, for references). Analogously considered, the following two kinds of existence of cytochrome b^ might be assumed. In (1), 'cytochrome 60' having no flavin and no enzymic activity, exists 562 Discussion in a state in which it has its own protein configuration separate from the dehydro- genase moiety of bakers' yeast lactate dehydrogenase, and forms a protein complex with the dehydrogenase moiety to be physiologically active. In (2) bakers' yeast lactate dehydrogenase itself holds a b-type haem and a flavin. Then, if this enzyme is treated at a higher pH or some other drastic conditions, the enzyme protein is modified to liberate the flavin but not the b-type haem. The two models of bakers' yeast lactate dehydrogenase may be considered to be as shown in Fig. 2 of our paper (Horio et al., this publication, p. 557), which we have already discussed. Here I would like to recall again the case of succinate dehydrogenase. As is well known, the extraction procedure of the enzyme is rather complex. The treatment of the starting source with organic solvent such as n-butanol (Morton, Nature, Lond. 166, 1092, 1950) causes a serious decrease of total activity of the enzyme. However, the enzyme can be easily extracted after this treatment, but not before the treatment. Slater discussed with Singer at the International Symposium on Enzyme Chemistry held in Japan (Proceedings of the International Symposium on Enzyme Chemistry, 1957) this point. The activity of succinate dehydrogenase to phenazine metho- sulphate, ferricyanide, methylene blue, etc., changes during purification steps. Slater questioned the inter-relationships between the activity-changes and the iron bound on the enzyme. Even in a most crude state, the solubilized succinate dehydrogenase cannot join with the purified cytochrome b. These things recall the same question as for bakers' yeast lactate dehydrogenase: what is the native succinate dehydrogenase and what is cytochrome 6? From this point of view, the activity changes of succinate dehydrogenase is quite similar to the activity changes of bakers' yeast lactate dehydro- genase. At the same time, there might be a great doubt about whether or not so-called 'cytochrome b' does exist in a free state independent from the 'physiologically native' succinate dehydrogenase. This same question might be raised in relation to diaphorase and cytochrome c^. In fact, during the purification of cytochrome c^ with the use of cholate, an intimate connection is observed between both components, and initially DPNH can reduce cytochrome c^. However, this enzymic reaction is broken as the purification procedure goes on, and purified diaphorase and cytochrome c^ do not react with each other. Very recently, Iwatsubo and Kubo (Medical School, University of Osaka, Osaka; personal communication) succeeded in demonstrating an interesting flavin-level reaction, using D-amino acid oxidase and xanthine oxidase which have been completely purified by them. In an anaerobic condition, in the presence of both enzymes, the electrons are transferred from D-amino acid to cytochrome c. This reaction may be demonstrated regardless of the use of apo- or holo-D-amino acid oxidase. In fact, under an experimental condition in which xanthine oxidase is perfectly stable and in which flavin is not liberated from the enzyme, the apo-enzyme of D-amino acid oxidase can consume oxygen in the presence of its substrate and xanthine oxidase. Their experiments may illuminate the flavin-cytochrome electron transferring system in respiration. BoERi: With reference to the observations of Horio and co-workers, and of Nygaard, we also have found that phenazine methosulphate is a good acceptor with flavocytochrome b^, and it may be used in an activity test. The test may be carried out manometrically, by following the oxygen consumption for reoxidation of phenazine methosulphate, as used for succinate dehydrogenase by Singer and Kearney. The addition of cyanide is unnecessary, as no hydrogen peroxide is formed. The test may also be performed spectrophotometrically by measuring the change in extinction at 387 m//. We use a molar extinction of 2-27 x 10^ for the oxidized dye, and of 1-62 x 10^ for the leuco compound. It is not necessary to use anaerobic cuvettes but it is necessary to use air-free solutions. Nomenclature of Cytochrome h^ and Derived Proteins Morton: The various terms used in the several papers related to yeast lactate dehydro- genase and cytochrome Z)^ calls for some comment, lest the confusion become too great. Cytochrome b^ 563 Historically, the name 'cytochrome 6,' was given by Dixon and co-workers (Bach, Dixon and Keilin, Nature, Land. 149, 21, 1942; Bach, Dixon and Zerfas, Biochem. J. 40, 229, 1946) to a haemoprotcin seen in yeast extracts, which was reduced in the presence of lactate and oxidized by cytochrome c and by oxygen. If we retain the prefix 'cytochrome' — given by Keilin to naturally-occurring intracellular haemoprotcin pigments capable of reversible oxidation and reduction — then the name 'cytochrome bo should be used for the intact, homogeneous protein containing haem and flavin groups which was first isolated and described by Appleby and Morton {Nature, Load. 173, 749, 1954). This crystalline protein has L(+)-lactate dehydrogenase activity. The evidence in favour of this view has been set out by Appleby and Morton (1954, loc. cit.; Biochem. J. 71, 492, 1959; and Biochem. J. 73, 539, 1959) and by Morton (Society of Biological Chemists, India, Silver Jubilee Souvenir, p. 177, 1955; Rev. pure appl. Chem. 8, 161, 1958). Although we hold the view that the polydeoxyribonucleotide (Morton, 1958, loc. cit.; Appleby and Morton, Biochem. J. 73, 539, 1959; Biochem. J. 75, 258, 1960), is an integral component of the naturally-occurring material, as yet there is no evidence that this polynucleotide has any direct function in relation to the enzymic activity. Hence the minimum requirement for lactate dehydrogenase activity is the flavohaemo- protein. Boeri refers to this flavohaemoprotein having enzymic activity as 'flavocytochrome b^, but there seems no good reason for the addition of 'flavo' other than to distinguish this material from derived proteins devoid of the flavin group. Okunuki and co-workers refer to 'cytochrome b^ as a haemoprotcin, devoid of L(+)-lactate dehydrogenase activity, which they isolated from yeast cells (Yamashita, Higashi, Yamanaka, Nozaki, Mizushima, Matsubara, Horio and Okunuki, Nature, Lond. 179, 959, 1957). The behaviour of this material as reported by Okunuki and co-workers (see Horio, this volume, p. 552 and p. 560), and our own observations (Morton, 1958: loc. cit., Morton, Armstrong and Appleby, this volume, p. 501; Armstrong, Coates and Morton, this volume, p. 571) indicate that this haemoprotcin is modified from intact cytochrome b^. The relationship of this material to intact cytochrome b^ is somewhat analogous to the relationship of the peptic (or tryptic) 'core' of heart-muscle cytochrome c to cytochrome c. I think that no one would refer to the 'core' as 'cytochrome c\ I therefore consider that the enzymically-inactive haemoprotcin derived from cytochrome 60 should not be called 'cytochrome b^ but some other name, possibly 'haemoprotcin 557 from cytochrome b.^, which would have the value of indicating the position of the a-absorption band and the relationship to cytochrome b.^, As suggested by the observations of Slonimski and colleagues (Slonimski and Tysarowski, C. R. Acad. Sci., Paris 246, 1111, 1958; Labeyri, Slonimski and Naslin, Biochim. biophys. Acta 34, 262, 1959) and of Nygaard (Biochim. biopliys. Acta 35, 212, 1959 and this volume, p. 544), it is likely that there is more than one 'lactate dehydro- genase of bakers' yeast', particularly if the yeast is grown under partially-anaerobic conditions. The term 'lactate dehydrogenase' refers to the protein which specifically activates lactate for dehydrogenation. Cytochrome ^2 's one such dehydrogenase (certainly the principal L(+)-lactate dehydrogenase of aerobic bakers' yeast) and the term 'cytochrome b^^ is therefore preferable to the less-specific 'lactate dehydrogenase of bakers' yeast' when the flavohaemoprotein is implied. The Substrate Specificity of Cytochrome h.^ BoERi: With reference to our non-crystalline preparations of cytochrome b^, and the crystalline preparation obtained by Appleby and Morton, the following points should be noted. (a) As found by Dickens and collaborators, our preparations dehydrogenate glycerate whereas the preparations of Morton and co-workers (Morton, Armstrong and Appleby, this volume, p. 501) do not. (b) Our preparations are inert to both isomers of malate, whereas the crystalline 564 Discussion preparations of Morton and colleagues are inhibited slightly by L-malate and strongly by D-malate (Morton et al., this volume, p. 501). It should be noted that both our preparation, and that of the Australian workers, fail to dehydrogenate reduced triphosphopyridine nucleotide (TPNH), whereas this is dehydrogenated by the preparation of Okunuki's school (Horio, Yamashita, Yaman- aka, Nozaki and Okunuki, this volume, p. 552). With reference to observations of Nygaard, in my experience the cytochrome 6, acts specifically on L(+)-lactate. Another enzyme of the yeast, present in anaerobic yeast, was shown by Slonimsky and his collaborators to act on d(— )-lactic acid. It is not a flavo-cytochrome, but a flavoprotein. Work is going on in our laboratory on this enzyme which we prefer to define, with a terminology different from that of Slonimsky, as a D-hydroxyacid dehydrogenase. Armstrong: The substrate specificity of crystalline cytochrome Zjj was shown by Appleby and Morton (Biochem. J. 73, 539, 1959) to include glycolate and a-hydroxybutyrate. The further studies of Armstrong and Morton (see Table 5, this volume, p. 512) show that a number of a-hydroxymonocarboxylic acids are substrates. Additional hydroxycarboxylic and certain other groups probably interfere with attachment if sufficiently close to the a-hydroxyl group. In D-malate, the hydroxyl group has the same configuration as in L-lactate if viewed from the /9-carboxyl end. Thus it is not surprising that D-malate is a much more powerful inhibitor of the reaction with L-lactate than is L-malate. One might expect that certain esters of malate might be substrates of cytochrome b.,. It would appear from the pH optima reported by Horio, Yamanaka, Yamashita, Nozaki and Okunuki (this volume, p. 552) that malate and TPNH are dehydrogenated by enzymes other than cytochrome b., in the preparation used by these workers. On the Cytochrome b Components in the Respiratory Chain in Yeast Chance: The following relates to the possible function of cytochrome 62 in the respiratory chain in yeast. In 1954 we were actively considering the possibility that the b com- ponent of yeast cells could be identified with components ranging from cytochrome b.^ Fig. 1. Spectrum of reduced cytochrome ^2 compared with that of cytochrome c at liquid air temperatures (RE-129a). 540 560 580 to peroxidase (Chance, 1954: In Mechanisms of Enzyme Action: Johns Hopkins Press, Baltimore, p. 399). We have, however, many new data to offer on this point, particu- larly low temperature data on cytochrome b., (Chance, Klingenberg and Boeri, Fed. Proc. 15, 231, 1956). It has been found that the low temperature spectrum of reduced Cytochrome bo 565 cytochrome b^, is easy to distinguish from that of cytochrome b of the bakers' yeast cells. The former is illustrated by Fig. 1 which shows that the 2 WITH /7-CHLOROMERCURIPHENYLSULPHONATE (PCMS) Previous studies (Appleby and Morton, Nature, Lond. 173, 749, 1954; Boeri and co-workers. Arch. Biochem. Biophys. 60, 463, 1956) have shown that mercurials such as PCMS are powerful inhibitors of the lactate-ferricyanide and lactate-ferricytochrome c reductase activities of cytochrome b^. On anaerobic incubation for about 15 min of intact, non-fluorescent cytochrome b^ in the standard buffer with PCMS at a final concentration of 1 x 10~* m, the solution showed an intense yellow fluorescence, while the colour changed from coral pink to a brownish pink. Spectroscopically the a- and /S-bands of the reduced haemoprotein component appeared unaltered. The material so treated sedimented in the ultracentrifuge as a single component, leaving a yellow non-sedimenting component behind. On exposure of the PCMS- treated mater