UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Electron transfer properties of flavin-modified and axial ligand variants of cytochrome c Twitchett, Mark Bradley 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-27263X.pdf [ 7.25MB ]
Metadata
JSON: 831-1.0088692.json
JSON-LD: 831-1.0088692-ld.json
RDF/XML (Pretty): 831-1.0088692-rdf.xml
RDF/JSON: 831-1.0088692-rdf.json
Turtle: 831-1.0088692-turtle.txt
N-Triples: 831-1.0088692-rdf-ntriples.txt
Original Record: 831-1.0088692-source.json
Full Text
831-1.0088692-fulltext.txt
Citation
831-1.0088692.ris

Full Text

E L E C T R O N T R A N S F E R PROPERTIES O F FLAVIN-MODIFIED A N D A X I A L LIGAND VARIANTS O F C Y T O C H R O M E C by Mark Bradley Twitchett B.Sc., The University of East Anglia, 1992 M.Sc , The University of East Anglia, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF G R A D U A T E STUDIES T H E DEPARTMENT OF BIOCHEMISTRY AND M O L E C U L A R BIOLOGY We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA February 1998 ©Mark Bradley Twitchett In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. 1 further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of B i Q C M E M l S T ^ A*»t> M O L H c U L A g . r o i O L O S Y The University of British Columbia Vancouver, Canada Date \ S T n FEfefeOAK^ DE-6 (2/88) A B S T R A C T Electron transfer (ET) reactions of metalloproteins are central to many biological processes. This fact combined with the simplicity of the chemistry involved makes this type of reaction amenable to detailed experimental and theoretical analysis. The current work employs two novel forms of mitochondrial cytochrome c to investigate critical structural factors that dictate the efficiency of intermolecular and intramolecular ET reactions of this protein and that have implications for metalloproteins in general. The first of these approaches involved the use of a new technique to study intramolecular E T reactions in which a combination of site-directed mutagenesis and chemical modification was used to construct a family of synthetic flavocytochromes c. Previous studies by Tollin and Cusanovich have used flavins extensively to investigate intermolecular ET reactions of electron transfer proteins, but flavins have not been used previously as donors in the study of intramolecular E T reactions. In the present work, flash photolysis was used to study the intramolecular E T kinetics of four cytochrome variants bearing a flavin group at position 8, 39, 62, or 85. To complement this work, the theoretical models of Beratan, Betts and Onuchic and of Siddarth and Marcus were used to identify the theoretically optimal ET pathways and donor-acceptor electronic coupling for each of these derivatives. A major finding derived from the kinetic studies of these synthetic flavocytochromes is that the reorganization energy for electron transfer, X, of the flavin center is 0.7(2) eV. In addition, the theoretical models for these reactions demonstrate that the correlation of the rate constants for intramoleculer E T observed can be understood in terms of the nature and length of the E T pathway between the donor and acceptor centers. The second approach described in this thesis involved the study of three cytochrome c variants ii in which the Met80 ligand to the heme iron was replaced with an alanyl residue. Analysis of the intermolecular reduction of these variants by Fe(EDTA) 2' was undertaken to gain insight into the role of the Met80 ligand and two other active site residues in regulating the electron transfer reactivity of the cytochrome. These variants exhibited a remarkable range in reactivity with this reductant as represented by the apparent electrostatics-corrected self-exchange rate constants, £ 1 1 c o r r , exhibited in this reaction (pH 7, u = 0.1 M , 25 °C): Met80Ala, kncm = 1.1 M"1 s"1; Met80Ala/Tyr67Phe, k^" = 3.4 x 102; and Met80Ala/Phe82Ser, knco" = 4.3 * lO - 2 M - 1 s"1. Possible explanations of these results are discussed. iii T A B L E O F C O N T E N T S ABSTRACT ii T A B L E OF CONTENTS iv LIST OF TABLES ix LIST OF FIGURES xi ABBREVIATIONS xv ACKNOWLEDGMENTS xvii 1. INTRODUCTION 1.1 Overview 1 1.2 Structural Aspects of Cytochrome c 1 1.3 Physiological Aspects of Cytochrome c 8 1.4 Factors Influencing Reduction Potential 10 1.5 Flavins and Flavoproteins 12 1.6 Remodeling of Heme Protein Active Sites 16 1.7 Electron Transfer Kinetics 20 1.7.1 Overview 20 1.7.2 Intermolecular Electron Transfer 20 1.7.2.1 Theoretical Studies 20 1.7.2.2 Intermolecular Electron Transfer Reactions of Cytochrome c 25 1.7.3 Intramolecular Electron Transfer 29 1.7.3.1 Theoretical Studies 29 iv 1.7.3.2 Intramolecular Electron Transfer Reactions of Cytochrome c 33 1.8 Objectives of this Dissertation 38 2. EXPERIMENTAL 2.1 Yeast Expression System 40 2.1.1 Mutagenesis Techniques 40 2.1.2 Protein Preparation 40 2.2 Bacterial Expression System 43 2.2.1 Mutagenesis Techniques 43 2.2.2 Protein Preparation 44 2.3 Flavin Preparation 47 2.4 Thiol Assay & Protein Modification 47 2.5 Tryptic Digestion & HPLC Analysis 49 2.6 Spectroscopic Characterization 50 2.6.1 Electronic Absorption Spectroscopy 50 2.6.1.1 General Analysis 50 2.6.1.2 Ligand Binding 50 2.6.1.3 pH Titrations 51 2.6.2 Fluorescence Spectroscopy 51 2.6.3 Circular Dichroism Spectroscopy 51 2.6.4 Fourier Transform Infrared Spectroscopy 52 2.7 Electrochemical Characterization 52 2.7.1 Cyclic Voltammetry 52 2.7.2 Spectroelectrochemical Techniques 54 v 2.8 Kinetic Experiments 57 2.8.1 Flash Photolysis Techniques 57 2.8.2 Fe(EDTA)2" Reduction Kinetics 58 2.9 Molecular Modeling & Pathway Calculations 59 3. RESULTS 3.1 Cysteine Variants & Flavin Modification 60 3.1.1 Protein Preparation 60 3.1.3 Flavin Synthesis 60 3.1.3 Cysteine Accessibility & Flavin Modification 62 3.1.4 Tryptic Digestion and HPLC Analysis 64 3.1.5 Electronic Absorption Spectroscopy 64 3.1.6 Circular Dichroism Spectroscopy 68 3.1.7 Fluorescence Spectroscopy 68 3.1.8 Electrochemistry 72 3.1.9 Flash Photolysis... 87 3.1.9.1 Intermolecular Electron Transfer Kinetics 87 3.1.9.2 Intramolecular Electron Transfer Kinetics 90 3.1.10 Structure Modeling 94 3.1.11 Electron Transfer Pathway Calculations 94 3.2 Axial Ligand Variants 100 3.2.1 Protein Expression and Purification 100 3.2.2 Electronic Absorption Spectroscopy 100 vi 3.2.2.1 Ligand Binding 100 3.2.2.2 pH Titrations 101 3.2.3 Circular Dichroism Spectroscopy 104 3.2.4 Electrochemistry 105 3.2.5 Ligand Binding Kinetics 105 3.2.6 Autoxidation 110 3.2.7 Fourier Transform Infrared Spectroscopy 110 3.2.8 Kinetics of Axial Ligand Variant Reduction by Fe(EDTA)2" 113 4. DISCUSSION 4.1 Overview 121 4.2 Cysteine Variants & Flavin Modification 122 4.2.1 Protein Characterization 122 4.2.2 Electron Transfer Kinetics 128 4.2.2.1 Intermolecular Electron Transfer Reactions 128 4.2.2.2 Intramolecular Electron Transfer Reactions 132 4.3 Axial Ligand Variants 138 4.3.1 Protein Characterization 138 4.3.2 Electron Transfer Properties 143 4.4 Concluding Remarks 150 5. REFERENCES 152 APPENDIX A 165 APPENDIX B 171 vii APPENDIX C 174 APPENDIX D 176 v i i i LIST OF TABLES 2.1 Mutagenic oligonucleotides for yeast expression system 40 2.2 Mutagenic oligonucleotides for E. coli expression system 44 3.3 Bimolecular rate constants for the reaction of cytochrome c Cys variants with 4,4-dithiodipyridine 62 3.4 Electronic absorption maxima and molar absorbances for unmodified & flavin modified cytochrome c variants 67 3.5 Fluorescence emission maxima & relative intensities for free and attached 7-acetyl-10-methylisoalloxazine 68 3.6 Electrochemical thermodynamic properties of wild-type and variant cytochromes 74 3.7 pKt values for residue 39 in the oxidized (pK0) & reduced ( p . r Q states 76 3.8 Potentials for 7-acetyl-10-methylisoalloxazine ontained from spectroelectrochemical experiments 78 3.9 Thermodynamic parameters of the oxidation-reduction equilibria of the flavin-modified cytochromes 83 3.10 pK3 values for residue 39 in the oxidized (pK0) & reduced (pKr) states of the flavin-modified cytochromes 84 3.11 Potentials for the flavin redox center of modified proteins calculated from spectroelectrochemical experiments 84 3.12 Rate constants for the reduction of cytochrome c (Fe3+) variants by various flavins 90 3.13 Rate constants and activation parameters for the reduction of cytochrome c (Fe3+) ix variants by various flavins 90 3.14 Observed rate kinetics, activation parameters & reorganization energies for the intramolecular electron transfer of flavin modified cytochrome c derivative 91 3.15 Amino acid residues selected by A l search and experimental and theoretical electronic coupling values 95 3.16 Absorption maxima and molar absorbances of cytochrome c axial ligand variants 102 3.17 Reduction potential values for the axial variants 105 3.18 Kinetic and thermodynamic parameters for the reduction of the axial variants by Fe(EDTA)2" 116 x LIST O F FIGURES 1.1 Amino acid sequence of yeast iso-1 -cytochrome c 3 1.2 The polypeptide fold of yeast iso-l -cytochrome c 4 1.3 Schematic drawing of iron protoporphyrin IX 5 1.4 The physiological electron transfer partners of cytochrome c 9 1.5 Structures of the vitamin riboflavin & the derived flavin coenzymes and the oxidation states of the flavin coenzymes 13 1.6 Model of catalytic cycle of flavocytochrome b2 15 1.7 Structural comparison of the active sites of the cyanide adduct of Met80Ala cytochrome c & oxy sperm whale myoglobin 19 1.8 Potential energy curves representing an electron transfer reaction. 23 1.9 The dependence of electron transfer rate on free energy 31 1.10 Electron transfer pathways and correlations of maximum electron transfer rates with distance in Ru(HisX) modified cytochromes c 37 2.11 Physical map of pING4 for the expression of yeast iso-l -cytochrome c in yeast 41 2.12 D N A sequence of the Smal-Hindlll fragment of pING4 42 2.13 Physical map of pBPCYCl(wt)/3 for the expression of yeast wo-1-cytochrome c in bacteria 45 2.14 D N A sequence of the Kpnl-Hindlll fragment of pBPC YC1 (wt)/3 46 2.15 Outline of synthesis of 7a-bromoacetyl-10-methylalloxazine 48 2.16 A schematic diagram of the cyclic voltammetry cell 53 2.17 Schematic representation of the optically-transparent thin layer electrode cell 55 xi 3.18 NMR spectra of 7-acetyl-10-methylalloxazine and 7a-bromoacetyl-10-methylalloxazine 61 3.19 Thiol assay kinetic traces demonstrating solvent accessibility of cysteine residues 63 3.20 FPLC elution profiles monitoring flavin attachment to His39Cys cytochrome c 65 3.21 HPLC analysis of tryptic peptides for wild type cytochrome c, the His39Cys variant, and the flavin modified His39Cys variant 66 3.22 The electronic absorption spectra of free 7-acetyl-10-methylisoalloxazine, reduced Thr8Cys cytochrome c and flavin modified reduced Thr8Cys cytochrome c 69 3.23 CD spectra of Thr8Cys cytochrome c and flavin modified Thr8Cys cytcohrome c 70 3.24 Fluorescence emission spectra of the free 7-acetyl-10-methylisoalloxazine and modified F£is39Cys cytochrome c variant 71 3.25 Cyclic voltammograms for Thr8Cys cytochrome c, 7-acetyl-10-methylisoalloxazine and flavin modified Thr8Cys cytochrome c 73 3.26 The temperature dependence of reduction potentials for cytochrome c cysteine variants 75 3.27 The pH dependence of reduction potential for cytochrome c cysteine variants 77 3.28 The pH dependence of the three reduction potentials of 7-acetyl-10-methylisoalloxazine 79 3.29 Spectroelectrochemical titration of 7-acetyl-10-methylisoalloxazine 80 3.30 The temperature dependence of reduction potentials of heme and flavin centers of flavin modified cytochrome c variants and free 7-acetyl-10-methylisoalloxazine 82 3.31 Dependence of the reduction potentials for heme centers of flavin-modified cytochromes on pH 85 xii 3.32 Spectroelectrochernical titration of flavin modified Thr8Cys cytochrome c 86 3.33 The dependence of kobs on [ferricytochrome c] in the reduction of cytochrome c by flavin derivatives 88 3.34 Eyring plots for the bimolecular reduction of horse heart and yeast cytochromes by flavin derivatives 89 3.35 Eyring plots for the intramolecular reactions of flavin modified cytochromes c 92 3.36 Dependence of koba on - A G 0 for the reduction of cytochromes c by the covalently attached flavin 93 3.37 Energy-minimized structure of flavin modified Thr8Cys cytochrome c and proposed electron transfer pathways 96 3.38 Energy-minimized structure of flavin modified His39Cys cytochrome c and proposed electron transfer pathways 97 3.39 Energy-minimized structure of flavin modified Asn62Cys cytochrome c and proposed electron transfer pathways 98 3.40 Energy-minimized structure of flavin modified Leu85Cys cytochrome c and proposed electron transfer pathways 99 3.41 Electronic absorption spectra of oxidized and reduced cytochrome c axial ligand variants 103 3.42 The electronic absorption spectra of Met80Ala ferricytochrome c as a function of pH 106 3.43 Thermal denaturation of cytochrome c axial ligand variants 107 3.44 Spectroelectrochernical titration of Met80Ala/Tyr67Phe cytochrome c 108 3.45 Transient absorption kinetics for the recombination of CO with Met80Ala xiii cytochrome c following photodissociation 109 3.46 Absorption spectra following the autoxidation of Met80Ala cytochrome c I l l 3.47 Infrared spectra (FTIR) of ferrous carbonyl derivatives of cytochrome c axial ligand variants 112 3.48 Absorption spectra monitoring the reduction of Met80Ala cytochrome c by Fe(EDTA) 2' 117 3.49 Absorption spectra monitoring the reduction of Met80Ala/Tyr67Phe cytochrome c by Fe(EDTA) 2' 118 3.50 Dependence of the observed rate constants for reduction of ferricytochrome c axial ligand variants on Fe(EDTA) 2' concentration 119 3.51 Eyring plots for the Fe(EDTA)2" reduction of ferricyochrome c axial ligand variants 120 4.52 Structural diagrams of the region about the cysteine mutation sites in the Thr8Cys, His39Cys, Asn62Cys and Leu85Cys cytochrome c variants 123 4.53 Structures of flavins involved in intermolecular electron transfer reactions with cytochromes c 129 4.54 Correlations of maximum ET rates in flavin (CysX) modified cytochromes c and Ru(HisX) modified cytochromes c with distance 134 4.55 Space filling representations of cytochromes c showing heme solvent accessibility 140 xiv ABBREVIATIONS A angstrom AI artificial intelligence bpy bipyridyl CD circular dichroism D N A deoxyribonucleic acid D T T dithiothreitol E D T A Ethylenediaminetetracetate ESR electron spin resonance E T electron transfer eV electron volt FAD flavin adenine dinucleotide FC Franck-Condon Fl- flavin-modified flox flavin, oxidized flred flavin, reduced A s qH- flavin, semiquinone F M N flavin mononucleotide FPLC fast protein liquid chromatography H.S. high-spin HPLC high performance liquid chromatography L.S. low-spin mV millivolt N A D H nicotinamide adenine dinucleotide nm nanometer N M R nuclear magnetic resonance O T T L E optically transparent thin-layer electrode SCE saturated calomel reference electrode SHE standard hydrogen electrode T F A trifluoroacetic acid T L C thin-layer chromatography T M L trimethyllysine TPCK N-tosyl-L-phenylalanine chloromethyl ketone U V ultraviolet xvi A C K N O W L E D G M E N T S I would like to acknowledge the many people who have helped and supported me throughout my studies. I would especially like to thank my supervisor Grant Mauk, for his support and guidance. Thank you for everything. Heart warm thanks must also go to all the members of the Mauk lab, both past and present: Brent, Cameron, Christie, Dean, Emma B., Emma L. , Fred, Kimphry, Lianglu, Marcia, Mark, Matt, Ralf, Sue, Susanna, and Tom. Many thanks to my committee members, Gary Brayer and Steve Withers, for reviewing my thesis. I must give a special thank you to Prabha Siddarth, both for her helpful discussions and for her work involving the molecular modeling and the pathway calculations. Thanks also to all those fellow biochemists who have become friends along the way. On a more personal level I would like to thank my family, especially mum, dad, nan and gramps, for their loving support and understanding in my decisions to pursue my goals where ever they took me. I'm sorry this usually meant taking me thousands of miles away from home. Nevertheless, you were always there for me, and it was wonderful to know that. Finally, I wish to thank my wife and best friend Sharon, for her support, patience and above all, love. I just want you to know that I could not have succeeded alone. xvii In memory of Lil. xviii 1. INTRODUCTION 1.1 Overview Electron transfer reactions are central to many biological processes such as respiration, photosynthesis and xenobiotic detoxification. As a result, theoretical and experimental studies concerning such reactions have been extensive. The mitochondrial electron transfer protein cytochrome c has frequently been used in this body of work because its structure and functional properties have been well characterized, it is thermodynamically stable and readily available, and it is readily amenable to chemical modification or to site-directed mutagenesis (Pettigrew & Moore, 1987; Moore & Pettigrew, 1990; Scott & Mauk, 1996). While the wealth of literature concerning biological electron transfer reactions has led to a general understanding of the physical factors that dictate electron transfer within proteins such as cytochrome c, our understanding of the relative importance and quantitative contributions of these factors is incomplete. As described later, the work presented in this dissertation addresses some of these issues through investigation of yeast iso-l-cytochrome c derivatives that have been produced by site-directed mutagenesis and chemical modification. Before considering this work in detail, however, it is useful to review briefly the literature that provides the basis for the present studies. 1.2 Structural Aspects of Cytochrome c The family of eukaryotic cytochromes c is one of the most extensively examined groups of proteins. The amino acid sequences of approximately 100 species have been determined (Cutler et al., 1987; Moore & Pettigrew, 1990) and three-dimensional structures of six species of mitochondrial cytochromes c have been elucidated: Saccharomyces cerevisiae iso-\ (Louie & Brayer, 1990; 1 Berghuis& Brayer, 1992) and iso-2 (Murphy etal., 1992), horse heart (Dickerson al., 1967,1971; Bushnell et al., 1990), tuna (Takano & Dickerson, 1981a,b), rice (Ochi etal, 1983), and bonito (Tanakaef a/., 1975). In general, mitochondrial cytochromes c are small, soluble, basic proteins composed of a single polypeptide chain of 103 to 113 residues. A relatively high degree of sequence similarity is observed between these cytochromes, with 23 residues being invariant among 96 of the known sequences. Of the remaining residues only 14 are highly variable (variability index >15 (Wu & Kabat, 1970; Moore & Pettigrew, 1990)). The amino acid sequence for yeast iso-l -cytochrome c, the protein studied in this work, is given in Figure 1.1. The positions of the invariant and highly variable residues are indicated in this diagram. All c-type cytochromes possess an iron porphyrin prosthetic group as exemplified by the crystal structure of yeast cytochrome (Louie & Brayer, 1990) (Figure 1.2). The prosthetic group is covalently bound to cysteine residues of the polypeptide chain through two thioether linkages by the post-translational attachment of heme (Thony-Meyer, 1997). The two cysteine residues are located at position 14 and 17 and are part of the characteristic sequence, Cys-X-X-Cys-His. The His 18 residue provides one of the two axial ligands. The heme iron is coordinated to the e2 nitrogen of this residue. The second axial ligand to the iron is provided by the sulfur atom of Met80. The heme group (Figure 1.3) is 'saddle' shaped due to the constraints caused by the two covalent linkages through pyrrole rings B and C. The remaining two pyrrole rings, A and D, each bear a single propionate group and each of these groups is buried within the protein structure. The majority of the heme group is, in fact, buried within the protein matrix, and only five heme atoms (CMD, CHD, CAC, CBC, and CMC) are exposed to the surrounding solvent. These five atoms constitute 9.5% of the heme surface area (Brayer & Murphy, 1996). 2 -5 Thr Glu Phe Lys Ala 1 Gly Ser Ala Lys Lys Gly Ala Thr Leu 10 Phe 11 Lys Thr Arg Cys Leu Gin Cys His Tru-20 Val 21 Glu Lys Gly Gly Pro His Lys Val d y 30 Pro 31 Asn Leu His Gly He Phe Gly Arg His 40 Ser 41 Gly Gin Ala Glu Gly Tyr Ser Tyr Thr 50 Asp 51 Ala Asn lie Lys Lys Asn Val Leu Trp 60 Asp 61 Glu Asn Asn Met Ser Glu Tyr Leu Thr 70 Asn 71 Pro Lys Lys Tyr He Pro Gly Thr Lys 80 Met 81 Ala Phe Gly Gly Leu Lys Lys Glu Lys 90 Asp 91 Arg Asn Asp Leu He Thr Tyr Leu Lys 100 Lys 101 Ala Cys 103 Glu Figure 1.1 Amino acid sequence of yeast /so-1-cytochrome c (Narita & Titani, 1969). Residues shown in bold are invariant among 96 known mitochondrial sequences and italicized residues have a variability index of greater than 15 (Moore & Pettigrew, 1990). Lysine 72 is trimethylated. Residue numbering is based on the sequence of tuna and horse cytochrome c. 3 Figure 1.2 The polypeptide fold of yeast iso-l -cytochrome c. The heme and axial ligands Hisl8 and Met80 are shown in bold (Louie & Brayer, 1990). 4 Figure 1.3 Schematic drawing of iron protoporphyrin IX. The atoms and the four pyrrole rings are labelled according to the Protein Data Bank nomenclature (Bernstein et al, 1977). Pyrrole rings C and D define the more exposed heme edge of cytochrome c and the porphyrin ring is covalently bound to the protein through two thioether linkages (CAB to CysHand CACto Cysl7). 5 The polypeptide chain is predominantly oc-helical with -45% of the amino acid residues involved in 5 a-helical segments (residues 2-14, 49-55, 60-70, 70-75 & 87-102). Other secondary structural elements include five type II p-turns. These p-turns account for the conservation of 5 glycyl residues (23, 34, 37, 45 & 77) within the cytochrome c family because only a hydrogen atom can be accommodated at residue Rj+2 of each p-turn. The internal residues surrounding the heme group are bulky, hydrophobic groups. Conversely a large proportion of the exposed surface residues are charged. The yeast iso-1-cytochrome possesses sixteen lysyl residues, three arginyl residues, four histidyl residues and a pi value of 10 (Kim et al, 1980). The charged residues are distributed around the surface in an asymmetrical manner with the positive lysyl residues occupying positions around the exposed heme area and on the back and top of the protein (with respect to the protein orientation shown in Figure 1.2). The basic regions are thought to be important for the interaction with other electron transfer proteins and to facilitate electron transfer reactions of the cytochrome c with such proteins. Complexes of this type are discussed in further detail in section 1.3 (Page 8). Experimental evidence that the two oxidation states of cytochrome c have different structural features has been presented for over 40 years. As early as the late 1950s it was demonstrated that ferrocytochrome c is more resistant to proteases, specifically subtilisin and trypsin, than is ferricytochrome c (Nozaki et al, 1958, Yamanaka et al, 1959). Since that time, a variety of techniques including thermal and chemical denaturation (Butt & Keilin, 1962; Dickerson & Timkovich, 1975) and one-dimensional NMR studies (Moore 1983; Williams etal, 1985) have been used to establish further the existence of conformational structural differences. For a recent review of these studies the reader is referred to Margoliash & Schejter (1996). Information on the oxidation state linked structural changes remained elusive until the 6 publication of the crystal structures of the oxidized and reduced forms of cytochromes c isolated from tuna (Takano & Dickerson, 1980, 1981a,b) and yeast (Berghuis & Brayer, 1992). The latter structures are discussed here since the yeast iso-1 -cytochrome c is the protein studied in this thesis. Only small localized perturbations in the positions of the polypeptide backbone atoms occur with a change in oxidation state. The majority of these differences occur in the polypeptide region which packs against the left side of the heme group (with respect to the protein orientation shown in Figure 1.2). These changes are thought to stabilize the positive charge on the oxidized heme iron (See section 1.4, Page 10). The structural changes that accompany oxidation include the loss of a hydrogen bond between Met80 SD and Tyr67 OH, a large displacement of the internal water Watl66 (1.7A change) and a change in the hydrogen bond network involving Wat 166, Asn52 and the heme propionate on pyrrole A. This latter conformational change is also proposed to be involved in the oxidation state dependence of the His39 pKa owing to the proximity of the heme propionate group (Moore et al., 1984). Because this His residue contributes to the pH dependence of the reduction potential exhibited by/so-1-cytochrome c, a discussion on this topic is presented in section 1.4 (Page 10). Recently the NMR solution structure of both the oxidized and reduced yeast /so-1-cytochromes c have been determined (Baistrocchi etal., 1996; Banci etal., 1997). These structures show that the secondary structure features are similar in both oxidation states. The oxidized protein exhibits greater flexibility in the residue segments 14-26 and 75-82. This flexibility is thought to be responsible for the observed increase in the rate of rotation of the Tyr67 side chain upon oxidation. In addition, the solution structures show a change in the hydrogen bond network between the heme propionate on pyrrole A and residue Gly41. This change is consistent with the corresponding crystal structures. These subtle changes observed in the yeast iso-1 -cytochromes c solution structures differ 7 from those observed in the horse heart cytochromes c. Comparison of the horse heart protein structures reveal relatively large changes upon oxidation. These changes include a disruption of a-helices and a significant difference in the heme solvent accessibility (Qi etal., 1996). 1.3 Physiological Aspects of Cytochrome c The major function of mitochondrial cytochrome c is to transfer electrons along the respiratory electron-transport chain. In this process, cytochrome c acts as a reversible, single electron carrier which shuttles electrons between ubiquinol-cytochrome c oxidoreductase (complex III) and cytochrome c oxidase (CcO) (complex IV) within the inter-membrane space (Figure 1.4). The ability of cytochrome c to transfer electrons efficiently between these two membrane bound electron transfer partners is dependent upon both a finely tuned reduction potential and the formation of reactive protein-protein complexes. In addition to this central role, cytochrome c also interacts with other mitochondrial electron transfer proteins (Figure 1.4) (See review in Pettigrew & Moore, 1987). These interactions include complex formation with sulphite oxidase (liver) and cytochrome c peroxidase (yeast, bacteria), both of which are involved in detoxification pathways, and flavocytochrome b2 (yeast lactate dehydrogenase) and cytochrome b5, which allow cytochrome c to channel electrons from two other pathways to cytochrome c oxidase. As mentioned previously, the surface of the protein surrounding the partially exposed heme edge of cytochrome c contains a large number of positively charged lysyl and arginyl residues. This region of the protein surface has been shown by experimental (e.g. Speck et al, 1979, 1981; Rieder &J3osshard, 1980; Speck & Margoliash, 1984; Pelletier& Kraut, 1992;Northrupe/(3/., 1993; Mauk etal, 1995) and by computational techniques (e.g. Poulos & Kraut, 1980; Mauk etal, 1986; 8 CcR CcO • i l l i l 1111111 11111111111111 Inner membrane 0 2 + 4 H + 2 H 2 0 Figure 1.4 The physiological electron transfer partners of cytochrome c (Derived from Pettigrew & Moore, 1987). Abbreviations: CcR, ubiquinol cytochrome c oxidoreductase; CcO, cytochrome c oxidase; Cc, cytochrome c; CcP, cytochrome c peroxidase; SO, sulfite oxidase; b2, flavocytochrome b2; b5, cytochrome b5. The arrowheads denote direction of electron transfer. 9 Pettigrew & Moore, 1987; Northrup et al, 1988; Tegoni et al, 1993) to be important in the formation of all these protein-protein complexes. The most extensively studied complexes of this type are the cytochrome c-cytochrome c peroxidase (Nocek etal, 1996) and cytochrome c-cytochrome b5 (Mauk et al, 1995) complexes. In both these systems, the protein complexes appear to be stabilized by charge complementarity between the interacting protein surfaces (Salemme 1976; Mauk etal, 1986; Eltis etal, 1991; Northrup etal, 1993; Poulos& Kraut 1980; Pelletier& Kraut, 1992). Nevertheless, despite this ability to form closely packed complexes, the heme groups associated with each protein involved are still separated by an appreciable distance (7-18 A heme edge to heme edge distance) that confirms the important involvement of long range electron transfer in such systems. 1.4 Factors Influencing Reduction Potential For optimal transfer of electrons, mitochondrial cytochromes c possess midpoint reduction potentials which fall approximately midway between the respiratory chains initial electron donor coupling process (NAD7NADH couple, -320 mV) and final electron acceptor coupling process (0 2 /H 2 0 couple, +820 mV) (Pettigrew & Moore, 1987). This requirement leads to the conservation of the reduction potentials of these cytochromes c such that variations are limited to ~±20 mV from an average value of 270 mV. The reduction potentials of heme containing proteins are dependent upon a number of factors that include the nature of the porphyrin, the axial ligands and the dielectric environment surrounding the heme (Moore & Williams, 1977; Moore et al, 1986; Moore & Pettigrew, 1990). The influence of the medium surrounding the heme center is substantial. This influence is exemplified by the work of Kassner (1972, 1973) that shows the reduction potential of bipyridine mesoheme methyl ester is 300 mV higher in benzene than in water. Similarly, if mitochondrial cytochrome c is compared to N-10 acetyl methionine cytochrome c heme octapeptide the reduction potential value for the intact protein is shown to be over 300 mV higher (Harbury et al, 1965). These variations are related to the ability of the surrounding dipoles and charges to accommodate the change in the overall charge on the heme. As such, studies have been carried out comparing the amount of heme exposure with reduction potential values within different proteins (Stellwagen, 1978; Schlauder & Kassner, 1979). Unfortunately, this analysis suffers from oversimplification because the nature of the porphyrin and the axial ligands are not considered. Although all mitochondrial cytochromes c contain a His and a Met residue as axial ligands, a number of heme proteins are known in which the sixth ligand site, normally occupied by the Met, is either vacant or is occupied by a second His group. Coordination environments of this type are found in such proteins as myoglobin (E m = +61 mV, Lim, 1990), cytochrome c peroxidase (E m = -190 mV, Conroy et al, 1978), cytochrome bs (E m = +5 mV, Reid, 1984) and class III bacterial cytochromes c (E m = -400 - -100 mV, Moore & Pettigrew, 1990). In addition the effect of substituents on the porphyrin ring must be considered. These effects are generally small as shown by the reduction potential differences between proteins containing c-type heme and protoheme IX (20-40 mV difference) (Williams, 1959; Falk, 1964; Barker et al, 1993). Of further note are the factors that dictate the pH dependence of the cytochromes reduction potential. For /so-1-cytochrome c the protein region surrounding the heme propionate on pyrrole A is thought to be responsible for the observed pH dependence (Moore et al, 1984). This region includes residues Arg38 and His39. Arg38 is known to undergo an oxidation-state conformational change (Berghuis & Brayer, 1992) and this change may be responsible for the oxidation-state linkage of the pATa of the nearby His39 residue. The effect of this localized electrostatic change is to reduce the reduction potential of the protein by -23 mV as the pH of the aqueous environment passes 11 through the histidine pKa value. Indeed, if this His39 residue is modified, the potential of the cytochrome in independent of pH (Moore etal., 1984). Thus this pH dependence provides a valuable indicator of perturbations of this region of the cytochrome. 1.5 Flavins and Flavoproteins Flavoproteins are versatile biological electron transfer agents that are involved in catalysis of a wide variety of reactions that include substrate dehydrogenation and the activation of dioxygen (Massey & Hemmerich, 1982; Massey & Ghisla, 1983; Stankovich, 1989). Flavoproteins usually contain either flavin mononucleotide (FMN) or flavin adenine dinucleotide (FAD) (Figure 1.5), with the flavin generally being bound tightly but non-covalently to the protein. A small number of flavoproteins are found in which the flavin cofactor is covalently bound to the protein (Steenkamp etal, 1978; Mclntire etal, 1985), through a cysteine, histidine or tyrosine residue. Regardless of the binding interactions the reactive center of the F M N or FAD is the isoalloxazine moiety of the flavin prosthetic group (Figure 1.5). This ring system provides the flavin with the ability to transfer either a single electron or to transfer two electrons simultaneously. The protein environment of the flavin group controls the kinetics and thermodynamics of the electron transfer properties of the flavin and, therefore, determines whether the flavin engages in a one or two electron transfer process. The flavin can exist in the oxidized, semiquinone and fully reduced oxidation states (Figure 1.5). These three oxidation states are spectroscopically distinct. The oxidized, semiquinone and fully reduced forms are yellow, blue or red, and colorless, respectively. The existence of two distinct semiquinone forms is related to the protonation state of the flavin radical (pKa = ~ 8-9). With flavins free in solution, these semiquinone radicals are in fast equilibrium with the oxidized and reduced forms, so only a small proportion of the flavin occurs in the radical form (Massey & Hemmmerich, 12 Isoalloxazine ring O O CH,- -O—P—O—P—O I I O- CH-Riboflavin (B2) OH OH • Flavin mononucleotide (FMN)-Flavin Adenine Dinucleotide (FAD) NHj CH C H . ^ O H + e- C H 3 L *—1 R H+ e- C H 3 C H Quinone (Oxidized) Semiquinone Hydroquinone (Fully reduced) Figure 1.5 (Top Panel A) Structures of the vitamin riboflavin and the derived flavin coenzymes, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD). (Bottom Panel B) Oxidation states of flavin coenzymes; oxidized, semiquinone and fully reduced. 13 1980; Draper & Ingraham,1968). This behavior is a consequence of the thermodynamic instability of the semiquinone as demonstrated by the two half-reaction reduction potentials of flavin mononucleotide (FMN), (E,°, fl o x-fl s qH = -238 mV and E 2°, fl^H-flredH2 = -172 mV) (Draper & Ingraham, 1968). In contrast, when this flavin is bound within a protein environment, the reduction potentials change. This change is due to the ability of the protein to stabilize the radical and fully reduced forms of the flavin. An excellent example of this stabilization is displayed by the enzyme flavocytochrome b2. This protein is found in the mitochondrial inter-membrane space and is responsible for the coupling of L-lactate dehydrogenation to the reduction of cytochrome c (Figure 1.3). Flavocytochrome b2 contains both a flavin binding domain and a heme binding domain that are linked by a short peptide hinge (Xia & Mathews, 1990). Upon reduction of the F M N group to the fully reduced state by the two electrons generated from the L-lactate dehydrogenation, the reduction potentials associated with the flavin group are shifted to +71 mV (E°, Efl o x-Efl s q') and -133 mV (E 2°, Efljq'-EflredH) (Tegoni et al, 1986). These values result in the stabilization of the semiquinone form and ensure that only a single electron is transferred to the heme group (E 0 = -18 mV) (Figure 1.6). The rate constant of electron transfer from the flavin to the flavocytochrome b2 heme is >1500 s"1 (Chapman et al., 1994; Hazzard et al, 1994) and reflects the moderate thermodynamic driving force of this reaction (100 mV). The reduced heme group then rapidly donates this electron to a cytochrome c molecule. The transfer of the second electron from the flavin semiquinone to the heme group of the flavocytochrome b2 is impeded because the potentials of the two centers are now +71 mV (Eflox-Eflsq") and -18mV (heme). For the second electron to be transferred, it is believed that the bound pyruvate product must dissociate from the enzyme. The F M N potentials in the pyruvate-free enzyme shifts to -16 mV (Eflox-Eflsq") and -60 mV (Efl^'-Efl^H), thus allowing the transfer of the 14 Efl o x /Ef l s Efl s q/Efl, E° red heme -16mV -60mV -16mV F1H Lactate F1H Fl H-Pyruvate Ef i o x /Ef l s q =+71mV E f l ^ / E f U =-133mV heme = -18mV Fl H-Pyruvate Pyruvate ^ C c Figure 1.6 Model of catalytic cycle of flavocytochrome b2. Numbers denote order in which steps in cycle proceed. Reduction potential values given are for the flavin and heme redox centers at that particular step in the cycle. Abbreviations: Fl, flavin mononucleotide; H, flavocytochrome Z>2-heme; Cc, cytochrome c,electron (Modified from Daff et al, 1996). 15 second electron to the heme group (E 0 = -16±5 mV). This second electron transfer reaction is relatively slow (120 s'1) and is thought to be the rate limiting step of the flavocytochrome b2 catalytic cycle (Daff et al., 1996). This process clearly demonstrates the ability of flavocytochrome b2 to mediate electron transfer between a one electron and a two electron process. Numerous studies have been conducted to probe the active sites of flavoproteins and investigate the factors that regulate the reduction potential of the bound flavin. For a recent review of these studies see Mayhew et al. (1996). In addition to this body of work, flavins have been attached to various proteins that are otherwise devoid of such a cofactor. These synthetic flavoproteins have enabled the investigation of such processes as substrate oxidation/reduction (Levine & Kaiser, 1978; Hilvert et al., 1988; Mihara et al., 1993, Shumyantseva et al., 1996) and D N A repair (Carell & Butenandt, 1997). From studies of this type, it is apparent that flavin modification of proteins can introduce novel and useful functional attributes that may offer theoretical insight into existing flavin mediated processes. 1.6 Remodeling of Heme Protein Active Sites As with flavins, the protoheme IX prosthetic group is found in a large number of proteins and participates in a wide variety of biological functions. These functions include electron transfer (cytochromes), oxygen storage and transport (myoglobins and hemoglobins), substrate oxidation (peroxidases) and substrate hydroxylation (mono-oxygenases). This functional diversity results from the protein environments in which the heme prosthetic group is located (e.g. Dawson, 1988). Consequently, an understanding of the manner in which these proteins regulate the reactivity of the heme is of fundamental importance. One approach to evaluating the effect of environment on heme reactivity has been to introduce structural features of one heme protein into the active site of another 16 by site-directed mutagenesis. Examples of such studies include the engineering of the active site of myoglobin to mimic the structural attributes of cytochrome b} and catalase by replacement of the distal residue Val68 with His (Lloyd et al, 1995) and the proximal His93 residue with tyrosine (Hildebrand etal, 1995), respectively. Cytochrome c has also proven to be amenable to studies of this type. Initial studies involving the alteration of the axial ligands to the heme iron produced cytochromes of varying functionality (Hampsey et al, 1986; Sorrell et al, 1989). Replacement of the Hisl8 axial ligand was particularly intriguing but unfortunately these variants could not be obtained in sufficient quantity to permit detailed spectroscopic studies, so the coordination environment of the Hisl8Arg variant remains equivocal. This problem was overcome to an extent by the technique of chemical semi-synthesis. This strategy was pioneered by Offord and Wallace (Rees & Offord, 1976; Wallace, 1993) and allows the production of reasonable quantities of modified protein. Recent studies (Raphael & Gray, 1989; Raphael & Gray, 1991; Wallace & Clark-Lewis, 1992) have concentrated predominantly on the replacement of the axial Met80 ligand with both naturally occurring residues (His, Leu, Cys, and Ala) and with non-coded amino acids (selenomethionine, cyanoalanine and norleucine). Of particular interest among these variants was the Met80Cys cytochrome c. This variant exhibited a shift in reduction potential of well over 600 mV to -390 mV. This value is remarkably similar to the potential of cytochrome P-450 (-360 to -170 mV) (Huang et al, 1986) in which the axial ligands are a Cys residue and a water molecule. A second axial ligand cytochrome variant exhibiting notable characteristics possessed the Met80Ala substitution. This protein displayed spectroscopic properties of a penta-coordinated, high-spin heme and displayed the ability to bind oxygen in a manner similar to myoglobin (Wallace and Clark-Lewis, 1992). Interestingly these authors reported a reduction potential of 185mV for this 17 variant. Unfortunately this value was determined by a single point equilibrium technique due to limited quantities of material (Wallace and Clark-Lewis, 1992) and was thus subject to a large experimental error. Further work on this variant was pursued by various groups and led to improved yields from both semisynthetic techniques (horse heart cytochrome c) (Bren and Gray, 1993 a, b) and site-directed mutagenisis (Yeast iso-l-cytochrome c) (Lu et al, 1993). This achievement enabled extensive spectroscopic characterization of this variant (Banci etal, 1995b; Bren etal, 1995; Bren, 1996) and culminated in the elucidation of the three-dimensional solution structure of the cyanide adduct of the protein (Banci et al, 1995a) (Figure 1.7). These studies all demonstrated that the Met80Ala cytochrome c variant could bind many small exogenous ligands. The oxygen and carbon monoxide derivatives of the reduced variant actually possessed greater ligand affinities than myoglobin. Specifically, Met80Ala was observed to bind 0 2 with an affinity o£K= 2.6 ±1 .4 uM"1, in comparison to Mb which bound with a value ofK= 0.4-1.2 uM"1 (Antonini and Brunori, 1971). Moreover the Met80Ala variant was found to auto-oxidize with a rate constant of kox = 0.01 h'1 (Bren and Gray, 1993a), a value much lower than that observed for Mb (kox = 0.08-0.22 h"1) (Brown and Mebine, 1969). It was suggested that the relatively high affinity for dioxygen is attributable to a hydrogen-bond involving the ligand and a nearby tyrosyl residue (Tyr67) (Figure 1.7). A similar stabilizing interaction is observed in Ascaris hemoglobin (De Baere et al, 1994, Kloek et al, 1994). The involvement of this Tyr residue was also implicated in the reduction of the pKa of the coordinated water to a value of 6.5, 3 pK units lower than that for myoglobin. Of further note is the suggestion that the residue Phe82 may play a role in the stabilization of the oxygenated derivative. This stabilization may result from favorable interactions between the Phe residue and the heme bound dioxygen group. A stabilizing effect of this type is seen in the 18 Figure 1.7 Structural comparison of the active sites of the cyanide adduct of Met80Ala cytochrome c (Banci et al, 1995b) and oxy sperm whale myoglobin (Phillips, 1980). The hydrogen bond between the ligand and residue Tyr67 in Met80Ala cytochrome c is denoted by the dotted line. 19 myoglobin variant Leu29Phe (Carver et al, 1992). In addition to these active site interactions, the ligand binding site is considered to be protected from the exterior solvent by the surrounding protein matrix. These steric factors decrease the ability of the bound ligand to dissociate from the protein into the surrounding solvent (Bren and Gray, 1993). 1.7 Electron Transfer Kinetics 1.7.1 Overview Electron transfer reactions are the simplest form of chemical reactions. As such they are amenable to both in depth theoretical description and to detailed mechanistic evaluation. Electron transfer reactions can be divided into two categories, namely intermolecular and intramolecular. Intermolecular reactions involve the transfer of electrons between donor and acceptor sites located in two separate molecules. Intramolecular reactions involve electron transfer between donor and acceptor sites within a single molecule. For the purpose of this thesis, the introduction to the theory and associated experimental work will be organized under these two subheadings. 1.7.2 Intermolecular Electron Transfer 1.7.2.1 Theoretical Studies Intermolecular electron transfer reactions can be classified as either inner-sphere or outer-sphere. Inner-sphere reactions generally involve the transfer of the electron via an intermediate in which the reactants share a bridging ligand in their inner coordination shells. In contrast, outer-sphere reactions involve electron transfer without modification of the ligands coordinated. Outer-sphere reactions are more biologically relevant since electron transfer between proteins generally occurs through weak through-space interactions. 20 Bimolecular outer-sphere electron transfer reactions can be represented by the following expression (Scott et al, 1985) D + A ( D - - A ) — ( D + - - A " ) <-^-> D + + A" [1] where D and A are the electron donor and acceptor respectively. Initially, the donor and acceptor reagents form the non-covalent precursor complex (D—A). The electron transfer then produces the successor complex (D+—A') which then dissociates to generate the separate oxidized and reduced products. The rate of electron transfer for these intermolecular reactions is therefore not only dependent upon the factors that dictate ket but also on those which affect the formation of the precursor complex. These latter factors include diffusion rates, electrostatic interactions and hydrophobic interactions. In many cases, these factors are major determinants of the overall rate of reaction. To aid in understanding the concepts involved in the electron transfer reaction (£ e t), the potential energy states of the nuclei in the complex can be visualized as two dimensional curves (Figure 1.8). These simplified curves represent the free energy as a function of nuclear coordinates of the precursor and successor complexes. The curves are parabolic because the assumption is made that the participating chemical bonds obey Hooke's law and act as simple harmonic oscillators. To satisfy the Franck-Condon principle, as applied to electron transfer reactions (Libby, 1952), the nuclear coordinates must distort to a conformation that is energetically common to both potential surfaces. This point is the intercept of the two curves. The energy difference at this point (AE) is a measure of the electronic interactions between the donor and acceptor. For strong 21 interactions, the splitting (AE) is large and creates a high probability for the conversion of the precursor to the successor complex. At the limit of this splitting, A E becomes sufficiently large to cause unit probability for electron transfer. This electron transfer reaction is an example of an adiabatic reaction. Although these adiabatic reactions are dominant in small molecule electron transfer reactions, the majority of biological electron transfers are likely to occur via non-adiabatic behavior (Scott et al, 1985). This non-adiabatic dominance in biological systems is attributed to the weak electronic interactions between the donor and the acceptor centers that result from the relatively large distances that separate these two centers. Thus, the electronic coupling energy A E becomes small, and the probability that non cross-over will occur is increased. To reach this transition point for electron transfer (the intercept), the system must have sufficient vibrational energy to overcome the activation barrier (AG*). According to transition-state theory, this energy barrier dictates the rate of cross-reaction (k12), as defined by the following expression: kl2 = RT/Nh exp <*at/RT) [2] where the term RT/Nh represents the vibrational frequency of the reactants. The activation free energy (AG*) is dependent upon the reaction free energy (AG 0 ) and the reorganization free energy (X). This relationship is defined by the following equation (Marcus, 1965): AG* - W+ (1/4X) x (A + AG 0 ' ) 2 [3] The reorganization free energy (A) is the energy required to bring the nuclear coordinates of the 22 ( A — D ) (A +—D") (A—-D) (A +—D") Figure 1.8 Potential energy curves representing an electron transfer reaction ( A - - D represents the precursor complex, A + —D" represents the successor complex, A G 0 is the driving force, A G * is the activation barrier, and A E is the energy gap separating the mixed and split potential energy curves). 23 precursor complex to those of the transition state to accommodate the transfer of the electron. This energy is composed of two components, the inner reorganization energy (X-^ and the outer reorganization energy (Xa) (Marcus & Sutin, 1985). The inner reorganization energy is associated with changing bond lengths and bond angles, and the outer reorganization energy is associated with surrounding solvent dipole orientation. In Equation 3 A G 0 ' is the free energy difference between the precursor and the successor complex. This energy is dependent upon the free energy difference between the fully separated reactants and products (AG 0) (see Figure 1.8). It is also dependent upon the work terms v/ and wp which are the energies required to bring together and separate the reactant and product molecules, respectively. These latter terms are predominantly associated with the electrostatic interactions between charges species and therefore can generally be neglected when one of the reactants is of neutral charge. The approach outlined above to predict electron transfer rate constants is highly dependent upon the understanding of the structural and dynamic properties of the system. Since this knowledge is often not available, an alternative theoretical model was therefore required. This classical model, developed by Marcus, predicts cross-reaction rate constants for electron transfer reactions between substitutionally-inert coordination complexes using the corresponding parameters for the self-exchange reactions of the individual reactants. This relationship is generally referred to as relative Marcus theory, and it assumes that the activation processes associated with the cross-reaction are the same as for the individual self-exchange reactions. This relationship is defined by Equation 4. kn = Akn*k22xKnxf, [4] The parameters ku and k22 are the self-exchange rates of the reactants, and K12 is the equilibrium 24 constant for the cross-reaction. For adiabatic and uniformaly non-adiabatic reactions the term/is equal to unity. This relationship has been used in a somewhat different perspective to study bimolecular electron transfer reactions of metalloproteins and substitutionally-inert metal complexes (Wherland and Gray, 1976). In the Wherland-Gray analysis, the apparent self exchange rate constant exhibited by a protein in a reaction with a small molecule reagent is calculated from the experimentally determined cross reaction rate constant (ku), the self-exchange rate constant of the inorganic reagent (k22) and the thermodynamic driving force of the reaction (AG°). The resulting value (kueorT) is also corrected for electrostatic work terms involved in the precursor complex formation and the successor complex dissociation. This value provides information concerning the mechanism by which the protein and small molecule reagent react. 1.7.2.2 Intermolecular Electron Transfer Reactions of Cytochrome c Cytochrome c, with its well defined structures and chemical properties, has proven to be an excellent protein for the study of the factors dictating bimolecular electron transfer rates. Extensive research has been carried out in this field utilizing an array of small inorganic and organic molecules acting as electron donors and acceptors. These reactants include Fe(EDTA)2" (Hodges etal., 1974), Co(phen)33+ (McArdle et al., 1977), Ru(NH3)5py (Cummins & Gray, 1977), Co(dipic)2 (Mauk & Gray, 1979), quinol semiquinones (Rich & Bendall, 1980) and flavin semiquinones (Tollin et al., 1986). Of particular importance to the work in this thesis are the studies involving the flavin semiquinones and the Fe(EDTA)2". The use of flavins in the study of bimolecular electron transfer reactions of metalloproteins has developed primarily through the efforts of Tollin & Cusanovich (Jung and Tollin, 1981; Ahmad 25 etal, 1981, Simondsen and Tollin, 1983). This strategy is based on the fact that the flavin triplet state (3F1) can be populated in a few nanoseconds with a high quantum efficiency by inter-system crossing from the first excited singlet state ('Fl). This triplet state is a strong oxidant with the ability to remove a proton or electron from a donor such as EDTA. This process produces the neutral flavin semiquinone, a kinetically unstable, strong reductant which can transfer an electron to an electron transfer protein. The E D T A radical formed in this process undergoes rapid decarboxylation to generate a second neutral flavin semiquinone with the E D T A fragmenting to produce stable products (Traber et al, 1982). In the absence of an electron acceptor, the flavin semiquinones will disproportionate to produce the oxidized and fully reduced flavin species. Consequently, for protein reduction to predominate in the reaction kinetics, the protein concentration must be sufficiently high to prevent semiquinone disproportionation. The use of structurally varied flavin derivatives in studies of this type has allowed the investigation of the steric, electrostatic, and electrochemical properties of the flavin on the rate of electron transfer (Tollin et al, 1986; Cusanovich et al, 1987). Similarly, knowing that the flavin semiquinone transfers the electron to the exposed heme edge of the protein (Meyer et al, 1984; Prysiecki et al, 1985), it has been possible to probe the environment around this heme edge. An excellent example of this approach is the reaction of F M N with the cytochrome c 2 from Rhodospirillum rubrum (Meyer et al, 1984). Despite this protein having a net charge of zero, the reactivity of this protein with a variety of flavins clearly indicates that the heme edge is located within a positive electrostatic surface. This importance of asymmetrical charge distribution around the protein surface is also apparent with the c-type cytochrome from R. spheroides (Weber & Tollin, 1985). A more subtle study involves the differing steric factors associated with residues that surround the exposed heme edge (Meyer et al, 1984), in which a clear correlation between the experimental 26 rate constants and the distance of the heme to the protein surface was reported. The use of flavins has also been extended to investigation of protein-protein electron transfer reactions. These studies have included analysis of the complexes formed by cytochrome c with cytochrome c oxidase (Ahmad et al, 1982), cytochrome c peroxidase (Hazzard et al, 1987, 1988), flavodoxin (Hazzard etal, 1986) and cytochrome b5 (Eltis et al, 1988). This work has aided in the understanding of the electrostatic interactions and dynamic motions within these electron transfer complexes. Small, substitutionary inert inorganic reagents have also been used extensively in the study of intermolecular electron transfer reactivity of metalloproteins. Fe(EDTA)2" has been used extensively and has proven useful in studying not only cytochrome c (Wherland & Gray, 1976; Cummins & Gray, 1977; Holwerda et al, 1980; Rafferty et al, 1990), but also other heme proteins such as cytochrome b5 (Reid & Mauk, 1982), cytochrome c 5 5 1 (Coyle & Gray, 1976) and myoglobin (Mauk & Gray, 1979; Lim & Mauk, 1985). The use of the relative Marcus formalism to derive the electrostatic-corrected self-exchange rate constants (knco") enables the electron transfer ability of these various proteins to be compared. In addition, this approach facilitates the comparison of variants of the same protein. The work of this type with cytochrome c and numerous variants has led to the suggestion that electron transfer reactions with Fe(EDTA)2* are dependent upon two major factors. These factors are the orientation and proximity of the Fe(EDTA) 2' molecule relative to the heme and the reorganization energy of the cytochrome. To understand the importance of the Fe(EDTA)2" orientation and proximity, the nature of this reagent must be considered. Fe(EDTA)2" has the ability to present two faces towards the cytochrome heme region, a hydrophilic face composed of the E D T A carboxyl groups, and a hydrophobic face consisting of the ethylene backbone. The more favorable 27 orientation for electron transfer would place the hydrophilic face against the protein to allow overlap of the %-n carboxyl orbitals with the corresponding heme orbitals. This orientation is thought not to occur owing to the hydrophobic nature of the heme crevice environment. The hydrophobic nature of this crevice causes the Fe(EDTA)2" to present its hydrophobic surface. Electron transfer studies involving cytochrome c variants with mutations at the Phe82 residue have supported this hypothesis in that more polar residues at this position result in enhanced electron transfer reactivity (Rafferty et al., 1990). This work also demonstrates that as the residue at position 82 becomes less bulky the rate of electron transfer increases presumably because Fe(EDTA)2"is able to approach the heme in such variants more closely. The reorganization energy of the protein also has an integral role in controlling electron transfer rates. Cytochrome c, unlike myoglobin, does not undergo an oxidation state linked change in coordination environment. Therefore, the reorganization energy of this protein is relatively low because no bonds have to be broken and changes in bond lengths are minimal. These characteristics are reflected in the kncon values for cytochrome c and myoglobin that are exhibited in the reactions of these proteins with Fe(EDTA) 2'. These values are 6.2 M ' V 1 (Wherland and Gray, 1976) and 0.02 M ' V 1 (Lim, 1990), respectively. Indeed a recent study of the myoglobin variant Val68His, which is six coordinated in both oxidation states (Lloyd etal., 1995), has demonstrated that the kncotT value increases by over one order of magnitude (Harris et al., 1997). This observation confirms the importance of the oxidation state linkage of coordination number on electron transfer reactions involving heme proteins. Replacement of several internal amino acids in cytochrome c lead to varying effects on the electron transfer reactivity of the protein with Fe(EDTA) 2' (Rafferty, 1992). These variations in reactivity are presumably brought about by changes in the reorganization energy of the protein. The 28 variants exhibiting this type of altered reactivity all possess changes in the hydrogen bond network involving an internal water molecule, Wat 166, which has been implicated in the oxidation state-linked conformational change of cytochrome c. Of particular note is the Tyr67Phe variant which exhibits a ten fold increase in reactivity towards Fe(EDTA) 2'. This variant contains a second internal water molecule in place of the hydrogen bond between the tyrosyl hydroxide and the axial Met80 sulfur atom (Berghuis et al., 1994). This structural change is thought to lower the reorganization barrier because movement of this water requires less energy than movement of the side chains involved in the hydrogen bond network found in wild-type cytochrome c. 1.7.3 Intramolecular Electron Transfer 1.7.3.1 Theoretical Studies Long-range electron transfer within proteins can be considered to be a weak non-adiabatic interaction. This situation is a consequence of the steric constraint of the donor and acceptor so that they are unable to interact directly. The rate constants associated with this type of non-adiabatic electron transfer process are described by Fermi's golden rule (Equation 5) (Marcus and Sutin, 1985). ^ t = 2u/hxH A B 2 x(FC) [5] This expression contains both an electronic and a nuclear term, with the nuclear element being the Franck-Condon factor (FC). As with intermolecular electron transfer reactions, the application of the Franck-Condon principle to these reactions is based on the assumption that the nuclei of the donor and acceptor act as simple harmonic oscillators, and can be regarded to be the sum of the overlap of the vibrational wavefunctions of the donor and acceptor centers. The classical 29 expression describing this overlap includes the driving force (-AG0) and the reorganization energy (A) associated with the reaction as shown below (Equation 6): FC = (4:iAkBT)- ,/j exp[-(X+AG°)2/4AkBT] [6] Examination of this expression reveals a parabolic dependence of In ket against driving force (-AG0) such that the electron transfer rate constant increases with increasing driving force to a point where X is equal to - A G 0 . The rate constant then decreases as the driving force value increases further. This decrease at higher driving force is referred to as the Marcus inverted region and is visualized in Figure 1.9. The second element of Fermi's golden rule, the electronic factor, states that the rate of electron transfer is proportional to the square of the electronic coupling of the donor and acceptor molecules ( H ^ 2 ) . This coupling is a function of the electronic wavelength of the medium separating the donor and acceptor and as such decays exponentially with distance, (d-d0) (Equation 7): H A B = H ^ e x p ^ 2 [7] H A H 0 is the electronic coupling at d=d0 where the donor and acceptor are in direct contact and P is the exponential decay constant. P is a coefficient that reflects the nature of the medium through which the electron must travel. Typical values for this coefficient for proteins have been suggested to be around 1.4 A(Moser et al, 1992) although some experimental evidence has shown the existence of variations in P within a protein matrix (See section 1.7.3.2, Page 33 ) . Various approaches have been developed to account for the heterogeneous nature of the protein medium in an attempt to calculate the electronic coupling more accurately (Onuchic & Beratan, 1990; Christensen etal, 1990 , 1992; 3 0 Figure 1.9 The dependence of electron transfer rate on free energy. The three sets of potential energy curves demonstrate the effect of varying the driving force on the activation barrier for electron transfer (-AG 0 is the thermodynamic driving force, -AG* is the activation barrier and A is the reorganization energy). 31 Kuki, 1991; Goldman, 1991; Siddarth & Marcus, 1990, 1992). These approaches provide strategies to represent the overlap of the molecular orbitals of the donor, acceptor and protein medium. One of the most extensively used model is that developed by Beratan, Betts, and Onuchic. (Beratan etal, 1990, 1991). This approach distinguishes through atom (covalent and hydrogen bond) from through space interactions which could be potentially involved in an electron transfer pathway. A separate P value is assigned to each type of connection and calculates a dominant pathway with a graph-search algorithm based on the overall electronic coupling values. An alternative theoretical approach to describe the heterogeneous nature of the intervening medium has been developed by Siddarth and Marcus (Siddarth & Marcus, 1990,1993a), through use of one-electron extended Huckel theory (Hoffman, 1963; Larsson, 1981) to calculate the electronic coupling between the donor and the acceptor so that any constructive or destructive interference between various orbitals can be accounted for. Unfortunately, even for a small protein such as cytochrome c, calculating the electronic coupling associated with all the atoms would require the use of a huge matrix (Siddarth & Marcus, 1993a, b). To overcome this problem a subset of amino acid residues which are important in mediating the electron transfer are initially selected by an artificial intelligence (AI) search. This computationally more manageable set of residues can then be used to calculate the electronic coupling between the donor and acceptor. Comparison of the two models mentioned above reveals that the latter approach is theoretically more advanced. Firstly, the method of Beratan, Betts, and Onuchic assigns a uniform value for the decay factor across any bond. In the method of Siddarth and Marcus the electronic coupling between pairs of atoms is calculated explicity, therefore accounting for the different nature of each atom. A second difference is that the two methods utilize different search procedures. The method of Siddarth and Marcus uses a broader set of atoms to start the search, while the Beratan, 32 Betts, and Onuchic method finds a single path between the donor and acceptor and then searches the region surrounding the acceptor for alternative pathways. 1.7.3.2 Intramolecular Electron Transfer Reactions of Cytochrome c Experiments providing information regarding long-range electron transfer generally involve proteins for which structural information is available to permit determination of the distance between the donor and acceptor centers. It is not surprising that cytochrome c features predominantly in such studies. The major approach employed in these studies is to attach a small pendant molecule to a surface amino acid residue that can act as an electron donor or acceptor. The attached group must be stable and non-labile in both oxidation states, and the attachment procedure should not perturb the structural or functional properties of the protein. With these considerations in mind, tethered coordination complexes have been used extensively in such studies. A comprehensive review concerning this literature is given by Scott (1996). The first work of this type involved tethered ruthenium complexes and was reported independently in 1982 by the groups of Gray and Isied (Isied et al, 1982; Winkler et al, 1982; Yocom et al, 1982). In these studies the imidazole side chain of the residue His33 of horse heart cytochrome c was used to attach a Ru(NH 3) 5 group to the protein surface. These early ruthenium ammine complexes were devoid of excited-state properties necessary for photo-induced electron transfer. For this reason pulse radiolysis and a separate photogenerated Ru(bpy)32+' molecule were used to initiate electron transfer through rapid reaction with the tethered ruthenium complex. A schematic representation of this type of reaction is given in Scheme 1. 33 Ru(bpy)32+ hv Ru(bpy)32+- + Ru(TII)-Fe(m) > Ru(n)-Fe(IIJ) K f Ru(m>Fe(II) Scheme 1 Scheme 1 is a simplification of the possible reactions and assumes the presence of a sacrificial donor, such as EDTA, to scavenge the Ru(bpy)33+ produced in the reaction. Subsequent studies involved zinc-substituted porphyrin derivatives of heme proteins than can be photoexcited to a reducing triplet state and can donate an electron to the attached ruthenium complex (Peterson-Kennedy et al, 1984; Meade et al, 1989). Later work demonstrated that the attachment of ruthenium complexes containing the bipyridine ligands could lead to direct electron transfer from the attached donor to the protein heme iron following flash photolysis (Chang et al, 1991; Millett & Durham, 1991). Variations in the ligands of the ruthenium and other metal complexes of this type led to the development of numerous modification procedures involving a variety of protein surface residues. These strategies included modifications of lysyl residues (Pan et al, 1988; Durham et al, 1989; Scott et al, 1991, Gorren etal, 1992), glutamic and aspartic residues (Conrad & Scott, 1989), cysteinyl residues (Geren etal, 1991) and tyrosyl residues (Lui, 1994). Cytochromes c modified in this manner exhibit electron transfer rate constants ranging from ~1 s 1 (Isied, 1990, Scott et al, 1991) to ~107 s"1 (Durham et al, 1989, Millett & Durham, 1991). 34 Such a wealth of experimental data has enabled the investigation of the factors theoretically predicted to dictate the rate of electron transfer within proteins (Marcus & Sutin, 1985). As outlined in equations 5, 6, and 7, the rate of electron transfer is dependent upon three major factors: the electronic coupling between the donor and acceptor (H^), the driving force associated with the reaction (-AG0) and the reorganization energy of the system (X). The electronic coupling component, which is a function of the distance and the protein medium intervening between the donor and acceptor has been extensively studied,. Work of this nature has been significantly aided by the advent of site-directed mutagenesis (Zoller & Smith, 1983, 1984). Initial work on this subject involved modified sperm-whale myoglobin with ruthenium complexes attached at histidines 12,48, 81, and 116 (Lieber etal., 1987, 1988) that led to correlation of ket with the donor-acceptor distance. From these studies a value of-0.9 A"1 was derived for the exponential decay constant (p) associated with the protein medium. A short time later, similar studies were undertaken with cytochrome c (Bowler et al., 1989; Therienetal., 1990; Change/al., 1991; Wuttkeetal., 1992; Casimiro etal, 1993;Karpishine/a/., 1994). Pendant ruthenium complexes were again attached at surface histidines at positions 33, 39, 54, 58, 62, 66, 72, and 79 (Figure 1.10). A number of these histidines had to be introduced by mutagenesis techniques. Comparison of kmax (ket when - A G 0 = X) with d-3 (the donor-acceptor distance minus the van der Waals contact of 3A) showed a definite correlation between the two factors (Figure 1.10). For reference the experimental date was compared to a line which utilized a p value of 1.4 A. This value was derived from studies that suggesting the protein matrix could be treated as a homogenous medium (Moser & Dutton, 1992; Moser et al, 1992). From the experimental rate constants it was found that the electronic coupling was actually weaker than that corresponding to a purely covalently linked system. This work therefore provided experimental 35 evidence for the necessity of theoretical models to account for the heterogeneity of proteins. Indeed when the model developed by Beratan, Betts, and Onuchic was applied to these experimental systems the calculated 'best-pathways' yielded effective o-tunneling lengths (ol) which correlated well with the maximum electron transfer rates (Figure 1.10) (Beratan et al, 1992; Onuchic et al, 1992). Similarly, the more sophisticated model of Siddarth & Marcus (Figure 1.10) also provided excellent agreement between experimental and calculated electronic coupling energies. (Siddarth & Marcus, 1993a, b). In addition to the dependence of electron transfer rates on electronic coupling, the influence of the driving force (-AG0) on the rate of electron transfer has also received a great deal of attention. Systems have been studied in which the driving force ranges from as low as 0.1 V (Isied et al, 1982) to as high as 1.9 V (Mines et al, 1996) well into the Marcus inverted region. Two particularly noteworthy examples of such studies involve cytochromes c labeled at position His33 (Mines etal, 1996) and a mutated His39Cys (Fairris et al, 1996) site. For the modified His33 cytochrome c the driving force was varied from 0.54 to 1.89 V simply by altering the ligands of the ruthenium complex, and the rates of electron transfer exhibited by these proteins clearly demonstrated the parabolic nature of the dependence on driving force (-AG0). The rates associated with the larger - A G 0 values decreased in a manner consistent with the existance of the Marcus inverted region. Studies of this type also enabled estimates of the systems reorganization energy (X) to be made. Early experimental work involving ruthenium-ammine modified cytochromes c gave a reorganization energy value of 1.19 eV (Meade et al. ,1989). Further studies corroborated this value, with a general convergence of opinion on a value of ~1.2 eV for the reorganization energy of ruthenium-ammine modified cytochromes c (Onuchic et al., 1992). Later work involving ruthenium-bpy complexes predicted reorganization energy values ranging from 0.8 to 1.0 eV (Mines etal, 1996; 36 Figure 1.10 (Top Panel A) Electron transfer pathways calculated from Beratan, Betts, & Onuchic theoretical model in Ru(HisX) modified cytochromes c (X=33, 39, 62, 72) (Modified from Beratan et al, 1992). Insets a and b show 'best-pathways' as calculated using Siddarth & Marcus model for Ru(His33) and Ru(His72) cyt c respectively (Modified from Siddarth & Marcus, 1993a). (Bottom Panel B) Correlations of maximum E T rates in Ru(HisX) modified cyts c with (d-3) (a) and al (b) (Modified from Wuttke et al, 1992 and Casimiro etal, 1993). 37 Fairris et al, 1996). These values were approximately one third lower than the estimates involving ruthenium-ammine modified cytochromes c. This observation was consistent with the smaller reorganization energy expected for the ruthenium-bpy complexes (Brown & Sutin, 1979). 1.8 Objectives of this Dissertation The objective of this thesis was to investigate the factors that dictate both intermolecular and intramolecular electron transfer within proteins through the study of two novel types of cytochrome c. The first of these approaches involved the use of a new technique to study intramolecular electron transfer reactions in which a flavin moiety was attached to the protein surface. Previous studies by Tollin, Cusanovich and co-workers have used flavins extensively to investigate intermolecular E T reactions, but the use of flavins as electron donors in intramolecular electron transfer reactions has not been reported. The major advantage of this new technique is that the position of the flavin electron donor at the moment of electron transfer is known from the specificity of the modification chemistry. Flavin-modified cytochromes also permit the assessment of the contributions of the unique properties of the flavin donor to the kinetics of electron transfer. Specifically, these novel cytochrome derivatives provide a means by which the reorganization energy of the flavin can be estimated. In principle, the insight gained from these studies should be of value in understanding the basic features of electron transfer reactions associated with naturally occurring flavocytochromes such as yeast flavocytochrome Z>2 and cytochrome P450BM3. The kinetic study of four such flavin modified cytochrome c variants, namely Thr8Cys, FIis39Cys, Asn62Cys and Leu85Cys, was undertaken. To complement these results the theoretical 38 models of Beratan, Betts, & Onuchic and Siddarth & Marcus were utilized. These models enabled the calculation of the best pathways and the donor-acceptor electronic coupling. In addition a variety of spectroscopic and electrochemical techniques were used to investigate the structural and functional properties of these synthetic flavocytochromes. The second system described in this thesis involved the study of three axial ligand variants of cytochrome c, namely Met80Ala, Met80Ala/Tyr67Phe and Met80Ala/Phe82Ser. The analysis of the intermolecular electron transfer reactions of these three variants was undertaken by studying their reaction with the reductant Fe(EDTA) 2'. This work enabled the investigation of the electron transfer reactivity of a ligand-binding coordination environment similar to that of myoglobin at the active site of cytochrome c. Again, to aid in understanding the kinetic results various techniques were employed to study the functional properties of these variants. These studies included analysis of thermal stability, ligand binding, and electrochemical properties. Some of the work described in this thesis has been published (Twitchett et al, 1997). The work involving the energy-minimized structures of the flavin-modified variants and the electron transfer pathway calculations using the Siddarth-Marcus model were done by Dr. P. Siddarth. 39 2. E X P E R I M E N T A L P R O C E D U R E S 2.1 Yeast Expression System 2.1.1 Mutagenesis Techniques The expression of the plasmid (pING4) containing the gene CYC1 for yeast iso-\-cytochrome c and the oligonucleotide-directed mutagenesis techniques used to construct the Thr8Cys and His39Cys variants have been described previously (Pielak et al, 1985; Inglis et al., 1991: Zoller & Smith, 1984; Kunkel, 1985). A plasmid map of pING4 is given in Figure 2.11, and the sequence of the CYC1 gene is given in Figure 2.12. The oligonucleotides used in preparing these variants (Table 2.1) were synthesized with a modified Applied Biosystems 3 80A DNA synthesizer at the UBC Nucleic Acid and Protein Synthesis laboratory. Each mutant gene was sequenced in its entirety after mutagenesis to confirm that no additional mutations had been introduced. The plasmid encoding the gene for the Leu85Cys variant was prepared previously by G. Guillemette in the laboratory of Prof. M . Smith at the University of British Columbia. All variants also possessed the Cysl02Thr mutation to avoid complications of rapid auto-reduction and protein dimerization (Cutler etal, 1987). Table 2.1 : Mutagenic oligonucleotides for yeast expression system* Thr8Cys 5 ' -CT-AGT-CTT-GAA-AAG-ACA-AGC-ACC-TTT-CTT-A-3 ' His39Cys 5 ' -GC-TTG-ACC-AGA-GCA-TCT-GCC-AAA-GAT-A-3' * The 5' end was phosphorylated. The underlined bases indicate the mutation site 2.1.2 Protein Preparation Mutant CYC1 genes were transformed into yeast strain GM3C-2, and cultures were grown on a YP medium (bactopeptone 2% w/v, yeast extract 1% w/v, glycerol 3% v/v, and lactate 1% v/v) 40 EcoRI Figure 2.11 Physical map of pING4 for the expression of yeast /so-1-cytochrome c in yeast. The genes encoding cytochrome c (CYCT), the P-lactamase (Amp1), the leu2 selectable marker and the origin of replication of the yeast '2u* circle are shown. 41 Smal UAS 1 CCCGGGAGCAAGATCAAGATGTTTTCACCGATCTTTCCGGTCTCTTTGGCCGGGGTTTACGGACGATGACCGA GGGCCCTCGTTCTAGTTCTACAAAAGTGGCTAGAAAGGCCAGAyAAACCGGCCCCAAA'rGCCTGCTACTGGCT -380 * * * * * * UAS 2 AGACCAAAGCGCCAGCTCATTTGGCGAGCGTTGGTTGGTGGATCAAGCCCACGCGTAGGCAATCCTCGAGCA TCTGGTTTCGCGGTCGAGTAAACCGCTCGCAACCAACCACCTAGTTCGGGTGCGCATCCGTTAGGAGCTCGT * * * * * * * GATCCGCCAGGCGTGTATATAGCGTGGATGGCCAGGCAACTTTAGTGCTGACACATACAGGCATATATATAT CTAGGCGGTCCGCACATATATCGCACCTACCGGTCCGTTGAAATCACGACTGTGTATGTCCGTATATATATA * * * * * * * * GTGTGCGACGACACATGATCATATGGCATGCATGTGCTCTGTATGTATATAAAACTCTTGTTTTCTTCTTTT CACACGCTGCTGTGTACTAGTATACCGTACGTACACGAGACATACATATATTTTGAGAACAAAAGAAGAAAA * * * * * * * CTCTAAATATTCTTTCCTTATACATTAGGTCCTTTGTAGCATAAATTACTATACTTCTATAGACACGCAAAC GAGAT T T AT AAGAAAG GAAT AT GT AAT C CAG GAAAC AT C GT AT TT AAT GAT AT GAAGAT AT CT GT GC GT T T G * * * * * * * -5 10 M T E F K A G S A K K G A T L F_ ACAAATACACACACTAAATTAATAATGACTGAATTCAAGGCCGGTTCTGCTAAGAAAGGTGCTACACTTTTC TGTTTATGTGTGTGATTTAATTATTACTGACTTAAGTTCCGGCCAAGACGATTCTTTCCACGATGTGAAAAG * * i * 20 * 40 20 30 _K T R C L Q C H T V E K G G P H K V G P JV L H G_ AAGACTAGATGTCTACAATGCCACACCGTGGAAAAGGGTGGCCCACATAAGGTTGGTCCAAACTTGCATGGT TTCTGATCTACAGATGTTACGGTGTGGCACCTTTTCCCACCGGGTGTATTCCAACCAGGTTTGAACGTACCA * 60 * 80 * 100 * 120 40 50 _I F G R H S G Q A E G Y S Y T D A N J K K JV V L_ ATCTTTGGCAGACACTCTGGTCAAGCTGAAGGGTATTCGTACACAGATGCCAATATCAAGAAAAACGTGTTG TAGAAACCGTCTGTGAGACCAGTTCGACTTCCCATAAGCATGTGTCTACGGTTATAGTTCTTTTTGCACAAC * 140 * 160 * 180 * 60 70 80 _W D E JV JV M_S E Y L T JV P K K Y J P G T K M_ A F_ TGGGACGAAAATAACATGTCAGAGTACTTGACTAACCCAAAGAAATATATTCCTGGTACCAAGATGGCCTTT ACCCTGCTTTTATTGTACAGTCTCATGAACTGATTGGGTTTCTTTATATAAGGACCATGGTTCTACCGGAAA 200 * 220 * 240 * 260 90 100 _G G L K K E K_D R JV D L I T Y L K K A T E 0_ GGTGGGTTGAAGAAGGAAAAAGACAGAAACGACTTAATTACCTACTTGAAAAAAGCCACTGAGTAAACAGGC CCACCCAACTTCTTCCTTTTTCTGTCTTTGCTGAATTAATGGATGAACTTTTTTCGGTGACTCATTTGTCCG * 280 * 300 * 320 * CCCTTTTCCTTTGTCGATATCATGTAATTAGTTATGTCACGCTTACATTCACGCCCTCCCCCCACATCCGCT GGGAAAAGGAAACAGCTATAGTACATTAATCAATACAGTGCGAATGTAAGTGCGGGAGGGGGGTGTAGGCGA * * * * * * * CTAACCGAAAAGGAAGGAGTTAGACAACCTGAAGTCTAGGTCCCTATTTATTTTTTTATAGTTATGTTAGTA GATTGGCTTTTCCTTCCTCAATCTGTTGGACTTCAGATCCAGGGATAAATAAAAAAATATCAATACAATCAT * * * * * * * * TTAAGAACGTTATTTATATTTCAAATTTTTCTTTTTTTTCTGTACAGACGCGTGTACGCATGTAACATTATA AATTCTTGCAATAAATATAAAGTTTAAAAAGAAAAAAAAGACATGTCTGCGCACATGCGTACATTGTAATAT * * * * * * * H i n d i I I CTGAAAACCTTGCTTGAGAAGGTTTTGGGACGCTCGAAGGCTTTAATTTGCAAGCTT GACTTTTGGAACGAACTCTTCCAAAACCCTGCGAGCTTCCGAAATTAAACGTTCGAA * * * * 600 Figure 2.12 D N A sequence of the Smal-Hindlll fragment of pING4 (Inglis etal., 1991). The CYC1 gene (1-330) encodes yeast iso-1 -cytochrome c (one-letter code symbol for amino acids and sequence number in italics; Smal and Hindlll restriction sites, and the upstream activating sequences (UAS) are indicated and underlined). 42 in a 50 L fermentor as described previously (Rafferty, 1992). Cell lysis and preliminary protein purification were then carried out following established techniques (Cutler et al, 1987). Final purification was achieved by ion exchange chromatography using an FPLC system fitted with a Mono-S HR 10/10 cation exchange column (Pharmacia). Protein solutions (1 mL, -15 mg/mL) were eluted with 20 mM sodium phosphate buffer (pH 7.2) using a linear salt gradient from 250 to 450 mMNaCl (Rafferty, 1992). 2.2 Bacterial Expression System 2.2.1 Mutagenesis techniques To produce the functional cytochrome c variant Asn62Cys and the three non-functional variants Met80Ala, Tyr67Phe/Met80Ala and Tyr67Phe/Met80Ala/Phe82Ser, site directed mutagenesis techniques (Zoller & Smith, 1984; Kunkel, 1985) were used with anE. coli cytochrome expression system described recently (Pollock et al, 1997). A map of the plasmid pBPCYCl(WT)/3 used for expression is given in Figure 2.13. The sequences of the CYC1 (yeast iso-l-cytochrome c) and CYC3 (cytochrome c heme lyase) genes are given in Figure 2.14. The oligonucleotides used in preparing the four variants, again synthesized at the UBC Nucleic Acid and Protein Synthesis laboratory, are given in Table 2.2. All variants contained the mutation TML72Lys because the bacterial expression system lacks the enzyme that is required to trimethylate Lys72. As with the cytochromes expressed in yeast, all cytochromes expressed in.E. coli also possessed the Cysl02Thr substitution. 43 Table 2.2 : Mutagenic oligonucleotides for E. coli expression system* L62C 5 ' -A-CTC-TGA-CAT-GTT-ACA-TTC-GTC-CCA-CAA-3' M80A 5' -CCC-ACC-AAA-GGC-CGC-CTT-GGT-ACC-AGG-3' Y67F/M80A 5' -TGG-GTT-AGT-CAA-GAA-CTC-TGA-CAT-GT-3' Y67F/M80A/F82S 5' -CTT-CAA-CCC-ACC-AGA-GGC-CGC-CTT-GG-3' * The 5' end was phosphorylated. The underlined bases indicate the mutation site The double variant Met80Ala/Phe82Ser was produced by digestion of the wild type and Tyr67Phe/Met80Ala/Phe82Ser pBPCYCl/3 plasmids with the restriction enzyme (Gibco BRL), which cut just upstream of the Met80Ala mutation site. The small fragment containing the mutations Met80Ala/Phe82Ser was then ligated with the larger fragment from the wild-type plasmid digest, as described elsewhere (Maniatis, 1989) to produce the desired variant. 2.2.2 Protein Preparation Freshly transformed E. coli (HB2151) colonies were used to initiate small scale cultures (3 mL Superbroth (Sambrooke/a/., 1989) and 100 mg/L ampicillin). These cultures were then used to inoculate large scale fermentation growths (Chemap FZ000 fermentor) as outlined in Pollock et al. (1997). The resulting culture was harvested with a Sharpies continuous-flow centrifuge. Cell lysis was performed as described previously (Hildebrand, 1996). The resulting supernatant solution was brought to 35% saturation by the slow addition of solid ammonium sulphate. The pH of the solution was monitored and kept at pH 7 by the addition of small amounts of 1M sodium hydroxide solution. To remove the resulting precipitate, the solution was centrifuged (Sorvall GS-3 rotor, 7000 rpm, 30 mins), and the supernatant was recovered and dialyzed against distilled water (20 L, 4°C). Preliminary protein purification was then carried out following established techniques (Cutler et al., 1987). Final 44 Figure 2.13 Physical map of pBPCYCl(wt)/3 for the expression of yeast iso-1 -cytochrome c in bacteria. The genes encoding cytochrome c (CYC1), cytochrome c heme lyase (CYC3) and the p-lactamase (Ampr) are shown. Promoters are shown as open triangles. 45 Kpnl G G T A C C C A T C TCGGTAGTGG GATAGGACGA G T T A A C C A C C A T C A A A C A G G A T T T T C G C C T G C A A C T C T C T CAGGGCCAGG CGGTGAAGGG AA GA AAAACC ACCCTGGCGC C C A A T A C G C A A T T A A T G C A G CTGGCACGAC A G G T T T C C C G T T A A T G T G A G TTAGCGCGAA T T G A T C T G G T C C A A T G C T T C TGGCGTCAGG CAGCCATCGG T C A C T G C A T A A T T C G T G T C G CTCAAGGCGC G A C A T C A T A A CGGTTCTGGC A A A T A T T G T G T C G T A T A A T G T G T G G A A T T G TGAGGGGATA G A A T T C A A G G C C G G T T C T G C TAAGAAAGGT T G C C A C A C C G TGGAAAAGGG TGGCCCACAT GGCAGACACT CTGGTCAAGC TGAAGGGTAT GTGTTGTGGG A C G A A A A T A A CATGTCAGAG GGTACCAAGA T G G C C T T T G G TGGGTTGAAG T A C T T G A A A A A A G C C A C T G A GTAAACAGGC A G C T A G C T T T T C A G T A A T T A T C T T T T T A G T T T G C T A A C T T T A T A G A T T A C A A A A C T T A G G T G G G T T G G T T TTGGGCAGAT CAAAAAACTA C A T C C A T G T C AGGGTGCCCA GTCATGCACG A G T G C C C C G T TATGCAGGGA GATAACGATA T G G C A G C A T C CAAACAGCCT GGCCAAAAGA G C A T C C C C A A GAGTCCAGAC AGTAACGAGT A C A A T G C T A T GGTTAGAAAG GGCAAGATTG T G G A G T C C A T GGTGCAGGTC C A C A A C T T T C AATGGGAAAA A C C G C A C A C A GATGAAAGCC GGAAACCGGG C G T A T T G A G C C C T C G T G C T C C G T C C C A T T T TAGCCAAGAA C T A C C A T T C G AGCGCAAAGC GGAACAACAA C C T C C A A C C T ACGGAGGGCC CGACGACGAA AACGGAATGC T A G A T A G T C T A G A C A A T G C T AAGGACCGGA G T C C G T C C T C T T C G T C C T C C G C C C C T T A A A T A G C C A A G T A A G A A A T A A T G A T G T C C T A G T A T A T T T T T C C A C C T T A T T A T CACAAGGTGC T A C C T T C A G G GGTACGATAC A T T C T G T G C T AGAGGCTACA T T A C T G A T T T GGGAAATTTC A G C C A A G C T T Hindm TACCGAAGAC A G C T C A T G T T A T A T C C C G C C GCTGGGGCAA ACCAGCGTGG A C C G C T T G C T C A A T C A G C T G T T G C C C G T C T CACTGGTGAA A A C C G C C T C T CCCCGCGCGT T G G C C G A T T C ACTGGAAAGC GGGCAGTGAG CGCAACGCAA T T G A C A G C T T A T C A T C G A C T GCACGGTGCA AAGCTGTGGT A T G G C T G T G C AGGTCGTAAA A C T C C C G T T C T G G A T A A T G T T T T T T G C G C C A A A T G A G C T G T T G A C A A T T A A T C A T C C G G C A C A A T T T C A C ACAGGAAACA GACCATGACT G C T A C A C T T T T C A A G A C T A G A T G T C T A C A A AAGGTTGGTC C A A A C T T G C A T G G T A T C T T T T C G T A C A C A G A T G C C A A T A T CAAGAAAAAC T A C T T G A C T A A C C C A A A G A A A T A T A T T C C T AAGGAAAAAG ACAGAAACGA C T T A A T T A C C C C C T T T T C C T T T G T C G A T C G GATCCGGCCA A A G C T A G C T A A G T T T T T A C A C T T A G T T A A A AGGGTATCGA T A C T A T G A A T T C A C A A A A A A CGGGCAAAGA T A T T G G T G G G GCAGCAGTAT A G T C G T C G T C G T C G T C G C C A C C A T C C T C T G G A A T A A A C C C G C T G A A C A A T ATGCCGGAGT T G G A C T T G C C C G T T G A T C G G A C C A T C T C C A TCTGGGAGTA T C C T T C T C C A CAACAGATGT GCGGTAGCGG CGAAGTCGCC GAAGATGCAG TAAATGAAGG GTGCTGGCAG GAAGTGCTCG ACGTGCAGCC T A A G T T G C T G A A A T T C A T G G GCTGGATGCA CCTGTGCGGC C T A C T G T T T C ACAGGCACGA C T G G A T T G T A CTCCGAGGCG TCAAGGAAGT T A G A T A C G T C T T G G A T T T C T C T A C T T T C C A CGTGGATGTC C G T C C T G C C C T G A C C C G T T T C T T G G A C C G G A T G A T C T C G G T G A T A T A C A G CCAGCGTAAG T A C G T G T A A A GCAGCCACAA T C A A T T T C A C T T T T T C A T A T A C C T T T A T C T GTGCCACGGC GGTAAAAAAC GGCGACCACG GGGCTGACAG AGACACCCGT CCAAATTGGA A A T A T C A C T C G T C G A C C T G C Figure 2.14 D N A sequence of the Ajwil-//i/idni fragment of pBPCYCl(WT)/3. The CYC1 gene encoding yeast iso-1 -cytochrome c is underlined. The CYC3 gene encoding heme lyase is double underlined. The Kpnl and Hindlli restriction sites are also indicated. 46 purification was achieved by ion exchange chromatography with an FPLC system fitted with a Mono-S HR 10/10 cation exchange column (Pharmacia) as described above. Due to the addition of NaN0 3 to the growth medium, for the four axial ligand variants, the final purified protein was the NO bound derivative. To remove the axially bound NO and produce the corresponding ferricytochrome derivatives, each of these variants was oxidized with K3[Fe(CN)6], placed under vacuum and illuminated on ice, with stirring for approximately 6 hours. The light source was a Dyna-Lume 'Sun-Lite' 1 lamp containing a tungsten argon bulb (Scientific Instruments Inc., Skokie, II.). The lamp was placed at a distance of ~2 cm from the sample. 2.3 Flavin Preparation The synthesis of 7-acetyl-10-methylisoalloxazine and 7a-bromoacetyl-10-methylisoalloxazine was carried out as outlined in Levine & Keiser (1978). A schematic representation of the synthesis procedure is given in Figure 2.15. The final products were analyzed by thin layer chromatography using aluminum oxide pre-coated T L C plastic sheets (F 2 5 4 , Type E) (BDH Chemicals, Toronto) and a chloroform/methanol (9:1) eluent. The final products were also analyzed by nuclear magnetic resonance spectroscopy using a Bruker MSL-200 spectrometer (CF 3 C0 2 D used as solvent) to validate purity. 2.4 Thiol Assay & Protein Modification To determine the relative reactivity of the surface cysteine residues that were introduced by mutagenesis, thiol assays were carried out. The introduction of the Cysl02Thr mutation to the variants ensured that the only reactive thiol present was that introduced by site-directed mutagenesis. 4,4'-dithiodipyridine (Aldrich) solution (100 uL, 0.5 mM) was added to each protein (1 mL, sodium 47 H 2 S 0 4 , H N 0 3 , 0 °C /?-chloroacetophenone O CH, a v N O , 4-Chloro-3-nitroacetophenone O C H 3 N H 2 , EtOH, A CH, H Jd I J , H 2 , P t 0 2 , H C l a N H C H , NO, 4-Methylamino-3-nitroacetophenone I  O (Alloxan) C H 3 0 B r 2 , H O A c o 7-Acetyl-10-methylisoalloxazine 0 B r C B C H 3 O O 7a-Bromoacetyl-10-methylisoalloxazine Figure 2.15 Outline of synthesis of 7a-bromoacetyl- 10-methylisoalloxazine (Modified from Levine & Kaiser, 1978). The C4A atom in the 7a-bromoacetyl-10-methylisoalloxazine structure is indicated by an asterisk. 48 phosphate buffer, pH 7.0, 0.1 M , 25 °C, protein concentration = 0.01 mM), and the absorbance increase at 324 nm was monitored with a Cary Model 219 spectrophotometer interfaced to a microcomputer (OLIS, Bogart, GA). To accomplish protein modification, Cys containing cytochrome variants (2 mL, ~3mg/mL in sodium phosphate buffer, pH 7.2, 0.02 M) were mixed with a 5-fold molar excess of 7a-bromoacetyl-10-methylisoalloxazine that was dissolved in Me 2SO (100 uL), and the reaction mixture was incubated in darkness at room temperature for 18 hrs. A 5-fold molar excess of cysteine solution (50 uL, -30 mM) was then added to remove the Cys thiol C4A adduct (See Figure 2.15) (Levine & Keiser, 1978) prior to addition of a 5-fold molar excess of flavin and incubation for an additional 18 hrs. The modified cytochrome was eluted over a Sephadex G-25 gel filtration column (15.0 x 0.5 cm) to remove excess flavin and oxidized to generate the flavin ferricytochrome c species. Residual unmodified protein (<10%) was removed by ion exchange chromatography (Mono-S HR 10/10 cation exchange column, Pharmacia) using the elution conditions outlined in Section 2.1.2 (Page 43). 2.5 Tryptic Digestion & H P L C Analysis Tryptic digestion of the cytochrome c variants was performed by addition of two separate 12.5 uL samples of trypsin solution (1 mg TPCK-trypsin/mL of 0.001 M hydrochloric acid) to the protein (1 mg protein in 500 uL of 0.2 M ammonium bicarbonate). The reagents were incubated at 37 °C for 24 hrs (the second trpysin sample was after 16 hrs). Hydrochloric acid (100 uL of 1M) was then added to stop hydrolysis, and the sample was taken to dryness with a Savant Speed Vac concentrator (Model SVC-100H). Water (500 uL) was then added and the sample was redried. Each final dried, hydrolysed sample was dissolved in 100 uL 0.05% (v/v) T F A and stored at -20 °C. Peptide maps were prepared by reverse-phase HPLC performed with a Beckman System 49 Gold equipped with a diode array detector. For each peptide map, 20 uL of sample was injected onto a C-18 reversed-phase column (Alltech 4.6 x 250 mm) that was equilibrated with 0.05% TFA. A linear gradient was established from 0 to 60% acetonitrile in 0.05% TFA over 135 minutes with a flow rate of 1 mL/min. The gradient was then increased such that over the next 15 minutes the acetonitrile concentration rose from 60 to 75%. Detection was performed by monitoring the absorbance at 210 and 280 nm. 2.6 Spectroscopic Characterization 2.6.1 Electronic Absorption Spectroscopy 2.6.1.1 General Analysis Electronic spectra were recorded with either a Cary Model 219 spectrophotometer interfaced to a microcomputer (OLIS, Bogart, GA) or a Cary Model 3E spectrophotometer. Sample temperatures were kept constant at 25 °C with a circulating thermostated water bath connected to a water-jacketed cell holder. Spectra were recorded in 1 cm pathlength quartz cuvettes (1 mL or 3 mL volume). Oxidized cytochrome derivatives were produced by the addition of NH4[Co(dipicolinate)2] with subsequent removal with a small Sephadex G-25 (Pharmacia) desalting column (0.5 x 15.0 cm). Ferrocytochrome derivatives were produced with the addition of a small amount of solid sodium dithionite (J. T. Baker Chemicals, Phillipsburg, NJ) which was stored under vacuum. Excess sodium dithionite was then removed with the desalting column described above. 2.6.1.2 Ligand binding Reduced forms of the variants possessing the Met80Ala substitution were produced by the addition of solid sodium dithionite. Exposure of the resulting deoxygenated form of these proteins 50 to air, CO and NO produced the 0 2 , CO and NO reduced derivatives respectively. The oxidized form of the proteins was produced by the addition of excess K3[Fe(CN)6] solution followed by elution over a small Sephadex G-25 desalting column (Pharmacia, 0.5 x 10 cm). 2.6.1.3 pH titration The electronic spectra of the oxidized variants with the Met80Ala substitution were monitored as the pH of the solution was adjusted from pH 4 to 11 by incremental addition of 0.1 M NaOH. Fitting the resulting data to the Henderson-Hasselbach equation for a one-proton process (Scientist, Version 2.0, Micromath, Utah) enabled the determination of the pK3 of the distal water. 2.6.2 Fluorescence Spectroscopy Fluorescence emission spectra were measured with an S L M AMTNCO SPF-500C spectrofluorometer interfaced to an IBM-AT microcomputer. Samples were prepared in 0.1 M sodium phosphate buffer (pH 7.0 at 25 °C) with a protein concentration of ~4 uM. Samples were excited at 425 nm (2 nm bandwidth), and the emission spectrum was recorded from 450 to 700 nm (2 nm bandwidth). 2.6.3 Circular Dichroism Spectroscopy Protein samples were prepared in sodium phosphate buffer (0.1 M , pH 7.0) and passed through a 0.22 urn filter prior to recording the CD spectrum to give a final concentration of ~5 uM. Circular dichroism spectra were recorded with a Jasco Model J-720 spectropolarimeter equipped with a Neslab Model RS-2 remote sensor and a Neslab Model RTE-110 circulating water bath (Neslab Instruments Inc., New Hampshire). The spectropolarimeter was calibrated with ammonium-c/-51 camphor- 10-sulphonate (Aldrich Chem. Co.). Spectra were recorded for samples placed in a cylindrical, water jacketed quartz cell (0.1 cm path length, volume 200 uL, 25 °C), and the average of three scans from 190 to 250 nm was collected. Thermal stability studies were carried out by monitoring the ellipticity at 222 nm over the temperature range 30 to 80 °C (heating rate of 50°C/hr). The thermal denaturation curves were smoothed with the Jasco 720 software filter function, and the midpoint melting temperature (TjJ was determined from the first derivative of the curve. 2.6.4 Fourier Transform Infrared (FTIR) Spectroscopy FTIR spectra were recorded with a Perkin-Elmer System 2000 spectrometer equipped with a liquid nitrogen cooled mercury cadmium telluride detector. Protein samples were concentrated to approximately 3 mM (50 uL volume in 0.1 M sodium phosphate buffer pH 7.0). To obtain the carbonyl derivative, a few grains of sodium dithionite were added, and CO gas was passed over the sample for 10 minutes. The sample was loaded into a cell fitted with CaF 2 windows and placed into a water-jacketed cell holder (Specac Inc.) that was connected to a circulating water bath (Lauda Model RS3). Spectra were recorded at 25 °C between 1900 and 2000 cm"1 to a resolution of 2 cm'1. Background correction was achieved by subtracting the spectrum of reduced wild-type cytochrome c of similar concentration. 2.7 Electrochemical Characterization 2.7.1 Cyclic Voltammetry Direct electrochemistry was performed with a three-electrode, two compartment glass cell at a gold surface modified with 4,4'-dithiodipyridine (Aldrich) (Figure 2.16). The gold disk electrode was polished prior to modification with an alumina (0.3 um)/water slurry on Mastertex® 52 reference electrode hose barbs for water-jacketed reference electrode compartment glass support rod working electrode Pt wire connected to Pt braid counter electrode also, sidearm for Ar bubbling tube sample compartment (ca. 400 uL volume) Luggin capillary F i g u r e 2.16 A schematic diagram of the cyclic voltammetry cell. The working electrode is a polished gold disk (modified with 4,4'-dithiodipyridine for measurement of protein reduction potentials). The counter electrode is a platinum wire coiled around the sample compartment capillary. The reference electrode is a saturated calomel electrode. 53 polishing cloth (Beuhler, Lake Bluff, JX). The saturated calomel reference electrode (SCE) (Radiometer, REF401) was maintained at 25 °C and connected to the sample compartment by a Luggin capillary (0.1 mm). A platinum wire counter electrode was inserted around the capillary in the sample compartment. Electrochemical measurements were performed on 0.5 mL of 0.4 mM protein solution in buffer (u=0.1 M). The sample temperature was controlled with a thermostated, circulating water bath. Electrode potentials were controlled with an programmable potentiostat (Ursar Electronics, Oxford, U.K.). The current output was recorded with an interfaced microcomputer with software developed with Labview for Windows (National Instruments) by D. Hable (Blue Moon Technical Services, Vancouver). Midpoint potentials were measured from steady-state voltammograms recorded at a sweep rate of 20 mVs"1. The error in the reduction potentials was estimated to be ±2 mV (Rafferty, 1992). The calculated reduction potentials were converted to the standard hydrogen electrode (SHE) scale as described by Dutton (1978). Errors associated with the thermodynamic parameters were calculated from weighted linear least squares fits of the data to the Eyring equation (Equation [13], Page 72). 2.7.2 Spectroelectrochernical Techniques Potentiometric titrations were performed with an optically transparent thin-layer electrode (OTTLE) described by Reid et al, (1982) (Figure 2.17). The working electrode was a semi-transparent gold mini-grid (500 lines/inch) (Buckbee-Mears Co., Minneapolis, MN). A saturated calomel electrode (Radiometer, REF401) was used as the reference electrode along with a platinum wire counter electrode. The O T T L E cell was machined from Lucite plexiglass. The water-jacketed OTTLE cell holder was thermostated at 25 °C with a Lauda Model RC-3 circulating water bath. The applied potential was controlled by a E G & G Model 173 Potentiostat/Galvanostat. Protein 54 i t Figure 2.17 Schematic representation of the optically-transparent thin layer electrode (OTTLE) cell. The observation cell is shown in an exploded view. The working electrode a gold minigrid (500 lines/inch) attached to a copper wire. The reference electrode is a saturated calomel electrode. The counter electrode is a platinum/wire electrode. 55 concentrations were ~100 uM (0.1 M sodium phosphate buffer). [Ru(NH3)6]Cl3 (recrystallized from commercial material, Alfa), phenosafrinine (Sigma Chemical Co., St Louis, MO) and 2-hydroxy-1,4-naphthoquinone (Sigma) were used as mediators at a concentration of 10% of the final protein concentration. Spectra were recorded with a Cary 219 spectrophotometer as described above. Potentials were fitted to the Nernst equation with the program Scientist (Version 2.0, Micromath, Utah). The error associated with each potential was calculated from the linear least squares fits of the data to the Nernst equation. To obtain potentials for the flavoquinone / flavosemiquinone couple (Ej0) and the flavosemiquinone / flavohydroquinone couple (E 2°), the data were fitted to the Michaelis equation [8], and the values for the semiquinone formation constant (K) were calculated from equation [9], as described previously (Clark, 1960; Draper &Ingraham, 1968; Williamson &Edmondson, 1985). The error associated with each semiquinone formation constant was calculated from weighted least squares fits of the data to the equation [8]. E h = E m + (RT/2F)ln(l + u)/(l-u) + (RT/2F) In ([1 + y(l V ) ] 1 / 2 + u / [1 + Y ( l - u 2)] 1 / 2 - u) [8] K=4Qy + \ [9] The values for Ej° and E 2° could then be calculated (25 °C) from the semiquinone formation constant using E 2° - E j 0 = 0.05916 log A". The calculated reduction potentials were converted to the standard hydrogen electrode (SHE) scale as described by Dutton (1978). 56 2.8 Kinetic Experiments 2.8.1 Flash Photolysis Techniques Kinetic experiments were performed with a flash photolysis spectrophotometer composed of an optical bench and optical components obtained from OLIS (Bogart, Georgia) and a Phase-R DL1020 dye laser (Phase-R, New Durham, NH) operated under computer control (Blue Moon Technical Services, Vancouver). The configuration of the flash optics was set up so that the dye laser pulse and axis of the sample beam were nearly coincidental (17°), as outlined by Sawicki & Morris (1981). The laser was operated with a 3 x 10"4 M ethanol solution of Coumarin 440 dye (Exciton Inc., OH) which has an emission maximum at 440 nm (wavelength range 423-462). All solutions were placed into a modified cuvette (10 mm path, 3 mL volume) and de-aerated thoroughly with a vacuum line. All samples were kept in darkness prior to analysis. For bimolecular reactions, sample solutions consisted of 0.1 mM flavin (riboflavin and lumiflavin were obtained from Sigma; 7<x-bromoacetyl-10-methylisoalloxazine and 7-acetyl-10-methylisoalloxazine were prepared as described above), 5 mM E D T A (disodium salt) and variable concentrations of cytochrome c (5-50 uM) dissolved in 0.1 M sodium phosphate buffer (pH 7.0). Upon illumination, the photo-generated flavin triplet state reacted with the E D T A to form the flavin semiquinone (approximately 0.5% of total flavin initially present). Subsequent electron transfer was monitored as changes in the absorption at 550 and 557 nm. Protein concentrations (> 5 uM) were sufficiently high relative to flavin semiquinone concentrations to ensure that the reactions proceeded under pseudo first-order conditions and that flavin disproportionation did not compete with protein reduction. First order rate constants, kobs, were calculated with the program Scientist (version 2.0, Micromath, Orem, UT). The error associated with each first order rate constant was calculated from the standard deviation of the fit of the data to a first order rate equation. Second order rate constants, k2, were calculated from weighted linear least square fits of 57 kobs versus protein concentration. For uni-molecular kinetic experiments involving the flavin modified cytochrome c samples, solutions consisted of 5 mM E D T A and modified protein (50-150 uM) dissolved in 100 mM phosphate buffer (pH 7.0; other pH values were obtained by addition of small amounts of 1 M NaOH or 1 M HC1). The reaction sequence relevant to these measurements is outlined in Scheme 2. Protein reduction was again monitored at 550 and 557 nm, and the first order kinetic data were analyzed using Scientist. The error associated with each first order rate constant was again calculated from the standard deviation of the fit of the data to a first order rate equation. EDTA Scheme 2 2.8.2 Fe(EDTA) 2 Reduction Kinetics Kinetic experiments monitoring the reduction of cytochrome c variants by Fe(EDTA)2" were performed with an OLIS stopped-flow mixing system interfaced to an OLIS RSM-1000 rapid scanning monochromator (OLIS, Bogart, GA). Temperature was controlled with a Neslab Model RTE-111 thermostated circulating water bath. Reactant solutions were made anaerobic by bubbling with nitrogen gas which had been passed through two vanadium / amalgamated zinc towers (Meites & Meites, 1948) and through a buffer solution to remove traces of dissolved oxygen and to humidify 58 the purging gas. Fe(EDTA) 2' solutions were made up as described previously (Wherland etal, 1975) by anaerobic mixing of sodium phosphate buffer-EDTA solution and ferrous ammonium sulfate solution. Protein solutions (5 uM) were deaerated similarly (sodium phosphate pH 7.0, u = 0.1 M , 25 °C). Reactions were carried out under pseudo-first order conditions with a reductant excess of at least 20-fold. Kinetic data were analysed with the program Specfit (Spectrum Software Associates, Chapel Hill, NC). The error associated with each rate constant was calculated from the standard deviation of the fit of the data to the appropriate rate equation. Second order rate constants, k2, were calculated from weighted linear least squares fits of kobs versus Fe(EDTA) 2' concentration. The errors associated with the calculated thermodynamic parameters were obtained from the weighted linear least squares fit of the data to the Eyring equation (Equation [22]), page 115). 2.9 Molecular Modeling and Pathway Calculations Energy minimized structural models for the four cysteine containing cytochrome c variants and the corresponding flavin modified derivatives were calculated. The model structures were calculated from the published structure of yeast iso-\ -cytochrome c (Louie & Brayer, 1990) by side-chain substitution and molecular mechanics energy minimization with INSIGHT (Version 2.2.0; Biosym, San Diego, CA) by Dr. P. Siddarth at the University of British Columbia. Two theoretical models were utilized to calculate the most likely pathway(s) for electron transfer between the flavin and the heme. The first of these methods was that developed of Beratan, Betts, and Onuchic (Greenpath, Version 2.2.0) (utilized by the author). The second method was that developed by Siddarth and Marcus (utilized by Dr. P. Siddarth). The results obtained from the two methods are compared. 59 3. R E S U L T S 3.1 Cysteine Variants and Flavin Modification 3.1.1 Protein Preparation For the Thr8Cys and His39Cys cytochrome c variants expressed in yeast, the yields of protein varied from 80 to 180 mg per 50 L of culture. Two cytochrome bands were resolved on the CM-Sepharose ion exchange column. The component eluted first was deamidated cytochrome c (Brautigan etal., 1978), and the second component was the native protein. The final purification step involving the FPLC Mono S column also resolved two bands, the first of these was residual deamidated protein. Throughout the purification procedure, no evidence of dimerisation was observed suggesting that the DTT was sufficient to prevent intermolecular disulfide bond formation. For the Asn62Cys variant expressed in the E. coli system, the yield of bacterial cells was usually 4 g/L of culture from which approximately 5 mg/L of cytochrome could be isolated. Although cytochrome expressed in E. coli is not trimethylated at Lys72, the behavior of the protein expressed in bacteria during purification was identical to that of protein expressed in yeast. 3.1.2 Flavin Synthesis NMR spectra of the 7-acetyl-10-methylisoalloxazine and 7a-bromoacetyl- 10-methylisoalloxazine synthesized in this work are given in Figure 3.18. These spectra confirm the purity of the two compounds, with each spectrum exhibiting the five proton peaks expected. The chemical shifts are similar to those described by Levine & Keiser (1978) and include the characteristic shift of the acetyl protons from 8 3.03 to 5 4.78 upon bromination of the 7-acetyl-10-methylisoalloxazine compound. 60 o o C H 3 CH 3 * Aromatic H's I 1 1 1 i i l 10 8 6 4 2 0 6 (ppm) Figure 3.18 NMR spectra of 7-acetyl-10-methylisoalloxazine (Panel A) and 7a-bromoacetyl-10- methylisoalloxazine (Panel B). Peaks marked with an asterisk correspond to the reference compound DSS (2,2'-dimethyl-2-silapentane- 5-sulfonate). 61 3.1.3 Cysteine Accessibility and Flavin Modification Thiol assays were used to investigate the reactivity of the cysteine residues introduced into the four variants. The kinetic traces and relative second order rate constants from these assays are given in Figure 3.19 and Table 3.3 respectively. The cysteines of all four variants exfiibit greatly increased reactivity relative to the Cys 102 residue of wild-type cytochrome c. The thiol of the Thr8Cys variant was the most reactive. As the thiol of the His39Cys variant was the least reactive of the four, reaction times for the modification of this variant with the flavin were optimized as outlined below, and these conditions were used for modification of the other variants. Thus the assumption was made that the more accessible cysteines would be completely modified under the conditions known to modify the least reactive thiol quantitatively. Table 3.3 : Bimolecular rate constants (relative to WT cytochrome c) for reaction of cytochrome c Cys variants with 4,4-dithiodipyridine. Variants Relative Bimolecular Rate Constant (M^min"1) Thr8Cys 1100(100) His39Cys 65(5) Asn62Cys 300(6) Leu85Cys 200(20) WT(102Cys) 1 Modification of the cytochrome c variants with the flavin moiety was monitored by FPLC. Elution profiles observed after various reaction times for the His39Cys variant modification are shown in Figure 3.20. The initial ferricytochrome sample is slowly converted to flavin modified ferricytochrome which in turn is converted to the flavin modified ferrocytochrome. Complete modification was achieved after approximately 24 hours with quantitative conversion to flavin 62 0.50 0 10 20 30 40 50 Time (min) Figure 3.19 Thiol assay kinetic traces for wild type (WT) cytochrome c and the Thr8Cys, His39Cys, Asn62Cys, and Leu85Cys variants (sodium phosphate buffer, pH 7.0, u = 0.1 M , 25 °C). 63 modified ferrocytochrome c (elution profile e in Figure 3.20). After modification, thiol assays were repeated to confirm the absence of free cysteine residues. All modified variants showed no absorbance increase over a period of one hour following addition of 4,4'-dithiodipyridine (data not shown), indicating that complete modification had been achieved. 3.1.4 Tryptic Digestion and H P L C Analysis The site of flavin attachment was identified by tryptic digestion of the unmodified and modified proteins followed by separation of the peptides by reversed-phase HPLC. For each modified protein the position of the peptide containing the cysteine residue was shifted, indicating that modification had been achieved at this site. Representative peptide maps for the His39Cys cytochrome c variant are shown in Figure 3.21 along with the relative positions of the modified peptide peaks for the three other variants. The four peaks marked with an asterisk are associated with flavin degradation products formed during the hydrolysis reactions. This assignment was confirmed by subjecting a sample containing only free flavin to tryptic digestion and separation by the HPLC. These four peaks were observed at the same elution times as those seen with the flavin modified protein samples. The retention time of the fifth additional peak observed for the protein samples, varied for each protein sample and was thus identified as the peptide containing the modified cysteine residue. 3.1.5 Electronic Absorption Spectroscopy The four variants all exhibit electronic adsorption spectra similar to that of wild type iso-1 -cytochrome c. Absorption maxima and associated extinction coefficients are given in Table 3.4. The spectra of the flavin-modified proteins correspond closely to the sum of the spectra for unmodified 64 in o a 0 10 20 30 40 Elution volume (ml) Figure 3.20 FPLC elution profiles monitoring flavin attachment to His39Cys cytochrome c (sodium phosphate buffer pH 7.2, 0.02 M). Reaction times are as follows: (a) 0 hrs, (b) 6 hrs, (c) 12 hrs, (d) 18 hrs, (e) 24 hrs. Free cysteine was added to the reaction after 18 hrs. Peak A corresponds to unmodified Fe(III) cytochrome c, peak B corresponds to modified Fe(III) cytochrome c, and peak C corresponds to modified Fe(II) cytochrome c. Material eluting in the first five minutes corresponds to excess free flavin and free cysteine. 65 T9in£ , .co//WT 0 20 40 60 80 100 120 140 Retention Time (min) Figure 3.21 HPLC analysis (C-18 reverse phase) of tryptic peptides for wild type cytochrome c, the His39Cys variant and the flavin modified His39Cys variant. Shaded peaks A, B & C correspond to the peptide containing His39, Cys39 and modified Cys39 respectively. Also denoted by arrows are the positions of the mutated and modified residue containing peptides for the variants Thr8Cys (T8C), Asn62Cys (N62C) & Leu85Cys (L85C). The positional change for the T9 peptide in the cytochrome c expressed in Kcoli, which contains the TML72Lys mutation is also indicated. Peaks marked with an asterisk correspond to flavin degradation products. 66 Table 3.4 : Electronic absorption maxima and molar absorbances for unmodified and flavin modified cytochrome c variants.a Absorption maxima [nm(mM "'cm"*)] Protein Visible Soret U V Cys8Thrl02 Fe* 530(11) 410(106) 361(28) 278(23) Fe 2 + 550(28) 520(16) 415(129) 315(34) 274(32) Cys39Thrl02 Fe* 529 409 361 279 Fe 2 + 550 520 414 315 273 Cys62Thrl02 Fe* 529 409 361 279 Fe 2 + 550 520 415 315 274 Cys85Thrl02 Fe* 529 409 360 278 Fe 2 + 550 520 415 315 274 Cysl02 Fe* 529 410 361 278 Fe2 + 550 520 415 315 274 Modified Cys8Thrl02 Fe3 + 529(12) 408(125) 358(38) 275(52) Fe2 + 549(29) 519(17) 414(149) 310(50) 273(62) Modified Cys39Thrl02 Fe* 529 407 358 278 Fe 2 + 549 519 414 312 272 Modified Cys62Thrl02 Fe* 529 407 359 277 Fe2 + 549 519 414 313 273 Modified Cys85Thrl02 Fe* 529 406 358 274 Fe 2 + 549 519 413 313 271 a lOOmM sodium phosphate buffer (pH 7.0) 67 cytochrome c and the unreacted flavin as shown in Figure 3.22. 3.1.6 Circular Dichroism Spectroscopy The circular dichroism spectra of the four unmodified and modified variants are superimposable in both the visible and far U V regions (200 - 450 nm). Representative spectra for the unmodified and flavin modified Thr8Cys variant are shown in Figure 3.23. 3.1.7 Fluorescence Spectroscopy The intensity of fluorescence emission (X e x c i t a l i o n = 425nm) associated with the flavin was significantly quenched (93-95%) following attachment to the protein variants. The X^ emission for the attached flavins exhibited bathochromic shifts of 5-7 nm relative to the corresponding unattached flavin (Table 3.5). Fluorescence emission spectra of the flavin 7-acetyl-10-methylisoalloxazine (Acflavin) and the modified His39Cys cytochrome c variant are given in Figure 3.24. Table 3.5 : Fluorescence emission maxima and relative intensities for free and attached 7-acetyl-10-methylisoalloxazine (A e x c i t a t i o n = 425 nm) A , ^ emission / nm Relative intensity Acflavin 528(1) 100 Fl-Thr8Cys 535(2) 7 Fl-His39Cys 534(2) 6 Fl-Asn62Cys 533(2) 7 Fl-Leu85Cys 534(2) 5 68 300 400 500 Wavelength (nm) 600 Figure 3.22 The electronic absorption spectra o f free 7-acetyl-10-methylisoalloxazine (solid), reduced Thr8Cys cytochrome c (dashed) and flavin modified reduced Thr8Cys cytochrome c (dotted) ( lOOmM sodium phosphate buffer, p H 7.0, 25 °C). 69 Figure 3.23 CD spectra of Thr8Cys cytochrome c (solid) and flavin modified Thr8Cys cytochrome c (dotted) (0.1M sodium phosphate buffer, pH 7.0, 25 °C). 70 Wavelength (nm) Figure 3.24 Fluorescence emission spectra of the free flavin 7-acetyl-10-methylisoalloxazine (solid) and modified His39Cys cytochrome c variant (dotted) (0.1 M sodium phosphate buffer pH7.0 25 °C) (X e x c i t a t i o n = 425nm). 71 3.1.8 Electrochemistry The midpoint potentials for wild type cytochrome c and the four cysteine containing variants were determined by cyclic voltammetry. All reduction potentials determined for these proteins are given in Appendix A. A typical cyclic voltammogram is illustrated in Figure 3.25a. The midpoint potentials observed at pH 7.0 (25 °C, u = 0.1 M) are shown in Table 3.6. The reduction potential for native yeast iso-l -cytochrome c was 280(2) mV, which is in good agreement with the value of 281(2) mV obtained by Rafferty (1992). Under the same conditions, three of the variants exhibit potentials similar to that of wild type cytochrome (274 - 280 mV). The His39Cys variant, however, was found to have a potential 25 mV lower than that of the wild-type protein. The thermodynamic parameters for the oxidation-reduction equilibria of these cytochromes were derived from the temperature dependence of the potentials and the following relationship: A G 0 = A H 0 - TAS° [10] Knowing that A G 0 = -nFE m permits the derivation of the relationship for a one electron process: E m = T A S ° / F - A H ° / F [11] where F is Faraday's constant. As the reaction entropy change associated with the entire electrochemical cell (AS0) includes the reference standard hydrogen electrode (SHE), the entropy associated with the reference electrode must be considered with the expression: A S ^ A S V A S a * [12] Knowing the entropic change for the SHE half-cell is 15.6 eu (cal mor'K' 1) at 25 °C (Taniguchi et al., 1982), equation 12 can be rewritten: E m = T A S ° r e / F - ( A H ° + 4650)/F [13] 72 a b c 0.1 u A A E = 1 2 0 m V -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 E ( m V vs S H E ) F i g u r e 3.25 C y c l i c vo l tammograms for (a) T h r 8 C y s cytochrome c, (b) acetyl-10-methyl i soa l loxazine , and (c) f l av in modif ied T h r 8 C y s cytochrome c. R e d u c t i o n potentials (vs. S H E ) and peak separations are indicated ( 0 . 1 M sod ium phosphate buffer p H 7.0 2 5 ° C , sweep rate 20 mVs" 1 ) . 73 Therefore, the slope of a plot of E m vs T represents AS° r e / F and the y-intercept represents (AH° + 4650) / F. The temperature dependence of the reduction potentials for wild-type cytochrome and the four variants are shown in Figure 3.26, and the associated thermodynamic parameters are given in Table 3.6. All the variants exhibited parameters similar to that of wild-type cytochrome with the exception, again, of His39Cys. This latter variant exhibited a relatively large standard entropy of reduction (-16.3(3) eu, relative to -11.0(2) eu for WT). Table 3.6 : Electrochemical thermodynamic properties of wild-type and variant cytochromes Protein Em,7 A G A H 0 AS r c° (mV) (kcal/mol) (kcal/mol) (eu) WT iso-1 280(2) -6.4(2) -14.4(7) -11.0(2) Thr8Cys 278(2) -6.4(2) -14.0(5) -9.9(2) His39Cys 255(2) -5.9(2) -15.4(9) -16.3(3) Asn62Cys 280(2) -6.4(2) -13.7(4) -8.9(2) Leu85Cys 274(2) -6.3(2) -14.4(6) -11.7(2) The pH dependence of the midpoint potentials was also investigated as a further probe of the effects of substituting cysteinyl residues at four locations on the protein surface. This issue was of particular interest in regard to the His39Cys variant because the histidine at position 39 has been identified previously as the single titratable group (Moore et al, 1984) the pK3 of which is dependent on the oxidation state of the heme iron. This pH dependence of E m can be described by the following relationship: EinnH = E^o + (RT/nF) In (Kf+ [FT]) / (K0 + [FT]) [14] 74 300 240 i ' 1 ' 1 1 270 280 290 300 310 320 Temperature (Kelvin) Figure 3.26 The temperature dependence of the reduction potentials of cytochrome c cysteine variants (0.1 M sodium phosphate buffer, pH 7.0): (•), Wild type; (•), Thr8Cys; (A), His39Cys; (o), Asn62Cys; (•), Leu85Cys. For figures 3.26, and 3.27, the error in the measured value of each point (±2 mV) is approximately three times the size of the symbols. 75 The dissociation constants, Kt and K0, represent the oxidation state-linked titratable groups of the reduced and oxidized proteins respectively. Values for pK0, pKr and E^,, are obtained by fitting the measured potentials to this equation and the resulting values for the wild-type and variant proteins are given in Table 3.7 with the associated data shown in Figure 3.27. The values of pK0, and pKT for the wild-type cytochrome c are in good agreement with those reported previously (pK0 = 6.6, pKt = 7.2 Cutler et al, 1989). Of the four variants, the His39Cys variant is the only one to deviate appreciably in behavior from the wild-type protein. This variant exhibits pK0 and pKt values of 7.5(2) and 7.9(2) respectively, increases of 0.7 - 0.9 pK units. Table 3.7 : pKa values for residue 39 in the oxidised (p^0) and reduced (piQ states Protein E ^ (mV) pK0 pKt WT iso-l 295(2) 6.6(1) 7.0(1) Thr8Cys 295(2) 6.3(1) 6.7(1) His39Cys 262(2) 7.5(2) 7.9(2) Asn62Cys 289(2) 6.8(1) 7.1(1) Leu85Cys 288(2) 6.7(1) 7.1(1) The reduction potential for the 2 electron reduction of free 7-acetyl-10-methylisoalloxazine was also studied by cyclic voltammetry. A typical cyclic voltammogram for the flavin is illustrated in Figure 3.25b. The reduction potential of the flavin at pH 7.0 (25 °C, u = 0.1 M) is -113(2) mV. As outlined above, the thermodynamic parameters were calculated from the temperature dependence data, and the resulting values for these parameters were A G = 5.2(2) kcal/mol, A H 0 = -4.9(7) kcal/mol and AS r e° = -18.5(4) eu. The pH dependence of the reduction potential of 7-acetyl-10-methylisoalloxazine was also studied, and the values for the 2 electron transfer process ( E J are shown in Figure 3.28. The E m 76 240 7 pH 8 F i g u r e 3.27 T h e p H dependence o f reduction potentials o f cy tochrome c cysteine variants (0.1 M sod ium phosphate buffer, 25 °C) : ( • ) , W i l d type; ( • ) , T h r 8 C y s ; (A), H i s 3 9 C y s ; ( o ) , A s n 6 2 C y s ; ( • ) , L e u 8 5 C y s . The interpolated curves were calculated us ing equation 14. 77 vs pH curve can be divided into three linear sections that cover the pH ranges 5.0 to 7.0, 7.0 to 9.5, and 9.5 to 11.9, respectively. This behavior is expected for a flavin system in which the oxidized and the reduced forms have only one pKa value each. This type of system is amenable to manual fitting as outlined by Draper & Ingraham (1968). As outlined in the experimental chapter of this thesis, spectroelectrochernical studies were conducted to obtain reduction potentials for the one electron processes associated with the flavoquinone / flavosemiquinone couple (E,) and flavosemiquinone / flavohydroquinone couple (Ej). The results of these measurements were fit to the Michaelis equation. The potential values for the 2 electron process calculated from the spectroelectrochernical data obtained at various values of pH are in excellent agreement with the corresponding data obtained by cyclic voltammetry and are given in Table 3.8. A representative family of spectra obtained from the spectroelectrochernical titration of the flavin is shown in Figure 3.29 along with the corresponding Nernst plot. The semiquinone formation constants and the values obtained for E , and E 2 are also given in Table 3.8 and Figure 3.28. Table 3.8 : Potentials for 7-acetyl-10-methylisoalloxazine obtained from spectroelectrochemcial experiments. pH K A E E 2 (mV) (mV) (mV) (mV) 5.01 27(1) 0.10(4) 60(10) -3(5) 57(5) 7.00 -114(1) 0.06(2) 72(10) -150(5) -78(5) 9.02 -174(1) 0.05(2) 80(10) -214(5) -134(5) 11.04 -284(1) 0.06(2) 74(10) -321(5) -247(5) 78 Figure 3.28 The pH dependence of the three reduction potentials of 7-acetyl-10-methylisoalloxazine. E m (•) is the potential of the oxidized-fully reduced species, and E ! and E 2 (A) are the potentials of the oxidized-semiquinone and fully reduced-semiquinone systems respectively. The solid lines are the fitted curves obtained by the graphical method of Draper & Ingraham (1968). These fits were used to estimate values for pKn pKs and pK0. Reduction potentials are also shown for the flavin component of the modified cytochrome c: (0), Thr8Cys; (•), His39Cys; (o), Asn62Cys; (A), Leu85Cys. The error in the measured value of each point (±2 mV) is contained within the size of the symbols. The error in the calculated E ! and E 2 values (±5 mV) is indicated with error bars. 79 350 400 450 500 550 Wavelength (nm) F i g u r e 3.29 Spectroelectrochemical t i tration o f 7-acetyl -10-methyl isoal loxazine (0.1 M sod ium phosphate buffer, p H 7.0 25 °C) . A p p l i e d potentials, E a p p ( m V vs S H E ) : (a) 0.0, (b) -75.6, (c) -94.5, (d) -100.8, (e) -107.6, (f) -115.3, (g) -125.1 , (h) -132.6, (i) -141.2, Q) -151.5, (k) -310.3. T h e inset is a Nerns t plot calculated from the absorbance at 421 nm. 80 The graphical method of Draper & Ingraham (1968) was also employed to estimate the pKr, pK0 & pK, values for the flavin (Figure 3.28). Because E x and values were determined at just four pH values, the assumption was made that the pH dependence of the two single electron coupling processes (E x and E 2 ) was similar to that of the two electron process (E.J over the pH range studied. Previous studies have shown this assumption to be valid for riboflavin, F M N (Draper & Ingraham, 1968), and 8a-imidazolylflavins (Williamson al., 1985). The values obtained in the current analysis, pKt = 7.00(20), pK0 = 9.40(20) & pKs = 8.20(20), are similar to those reported previously for riboflavin and F M N (p^ = 6.25 pK0 = 10.00 pKs = 8.27 & pK, = 6.72 pK0 = 10.35 pKs = 8.55, respectively (Draper & Ingraham 1968)). Reduction potentials for the heme and flavin sites of the modified variants were also investigated with both cyclic voltammetry and spectroelectrochernical techniques. A typical cyclic voltammogram obtained for a flavin modified cytochrome is shown in Figure 3.25c. Potentials of the heme and flavin centers (pH 7.0, 25 °C, u = 0.1 M) of the four modified variants are given in Table 3.9. While no appreciable change was observed in the potentials of the heme iron (±4 mV), a decrease of 5-10 mV was observed for the covalently bound flavin centers. Notably, it was possible to achieve direct electrochemistry of the heme iron centers of the modified proteins without modification of the gold electrode surface with 4,4'-dithiodipyridine. This observation suggests that the attached flavin mediates electron transfer between the protein and the electrode surface. The temperature dependence of the reduction potentials associated with the heme and flavin centers was investigated to obtain the corresponding thermodynamic parameters. The resulting data are shown in Figure 3.30, and the associated parameters are summarized in Table 3.9. From these results, it is clear that flavin-modification causes no appreciable changes in the parameters observed for the heme iron center. In contrast, the enthalpies and entropies of the flavin centers 81 300 240 270 280 290 300 310 320 Temperature (Kelvin) W -130 -140 270 280 290 300 310 Temperature (Kelvin) 320 Figure 3.30 The temperature dependence, of heme and flavin center reduction potentials of flavin-modified cytochrome c variants and of free 7-acetyl-10-methylisoalloxazine (0.1 M sodium phosphate buffer, pH 7.0): (A) Heme centers of (•), Fl-Thr8Cys; (•), Fl-His39Cys; (•), Fl-Asn62Cys; (•), Fl-Leu85Cys. (B) Flavin centers of (•), free flavin; (A), Fl-Thr8Cys (•), Fl-His39Cys; (0), Fl-Asn62Cys; (o), Fl-Leu85Cys. The error in the measured value of each point (±2 mV) is approximately three times the size of the symbols. 82 undergo a significant change upon attachment to the protein ( A H 0 changes from -4.9(7) kca l /mol to -6.5(7) - -8.2(8) kca l /mol and A S ° r c changes from -18.5(4) eu to -25.0(3) - -28.9(3) eu). T a b l e 3.9 : Thermodynamic parameters o f the oxidat ion-reduct ion equil ibria o f the f lavin-modif ied cytochromes heme center flavin center E m , 7 AG AH° ASrc° Em,7 AG AH° ASr c° (mV) (kcal/mol) (kcal/mol) (eu) (mV) (kcal/mol) (kcal/mol) (eu) Fl-Thr8Cys 279(2) -6.5(2) -14.1(9) -10.0(3) 123(2) 5.7(2) -6.5(7) -25.0(3) Fl-His39Cys 257(2) -5.6(2) -15.0(6) -15.9(2) 115(2) 5.3(2) -8.2(8) -28.9(3) Fl-Asn62Cys 279(2) -6.4(2) -13.8(5) -9.3(2) 118(2) 5.4(2) -7.0(8) -25.8(3) Fl-Leu85Cys 272(2) -6.3(2) -14.7(5) -11.9(2) 115(2) 5.3(2) -7.4(9) -26.5(4) The p H dependent electrochemical behavior o f the modif ied proteins was also investigated. T h e pKa values for the titratable group, the protonat ion state o f w h i c h is l inked to the oxidat ion state o f the heme i ron (pKt and pK0), were determined as described above for the unmodif ied proteins. The resulting data are shown in F igure 3.31, and the values derived from these data are summarized in Table 3.10. A s can be seen f rom these results, all o f the cytochromes exhibit s imilar behavior i n these experiments w i t h the notable except ion o f the flavin-modified H i s 3 9 C y s variant. Th is modif ied variant is unique in that the potential o f the heme center is independent o f p H . The p H dependence o f the midpoint potentials o f the flavin centers o f the modif ied variants was also studied over the p H range 5.5 to 8.0. These data are summarized in F igure 3.28 (open symbols) to reveal p H dependent behavior w i t h the same general trend as that o f flavin that is not attached to the protein. 83 Table 3.10 : pA"a values for residue 39 in the oxidized (pK0) and reduced (pKT) states of the flavin-modified cytochromes Protein E^o (mV) p ^ ; pKr Fl-Thr8Cys 294(2) 6.4(1) 6.8(1) Fl-His39Cys 260(2) N A N A Fl-Asn62Cys 287(2) 6.8(1) 7.1(1) Fl-Leu85Cys 287(2) 6.5(1) 6.9(1) To obtain the semiquinone formation constants and the one electron couple potentials of the modified proteins, Ej and E 2 , spectroelectrochernical experiments were carried out. A representative family of spectra obtained for the flavin from such an experiment is shown in Figure 3.32. Reduction potentials (E m , E ! and E 2 ) and semiquinone formation constants (K) for the flavin centers of the four variants obtained from these measurements are given in Table 3.11. The values of K and thus the separation between the E t and E 2 potentials are similar to those associated with the free flavin, indicating that the attachment process does not perturb the relative stability of the semiquinone form. Table 3.11 : Potentials for flavin redox center of modified proteins calculated from spectroelectrochernical experiments. PH E m K A E E i E 2 (mV) (mV) (mV) (mV) Fl-Thr8Cys 7.0 -113(1) 0.04(2) 82(18) -153(9) -72(9) Fl-His39Cys 7.0 -119(1) 0.04(2) 84(18) -161(9) -77(9) Fl-Asn62Cys 7.0 -116(1) 0.06(2) 72(10) -152(5) -80(5) Fl-Leu85Cys 7.0 -116(1) 0.08(3) 64(12) -148(6) -84(6) 84 F i g u r e 3.31 Dependence o f the reduct ion potentials for the heme centers o f f l av in -modi f i ed cytochromes o n p H (0.1 M sod ium phosphate buffer, 25 °C) : ( • ) , F l - T h r 8 C y s ; (A), F l -H i s 3 9 C y s ; ( • ) , F l - A s n 6 2 C y s ; ( • ) , F l - L e u 8 5 C y s . T h e so l id l ines represent the fits o f the data to E q 15. T h e error i n the measured value o f each point ( ± 2 m V ) is approximate ly three t imes the size o f the symbols . 85 500 550 Wavelength (nm) Figure 3.32 Spectroelectrochernical titration of flavin modified Thr8Cys cytochrome c (0.1 M sodium phosphate buffer, pH 7.0 25 °C). Applied potentials, E a p p (mV vs SHE): (a) +40.0, (b) -25.0, (c) -40.2, (d) -55.1, (e) -70.4, (f) -85.0, (g) -99.9, (h) -115.1, (i) -129.8, 0) -145.0, (k) -160.3, (1) -239.8. Inset A is a Nernst plot calculated from the absorbance at 432 nm. Inset B shows the expanded spectra of the modified protein with a fully reduced (R) and the fully oxidized flavin (O) component. 86 3.1.9 Flash Photolysis 3.1.9.1 Intermolecular Electron Transfer Kinetics The dependences of the observed rate constants (&obs) for the reaction of cytochromes with various flavin semiquinone derivatives on protein concentration are shown in Figure 3.33, and the kobi values are tabulated in Appendix B. In each case, kobi varied linearly with protein concentration, and the linear fits to the data extrapolated within the uncertainty of the fit to the origin. The slopes of these plots provide the second order rate constants for these reactions (Equation 15):. *obS = kn [Cytochrome c] [15] The k12 values derived from these data are given in Table 3.12. The rate constant for the reduction of horse heart ferricytochrome c by riboflavin (0.48 x 108 M^s'1) is in good agreement with previously reported values (Meyer et al., 1984). Both 7-acetyl-10-methylisoalloxazine and 7a-bromoacetyl-10-methylisoalloxazine reduce the cytochrome more rapidly than does riboflavin (k12 = 0.63 x 108 and 0.58 x 108 M^s"1, respectively). The yeast cytochrome expressed in yeast and in E. coli is more rapidly reduced by 7-acetyl-10-methylisoalloxazine than is the horse heart protein. The cytochrome variants exhibit rate constants in these reactions that are within experimental error (~ ±10%) of the values observed for the wild-type protein. The temperature dependence of cytochrome reduction by flavins was measured for three bimolecular reactions. The activation enthalpies (AH*) and entropies (AS*) for these reactions were calculated by fitting the data to the Eyring equation: Ink l 2/T = A S * / R - A H * / R T + In& B /h [16] The activation parameters derived from these analyses are given in Table 3.13, and the corresponding Eyring plots are shown in Figure 3.34. 87 Figure 3.33 The dependence of kobs on [ferricytochrome c] in the reduction of horse heart cytochrome c with (—•—•), riboflavin; (• • o - •), 7-acetyl-lO-methylisoalloxazine; and ("•"), 7a-bromoacetyl-10-methylisoalloxazine; and for the reduction of yeast /so-l-cytochrome c (—•-), wild-type; ( - • - ) , Thr8Cys; (••-A--), His39Cys; (~0~), Leu85Cys; E. coli cytochrome c (• • * • •), WT; & (•"•--), Asn62Cys with 7-acetyl-10-methylisoalloxazine (0.1 M sodium phosphate buffer pH 7.0, 25 °C). The error in each point is estimated to be ± 5%. 88 F i g u r e 3.34 E y r i n g plots for the b imolecu la r reduct ion o f horse heart and yeast cytochromes by f l av in derivatives (0.1 M sod ium phosphate buffer, p H 7.0, 25 °C) : ( • ) , horse heart cy tochrome reduct ion b y r ibo f l av in ; (A), horse heart cytochrome reduct ion by 7-acetyl-10-methyl i soa l loxaz ine ; (o) , yeast iso-\-cytochrome reduct ion by 7-acetyl -10-methyl isoal loxazine . 89 Table 3.12 : Rate constants for the reduction of cytochrome c (Fe3+) variants by various flavins protein flavin kn xlO'8 (M' 1 s'1) horse heart cytochrome c riboflavin 0.48(5) horse heart cytochrome c 7-acetyl-10-methylisoalloxazine 0.63(4) horse heart cytochrome c 7a-bromoacetyl-10-methylisoalloxazine 0.58(3) iso-l -cytochrome c 7-acetyl-10-methylisoalloxazine 1.03(7) iso-1 -cytochrome c Thr8Cys 7-acetyl-10-methylisoalloxazine 1.02(10) iso-l -cytochrome c His39Cys 7-acetyl-10-methylisoalloxazine 0.99(9) wo-1-cytochrome c Leu85Cys 7-acetyl-10-methylisoalloxazine 1.10(10) E. coli cytochrome c 7-acetyl-10-methylisoalloxazine 1.05(8) E. coli cytochrome c Asn62Cys 7-acetyl-10-methylisoalloxazine 1.12(10) Table 3.13 : Rate constants and activation parameters for the reduction of cytochorme c (Fe3+) by various flavins protein flavin kn AH*, AS*, (M 'V 1 x 10'8) (kcal mol'1) (e.u.) horse heart riboflavin 0.48(5) 3.9(2) -10.4(1) horse heart 7-acetyl-10-methylisoalloxazine 0.63(4) 3.8(2) -10.5(1) iso-l - 7-acetyl-10-methylisoalloxazine 1.03(7) 3.3(3) -10.7(1) 3.1.9.2 Intramolecular Electron Transfer Kinetics The intramolecular electron transfer kinetics of the four flavin modified cytochrome c variants were also studied by laser flash photolysis techniques. Each variant was studied over a pH range of approximately 5-8 and over a temperature range of 284 to 308 K at pH 7.0. The results obtained at pH 7.0 and T = 298 K are given in Table 3.14, and the results obtained under all conditions studied are given in Appendix C. Activation parameters calculated from the temperature dependence data are also given in Table 3.15. If it is assumed that the weak dependence of TVl on 90 temperature can be neglected (Marcus & Sutin, 1985) and that the protein structure and are temperature independent, a plot of In k vs. T"1 has a slope of -(AG + A) 2 AUK B , thus enabling an estimation of the reorganization energy (X). The corresponding Eyring plots are shown in Figure 3.35, and the values of X derived from this analysis are given in Table 3.14. All four modified proteins exhibit AH* values of approximately 3.0(2) kcal mol'1 and large negative entropies of activation (AS* = -31 - -38 e.u.). The calculated reorganization energies (A) vary from 1.0(1) to 1.3(1) eV with an overall average value of 1.2(2) eV. Table 3.14 : Observed rate constants, activation parameters and reorganization energies for the intra-molecular electron transfer of flavin modified cytochrome c derivatives k a "•obs (s-1) AH* (kcal mol'1) AS* (e.u.) X (eV) Fl-Thr8Cys 4.41(22) x 103 3.2(2) -31(1) 1.3(1) Fl-His39Cys 1.29(7) x 103 3.0(2) -35(1) 1.2(1) Fl-Asn62Cys 2.46(12) x 102 3.0(2) -38(1) 1.0(1) Fl-Leu85Cys 8.51(42) x 102 3.0(2) -35(1) 1.3(1) a p H 7 . 0 T = 2 5 ° C Knowing X for each variant, it is possible to estimate the corresponding electronic coupling parameters, H ^ by fitting the experimental driving-force dependent electron transfer rate constants for each variant to Eq. 5 by least squares analysis. The resulting plots of In ket vs - A G 0 are shown in Figure 3.36. From these fits, an experimental estimate for the electronic coupling values (HAB) associated with the electron transfer process can be obtained. Comparison of these values (Table 3.15) and the H ^ values obtained from the pathway calculations is considered below. The value for the reorganization energy of cytochrome c (0.5 eV) has been derived from analysis of the self-exchange reaction of this protein (Marcus & Sutin, 1985). The reorganization energy for the 91 Figure 3.35 Eyring plots for the intra-molecular reactions of flavin modified cytochrome c (0.1 M sodium phosphate buffer, pH 7.0, 25 °C): (A), Fl-Thr8Cys; (o), Fl-His39Cys, (o), Fl-Asn62Cys; (•), Fl-Leu85Cys. The error for each point is contained within the size of the symbol. 92 Figure 3.36 Dependence of kobs on - A G 0 for the reduction of cytochromes c by the covalently attached flavin (sodium phosphate buffer, 25 °C): (*), Fl-Thr8Cys; (o) Fl-His39Cys; (o), Fl-Asn62Cys ;(•), Fl-Leu85Cys. The solid lines are the best fits to equation 5. The shaded area represents the uncertainty in the fit when using the range of values for X obtained from the temperature dependence data. A representative absorption transient and fit are shown in the inset. The error for each point is contained within the size of the symbol. 93 flavin attached to the cytochrome can then be determined by simply subtracting the protein reorganization energy from the overall reorganization energy of the system. In this way the reorganization energy of the flavin center is calculated to be 0.7(2) eV. 3.1.10 Structure Modeling To aid interpretation of the kinetic results, energy minimized structural models for the four variants and their corresponding flavin modified derivatives were calculated using the program INSIGHT (Biosym, CA) (In collaboration with Dr. P. Siddarth) as described in section 2.9 (Page 59). The calculations predict that the four cysteine variants all have structures similar to that of wild type iso-1 -cytochrome c, and upon subsequent modification with 7a-bromoacetyl-10-methylisoalloxazine, the protein structures still remain essentially unperturbed. Only small localized changes around the attachment sites are anticipated (Figures 3.37-3.40). For the modified Cys 8, 39, and 85 variants, the flavin appears to be oriented away from the protein surface into the surrounding solvent, suggesting that electron transfer will only occur through the covalent linkage with the cysteine residue. The model for the modified Cys 62 variant is somewhat different and suggests that the flavin is oriented closer to the protein surface. Aside from the covalent attachment to residue 62, the flavin moiety in his derivative is predicted to be close proximity to two other residues, Glu -4 (d [flav62 N5 - Glu-4 Oel] = 2.7A) and Glu 66 (d [flav62 C9 - Glu66 02] = 2.1 A). This proximity may lead to the possibility of electron transfer through space from the flavin to these two residues. Such a mechanism must be considered in subsequent kinetic analysis and electron transfer pathway modeling. 3.1.11 Electron Transfer Pathway Calculations The best single pathway between the flavin donor and the heme acceptor for each 94 modified variant calculated with the Greenpath program is illustrated in Figures 3.3 7-3.40 (top panel). Artificial intelligence (Al) searches were also carried out to identify the most likely set of residues involved in electron transfer pathways from the donor to acceptor center (In collaboration with Dr. P. Siddarth). These searches produced several pathways for each variant, and the amino acid residues identified in this process are given in Table 3.15. Figures 3.37-3.40 (bottom panel) illustrate these pathways and the positions of the critical residues within each structure. Using only the amino acid residues identified by the A l search, quantum mechanical calculations of the electronic coupling matrix element were performed. Because extended Hiickel theory was used to calculate FL^ (Table 3.15), the uncertainty in the values of is large. In addition these calculations assume that both the protein structure and the electronic coupling within this structure are independent of temperature. The theoretical values of FL^ are in good agreement with those calculated from the electron transfer kinetics (section 3.1.8.2; Table 3.15). Table 3.15 : Amino acids residues selected by A l search and experimental and theoretical electronic coupling values Modified protein Thr8Cys His39Cys Asn62Cys Leu85Cys Cys 14 Ser40 Glu66 Cysl4 Phe 10 Gly41 Tyr67 Argl3 Amino acid pathway Lys 11 Asn52 Met80 Cysl7 set Ala7 Val57 Met64 Leu 15 Leu9 Leu58 Asn63 Phe 10 Argl3 Trp59 Leu9 experimental FL^ 0.08(3) 0.04(2) 0.01(1) 0.04(2) theoretical FL^ 0.07 0.03 0.02 0.04 95 F i g u r e 3.37 E n e r g y - m i n i m i z e d structure o f f l av in modi f i ed T h r 8 C y s cy tochrome c. (a) E l e c t r o n transfer pathway proposed by Greenpath program (thick l ines; dashed l ines denote hydrogen bonds), (b) Residues invo lved i n electron transfer pa thway as selected b y A I -superexchange mode l ( thick lines). 96 F i g u r e 3.38 E n e r g y - m i n i m i z e d structure o f f l av in modi f i ed H i s 3 9 C y s cy tochrome c. (a) E l e c t r o n transfer pathway proposed by Greenpath program ( thick l ines; dashed l ines denote hydrogen bonds), (b) Residues invo lved i n electron transfer pathway as selected by A I -superexchange mode l (thick lines). 97 F i g u r e 3.39 E n e r g y - m i n i m i z e d structure o f f l av in modi f ied A s n 6 2 C y s cy tochrome c. (a) E l e c t r o n transfer pathway proposed by Greenpath program (thick l ines; dashed l ines denote hydrogen bonds), (b) Residues invo lved i n electron transfer pathway as selected b y A I -superexchange mode l ( thick lines). 98 Ala81 a F i g u r e 3.40 E n e r g y - m i n i m i z e d structure o f f l av in modi f ied L e u 8 5 C y s cy tochrome c. (a) E l e c t r o n transfer pathway proposed by Greenpath program (thick l ines; dashed l ines denote hydrogen bonds), (b) Res idues invo lved in electron transfer pathway as selected b y A I -superexchange mode l ( thick lines). 99 3.2 Axial Ligand Variants 3.2.1 Protein Expression and Purification The yield of purified cytochrome c variants obtained from recombinant E. coli varied from <0.1 mg/L to 3.0 mg/L of culture. The expression of the Met80Ala variant was the most efficient, and expression of the Tyr67Phe/Met80Ala/Phe82Ser variant was the least efficient. Expression of the triple variant Tyr67Phe/Met80Ala/Phe82Ser was insufficient to permit spectroscopic characterization of the protein, so this variant will not be considered further. The efficiency of protein expression correlated approximately with the affinity of the protein for binding of NO to the heme iron. Indeed, if no NaN0 3 supplement was added to the growth medium, then expression of the axial ligand variants was essentially suppressed, suggesting that the binding of NO to the heme is necessary for expression of variants with the Met80Ala substitution. 3.2.2 Electronic Absorption Spectroscopy 3.2.2.1 Ligand Binding The electronic absorption maxima of several derivatives of Met80Ala, Met80Ala/Tyr67Phe and Met80Ala/Phe82Ser variants are given in Table 3.16. Representative spectra of the Fe(III) and Fe(II) derivatives are shown in Figure 3.41. All three variants exhibit spectra that are consistent with a heme site that can bind an exogenous ligand at the sixth coordination position. As a result, the spectra of these ferricytochrome variants resemble the corresponding spectra of myoglobin. Overall, the maxima of the Fe(III) derivatives all exibit a blue-shift of 5-10 nm relative to myoglobin. This difference can be attributed to the presence of a c-type heme in the cytochrome variants and a 6-type heme in myoglobin. As discussed below (Section 3.2.2.2, page 101) spectra of the ferricytochrome variants at pH 7.0 differ from the corresponding myoglobin spectrum owing to the variation in pKa 100 of the distally coordinated water molecule. Consequently, the relative amounts of high-spin and low-spin heme iron at pH 7 vary from protein to protein. As expected, replacement of Met80 eliminates the 695 nm band from the spectrum of these ferricytochrome variants. In contrast to the spectra of the ferricytochrome variants, spectra of the ferrocytochrome variants differ markedly from the spectrum of deoxy myoglobin. The Soret bands of the Met80Ala and Met80Ala/Tyr67Phe ferrocytochromes are composed of two components, a major peak with a maximum at -411 nm and a minor peak giving rise to a shoulder at 430 nm. These spectra compare to the spectrum of myoglobin which has a Soret maximum at 434 nm that is characteristic of a high-spin ferrous heme iron center (Makinen et al, 1983). This observation has been reported previously for the yeast cytochrome c His39Asn/Met80Ala/Cysl02Ser variant (Bren, 1996). These authors attributed the two Soret maxima of this variant to five-coordinated, high-spin (430 nm) and six-coordinated, low-spin (-411 nm) components. The corresponding spectrum of the Met80Ala/Phe82Ser ferrocytochrome variant is somewhat different with a sharp Soret band at 41 lnm and that lacks the secondary maximum at ~430nm. This observation that the Met80Ala/Phe82Ser variant heme exists as a low-spin species upon reduction is further indicated by the visible region of the spectrum in which the a (519 nm) and P (549 nm) bands are clearly resolved. The spectra of the Fe(II) CO and 0 2 derivatives of all three variants exhibit features similar to those of the corresponding myoglobin derivatives but with the 5-10nm blue-shift discussed above. This observation suggests that the sixth ligand bound to the reduced variants is sufficiently labile to be replaced easily by either CO or 0 2 . 3.2.2.2 pH Titrations As previously reported (Bren, 1996), the pK3 of the distal water ligand of the Met80Ala variant is 6.2(1), a value much lower than that for wild type horse heart myoglobin (pATa = 8.9) 101 T a b l e 3.16 : Abso rp t i on maxima (nm) and molar absorbances" ( rnM^cm" 1 ) o f cy tochrome c axial l igand variants ( S o d i u m phosphate buffer, p H 7.0, u = 0 . 1 M , 2 5 ° C ) Pro te in Soret Visible/near-TR M e t 8 0 A l a Fe(III) Fe(III) N O Fe(II) deoxy Fe(II) o x y Fe(II) C O Fe(II) N O M e t 8 0 A l a / T y r 6 7 P h e Fe(III) Fe(III) N O Fe(II) deoxy Fe(II) o x y Fe(II) C O Fe(II) N O M e t 8 0 A l a / P h e 8 2 S e r Fe(III) Fe(III) N O Fe(II) deoxy Fe(II) o x y Fe(II) C O Fe(II) N O 406(122) 415.5 411(87); 435, sh b 408(113) 414(234) 411 405(107) 414.5 416(80); 435, sh 408.5(98) 413.5(182) 410 401.5 414 411.5 404 412.5 409 535(10) 529 514, sh 537(77) 538 527(10) 528 519, sh 538(12) 537 529 528 519 536 536 548(10) 533(13) 548.5(9) 531(8) 561 531.5 563(7) 562.5 570, sh 571(9) 555, sh 567 559(7) 560 567, sh 570(9) 5 5 3 , s h 566 620 560 549 570 555, sh 566 a M o l a r absorbances correspond to horse heart cytochrome c variants (Bren , 1996). b sh = shoulder peak. 102 F i g u r e 3.41 E lec t ron i c absorption spectra o f o x i d i z e d (dotted) and reduced (sol id) cy tochrome c ax ia l l igand variants (100 m M sod ium phosphate buffer, p H 7.0, 25 °C) . ( A ) M e t 8 0 A l a ; ( B ) M e t 8 0 A l a / T y r 6 7 P h e ; ( C ) M e t 8 0 A l a / P h e 8 2 S e r . 103 (Antonini & Brunori, 1971). The pH dependence of the spectra for the Met80Ala variant is given in Figure 3.42. The corresponding pKa values for the Met80Ala/Tyr67Phe and Met80Ala/Phe82Ser variants derived from similar titrations are 5.8(1) and 7.5(1), respectively. These latter results are somewhat surprising because it has been suggested previously that the low pATa observed for the Met80Ala variant is probably attributed to stabilization of the coordinated hydroxide ligand through hydrogen bonding to the Tyr67 residue (Bren et al., 1995). 3.2.3 Circular Dichroism Spectroscopy The U V - C D spectra for the three variants (data not shown) are all similar to that of wild-type /'so-1-ferricytochrome c, which suggests that the secondary structures of the variants are the same as that of wild-type cytochrome c. Thermal denaturation curves for these variants are given in Figure 3.43 along with the calculated first derivatives of these curves. The melting temperatures obtained from these data for the Met80Ala and Met80Ala/Tyr67Phe variants are 53(1) and 50(1) °C, respectively. These values are marginally lower than the value for wild-type iso-1 -ferricytochrome c (54°C). Interestingly the Met80Ala/Phe82Ser variant exhibits a melting curve composed of two transitions at 45(1) and 48(2) °C. This observation is consistent with the hypothesis that the protonation state of the heme-bound water affects the overall thermal stability of this variant. At pH 7.0 the Met80Ala/Phe82Ser variant is composed of both water bound and hydroxide bound species, with the majority of the protein possessing a heme with a bound water molecule. It is conceivable that the transition at 45(1) °C corresponds to the thermal denaturation of the water bound species with the lower melting temperature resulting from the additional charge on the heme core. A similar observation was reported by Bren (1996) for the yeast His39 Asn/Met80Ala/Cys 102Ser variant, albeit to a lesser extent, and a similar explanation was proposed for this result. 104 3.2.4 Electrochemistry The reduction potentials (vs SHE, Sodium phosphate pH 7.0, T = 25 °C) of the Met80Ala, Met80Ala/Tyr67Phe, and Met80Ala/Phe82Ser variants (u= 0.1 and 0.5 M) determined by spectroelectrochernical methods are given in Table 3.17. These values are all significantly lower than the value for wild-type yeast cytochrome c (290 mV) as expected for variants in which the axial methionine ligand has been removed. A representative spectroelectrochernical titration of the Met80Ala/Tyr67Phe variant is shown in Figure 3.44. Table 3.17 : Reduction potential values for the axial variants Met80Ala, Met80Ala/Tyr67Phe, and Met80Ala/Phe82Ser (pH 7.0, 25 °C) p Met80Ala Met80Ala/Tyr67Phe Met80Ala/Phe82Ser 0.1 M -27(2) 26(2) -98(5) 0.5 M -45(2) 11(2) -109(5) 3.2.5 Ligand Binding Kinetics A representative transient absorption trace obtained for binding of CO to the Met80Ala variant is shown in Figure 3.45. Rate constants for CO recombination for the three variants derived from such measurements are similar to those observed for wild type horse heart myoglobin (kon = 5.0 x 10s M ' V 1 ) . These rate constants are 2.2(3) x 105 M ' V 1 for Met80Ala, 4.7(3) x 10s M ' V 1 for Met80Ala/Tyr67Phe and 9.5(4) x 10s M ' V 1 for Met80Ala/Phe82Ser. These results are in contrast to the observations of Bren & Gray (1993a) for the semisynthetic Met80Ala horse heart cytochrome c variant in which an anomalously slow CO recombination (8.7(8) x 103 M 'V 1 ) was recorded. These authors also reported that the His39Asn/Met80Ala/Cysl02Ser yeast cytochrome variant failed to exhibit detectable CO recombination upon photolysis. 105 300 400 500 600 700 Wavelength (nm) Figure 3.42 The electronic absorption spectra of Met80Ala ferricytochrome c as a function of pH (0.1 M sodium phosphate buffer, 25 °C). The dependence of absorbance at 621 nm on pH was fit to a single deprotonation process (Inset). 106 40 45 50 55 60 Temperature (°C) F i g u r e 3.43 T h e r m a l denaturation o f cytochrome c ax ia l l igand variants (0.1 M sod ium phosphate buffer, p H 7.0, heating rate o f 50 °C/hr ) . ( A ) T h e r m a l denaturation curves. ( B ) F i rs t derivatives o f denaturation curves. 107 0.3 j I i__ i i i i i i i 300 400 500 600 700 Wavelength (nm) Figure 3.44 Spectroelectrochemical titration of Met80Ala/Tyr67Phe cytochrome c (0.5 M sodium phosphate buffer pH 7.0 25 °C). Applied potentials, E a p p (mV vs SHE): (a) +24.00, (b) +59.8, (c) +45.2, (d) +30.1, (e) +15.0, (f) 0.0, (g) -14.9, (h) -30.0, (i) -44.8, (j) -60.2, (k) -200.0. A Nernst plot calculated from the absorbance values at 405 nm is provided in the inset. 108 -0.015 -0.002 0.000 0.002 0.004 0.006 0.008 Time (s) F i g u r e 3.45 Transient absorption kinet ics (monitored at 414 nm) for the recombina t ion o f C O w i t h M e t 8 0 A l a cytochrome c f o l l o w i n g photodissociat ion (0.1 M sod ium phosphate buffer p H 7.0, 25 °C) . T h e smooth l ine is the fit to a single exponential fit. T h e residuals are shown i n the upper panel . 109 3.2.6 Autoxidation The rate constants for autoxidation of the oxygenated derivatives of the Met80Ala, Met80Ala/Tyr67Phe, and Met80Ala/Phe82Ser variants are 0.02(1), 0.13(1), and 0.45(1) h'1 respectively (0.1M sodium phosphate buffer, pH 7.0,25 °C). Representative spectra obtained during autoxidation of the Met80Ala variant are given in Figure 3.46. The rate constant for the autoxidation of the Met80Ala variant is similar to that reported by Bren & Gray (1993 a) for horse heart Met80Ala (km = 0.01-0.04 h"1), but it is greater than the rate constant reported for the Fus39AsnMet80Ala/Cysl02Seryeastcytochromecvariant(0.003(l)h-1)(Bren, 1996). Both double variants studied in the present work exhibit increased autoxidation rates relative to the Met80Ala variant, indicating that both residues Tyr67 and Phe82 are partially responsible for the slower rate observed with the Met80Ala single variant. 3.2.7 Fourier Transform Infrared (FTIR) Spectroscopy To investigate the effects of replacing Met80 and adjacent residues on ligand binding to the heme iron further, FTIR spectra of the carbonyl derivatives of these variants were examined (Figure 3.47). For the Met80Ala variant, CO stretching frequencies, u c o , were observed at 1956 cm'1 (predominant peak) and 1969 cm'1 (shoulder). The corresponding spectrum for the Met80Ala/Tyr67Phe double variant shows a single peak at 1956 cm"1 with no evidence of the shoulder. The CO derivative of the Met80Ala/Phe82Ser variant also exhibits a single maximum, but u r n is shifted to 1960 cm"1. 110 300 400 500 600 Wavelength (nm) Figure 3.46 Absorption spectra following the autoxidation of Met80Ala cytochrome c (0.1 M sodium phosphate buffer pH 7.0, 25 °C). The inset shows a plot of absorbance values at 500 and 540 nm against time with the data fit to a single exponential model. I l l F i g u r e 3.47 Infrared spectra ( F T I R ) o f ferrous carbonyl derivatives o f cy tochrome c ax ia l l igand variants ( 0 . 1 M sod ium phosphate buffer, p H 7.0, 25 °C) . 112 3.2.8 Kinetics of Axial Ligand Variant Reduction by Fe(EDTA) 2 The kinetics of reduction of the three axial ligand variants by Fe (EDTA)2" were studied to provide initial assessment of the effects of these modifications on the electron transfer reactivity of the cytochrome. A representative family of rapid-scan spectra obtained for reduction of the Met80Ala variant by Fe(EDTA)2" is shown in Figure 3.48. The fit of the absorbance transient at 434.6 nm to a single exponential function is shown in the inset. The observed rate constants, kobs, obtained in this manner are tabulated in Appendix D. Unlike the Met80Ala variant, the Met80Ala/Tyr67Phe variant exhibits biphasic reduction kinetics as shown by the set of spectra and inset given in Figure 3.49. From these results, it appears that this protein is initially reduced to an Fe(II) intermediate with the spectroscopic properties of a low-spin heme iron. This fast reduction is followed by a slower, second phase in which the reduced low-spin form converts to a reduced high-spin/low-spin form. The initial phase is dependent upon the concentration of the reducing agent and presumably represents the actual reduction of the heme iron. On the other hand, the slower phase is independent of Fe(EDTA)2" concentration and exhibits a rate constant of 0.83(10) s"1. The rate constants observed for the first phase were, therefore, used in subsequent evaluation of electron transfer behavior of this variant. The observed rate constants obtained in this manner are also tabulated in Appendix D. Whereas both the Met80Ala and Met80Ala/Tyr67Phe cytochrome c variants were reduced to completion, the Met80Ala/Phe82Ser protein was only partially reduced under the experimental conditions employed. This observation is a consequence of the low reduction potential of this variant (E m = -109 mV) and the relatively high reduction potential of Fe(EDTA) 2" under these conditions (+95 mV) (Reid, 1984). Consequently, to obtain values for the data were fit to a pseudo-first order reversible rate equation (Cummins & Gray, 1977; Reid etal, 1986) as outlined below (Eq. 17, 113 18, and 19): A = AA <s + exp(fcobs/3-t) -1 + A [17] | _ l + aexp(fcobj3-t) a = Kai+l [18] cq [19] where A is the absorbance at time t, A A is the total absorbance change, and A M is the final absorbance. The equilibrium constant for the reaction (Keq) is determined from the following expression: where A ( is the initial absorbance, A„ is the final absorbance and A f is the final absorbance had the reaction gone to completion. A f was determined by reducing the protein completely with a small amount of solid sodium dithionite. The observed rate constants obtained in this manner are tabulated in Appendix D . The dependence of kobs on [Fe(EDTA) 2"] for all three variants is given in Figure 3.50. A l l three plots are linear over the Fe(EDTA) 2 " concentration range used. The second order rate constants (kl2) were determined from the gradient of these plots as defined by the following expression A . - A A - A [20] 114 *obs = *i2 [Fe(EDTA)2"] [21] The resulting second order rate constants obtained from these analysis are given in Table 3.18. To account for differences in thermodynamic driving forces and electrostatics on the observed rate constants, the electrostatics corrected self-exchange rate constants exhibited by the proteins in these reactions (£a c o r r ) were calculated with the relative Marcus formalism as developed by Wherland and Gray (1976). These values are also given in Table 3.18. The values of £ n c o r r obtained in this manner fortheMet80Ala(l. 11 M'V 1 ) , Met80Ala/Tyr67Phe(343 M ' V 1 ) and Met80Ala/Phe82Ser (0.04 M ' V l) variants exhibit remarkable range. The corresponding values reported for wild-type yeast cytochrome c (10.9 M ' V 1 ; Rafferty etal, 1990) and horse heart myoglobin (0.02 M ' V 1 ; Lim, 1990) provide useful points of reference for comparison. The activation parameters for these reactions were also determined from the temperature dependence of the second order rate constants with the Eyring equation: The resulting Eyring plots are given in Figure 3.51, and the corresponding enthalpies and entropies of activation are summarized in Table 3.18. The low-spin to high-spin interconversion process exhibited by the Met80Ala/Tyr67Phe variant was also dependent on temperature, and the Eyring plot (Fig 3.51) and activation parameters for this process (Table 3.18) are included in this analysis. In (Ar12/T) = In (£ B/h) + (AS*/R) + (AH*/RT) [22] 115 Table 3.18 : Kinetic and thermodynamic parameters for the reduction of the axial variants Met80Ala, Met80Ala/Tyr67Phe, and Met80Ala/Phe82Ser by Fe(EDTA)2". Protein kn E m kucon AH* AS* AG* ( M V ) (mV) (M's 1) (kcal mol1) (e.u.) (kcal mol1) wild-type cytochromea 7.2 x 104 290(2) l.lxlO 1 3.5(2) -24.7(8) -Met80Ala 26(1) -45(2) 1.1 9.7(4) -19.5(8) 15(1) Met80Ala/ Tyr67Phe reduction 1.4(1) xlO3 11(2) 3.4xl02 7.5(3) -18.7(8) 13(1) L.S. -H.S. - - 12.2(5) -10.3(6) 15(1) Met80Ala/ Phe82Ser 1.4(2) 109(5) 4.3xl0"2 9.2(5) -27.3(9) 17(2) horse heart myoglobinb 22.5(5) 61(1) 2 x 10-2 12(l)c -13(5)c -a pH6.0; Rafferty(1992). bLim(1990). cCassatefa/. (1975). 116 Figure 3.48 Absorption spectra monitoring reduction of Met80Ala cytochrome c by Fe (EDTA)2"(u = 0.5 M sodium phosphate buffer, pH 7.0, 25 °C). The spectra shown were recorded over the initial 30 s of the reaction. Inset shows the change in absorbance monitored at 434.6 nm. The smooth line represents the non-linear least squares fit to the data. 117 400 450 500 550 600 Wavelength (nm) F i g u r e 3.49 A b s o r p t i o n spectra moni tor ing the reduct ion o f M e t 8 0 A l a / T y r 6 7 P h e cy tochrome c b y F e ( E D T A ) 2 " ( u = 0.5 M sod ium phosphate buffer p H 7.0, 25 °C) . ( A ) T h e spectra show the in i t i a l reduct ion and were recorded over the first 500 ms o f the reaction. ( B ) T h e spectra show the l o w - s p i n to h igh-spin convert ion and were recorded over the react ion per iod o f 0.5 to 5 s. ( C ) S ingu la r value decompos i t ion analysis results o f the reaction. T h e calculated spectra are shown for (I), o x i d i z e d protein; (II) reduced protein l o w - s p i n intermediate; (HI) reduced protein h igh-spin final species. Inset in panel A shows the change i n absorbance moni tored at 548 nm. The smooth l ine represents a bi-phasic non-l inear least squares fit to the data. 118 0.5 -0 . 4 -0.3 -0 . 2 -0.1 -0 -25 -20 -15 -10 -5 -0 -0.025 -0.020 -0.015 -0.010 -0.005 -oL [Fe(EDTA) ] (mM) F i g u r e 3.50 Dependence o f the observed rate constants for reduct ion o f ferr icytochrome c ax ia l l igand variants o n F e ( E D T A ) 2 " concentration (u = 0.5 M sod ium phosphate buffer p H 7.0, 25 °C) : ( A ) , M e t 8 0 A l a ; (B) , M e t 8 0 A l a / T y r 6 7 P h e ; ( C ) , M e t 8 0 A l a / P h e 8 2 S e r . 119 Figure 3.51 Eyring plots for the Fe(EDT A) 2' reduction of ferricytochrome c axial ligand variants (u = 0.5 M sodium phosphate buffer pH 7.0, 25 °C). (•), Met80Ala; (•), Met80Ala/Tyr67Phe reduction; (o), Met80Ala/Tyr67Phe low-spin to high-spin conversion (•), Met80Ala/Phe82Ser. 120 4. DISCUSSION 4.1 Overview The work reported in this dissertation addresses two basic issues concerning the factors that dictate the electron transfer reactivity of cytochrome c. The work with flavin-modified cytochromes considers the effect of electron donor structure on the kinetics o f intramolecular electron transfer. In previous studies of electron transfer proteins into which additional oxidation-reduction centers have been introduced, the new reactive centers have invariably been coordination complexes. Sites of this type are usually highly charged in both oxidation states. The use of flavins as protein modifying reagents in the current work is a significant departure from these other studies. A s such, the synthetic flavocytochromes can be regarded in a simple sense as models for naturally-occurring flavocytochromes that are generally less amenable to detailed studies o f the type undertaken here owing to their greater structural and functional complexity and, in many cases, owing to limited information concerning their structures. The studies o f cytochrome variants in which the Met80 axial ligand has been replaced with an alanyl residue extend previous studies of the Met80Ala variant to consider the effects of removing this critical residue on the electron transfer properties of the cytochrome. In addition, new insight is provided concerning the effects of the Met80Ala substitution on ligand binding properties o f the protein through characterization of variants in which additional active site residues believed to be important in controlling the thermodynamics o f ligand binding to the Met80Ala variant have been replaced. 121 4.2 Cysteine Variants and Flavin Modification 4.2.1 Protein Characterization Spectroscopic and electrochemical characterization of the cytochrome c variants and the corresponding modified proteins was essential to interpret the kinetic data. The fact that no appreciable change is observed in either the electronic or CD spectra of the four cysteine variants relative to the corresponding spectra of the wild type cytochrome is indicative that there is little structural change to both the heme environments or the secondary structure of the variants. This observation is further supported by the molecular modeling results which indicate only small localized structural perturbations around the individual mutation sites, as shown in Figure 4.52. The validity of these energy-minimized structures is supported by comparing the calculated structure for Leu85Cys cytochrome c with the crystal structure of the same variant that was elucidated by Lo (1995). This comparison around the mutation site, as shown in Figure 4.52, clearly exhibits a good correlation between the two structures. Modification of the variants with the flavin moiety to the exposed surface cysteinyl residues was shown to be successful by electronic absorption spectroscopy. Results from the peptide mapping analysis confirmed this site of attachment, although the hydrolysis procedure appeared to destroy the majority of the linkage between the flavin and the cysteine residue. The fluorescence data also indicated the attachment of the flavin to a cysteine residue. Flavin fluorescence has been shown to be significantly quenched by coupling to sulfur containing amino acids (Penzer & Radda, 1967) as shown by all the modified proteins. The A m a x e n m bathochromic shift is also consistent with substituted flavins (Falk & McCormick, 1976) and supports the hypothesis that the attached flavin molecule is still within aqueous surroundings and not partially buried within the protein structure. This conclusion further supports the energy-minimized structures of the modified proteins. Further evidence that the 122 Thr8Cys cytochrome c His39Cys cytochrome c Asn62Cys cytochrome c Leu85Cys cytochrome c F i g u r e 4.52 Structural diagrams of the region about the cysteine mutation site in the Thr8Cys, His39Cys, Asn62Cys & Leu85Cys cytochrome c variants. (A) The structure of the wild-type protein (thin lines) is superimposed on the energy minimized-structure of the of the variants (thick lines). ( B ) The crystal structure of the L85C variant (thin lines) is superimposed on the energy minimized-structure of the Leu85Cys variant (thick lines) (Crystal structure of Leu85Cys cytochrome c taken from Lo et al, 1995). 123 modification process does not cause major alterations to the environments of the flavin or heme is provided by both the electronic and CD spectra of the modified proteins. The CD spectra for the unmodified and flavin modified proteins are superimposed in both the visible and far U V regions, indicating that neither the protein secondary structures nor the heme environments have been altered. Electrochemical analysis of the separate protein and flavin components, and the modified proteins, was carried out for two reasons. First, electrochemical analysis was performed to investigate any alterations to the reduction potentials of the variants and the flavin. Such a change would be indicative of a variation in the surrounding environment of the heme or the flavin centers. Second, electrochemical analysis was performed to calculate the individual driving forces associated with each electron transfer reaction studied. The fact that no appreciable change in the reduction potentials of the variants Thr8Cys and Asn62Cys was observed relative to wild type indicates that the heme environments of these proteins was unperturbed, which is expected since these variants possess surface residue substitutions that are somewhat removed from the heme center. The decrease of 6(2) mV in the reduction potential of the Leu85Cys variant is marginally greater than the change in value reported by Lo (1995) which showed a decrease of 2(3) mV. The crystal structure for this variant indicates that the percentage of heme surface area exposed to solvent (9.1 %) is similar to that in wild-type cytochrome c (9.0 %) and the sulfur atom (7.1 A) and a new water molecule (9.1 A) are distant from the heme. Therefore, it is unlikely that this decrease in reduction potential is caused by a change in the polarity of the heme environment. To investigate this issue further, it is necessary to analyze the thermodynamic parameters associated with the dependence of the reduction potential on temperature, as discussed below. Similarly, to address the factors which produce the large (25 mV) decrease in potential observed for the His39Cys variant, the enthalpic and entropic parameters must be studied (Table 3.6). 124 Consideration of these values for the Leu85Cys variant reveals that the enthalpy of reduction of this variant is the same as that for wild-type cytochrome c ( A H 0 = -14.4(6) kcal/mol). As the enthalpy for this equilibrium has been associated with the polarity of the heme environment (Kassner, 1973) it may be argued that the Leu85Cys mutation does not perturb the heme environment and that the decrease in reduction potential must result from other factors. These factors include the alteration of side chain packing within the hydrophobic core and changes in the solvent reordering around the polypeptide chain, both of which are associated with the entropy of reduction (Taniguchi et al., 1980; Murphy, 1993). The small negative change in this entropic value for the Leu85Cys variant (AS 0 = -11.7(2) eu) relative to that for wild-type cytochrome (AS 0 = -11.0(2) eu) indicates that these conformational and solvation differences between the two oxidation states of the protein have been increased to favor the oxidized state and lower the reduction potential. Indeed, changes as small as 2 to 3 eu have been shown to lower the reduction potentials of cytochrome variants by as much as 40 mV (Rafferty, 1992), so it is reasonable to assume that a 0.7 eu decrease in the standard entropy is sufficient to produce a decrease of 6 mV in potential. The 25 mV decrease in reduction potential for the His39Cys variant can also be attributed to an entropic origin (AS° = -16.3(3) eu). For such a large decrease in entropy, relative to that for the wild-type cytochrome, a greater change than the observed 25 mV is anticipated, but this entropic change is partially compensated by a decrease in the standard enthalpy (AH 0 = -15.4(9) kcal/mol for the His39Cys variant; A H 0 = -14.4(7) kcal/mol for wild-type cytochrome). As noted earlier, His39 has been previously identified as the single titratable group with an oxidation-state linked pK3 (Robinson et al., 1983). This identification is consistent with the reduction potential pH dependence of cytochromes c isolated from Rhodapseudomonas viridis and Candida krusei, both of which have a histidine residue at position 39 (Moore et al, 1984). Conversely, horse 125 heart cytochrome c has a lysine residue at position 39 and the potential of this protein is independent of pH between pH 4 and 9. The pKa values for the Thr8Cys, Asn62Cys, and Leu85Cys variants (Table 3.7) are all similar to that of wild-type cytochrome c, within experimental error, and all show a difference between oxidation state pK3s of 0.3-0.4. These results indicate that the mutations do not perturb the His39 environment, and thus the weak interactions between the heme and the histidine residue which are responsible for this oxidation state linked change in the pK3 (Robinson et al., 1983) are unaffected. This result is expected because these three mutations are all situated at positions distant from the His3 9 region. The pK3 values associated with the His3 9Cys variant are approximately 1 p^ a unit higher than those for the wild-type protein, with pK°x = 7.5(2) and pKpA = 7.9(2). This finding is consistent with the hypothesis that a Cys residue at position 39 titrates with an oxidation-state linked pK3. The fact that the difference between these two pKa values remains 0.4 is indicative that the weak interactions which exist between the heme and His39 are still present in the cysteine variant. The electrochemical properties of the free 7-acetyl-10-methylisoalloxazine are consistent with a flavin in which the oxidized and reduced forms each exhibit a single pK3. The observed reduction potential of -113(2) mV is approximately 90 mV higher than the corresponding values of riboflavin and F M N (Draper & Ingraham, 1968). This increase in reduction potential is indicative of the electron withdrawing nature of the substituted acetyl group at position C-7, which is expected to stabilize the reduced flavin species. The values of the semiquinone formation constants for this flavin are similar to those exhibited by other flavins which have been analyzed in this manner and indicate that the acetyl group at position 7 does not perturb the semiquinone formation in any way. The reduction potential, at pH 7.0, for the heme component of each variant is essentially identical to that of the unmodified variant, indicating that the modification does not perturb the heme 126 environment. Similarly, the thermodynamic parameters of the oxidation-reduction equilibrium of each variant are unaffected, providing additional evidence that the flavin attachment does not alter the electrostatic interactions or internal packing properties of the heme environment. On the other hand, the electrochemical behavior of the flavin changes moderately upon modification. The reduction potentials exhibit small decreases upon attachment of the flavin, with the flavin center of modified Thr8Cys cytochrome c having a value of-123(2) mV, 10 mV lower than the value for the free flavin. This result indicates that the oxidized state of the flavin is stabilized, possibly by slight alterations in the electron withdrawing capacity of the position C-7 attachment site. In addition, the thermodynamic parameters for this equilibrium are significantly perturbed upon modification. Both the standard entropy and enthalpy for the reaction decrease relative to those of the free flavin. This finding is probably a consequence of the close proximity of the attached protein. Although it appears that the flavin remains exposed to solvent on the surface of the protein, the oxidation-state linked solvent reorganization around the flavin is undoubtedly affected by attachment to the protein. Another interesting property of the flavin modified variants is the electrochemical behavior of the His39Cys variant. The pH dependence of the reduction potential of this flavin modified protein is abolished over the pH range of 5.5 to 8.0. This behavior is consistent with the fact that the titratable cysteine proton has been removed by the formation of the thioether bond. Similar observations have been reported for the Rhodopseudomonas viridis cytochrome c2, in which modification of His39 with diethylpyrocarbonate eliminates the pH-linked titratable group (Moore et al, 1984). These electrochemical results indicate that the donor and acceptor components of flavin-modified cytochromes c are generally quite similar to those of cytochrome c and the unreacted flavin. 127 4.2.2 Electron Transfer Kinetics 4.2.2.1 Intermolecular Electron Transfer Reactions The bimolecular electron transfer kinetics were studied primarily as a control for the analysis of the unimolecular kinetics, namely to investigate whether these intermolecular reactions could interfere with the intramolecular results. Nevertheless, the results of these studies are of interest in their own right in view of the use of such information by Tollin, Cuzanovich and their coworkers over the past several years. It has been previously established that the site of electron transfer in bimolecular reactions involving cytochromes c and flavin semiquinones is near the exposed heme edge region of the protein (Miller & Cuzanovich, 1974; Smith et al, 1980). ESR experiments have been interpreted as indicating that the reactive region of the flavin semiquinone species is between the N-5 and C-8 atoms of the central pyrazine ring and adjacent aromatic ring (See Figure 4.53) (Guzzo & Tollin, 1964). The rate constants for such reactions are controlled not only by the steric and electrostatic factors affecting the rate of diffusion but also by the thermodynamic driving force for the reaction, which is defined by the difference in reduction potentials of the protein and the flavin. Although the reduction potentials of 7-acetyl-10-methylisoalloxazine and 7cc-bromoacetyl-10-methylisoalloxazine are -100 mV higher than riboflavin, these flavins reduce horse heart cytochrome c much more rapidly than does riboflavin. This observation reflects the steric effect of the ribose moiety (-[CH 2(CHOH) 3CH 2OH]) of riboflavin which is found at position N-10 (Figure 4.53). The rate constants for the reaction of the two synthetic flavins with ferricytochrome c are comparable to that of lumiflavin (Ahmad et al, 1981) (k2 = 0.62x108 s"1). All three of these flavins possess a methyl group substituted at position N-10. A second factor which may contribute to the more facile reduction of cytochrome c by the synthetic flavins is the addition of the acetyl group at 128 7-acetyl -10-methyl isoal loxazine 7a -b romoace ty l - l O-methyl isoal loxazine F i g u r e 4.53 Structures o f f lavins i nvo lved in inter-molecular electron transfer reactions w i t h cytochromes c. 129 position C-7. Although the size of the acetyl group is greater than that of a methyl group, thus increasing steric restraints, the acetyl group may allow for increased derealization of the electron density from the isoalloxazine ring, thereby increasing the reactive surface of the flavin which is able to interact with the protein. The effect of the change in the pK values for the semiquinones must also be considered. The pKa for riboflavin is 8.3 (Draper & Ingraham, 1985) while for 7-acetyl-10-methylisoalloxazine and 7a-bromoacetyl-lO-methylisoalloxazine the pKs values are both 8.2. Thus a higher concentration of the more reactive anionic semiquinone form, albeit small, will be present with the synthetic flavins at pH 7.0. The difference in rate constants between 7-acetyl-10-methylisoalloxazine and 7a-bromoacetyl-lO-methylisoalloxazine is most likely due to the increased steric hindrance caused by the larger bromine atom. The electronegative properties of the bromine will also alter the charge distribution in the flavin, by increasing electron density within the benzene ring and thus causing a less favorable interaction with the non-polar protein heme region, as has been observed with chlorinated flavins (Ahmad et al, 1981). From this discussion it is apparent that the factors that contribute to diffusion have a greater influence on the reaction of the flavin and the cytochrome than does the driving force of the reaction. The results of this bimolecular kinetic analysis also show that the rate constant for 7-acetyl-1 O-methylisoalloxazine reduction of horse heart cytochrome c is nearly half that of the reaction of the same flavin with yeast iso-l- cytochrome expressed in yeast or E. coli. This result can be correlated with the proportion of the heme group that is exposed to solvent and thus available for direct interaction with the flavin. For horse heart cytochrome c the surface of the heme exposure to solvent is 34.7 A 2, which constitutes 0.7% of the protein surface, while for yeast /'so-1-cytochrome c these values are 44.4 A 2 and 0.9% respectively (Brayer & Murphy, 1996). To date no crystal structure of the yeast cytochrome expressed in E. coli has been obtained, but the present kinetic 130 results suggest that his feature of the yeast protein structure will be the same regardless of the organism in which it is expressed. Also, as noted by Meyer etal. (1984) the side chains of residues 16, 28, and 81 project into solution from the region around the heme crevice. For horse heart cytochrome, these residues are Gin 16, Thr28 and Ile81 whereas for yeast iso-l-cytochrome c, these residues are Gin 16, Val28 and Ala81. These same residues are found in the yeast Candida krusia cytochrome c, which has been shown previously to react with F M N approximately 1.25 times faster than does horse heart cytochrome c, owing to decreased steric hindrance resulting from substitution of Ile81 with Ala. Therefore, the expected rate constants for yeast wo-1-cytochrome c should be in the region of (0.63 x 108) x 1.25 x (0.9/0.7) or 1.01 x 108 M " V . This value is in excellent agreement with the experimental data results (iso-l -cytochrome c expressed in yeast, kn = 1.03(7) x 108 M ' V 1 ; /so-1-cytochrome c expressed inE. coli, k12 = 1.05(8) x 108 M 'V 1 ) . Finally, the rate constants for flavin reduction of the Cys-containing variant are within experimental error of those for the wild type cytochrome. This result is not surprising because the cysteine residues are not near the heme crevice and should not alter the steric characteristics for reaction of the flavin with the active site heme. The activation parameters for the protein reduction by various flavins are consistent with previous reports for the reaction of cytochromes c with small electron transfer reagents (Rafferty et al, 1996) (/so-1-cytochrome c with Fe(EDTA) 2' AH* =3.5 kcal mol'1, AS* = -24.7 e.u. & iso-l-cytochrome c with Co(phen)33+ AH*= 11.4 kcal mol"1, AS* = -5.5 e.u.). The low activation enthalpies are indicative of the large driving force for the reaction and the low electrostatic repulsion between reactants that result from the neutrality of the flavin semiquinone. The importance of the flavin orientation for efficient electron transfer gives rise to the low activation entropies because the number of degrees of freedom for the collisional complex formation has been reduced. The activation parameters for the three bimolecular reactions involving the flavins are similar. This similarity 131 indicates that the mechanism for the precursor complex formation involved in these reactions is, as expected, essentially the same. 4.2.2.2 Intramolecular Electron Transfer Reactions The unimolecular electron transfer kinetic studies exhibit a temperature dependence that is consistent with the results observed for ruthenium modified cytochromes. The reorganization energies (A) derived from the temperature dependence of these reactions are in the range of RuA4L2/cytochrome c systems studied previously (0.8-1.3 eV) (Isied et al., 1984; Nocera et al., 1984). The overall reorganization energy for the flavoproteins (1.2(2) eV) leads to a flavin donor center reorganization energy of 0.7(2) eV when the value of 0.5 eV is used for the reorganization energy of the cytochrome c (Marcus & Sutin, 1985). This seemingly high value for X for the flavin is a combination of both the Xm and A o u , for the flavin and the surrounding solvent. The magnitude of X for the flavin can be justified by considering the large energetic barrier for distortion of the equilibrium nuclear geometries of organic molecules in aqueous media (Dutton & Moser, 1994). In addition a recent study by Millet and co-workers has estimated the reorganization energy of cytochrome c to be 1.5 eV, thus giving the protein component a value of 0.75 eV (Fairris et al, 1996), which leads to a lower X value of 0.45(20) eV for the flavin component. To investigate the correlation between the intramolecular electron transfer rate constants and the electronic coupling of the donor and acceptor centers the k^ and experimental values for each variant were calculated. The energy minimized molecular models for the modified proteins were also utilized to search for possible relationships between coupling factors and electron transfer rates. Numerous theories have been proposed to describe electronic coupling within protein structures. The simplest of these models assumes a homogeneous medium between the donor and acceptor and leads 132 to an exponential decrease in the rate of electron transfer with increasing donor-acceptor separation. As outlined in the introduction, studies have shown that although this hypothesis works well for most systems, it does not account adequately for electron transfer pathways involving non-covalent components. Indeed if the kmax values for the flavin modified cytochromes c are plotted against the simple donor-acceptor separation distance (d-3) the experimental data fall below a reference line that is based on the assumption that P is 1.4 A (Figure 4.54). This result, therefore, suggests that any model used to represent the results of this work should account for the heterogeneic nature of the protein medium located between the flavin and the heme centers. If the model of Beratan and Onuchic is used, a better correlation between the calculated a-tunneling lengths and kmix is observed although the flavin-modified His39Cys variant deviates from this relationship. It is of interest to note that the k^ values for the flavin modified cytochromes studied in this work, with the exception of that for the modified F£is39Cys variant, lie above the line calculated for ruthenium modified cytochromes. This line is calculated using a value of P' = 0.73 A"1, and the one bond limit electron transfer rate, the rate when the donor and acceptor orbitals can overlap directly, is 1.7 x 1012 s"1 (Figure 4.54). This observation is indicative that the electronic coupling between the flavin donor and the protein heme acceptor is greater than that observed for the ruthenium modified cytochromes. Preliminary comparisons between the flavin- and ruthenium-modified proteins, in which the donor is attached to residue 39, suggest that the electronic coupling was actually weaker for the flavocytochrome, even after the consideration of the two additional covalent bonds involved in the flavoprotein pathway resulting from differing modes of attachment (Twitchett et al, 1997). However, analysis of the three other synthetic flavoproteins sheds doubt upon this initial conclusion. Indeed, kmax for electron transfer from a flavin attached to position 62 is slightly greater than k^ for the corresponding ruthenium-cytochrome despite the presence of an 133 0 0 5 10 (d- 3) / A 15 10 20 o l / A 30 40 F i g u r e 4.54 Corre la t ions o f m a x i m u m E T rates i n f l av in ( C y s X ) mod i f i ed cytochromes c ( o ) & R u ( H i s X ) modi f i ed cytochromes c ( • ) w i t h (d-3) (a) and o l (b) ( R u ( H i s X ) modi f i ed cytochromes c data taken f rom Wut tke et al, 1992 and C a s i m i r o et al, 1993). (a) S o l i d reference l ine has a slope o f 1.4 A"1 and intercept o f 1 x 1 0 1 3 (b) S o l i d reference l ine (best fit to R u data) has a slope o f 0.73 A"1 and intercept o f 3 x 10 1 2 ; Dot ted reference l ine (best fit to f l av in data) has a slope o f 0.73 A - 1 and intercept o f 1 x 10 1 3 . 134 additional covalent bond in the pathway selected for the flavocytochrome. One possible explanation for this result is the likely motion of the flavin donor about its attachment site. The energy-minimized structures of the flavoproteins must be considered to be static and as such cannot take into consideration the fact that the flavin donor and the modified protein could conceivably attain a conformation in which electron transfer would be enhanced or retarded. A similar hypothesis was given as a possible explanation for the poor correlation observed between ket and donor-acceptor distance in cytochromes c containing covalently attached cobalt cage complexes. For these cytochrome derivatives the rate constants varied over the narrow range of 1.0 to 3.2 s'1 despite distance dependence models that predicted a 200 - 2000 fold variation in ket (Conrad et al, 1992). In the present case interpretation of the kinetics would be facilitated by availability of information concerning electron transfer kinetics of cytochrome c with ruthenium attached to residues 8 and 85. If the rate constants for the flavocytochrome kinetics are fit to a line calculated with the assumption that P'= 0.73 A"1 as for ruthenated cytochromes, then the one bond limit electron transfer rate is ~1 x 1013 s"1 (Figure 4.54). This value closely approximates the rate constant assumed for an adiabatic electron transfer process and demonstrates a reasonable correlation between donor-acceptor distance and the rate of electron transfer. Furthermore, these rate constants demonstrate that for long distance electron transfer reactions within proteins the electronic coupling factor is dependent on the nature of the donor or acceptor. Nevertheless, discrepancies arise using this pathway model to describe the flavocytochrome c as illustrated by the difficulty of this approach with the behavior of the His39Cys variant. This discrepancy could be a consequence of the fact that although the version of the Beratan-Onuchic model used utilizes multiple pathways it does not consider interferences between these pathways. It should be noted that recent advances concerning the Beratan-Onuchic model have considered the nature of these coupling pathway interferences (Regan et al, 1993). 135 This pathway interference problem is addressed by the method of Siddarth and Marcus as the values calculated with this model exhibit an excellent correlation between the theoretically and experimentally derived electronic coupling values. This agreement suggests that the Siddarth-Marcus model resolves the discrepancies found in the distance dependence of the rates, thereby suggesting that interactions between pathways need to be considered in calculating electronic couplings as mentioned above. The values calculated by this method for the flavin modified cytochromes are greater than those calculated for the corresponding ruthenium-modified cytochromes as least for modifications at positions 39 and 62 (flavocytochrome c Cys39 = 0.03 cm'1; Ru cytochrome c His39 = 0.01 cm"1; flavocytochrome c Cys62 = 0.02 cm"1; Ru cytochrome c His62 = 0.002 cm'1) (Siddarth & Marcus, 1993a). This difference is undoubtedly a consequence of the difference in residues through which the flavin (Cys) and ruthenium (His) groups are attached to the cytochrome. If correct this difference in H ^ values between flavocytochromes and ruthenium modified cytochromes means that the electronic coupling between the flavin and heme centers is greater than between the ruthenium and heme centers. It must be emphasized, however, that the values obtained by this method are based on extended-Huckel theory and as such the absolute values are not as reliable as the relative values within a particular system. The electron transfer pathways identified for each flavoprotein by the Siddarth-Marcus and Beratan-Onuchic models merit consideration in an effort to understand why the results of the two analyses differ. For the modified Thr8Cys variant, which exhibits the largest ket of the four derivatives examined, Leu9, PhelO, and Cys 14 are implicated in coupling the flavin and heme centers by both models. The Siddarth-Marcus AI search, however, also indicates a role for Ala7, Lysl 1 and Argl3 in this process. As residues 7, 11, and 13 reside in an alpha helix, it is possible that this secondary structural element may enhance electronic coupling. The ordered structure of a helix combined with 136 the number of strong hydrogen bonds present in such structures may provide the basis for this property of an alpha helix. Similar observations have been made regarding a p-strand secondary structure element (Langen et al, 1995). These studies demonstrated that the ordered structure of P-strands led to enhanced electronic coupling between the donor-acceptor sites, more so than a-helical segments. Nevertheless, it is reasonable to assume that an alpha helix will provide enhanced coupling relative to a more unstructured portion of the protein structure (Winkler & Gray, 1997). The residues selected for the modified Leu85Cys variant by the AI search procedure do not correspond to those selected by the Beratan-Onuchic model. The residues identified by the AI procedure is a shorter, more direct path ( H ^ = 0.04 cm'1) and results in an electronic coupling similar to that of the modified His39Cys variant ( H ^ = 0.03 cm'1). The Beratan-Onuchic model fails to find a suitable H-bond or through-space interaction between the donor and acceptor sites in this case. Finally, the energy minimized structure of the flavin-modified Asn62Cys variant indicates that the flavin group is adjacent to residues Glu-4 and Glu66. This fact is considered by the AI search, which identifies residue Glu66 as a participant in donor-acceptor coupling. The only other discrepancy between the results of the Siddarth-Marcus and the Beratan-Onuchic analysis is that the Siddarth-Marcus analysis indicates a role for Met64 that is not identified by the Beratan-Onuchic model. Otherwise, both models agree that Asn63, Tyr76, and Met80 are involved in donor-acceptor coupling in this flavin-modified cytochrome derivative. For the flavin-modified cytochromes studied here, it is interesting to note that the Siddarth-Marcus analyses all identify structural elements as involved in donor-acceptor site coupling that enhances the efficiency of electron transfer relative to the pathways identified by the Beratan-Onuchic model. This observation is presumably the basis for the consistent difference between rate constants observed in the intramolecular electron transfer reactions studied here and those predicted by the 137 Beratan-Onuchic model (Figure 4.54). 4.3 Axial Ligand Variants 4.3.1 Protein Characterization The spectroscopic characterization of the Met80Ala, Met80Ala/Tyr67Phe and Met80Ala/Phe82Ser cytochrome c variants presented in this work expands on the characterization previously reported by the group of H. B. Gray on the Met80Ala and Met80Ala/Tyr67Phe variants. In addition to the elucidation of a number of completely new functional properties for the variants, several experiments were repeated to determine whether the cytochromes c isolated from E. coli exhibited characteristics different from that of the corresponding proteins isolated from yeast. The resulting characterization of these variants was a prerequisite for investigation of the electron transfer reactivity of these proteins that is discussed in the next section (4.3.2). All three axial ligand ferricytochrome variants display electronic absorption spectra that are indicative of a six-coordinate heme iron center in which a water molecule is axially coordinated at an exchangeable ligand binding site that is produced by the replacement of Met80 with an alanyl residue. This coordinated water molecule titrates with a pKa of 6.2 as observed by Band etal. (1995a) for the Met80Ala variant of horse heart cytochrome c produced by semisynthetic methods. This pKa is significantly lower in value than the corresponding pKa for wild-type horse heart myoglobin (pKa -8.93, Antonini & Brunori, 1971) and is lower than the pKa of 7.6 reported for free heme (Shack & Clark, 1947). Bren and colleagues (1995) speculated that the low pKa of the Met80Ala variant may result from stabilization of a coordinated hydroxyl group by proximity of the phenolic side chain of Tyr67 through hydrogen bond formation. The pKa values observed here for the Met80Ala/Tyr67Phe and Met80Ala/Phe82Ser double variants indicate that this hypothesis is incorrect. If the hydroxyl 138 group of Tyr67 stabilizes a coordinated hydroxyl ligand, then removal of this hydroxyl group should increase the pKa of the coordinated water molecule. However, the pKa of 5.8 exhibited by the Met80Ala/Tyr67Phe variant is even lower than that of the single variant. This decrease in pKa suggests that any interaction between Tyr67 and the bound hydroxide actually has a destabilizing effect, possibly due to steric and/or electrostatic interactions. Thus it is more likely that the decreased pKa exhibited by the Met80Ala variant relative to wild type myoglobin results from more general differences between the distal heme pockets of the two proteins. On the other hand, the pKa of the coordinated water molecule in the Met80Ala/Phe82Ser double variant is more than one pKa unit greater than that of the Met80Ala variant. The Phe82Ser substitution is expected to increase the polarity of the heme binding site by placing a more electronegative residue at position 82 and by introducing a solvent channel that increases the exposure of the heme prosthetic group to solvent (Louie et al, 1988). Indeed, use of the program SURFCV (Sridharan et al, 1995) to estimate the solvent accessibility of the heme prosthetic group in the three variants with the Met80Ala substitution considered in this work leads to the expectation that the heme group in the Met80Ala/Phe82Ser double variant is 35% more exposed to solvent than the heme groups of the Met80Ala or Met80Ala/Tyr67Phe variants. Nevertheless, the heme groups of these latter two proteins are about 50% more exposed to solvent than is the case for the wild-type protein (Figure 4.55). Interestingly, the degree of heme exposure to solvent correlates with the thermal stabilities of these proteins as represented by the melting temperature, Tm. This correlation probably results from the combined destabilizing effects of elimination of the Met80 axial ligand and the decreased hydrophobic character of the heme binding site through increased solvent accessibility and introduction of relatively hydrophilic residues. Further information concerning the effects of these substitutions on the environment of the 139 D F i g u r e 4.55 Space f i l l i n g representations o f cytochromes c showing heme solvent accessibi l i ty . A ) W i l d - t y p e B ) M e t 8 0 A l a C ) M e t 8 0 A l a / T y r 6 7 P h e D ) M e t 8 0 A l a / P h e 8 2 S e r (Left structures show side-on v i e w ; R igh t structures show top-side v iew) . C o l o r legend: R e d , heme; Y e l l o w , axia l residues 18 & 80; Orange, residue 67; Green, residue 82. 140 heme was sought from considering the stability of the oxygenated derivatives to autoxidation and the kinetics of CO recombination. Bren and Gray (1993 a) have previously reported that the Met80Ala variant of horse heart cytochrome c prepared by semi-synthesis is relatively stable to autoxidation. These authors attributed the stability of the oxygenated form of the variant to stabilizing interactions of the coordinated dioxygen with Tyr67 and Phe82 and to the protection of the ligand binding site by the overall protein fold. The 20-fold increase in the rate constant for autoxidation of the Met80Ala/Phe82Ser variant relative to that for the Met80Ala variant is consistent with this argument in that replacing Phe82 with a seryl residue significantly increases the hydrophilic character of the heme binding site (Louie et al. ,1988). Autoxidation of oxymyoglobin and oxyhemoglobin is generally regarded to occur through formation of the corresponding deoxygenated intermediate prior to oxidation of the heme iron. If this same mechanism operates for cytochrome variants with the Met80Ala substitution, then the solvent channel introduced by replacing Phe82 with Ser may promote autoxidation by providing a facile route for ligand dissociation that reduces the barrier to dissociation of coordinated dioxygen and promotes formation of the deoxygenated intermediate. The Met80Ala/Tyr67Phe double variant exhibits a smaller, six-fold increase in the rate constant for autoxidation relative to that of the Met80Ala variant. The simplest explanation for this observation is destabilization of the oxygenated derivative by the loss of a hydrogen-bond involving Tyr67 and the coordinated dioxygen. The results obtained in the present work concerning the rebinding of CO following flash photolysis are consistent with these conclusions. Most notably, the Met80Ala/Phe82Ser double variant exhibits the fastest rate of CO association of all three variants possessing the Met80Ala substitution as expected if the solvent channel created by the Phe82Ser substitution facilitates ligand diffusion into the ligand binding site of this variant. On the other hand, the current results are 141 distinctly different from those reported by Bren and Gray (1993a). These authors observed a rate constant for CO recombination with the Met80Ala variant that was two orders of magnitude lower than that observed in the present study. In addition, Bren (1996) failed to detect CO recombination with the His39Asn/Met80Ala double variant of yeast cytochrome c, a result for which no explanation is apparent. Although it is clear that much remains to be learned from pursuing thermodynamic studies of ligand binding to variants of this type, the results of the present studies, at least, are internally consistent, albeit of limited scope. FTIR analysis of the carbonyl derivatives of these variants provides another source of information concerning the active sites of these variants. Carbonyl derivatives of heme proteins frequently exhibit complex infrared spectra because the coordinated CO can often assume more than one orientation. As a result, multiple vibrational conformers can arise from a variety of steric and electrostatic interactions between the CO and residues in the distal heme pocket (Ray etal, 1994). In the case of the carbonyl derivative of the Met80Ala variant a major CO stretching band is observed at 1956 cm"1. A similar stretching frequency has been observed for carbonyl myoglobin arid interpreted to be characteristic of a CO ligand that is parallel to or slightly tilted from the normal to the heme plane (Ormos et al, 1988; Ray et al, 1994). This binding orientation is consistent with the three-dimensional solution structure of the cyanide derivative of the Met80Ala variant determined by ^ - N M R spectroscopy (Banci et al, 1995a). In this structure, the cyanide ligand is oriented perpendicular to the heme; the relatively large cavity in the distal heme pocket resulted in significantly reduced steric restraint. A shoulder at 1969 cm'1 is also observed in the spectrum of the carbonyl derivative of the Met80Ala variant. This conformer may result from a CO conformer in which interaction of the bound CO with the relatively electronegative Tyr67 hydroxyl group increases the C — O bond order. This interpretation is consistent with the absence of this shoulder in the spectrum 142 of the Met80Ala/Tyr57Phe carbonyl derivative. On the other hand, this shoulder is also absent from the spectrum of the Met80Ala/Phe82Ser carbonyl derivative despite the presence of Tyr67 in this protein. Perhaps the removal of the relatively bulky Phe82 residue permits greater mobility of the coordinated CO ligand so that interaction with the Tyr67 hydroxyl group no longer offers sufficient energetic advantage. The 4 cm"1 shift observed in the principal carbonyl stretching frequency of this variant can be attributed to these structural differences as well as to increased polarity and solvent accessibility of the coordinated CO. 4.3.2 Electron Transfer Properties All three axial ligand variants studied in this work exhibit reduction potentials that are significantly lower than that of wild-type yeast cytochrome c (290 mV, pH 6.0 (Rafferty, 1992)). This characteristic undoubtedly results at least in part from the loss of the electron accepting character of the axial Met80 ligand which normally contributes to the stability of the reduced form of the protein. In the case of the Met80Ala/Phe82Ser variant, the increased accessibility of the heme to solvent that results from replacement of Phe82 will help stabilize the oxidized cytochrome (Rafferty etal., 1990) as indicated by the fact that the double variant with this substitution exhibits a potential that is -65 mV lower than that of the Met80Ala variant. On the other hand the structural basis for the observation that the potential of the Met80Ala/Tyr67Phe double variant is higher than that of the Met80Ala single variant is more challenging to explain. The internal nature of the Tyr67Phe substitution should not result in any change in exposure of the heme prosthetic group to solvent. However, the removal of the phenolic hydroxyl group of the tyrosyl residue and introduction of the non-polar phenylalanyl residue will decrease the polarity of the heme environment. A decrease in the dielectric of the heme environment should increase the reduction potential of the heme iron, but the 143 anticipated magnitude of this effect is difficult to quantify. Although quantitative rationalization of the electrochemical behavior of the variants with the Met80Ala substitution is a challenging undertaking, knowledge of the reduction potentials of these variants is useful in the interpretation of the kinetics by which they are reduced by Fe(EDTA)2". Application of relative Marcus theory to these results as developed by Wherland and Gray (1976) permits estimation of the apparent self-exchange rate constant exhibited by the variants in their reaction with this reducing agent. In this analysis, the reactivity of the reducing agent is accounted for through knowledge of its self-exchange rate constant, the thermodynamic driving force of the reaction of each variant with Fe(EDTA)2" is accounted for through knowledge of the reduction potentials of the reactants, and the electrostatic contribution to precursor complex formation and successor complex dissociation is accounted for by modified Debye-Huckel theory. Note that in the present case, the use of a single reducing reagent and the minimal effect of the mutations on electrostatic properties of the variants should, in any case, make these contributions to the reduction kinetics essentially identical for each variant. In other words, the principal basis for differences in the electrostatics corrected self-exchange rate constant (kueon) for the variants in their reaction with Fe(EDTA) 2' is expected to be differences in the thermodynamic driving forces for each of the reactions, differences in the manner in which Fe(EDT A)2" interacts with the surface of each protein, and mutation-induced differences in the Franck-Condon activation barrier. The kucorr value derived from the reduction of the Met80Ala variant by Fe(EDTA)2* (1.1 M" 1 s"1) is intermediate between the corresponding values exhibited by wild-type yeast cytochrome c (10.9 M-'s^Raffertyera/., 1992;Rafferty, 1992)) and the value for myoglobin (0.02 M" 1 s'1 (Lim, 1990)). The low reactivity of metmyoglobin in reduction by Fe(EDTA)2" has been attributed previously to the larger reorganizational energies required for the oxidation state-linked change in coordination number 144 of this protein. This conclusion is supported by the observation that the apparent self-exchange rate constant of the Met80Ala variant in this reaction is about an order of magnitude lower than that of the wild-type protein. The fact that kncon for the Met80Ala variant is nonetheless two orders of magnitude greater than that of myoglobin suggests that other structural or electronic factors contribute significantly to the relative reactivity of these two proteins. A similar conclusion resulted from similar analysis of the Val68His variant of horse heart myoglobin. This variant has been shown to be six-coordinate in both oxidation states (Lloyd et al., 1995) and exhibits a kuco" of 0.25 M ' 1 s'1 during reduction by Fe(EDTA)2" (Harris et al., 1997). Although this value is about an order of magnitude greater than the corresponding value exhibited by wild-type myoglobin, it is significantly lower than that of cytochrome c or cytochrome b5 (Reid et al., 1986), the heme iron center of which also possesses a bisimidazole coordination environment in both oxidation states. At least two additional contributions to the difference in reactivity between wild-type cytochrome c and the Met80Ala variant should be considered. First, the electronic properties of the two proteins are non equivalent. Although the heme iron of the wild-type protein is low-spin in both oxidation states, the iron center of the oxidized variant is predominantly low-spin while the reduced variant is a combination of low-spin and high-spin species (Bren, 1996). Similar electronic considerations may contribute to the low reactivity of the Val68His variant of horse heart myoglobin relative to cytochromes c and b5 (Harris etal., 1997). Second, the hydrogen bonding network that involves Tyr67, Met80, and Wat 166 in the wild-type cytochrome (Louie & Brayer, 1990; Berghuis & Brayer, 1992) is clearly disrupted in the Met80Ala variant. Reduction of the wild-type protein involves formation of a hydrogen bond between Tyr67 and Met80 that cannot form in the variant. At present, the only three-dimensional structure available for the Met80Ala variant has been determined by NMR analysis of the cyanide adduct (Banci et al., 1995), so detailed knowledge of the 145 hydrogen-bonding characteristics of the active site of this protein and the oxidation state-linked dependence of these interactions remains to be elucidated. From these considerations, it is apparent that the reorganization barrier to electron transfer of the Met80Ala variant is a critical factor in eventual development of a rigorous understanding of the electron transfer reactivity of this protein and that determination of the structure of the protein in both oxidation states will be required for this achievement. The multiple variants with the Met80Ala substitution will be subject to many of these same considerations, but their behavior will differ from that of the Met80Ala variant insofar as these proteins experience further changes in internal hydrogen-bonding and introduction of internally-bound water molecules. In addition to the reorganizational barrier, the manner of interaction of the wild-type protein and the Met80Ala containing variant with Fe(EDTA)2" may differ from one another. The fact that the apparent self-exchange rate constant for the reaction of the Met80 Ala variant with this reducing agent is two orders of magnitude greater than that of myoglobin may indicate that Fe(EDTA)2" can interact more effectively with the heme group of the cytochrome variant. This assumption is reasonable considering the Met80Ala mutation increases the solvent exposure of the heme (section 4.3.1). Greater exposure of the heme should lead to more effective interaction of the reducing agent with the protein to result, effectively, in a decreased donor-acceptor separation. Comparison of the activation entropy of the Met80Ala variant in this reaction (AS* = -19.5(8) e.u.) with that of the wild-type protein (AS* = -24.7 e.u. (Rafferty et al, 1990; Rafferty, 1992)) may be interpreted as support for this conclusion. While the difference in activation entropy exhibited by the wild-type and Met80Ala variant may be related to the formation of the precursor complex, the activation enthalpies observed in these reactions are influenced at least in part by the differences in reorganization barrier or oxidation state-146 linked changes in coordination environment exhibited by the two proteins. The greater activation enthalpy exhibited in this reaction by the Met80Ala variant (AH* = 9.7(4) kcal mol"1) compared to that of the wild-type protein (AH* = 3.5 kcal mol"1 (Rafferty et al, 1990; Rafferty, 1992)) is consistent with the view that the variant undergoes more extensive structural change upon reduction. Notably, the activation enthalpy for reduction of metmyoglobin by Fe(EDTA)2" is even slightly greater (AH* = 12.1 kcal mol"1 (Lim 1990)). The reduction of the Met80Ala/Tyr67Phe double variant by Fe(EDTA) 2' displays biphasic kinetics. The initial, faster phase corresponds to the reduction of the heme iron center to produce a low-spin ferric species that is probably six-coordinate. The rate constant for the second, slower phase is independent of Fe(EDTA)2" concentration. This observation combined with the electronic spectra observed in the rapid-scan stopped flow kinetics experiments lead to the conclusion that this phase corresponds to the conversion of the low-spin species to a form of the protein that exhibits a spin equilibrium similar to that observed for the Met80Ala ferricytochrome. The fact that the low-spin ferricytochrome intermediate is not observed during reduction of the Met80Ala variant indicates that some structural feature of the Tyr67Phe substitution in the double variant stabilizes the low-spin ferricytochrome intermediate. Evidence for such stabilization is provided by the lower pATa value obtained for titration of the water molecule coordinated to the Met80 Ala/Tyr67Phe ferrocytochrome double variant relative to the value obtained for the Met80Ala single variant. Regardless of the nature of the basis for the formation of the low-spin intermediate, this feature of the protein provides an opportunity to compare the self-exchange rate constants, kncorT of cytochrome derivatives with differing coordination environments and to consider the resulting effects of these differences on the reorganization energies of each protein. The ! C o r T exhibited by the Met80Ala/Tyr67Phe variant during reduction by Fe(EDTA) 2' (343 147 M" 1 s"1) is an order of magnitude greater than the value exhibited by the wild-type protein and two orders of magnitude greater than that of the Met80Ala variant. Previous work has established that the Tyr67Phe substitution alone increases knco" significantly and concluded that this increase results primarily from a decrease in the reorganization barrier (Rafferty et al., 1990). The single variant has been shown by X-ray crystallographic structure determination to possess an additional water molecule near the heme in approximately the position occupied by the phenolic hydroxyl group of Tyr67 in the wild-type protein (Berghuis et al., 1994). Although it is possible that a related change occurs at the active site of the Met80Ala/Tyr67Phe variant, it is more likely that further changes occur as the result of the relatively substantial structural change introduced through the Met80Ala substitution. The activation parameters for reduction of the double variant provide some additional insight. The activation enthalpy determined for the fast, electron transfer, phase of the reaction of the double variant (7.5(3) kcal mol -1) is somewhat lower than the corresponding value for the Met80Ala single variant (9.7(4) kcal mol"1). This is expected because the enthalpic energy associated with the change in coordination number has been removed. On the other hand, the activation entropy for the fast phase for reduction of the double variant (-18.7(8) e.u.) and the Met80Ala variant (-19.5(8) e.u.) are comparable to each other. This observation indicates that the factors controlling the formation of the precursor complex are similar for both proteins. The apparent self-exchange rate constant of the Met80Ala/Phe82Ser double variant (kneon = 0.04 M ' 1 s'1) is surprisingly low. The greater exposure of the heme prosthetic group of this protein to solvent combined with the previous observation that the Phe82Ser substitution increases the apparent self-exchange rate constant (Rafferty et al., 1990) leads to the expectation that the reactivity of the protein with Fe(EDTA) 2' should be enhanced relative to that of the wild-type protein. Furthermore, spectroscopic evidence suggests that this protein remains six-coordinate in both 148 oxidation states, a characteristic that is also exhibited by the wild-type cytochrome (knco" = 10.9 M ' 1 s"1) and the Met80Ala/Tyr67Phe variant (kncon = 343 M ' 1 s'1). The reactivity of the Met80Ala/Phe82Ser variant is essentially the same as that of myoglobin (kncon = 0.02 M" 1 s"1) in this reaction. One possible contribution toward this behavior may be that at pH 7.0, the heme iron atom of the oxidized variant is predominantly high-spin, and the sixth ligand is probably a water molecule rather than the hydroxide ligand proposed for the Met80Ala and Met80Ala/Tyr67Phe variants. The inefficiency of electron transfer exhibited by the Met80Ala/Phe82Ser variant, therefore, could be explained by the combined effects of the change in spin state that occurs upon reduction of the protein and the change in axial ligation from a water molecule in the oxidized protein to an hydroxide group in the reduced protein. This model leads to the expectation of a large reorganization barrier for electron transfer by this protein which is consistent with the activation enthalpy for reduction by Fe(EDTA)2" (AH* = 9.2(5) kcal mol"1). This value is comparable to the corresponding values exhibited by other proteins that undergo oxidation state-dependent changes in coordination number (e.g., myoglobin, AH* =12.1 kcal mol'1; Met80Ala variant of cytochrome c, AH* = 9.7(4) kcal mol"1). For comparison, the activation enthalpy for reduction of wild-type ferricytochrome c, which is six-coordinate and low-spin in both oxidation states, by Fe(EDTA)2" is just 3.5 kcal mol'1 (Rafferty etal., 1992). In principle, this hypothesis could be evaluated by studying the kinetics of the reaction at pH 8.0 or greater in an effort to maintain hydroxide as the sixth ligand in both oxidation states. Unfortunately, at alkaline pH, the oxidized cytochrome undergoes a pH-linked conformational change in which one of two lysyl residues can bind to the heme iron (Theorell & Akesson, 1941; Ferrer et al., 1993; Rosell et al., 1997). This property of the protein renders kinetic studies relatively complex under these conditions. 149 4.4 Concluding Remarks The work described in this dissertation uses two novel types of cytochrome c variant to study some of the structural factors that are known to contribute to the efficiency of inter- and intramolecular electron transfer reactions. The Cys-containing variants permit site-specific chemical modification with flavin residues that provide a useful alternative to the transition metal complexes that have been used widely as protein modifying agents for introduction of new electron transfer centers into proteins. A major result arising from kinetic studies of the synthetic flavocytochromes produced from these variants is that the flavin center has a reorganization energy, A, of 0.7(2) eV. This value is greater than that exhibited by ruthenium centers in similar environments. In addition, the correlation of the rate constants for intramolecular electron transfer observed for the four synthetic flavocytochromes studied can be understood in terms of the nature and the length of the electron transfer pathway between the donor and acceptor centers. From this work the utility of future use of a variety of alternative electron donors (e.g., quinones) as protein modifying reagents for the study of intramolecular electron transfer reactions of proteins is apparent. The variants in which the axial Met80 ligand was replaced with an alanyl residue provided insight concerning the role of the heme iron coordination environment and precursor complex formation in determining the electron transfer reactivity of cytochrome c. In particular, these variants emphasize the influence of heme iron reorganization energy on the efficiency of intermolecular electron transfer reactions. Models are proposed that invoke a role for both changes in spin state and coordination number that may account for significant differences in reactivity observed for these variants. Rigorous evaluation of the validity of these models will require additional spectroscopic analysis and possibly crystallographic studies. However, the present work demonstrates clearly the remarkable potential that these variants offer for understanding the magnitudes of the contributions 150 of these fundamental factors to the efficiency of metalloprotein electron transfer reactions. 151 5 . R E F E R E N C E S A h m a d , I., Cusanovich , M . A . , T o l l i n , G . (1981) Proc. Natl. Acad. Sci. U.S.A. 7 8 , 6724-6728. A h m a d , I., Cusanovich , M . A . , T o l l i n , G . (1982) Biochemistry 2 1 , 3122-3128. A n t o n i n i , E . , B r u n o r i , M . (1971) Hemoglobin and Myoglobin in their Reactions with Ligands, N o r t h - H o l l a n d , Amsterdam. Ba is t rocch i , P . , B a n c i , L . , Ber t in i , I., Turano, P . B r e n , K . L . , Gray , H . B . (1996) Biochemistry 3 5 , 13788-13796. B a n c i , L . , Be r t in i , I., B r e n , K . L . , Gray , H . B . , Sompornpisut , P . , Turano, P . (1995a) Biochemistry 3 4 , 11385-11398. B a n c i , L . , Ber t in i , I., B r e n , K . L . , Gray , H . B . , Turano, P . (1995b) Chemistry and Biology 2 , 377-383. B a n c i , L . , Be r t in i , I., B r e n , K . L . , Gray , H . B . , Sompornpisut , P . , Turano, P . (1997) Biochemistry 3 6 , 8992-9001. B a r k e r , P . D . , Ferrer, J . C , Mylra jan , M . , Loeh r , T . M . , Feng , R . , K o n i s h i , Y . , Funk , W . D . , M a c G i l l i v r a y , R . T . A . , M a u k , A . G . (1993) Proc. Natl. Acad. Sci. USA 90, 6542-6546. Beratan, D . N . , Onuchic , J . N , Bet ts , J . N . , B o w l e r , B . E . , Gray , H . B . (1990) J. Am. Chem. Soc. 1 1 2 , 7915-7921. Beratan, D . N . , Betts , J . N , Onuchic , J . N . (1991) Science 2 5 2 , 1285-1288. Beratan, D . N . , Onuchic , J . N . , Wink le r , J . R . , Gray , H . B . (1992) Science 2 5 8 , 1740-1741. Berghuis , A . M . , Brayer , G . D . (1992) J . Mol. Biol. 2 2 3 , 259-276. Berghuis , A . M . , Guillemette, J . G . , Smith, M . , Brayer , G . D . (1994) J . Mol. Biol. 2 3 5 , 1326-1341. Be rns t e in , F . C , K o e t z l e , T . F . , Wi l l i ams , G . J . B . , M e y e r , E . F . , B r u c e , M . D . , Rodger , J . R . , Kennard , O . , Shimanouchi , T . , Tasmui , M . (1977)7 . Mol. Biol. 1 1 2 , 535-542. Bet ts , J . N . , Beratan, D . N , Onuchic , J . N . (1992)7 . Am. Chem. Soc. 1 1 4 , 4043-4046. Bje r rum, M . J . , Cas imiro , D . R . , Chang, I., D i B i l i o , A . J . , Gray , H . B . , H i l l , M . G . , Langen , R . , M i n e s , G . A . , S k o v , L . K . , Wink le r , J . R . , Wut tke , D . S. (1995) J. Bioenerg. Biomem. 2 7 , 295-302. B o w l e r , B . E . , Meade , T. J. , M a y o , S. L . , Richards, J. H . , Gray , H . B . (1989) J . Am. Chem. Soc. I l l , 152 8757-8759. Brautigan, D. L . , Ferguson-Miller, S., Margoliash, E . (1978) MethodsEnzymol. 5 3 , 130-139. Brayer, G. D.; Murphy, M . E. P. (1996) In Cyochrome c: A Multidisciplinary Approach, Scott, R. A., Mauk, A. G. (Eds), University Science Books: Sausalito, pp. 103-166. Bren, K. L. (1996) Ph.D. Dissertation, California Institute of Technology, Pasadena. Bren, K. L . , Gray, H. B. (1993a) J. Am. Chem. Soc. 1 1 5 , 10382-10383. Bren, K. L. , Gray, H. B. (1993b) J. Inorg. Biochem. 5 1 , 111. Bren, K. L., Gray, H. B., Band, L. , Bertini, I., Turano, P. (1995)./. Am. Chem. Soc. Ill, 8067-8073. Brown, W. D., Mebine, L. B. (1969) J. Biol. Chem. 2 4 4 , 6696-6701. Brown, G. M . , Sutin, N. (1979) J. Am. Chem. Soc. 1 0 1 , 883-892. Bushnell, G. W., Louie, G. V., Brayer, G. D. (1990) J. Mol. Biol. 2 1 4 , 585-595. Butt, W. D., Keilen, D. (1962) Proc. Royal Soc. Lond. B 1 5 2 , 429-458. Carell, T., Butenandt, J. {\991)Angew. Chem. Int. Ed. Engl. 3 6 , 1461-1464. Carver, T. E. , Brantley, R. E. Jr., Singleton, E. W., Arduini, R. M . , Quillin, M . L. , Phillips, G. N. Jr., Olson, J. S. (1992) J. Biol. Chem. 2 6 7 , 14443-14450. Casimiro, D. R., Richards, J. H. , Winkler, J. R., Gray, H. B. (1993) J. Phys. Chem. 97,13073-13077. Cassat, J. C , Marini, C. P., Bender, J. W. (197'5) Biochemistry, 1 4 , 5470-5475. Chang, I.-J., Gray, H. B., Winkler, J. R. (1991)7. Am. Chem. Soc. 1 1 3 , 7056-7057 Chapman, S. K., Reid, G. A., Daff, S., Sharp, R. E. , White, P., Manson, F. D. C , Lederer, F. (1994) Biochem. Soc. Trans. 2 2 , 713-718. Cheddar, G., Meyer, T. E . , Cusanovich, M . A., Stout, C. D., Tollin, G. (1989) Biochemistry 2 8 , 6318-6322. Christensen, H. E. M . , Conrad, L. S., Mikkelson, K. V., Nielsen, M . K., Ulstrup, J. (1990) Inorg. Chem. 29, 2808-2816. Christensen, H. E. M . , Conrad, L. S., Hammerstad-Pedersen, J. M . , Ulstrup, J. (1992) FEBSLett. 2 9 6 , 141-144. 153 Clark , W . M . (1960) In Oxidation-Reduction Potentials of Organic Systems, Wi l l i ams and W i l k i n s , Ba l t imore , M D , pp 107-203. Conrad , D . W . , Scott , R . A . (1989) J. Am. Chem. Soc. I l l , 3461-3463. Conrad , D . W . , Zhang , H . , Stewart, D . E . , Scott, R . A . (1992)7 . Am. Chem. Soc. 114, 9909-9915. C o n r o y , C . W . , T y m a , P . , D a u m , P . , E rman , J . E . (1978) Biochim. Biophys. Acta. 537, 62-69. C o y l e , C . L . , Gray , H . B . (1976) Biochem. Biophys. Res. Commun. 73, 1122-1127. Cummins , D . , Gray , H . B . (1977) J . Am. Chem. Soc. 99, 5158-5167. Cusanovich , M . A . , M e y e r , T . E . , T o l l i n , G . (1987) Adv. Inorg. Chem. 7, 37-91 . Cusanovich , M . A . , M e y e r , T . E . , T o l l i n , G . (1988) In Heme Proteins v o l . 7 o f Advances in Inorganic Biochemistry, E i c h o r n , G . L . , M a r z i l l i , J . G . (Eds) Elsevier , N e w Y o r k , pp 37-92. Cusanovich , M . A . , T o l l i n , G . (1996) In Cytochrome c: A Multidisciplinary Approach; Scott, R . A . , M a u k , A . G . (Eds) , Univers i ty Science B o o k s : Sausalito, pp. 489-513 . Cut ler , R . L . , Pie lak, G . J . , M a u k , A . G . , Smith, M . (1987) Protein Engineering 1, 95-99. Cut ler , R . L . , Dav i s , A . M . , Creighton, S., Warshel , A . , M o o r e , G . R . , Smith , M . , M a u k , A . G . (1989) Biochemistry 28, 3188-3197. Daff, S., Ingledew, W . J . , R e i d , G . A , Chapman, S. K . (1996) Biochemistry 35, 6345-6350. D a w s o n , J. H . (1988) Science 240, 433-439. D e B a e r e , I., Perutz , M . F . , K i g e r , L . , M a r d e n , M . C , Poyart, C . (1994) Z^oc . Natl. Acad. Sci. USA 91, 1594-1597. Dickerson , R . E . , K o p k a , M . L . , Borders , C . L . , V a m u m , J., We inz ie r l , J. E . (1967)7 . Mol. Biol. 29, 77-95. D i c k e r s o n , R . E . , Takano, T . , Eisenberg, D . , K a l l a i , O . B . , Samson, L . , Cooper , A . , Margo l i a sh , E . (1971)7 . Biol. Chem. 246, 1511-1532. D icke r son , R . E . , T i m k o v i c h , R . (1975) In The Enzymes, 3 rd edn, V o l 11. B o y e r , P . D . (Ed) , A c a d e m i c Press, L o n d o n , pp 397-547. Draper , R . D . , Ingraham, L . L . (1968) Arch. Biochem. Biophys. 125, 802-808. Dunfo rd , H . B . (1991) In CRC Peroxidases in Chemistry and Biology, Everse , J . , Everse , K . E . , Gr i sham, M . B . (Eds) C R C Press, B o s t o n pp 1-24. 154 Durham, B., Pan, L. P., Long, J. E . , Millett, F. (1989) Biochemistry 28, 8659-8665. Dutton, P. L. (1978) Meth. Enzymol. 54, 411-435. Dutton, P. L. , Moser, C. C. (1994) Proc. Natl. Acad. Sci. U.S.A. 91, 10247-10250. Eltis, L . , Mauk, A. G., Hazzard, J. T., Cusanovich, M . A., Tollin, G. (1988) Biochemistry 27, 5455-5460. Eltis, L . , Herbert, R. G., Braker, P. D., Mauk, A. G., Northrup, S. H. (1991)Biochemistry 30, 3663-3674. Fairris, J. L. , Wang, K., Geren, L. , Saunders, A. J., Pielak, G. J., Durham, B., Millett, F. (1996) Adv. in Chemistry Series, In Press. Falk, J. E . (1964) Porphyrins andMetalloporphyrins, Elsevier, New York. Falk, M . C , McCormick, D. B. (1976) Biochemistry 15, 646-653. Ferrer, J. C , Guillemette, J. G., Bogumil, R., Inglis, S. C , Smith, M . , Mauk, A. G. (1993)/. Am. Chem. Soc. 115, 7507-7508. Geren, L . , Hahm, S., Durham, B., Millett, F. (1991) Biochemistry 30, 9450-9457. Goldkorn, T., Schetjer, A. (1976) Arch. Biochem. Biophys. Ill, 39-45. Goldman, C. (1991) Phys. Rev. A43, 4500-4509. Gorren, A. C , Chan, M . L. , Scott, R. A. (1992) Bioconj. Chem. 3, 291-294. Guzzo, A. V., Tollin, G. (1964) Arch. Biochem. Biophys. 105, 380-386. Hampsey, D. M . , Das, G., Sherman, F. (1986)7 Biol. Chem. 261, 3259-3271. Harbury, H. A., Cronin, J. R., Fanger, M . W., Hettinger, T. P., Murphy, A. J., Myer, Y. P., Vinogradov, S. N. (1965) Proc. Natl. Acad. Sci. USA 54, 1658-1664. Harris, T. R., Rosell, F., Hilderbrand, D. P., Mauk, A. G. (1997) In Preparation. Hazzard, J. T., Cusanovich, M . A., Tainer, J. A., Getzoff, E. D., Tollin, G. (1986) Biochemistry 25, 3318-3328. Hazzard, J. T., Poulos, T. L . , Tollin, G. (1987) Biochemistry 26, 2836-2848. Hazzard, J. T., Moench, S. J., Erman, J. E. , Satterlee, J. D., Tollin, G. (1988)Biochemistry 21,2002-2008. 155 Hazzard, J. T., McDonough,, C. A., Tollin, G. (1994) Biochemistry 33, 13445-13454. Hildebrand D. P. (1996) Ph.D. Dissertation, University of British Columbia, Vancouver. Hildebrand, D. P., Burk, D. L. , Maurus, R., Ferrer, J. C , Brayer, G. D., Mauk, A. G. (1995) Biochemistry 34, 1997-2005. Hilvert, D., Hatanaka, Y., Kaiser, E . T., (1988) 7. Am. Chem. Soc. 110, 682-689. Hodges, H. L. , Holwerda, R. A., Gray, H. B. (1974)7 Am. Chem. Soc. 96, 3132-3137. Hoffman, R. J. (1963)7. Chem. Phys. 39, 1397-1405. Holwerda, R. A., Knoff, D. B., Gray, H. B., Clemmer, I'D., Crowley, R., Smith, I M . , Mauk, A. G. (1980)7. Am. Chem. Soc. 102, 1142-1146. Huang, Y. Y., Hara, T., Sliger, S., Coon, M . I , Kimura, T. (\9%6) Biochemistry 25, 1390-1394. Inglis, S. C , Guillemette, I G., Johnson, I A , Smith, M . (1991) Protein Eng. 4, 569-574. Isied, S. S. (1990) In Advances in Chemistry Series. Electron Transfer in Biology and the Solid State, Vol. 226, Johnson, M . K., King, R. B., Kurtz, D. M . , Jr., Kutal, C , Norton, M . L. , Scott, R. A. (Eds) ACS, Washigton, D.C. pp 91-100. Isied, S. S., Worosila, G., Atherton, S. J. (1982)7 Am. Chem. Soc. 104, 7659-7661. Isied, S. S., Kuehn, C , Worosila, G. (1984) 7. Am. Chem. Soc. 106, 1722-1726. Jung, J., Tollin, G. (1981) Biochemistry 20, 5124-5131. Karpishin, T. B., Grinstaff, M . W., Jonar-Panicucci, S., McLendon, G., Gray, H. B. (1994) Structure 2,415-422. Kassner, R. J. (1972) Proc. Natl. Acad. Sci. USA 69, 2263-2267. Kassner, R. J. (1973)7. Am. Chem. Soc. 95, 2674-2677. Kim, C-S., Kueppers, F., Dimaria, P., Farooqui, J., Kim, S., Paik, W. K. (1980)Biochim. Biophys. Acta. 622, 144-150. Kloek, A. P., Yang, J., Mathews, F. S., Frieden, C , Goldberg, D. E. (1994) 7. Biol. Chem. 269, 2377-2379. Kuki, A. (1991) Structure and Bonding 75, 49-83. Kunkel, T. A. (1985) Proc. Natl. Acad. Sci. USA 72, 488-492. 156 Langen, R., Chang, I.-J., Germanas, J. P., Richards, J. H. , Winkler, J. R., Gray, H. B. (1995) Science 268, 1733-1735. Larsson, S. (1981)7 Am. Chem. Soc. 103, 4034-4040. Levine, H. L. , Kaiser, E . T. (1978)7 Am. Chem. Soc. 100, 7670-7677. Libby, W. F. (1952)7. Phys. Chem. 56, 863-868. Lieber, C. M , Karas, J. L . , Mayo, S. L . , Albin, M . , Gray, H. B. (1987) "Long-Range Electron Transfer in Proteins," The Robert A. Welch Foundation Conference on Chemical Research XXXI pp 9-26. Lieber, C. M . , Karas, J. L. , Mayo, S. L. , Axup, A. W., Albin, M . , Crutchley, R. J., Ellis, W. R., Jr., Gray, H. B. (1988) In Trace Elements in Man and Animals, Vol. 6, Hurley, L. S., Keen, C. L . , Lonnerdal, B., Rucker, R. B. (Eds) Plenum Press, New York pp 23-27. Lim, A. R. (1990) Ph.D. Dissertation, University of British Columbia, Vancouver. Lim, A. R., Mauk, A. G. (1985) Biochem. 7. 229, 765-769. Lloyd, E . , Hilderbrand, D. P., Tu, K. M . , Mauk, A. G. (1995)7. Am. Chem. Soc. Ill, 6434-6348. Lo, T. P. (1995) Ph.D. Dissertation, University of British Columbia, Vancouver. Lo, T. P., Murphy, M . E . P., Guillemette, J. G., Smith, M . , Brayer, G. D. (1995) Protein Sci. 4,198-208. Louie, G. V., Brayer, G. D. (1990)7. Mol. Biol. 214, 527-555. Louie, G. V., Pielak, G. J., Smith, M . , Brayer, G. D. (1988) Biochemistry 21, 7870-7876. Lu, Y., Casimro, D. R., Bren, K. L. , Richards, J. H. , Gray, H. B. (1993) ProL Natl. Acad. Sci. USA 90,11456-11459. Lui, R.-Q. (1994) Ph.D. Dissertation, University of Arkansas, Little Rock. Makinen, M . W., Churg, A. K. (1983) In Iron Porphyrins, Lever, A. P. B., Gray, H. B. (Eds.) Addison Wesley, London, pp 141-235. Maniatis, T., Fritsch, E . F., Sambrook, J. (1989) Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory Press, New York. Marcus, R. A. (1965)7. Chem. Phys. 43, 679-701. Marcus, R. A., Sutin, N. (1985) Biochim. Biophys. Acta, 811, 265-322. 157 Margo l i a sh , E . Schejter, A . (1996) In Cytochrome c: A Multidisciplinary Approach; Scott , R . A . , M a u k , A . G . (Eds) , Univers i ty Science B o o k s : Sausalito, pp. 3-31. Massey , V . , Ghis la , S. (1983) In Biological Oxidations, Sund, H . , U l l r i c h , V . (Eds) Springer, B e r l i n pp. 114-139. Massey , V . , Hemmer ich , P . (1980) Biochem. Soc. Trans. 8, 246-250. Massey , V . , Hemmer ich , P . (1982) InFlavin and Flavoproteins, Massey , V . , Wil l iams, C . H . (Eds) E l sev i e r /Nor th H o l l a n d , N e w Y o r k pp. 83-96. M a t h e w s , F . S. (1985) Prog. Biophys. Mol. Biol. 45, 1-56. M a u k , A . G . , Gray , H . B . (1979) Biochem. Biophys. Res. Commun. 86, 206-210. M a u k , M . R . , M a u k , A . G . , Weber , P . C , Mat thews , J . B . (1986) Biochemistry 25, 7085-7091. M a u k , M . R . , M a u k , A . G . , M o o r e , G . R . , Nor thrup , S. H . (1995) Jour. Bioeneg. &Biomembr. 27, 311-330. M a y h e w , S. G . , O ' C o n n e l l , D . P . , O 'Fa r r e l l , P . A , Y a l l o w a y , G . N . , Geoghegan, S. M . (1996) Biochem. Soc. Trans. 24, 122-127. M c A r d l e , J . V . , Y o c o m , K . , Gray , H . B . (1977) J. Am. Chem. Soc. 99, 4141-4145. M c l n t i r e , W . , Singer, T . P . , Ameyama , M . , A d a c h i , O . , Matsushi tak, K . , Shinagawa, E . (1995) Biochem. J. 231, 651-654. M e a d e , T . J . , Gray , H . B . , Wink le r , J . R . (1989) J . Am. Chem. Soc. I l l , 4353-4356. Mei tes , L . , Mei t e s , T . (1984) Anal. Chem. 20, 984. M e y e r , T . E . , Watk ins , J . A . , P rzys ieck i , C . T. , T o l l i n , G . , Cusanovich , M . A . (1984) Biochemistry 23, 4761-4767. M i h a r a , H . , T o m i z a k i , K . - Y . , N i sh ino , N , Fug imoto , T . (1993) Chem. Lett. 1533-1536. M i l l e r , W . G . , Cuzanov ich , M . A . (1974) Bioelectrochem. Bioenerg. 1, 97-111. Mi l l e t t , F . , Durham, B . (1991) In Metal Ions in Biological Systems, V o l . 27 , Sigel , H . , Sigel , A . (Eds) M a r c e l Dekke r , Inc., N e w Y o r k pp.223-264 M i n e s , G . A . , B je r rum, M . J . , H i l l , M . G . , Cas imiro , D . R . , Chang , I., W i n k l e r , J . R . , Gray , H . B . ( 1 9 9 6 ) / . Am. Chem. Soc. 118, 1961-1965. M o o r e , G . R . (1983) FEBSLett. 161, 171-175. 158 Moore, G. R., Williams, R. J. P. (1977) FEBSLett. 79, 223-229. Moore, G. R., Pettigrew, G. W. (1990) Cytochromes c: Evolutionary, Structural and Physicochemical Aspects, Springer-Verlag, Heidelberg. Moore, G. R., Harris, D. E . , Leitch, F. A., Pettigrew, G. W. (1984) Biochim. Biophys. Acta 764, 331-342. Moore, G. R., Pettigrew, G. W., Rogers, N. K. (1986) Proc. Natl. Acad. Sci. USA 69, 2263-2267. Moser, C. C , Dutton, P. L. (1992) Biochim. Biophys. Acta 1101, 171-176. Moser, C. C , Keske, J. M . , Warncke, K., Farid, R. S., Dutton, P. L . (1992) Nature 355, 796-801. Murphy, M . E. P. (1993) Ph.D. Dissertation, University of British Columbia, Vancouver. Murphy, M . E. P., Nail, B. T., Brayer, G. D. (1992) 7. Mol. Biol. 227, 160-176. Narita, K., Titani, K. (1969)7. Biochem. 65, 226-241. Nocek, J. M . , Zhou, J. S., Deforest, S., Priyadarshy, S., Beratan, D. N , Onuchic, J. N., Hoffman, B. M . (1996) Chemical Reviews 96, 2459-2489. Nocera, D. G., Winkler, J. R., Yocom, K. M . , Bordignon, E , Gray, H.B. (1984)7. Am. Chem. Soc. 106, 5145-5150. Northrup, S. H. , Boles, J. O., Reynolds, J. C. (1988) Science 241, 67-70. Northrup, S. H. , Thomasson, K. A., Miller, C. M . , Barker, P. D., Eltis, L. D., Guillemette, J. G., Inglis, S. C , Mauk, A. G. (1993) Biochemistry 32, 6613-6623. Nozaki, M . , Mutzushima, H. , Horio, T., Okunuki, K. (1958)7 Biochem. 256, 673-676. Ochi, H. , Hata, Y., Tanaka, N., Kakudo, M . , Sakuri, T.,Achara, S., Morita, Y. (1983)7. Mol. Biol. 166, 407-418. Onuchic, J. N , Beratan, D. N. (1990)7. Chem. Phys. 92, 722-733. Onuchic, J. N , Beratan, D. N , Winkler, J. R., Gray, H. B. (1992) Annu. Rev. Biophys. Biomol. Struct. 21, 349-377. Ormos, P., Braunstein, D., Frauenfelder, H. , Hong, M . K., Lin, S.-L., Sauke, T. B., Young, R. D., (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 8492-8496. Pan, L P . , Durham, B., Wolinska, J., Millett, F. (1988)7. Am. Chem. Soc. 110, 7114-7118. 159 Pelletier, H. , Kraut, J. (1992) Science, 258, 1748-1755. Penzer, G. R., Radda, G. K. (1967) Q. Rev., Chem. Soc. 21, 43-48. Peterson-Kennedy, S. E. , McGourty, J. L. , Hoffman, B. M . (1984) J. Am. Chem. Soc. 106, 5010-5011 Pettigrew, G. W., Moore, G. R. (1987) Cytochromes c: Biological Aspects, Springer-Verlag, Heidelberg. * Phase-R Corporation, Box G-2 Old Bay Rd., New Durham, N H 03855. Phillips, S. E. (1980) J. Mol. Biol. 142, 531-534. Pielak, G. J., Mauk, A. G., Smith, M . (1985) Nature 313, 152-154. Pollock, W. B. R., Rosell, F. I., Twitchett, M . B., Dumont, M . E . , Mauk, A. G. (1997)Biochemistry, In Preparation. Poulos, T. L. , Kraut, J. (1980) J. Biol. Chem. 255, 10322-10330. Prysiecki, C. T., Tollin, G., Meyer, T. E. , Staggers, J. E . , Cusanovich, M . A. (1985) Arch. Biochem. Biophys. 238, 334-343. Qi, P. X., Beckman, R. A., Wand, A. J. (1996) Biochemistry, 35, 12275-12286. Rafferty, S. P. (1992) Ph.D. Dissertation, University of British Columbia, Vancouver. Rafferty, S. P., Pearce, L. L . , Barker, P. D., Guillemette, J. G., Kay, C. M . , Smith, M . , Mauk, A. G. (1990) Biochemistry, 29, 9365-9369. Rafferty, S. P., Guillemette, J. G., Berghuis, A. M . , Smith, M . , Brayer, G. D., Mauk, A. G. (1996) Biochemistry 35, 10784-10792. Raphael, A. L. , Gray, H. B. (1989) Protein, Struct., Fund., Genet. 6, 338-340. Raphael, A. L. , Gray, H. B. (1991) J. Am. Chem. Soc. 113, 1038-1040. Ray, G. B., Li , X. -Y. , Ibers, J. A., Sessler, J. L. , Spiro, T. G. (1994)7. Am. Chem. Soc. 116, 162-176. Rees, A. R., Offord, R. E. (1976) Biochem. Jour. 159, 467-479. Regan, J. J., Risser, S. M . , Beratan, D. N , Onuchic, J. N. (1993) J. Phys. Chem. 97, 13083-13088. Reid, L. S. (1984) Ph.D. Dissertation, University of British Columbia, Vancouver. 160 R e i d , L . S., M a u k , A . G . (1982) . / . Am. Chem. Soc. 104, 841-845. R e i d , L . S., Taniguchi , V . T., Gray , H . B . , M a u k , A . G . (1982) J. Am. Chem. Soc. 104, 7516-7519. R e i d , L . S., L i m , A . R . , M a u k , A . G . (1986) J . Am. Chem. Soc. 108, 8197-8201. R i c h , P . R . , Benda l l , D . K . (1980) Biochim. Biophys. Acta. 592, 506-518. Rieder , R . , Bosshard , H . R . (1980) J. Biol. Chem. 255, 4732-4739. Rob inson , M . N , B o s w e l l , A . P . , E l ey , C . S. G . , H u a n g , Z . - X . , M o o r e , G . R . (1983) Biochem. J. 213, 687-700. R o s e l l , F . I., Ferrer, J . C , M a u k , A . G . (1997) submitted to J. Am. Chem. Soc. Salemme, F . R . (1976) J . Mol. Biol. 102, 563-568. Sambrook, J . , Fr i tsch, E . F . , Mania t i s , T. (1989) Molecular Cloning: a Laboratory Manual, 2nd E d . , C o l d Spr ing Ha rbo r Labora tory Press, N e w Y o r k . S a w i c k i , C . A . , M o r r i s , R . J . (\9U)Meth. Enzymol. 76, 667-681. Schlauder, G . , Kassner , R . J. (1979)7 . Biol. Chem. 254, 4110-4113. Scott , R . A . (1996) In Cytochrome c: A Multidisciplinary Approach, Scott , R . A . ; M a u k , A . G . , Eds . , Univers i ty Science B o o k s : Sausalito, pp. 515-541. Scott , R . A . , M a u k , A . G . (1996) Cytochrome c: A Multidisciplinary Approach, Univers i ty Science B o o k s : Sausalito. Scott , R . A . , M a u k , A . G . , Gray , H . B . (1985) J . Chem. Ed. 62, 932-938. Scott , R . A . , Conrad , D . W . , Eidsness, M . K . , Gor ren , A . C , W a l l i n , S. A . (1991) InMetallons in Biological Systems, V o l . 27, Sigel , H . , Sigel , A . (Eds) M a r c e l D e k k e r , Inc. N e w Y o r k pp 199-222. Shack, J . , C la rk , W . M . ( 1 9 4 7 ) / . Biol. Chem. 171, 143-187. Shumyantseva, V . V . , U v a r o v , V . Y . , B y a k o v a , O . E . , A r c h a k o v , A . I. (1996) Biochem. Mol. Biol. Int. 38, 829-838. Siddarth, P . , M a r c u s , R . A . (1990) J . Phys. Chem. 94, 8430-8434. Siddarth, P . , M a r c u s , R . A . (1992) J. Phys. Chem. 96, 3213-3217. Siddarth, P . , M a r c u s , R . A . (1993a) J . Phys. Chem. 97, 2400-2405. 161 Siddarth, P . , M a r c u s , R . A . (1993b) J . Phys. Chem. 97, 13078-13082. Simondsen, R . P . , T o l l i n , G . (1983) Biochemistry 22, 3008-3016. Smith, M . B . , Stonehuerner, J . , A h m e d , A . J . ; Staudenmayer, N . , Mi l l e t t , F . (1980) Biocheim. Biophys. Acta 592, 303-313. Sorrel l , T . N . , M a r t i n , P . K . , B o w d e n , E . F . (1989) J . Am. Chem. Soc. I l l , 766-767. Speck, S. H . , Margo l i a sh , E . (1984) J . Biol. Chem. 259, 1064-1072. Speck, S. H . , Ferguson-Mi l le r , S., Osheroff, N . , Margo l i a sh , E . (1979) Proc. Natl. Acad. Sci. (USA) 76, 155-159. Speck, S. H . , K o p p e n o l , W . H . , Dethmers, J . K . , Osheroff, N . , Margo l i a sh , E . , Rajagopalan, K . V . (1981)7 . Bio. Chem. 256, 7394-7400. Sridharan, S., N i c h o l l s , A . , Sharp, K . A . (1995) J . Comp. Chem. 16, 1038-1044. Stankovich, M . T. , (1989) In Flavoproteins V o l . 1 Redox Properties of Flavins andFlavoproteins, C R C Press, Inc., F l o r i d a pp. 401-425. Steenkamp, D . J . , M c l n t i r e , W . , Kenney , W . C . (1978) J . Biol. Chem. 253, 2818-2824. Stel lwagen, E . (1978) Nature 275, 73-74. Sun, J . , Wishar t , J . F . , Gardineer, M . B . , C h o , M . P . , Isied, S. S. (\99S)Inorg. Chem. 34,3301-3309. Takano, T . , D icke r son , R . E . (1980) Proc. Natl. Acad. Sci. (USA) 77, 6371-6375. Takano, T . , D icke r son , R . E . (1981a) J . Mol. Biol. 153, 79-94. Takano, T . , D icke r son , R . E . (1981b) J . Mol. Biol. 153, 95-115. Tanaka, T . , Yamane , T . , Tsukihara , T . , Ash ida , T . , K a k u d o , M . (1975) J . Biochem. (Tokyo) 11,147-162. Taniguchi , V . T . , Sailasuta-Scott, N . , A n s o n , F . C , Gray , H . B . (1980) Pure & Applied Chem. 52, 2275-2281. Taniguchi , I., Toyosawa , K . , Yamaguch i , H . , Yasukauch i , K . (1982) Electroanal. Chem. 140, 187-190. Tegoni , M . , Janot., J . - M . , Labeyrie , F . (1986) Eur. J. Biochem. 155, 491-503. Tegon i , M . , Whi te , S. A . , Rousse l , A . , Ma thews , F . S., Cambi l lau , C . (1993)Proteins: Struct. Funct. 162 Genet. 16, 408-422. Theorel l , H , Akesson, A. (1941) J. Am. Chem. Soc. 63, 1804-1820. Therien, M . J . , Selmon, M . , Gray , H . J3., Chang, I. J . , Wink le r , J . R . (1990) J . Am. Chem. Soc. 112, 2420-2422. T h o n y - M e y e r , L . (1997)Microbiol. &Mol. Biol. Rev. 61, 337-376. T o l l i n , G . , M e y e r , T . E . , Cusanovich , M . A . (1986) Biochim. Biophys. Acta 853, 29-41 . Traber, R . , Kramer , H . E . A . , Hemmer ich , P . (1982) Biochemistry 21, 1687-1693. Twitchet t , M . B . , Ferrer, J . C , Siddarth, P . , M a u k , A . G . (1997)7 . Am. Chem. Soc. 119, 435-436. Wal lace , C . J . (1993) FASEB Jour. 7, 505-515. Wal lace , C . J . A . , C l a r k - L e w i s , I. (1992)7 . Biol. Chem. 267, 3852-3861. Weber , P . C , T o l l i n , G . ( 1 9 8 5 ) 7 Biol. Chem. 260, 5568-5573. Wher land , S, H o l w e r d a , R . A . , Rosenberg, R . C , Gray , H . B . (1975)7 . Am. Chem. Soc. 97, 5260-5262. Wher land , S., Gray , H . B . (1976) Proc. Natl. Acad. Sci. USA 78, 2950-2954. Wi l l i ams , G . , Clayden, N . J . , M o o r e , G . R . , Wi l l i ams , R . J . P . (1985) 7. Mol. Biol. 183, 447-460. Wi l l i ams , R . J . P . (1959) In The Enzymes 2nd Ed, B o y e r , P . D . , L a r d y , H . , M y r b a c k , K . (Eds) A c a d e m i c Press L o n d o n , pp 391-441. Wi l l i amson , G . , Edmondson , D . E . (1985) Biochemistry, 24, 7790-7797. W i n k l e r , J . R . , Gray , H . B . (1997) 7. Biol. Inorg. Chem. 2, 399-404. W i n k l e r , J . R . , N o c e r a , D . G . , Y o c o m , K . B . , Bordignon, E . , Gray , H . B . (1982)7 . Am. Chem. Soc. 104, 5798-5800. W u , T . T. , Kabat , E . A . (1970)7 . Exp. Med., 132, 211-250. Wut tke , D . S., Bjer rum, J . I , Wink le r , J . R . , Gray , H . B . (1992) Science 256, 1007-1009. X i a , Z . - X . , Ma thews , F . S. (1990)7 . Mol. Biol. 212, 837-863. Yamanaka , T . , Mu tzush ima , H . , N o z a k i , M . , H o r i o , T . , O k u n u k i , K . (1959)7 . Biochem. 46, 121-132. 163 Y o c o m , K . M . , She l ton , J . B . , Shelton, J . R . , Schroeder, W . A . , W o r o s i l a , G . , Isied, S. S., B o r d i g n o n , E . , G r a y H . B . (1982) Proc. Natl. Acad. Sci. USA 79, 7052-7055. Zol le r , M . J . , Smith, M . (1983) Methods Emymol. 100, 468-500. Zo l l e r , M . J . , Smith, M . (1984) DNA 3, 479-488. 164 APPENDIX A -Cyclic voltammetry results (Error in reduction potential values ±2 mV) Protein pH T(K) F° heme Fe 27Fe 3 + (mV vs SHE) A E h e m e (mV) ^ flavin Ox/Red (mV vs SHE) AEflavin (mV) 7-acetyl-10-methylisoalloxazine 4.93 298 - - 23 58 5.56 298 - - -14 62 6.16 298 - - -67 63 6.57 298 - - -85 62 7.05 278 - - -105 58 283 - - -108 56 288 - - -110 54 293 - - -112 55 298 - - -113 53 303 - - -116 52 308 - - -117 50 313 - - -119 50 7.54 298 - - -126 56 8.00 298 - - -141 56 8.51 298 - - -161 55 9.01 298 - - -174 54 9.55 298 - - -196 56 9.90 298 - - -201 52 10.46 298 - - -246 54 10.98 298 - - -274 50 11.68 298 - - -312 50 7a-bromoacetyl-10-methylisoalloxazine 4.93 298 - - -7 56 5.50 298 - - -44 57 165 Prote in P H T ( K ) ^ heme F e 2 7 F e 3 + ( m V vs S H E ) ( m V ) ^ flavin O x / R e d ( m V vs S H E ) AEflavij, ( m V ) 7 a-bromoacetyl-10-methylisoalloxazine 6.11 298 - - -73 55 6.57 298 - - -86 54 7.05 277 - - -101 56 284 - - -105 58 288 - - -107 54 293 - - -110 54 298 - - -113 53 303 - - -115 50 308 - - -118 50 7.54 298 - - -154 56 8.04 298 - - -185 55 8.52 298 - - -196 57 9.02 298 - - -206 52 9.51 298 - - -211 53 9.96 298 - - -216 54 10.46 298 - - -245 54 10.98 298 - - -273 54 11.68 298 - - -323 53 T h r 8 C y s cytochrome c 5.49 298 293 64 - -6.04 298 290 63 - -6.51 298 284 66 - -7.04 278 286 65 - -284 285 70 - -287 282 69 - -166 Prote in P H T ( K ) F ° ^ heme F e 2 7 F e 3 + ( m V vs S H E ) A E ^ ( m V ) v° flavin O x / R e d ( m V vs S H E ) AEjfcvi,, ( m V ) T h r 8 C y s cytochrome c 7.04 293 280 66 - -298 278 70 - -304 276 68 - -307 274 67 - -313 272 67 - -7.53 298 276 66 - -8.11 298 274 66 - -A c f l a v i n modif ied T h r 8 C y s cytochrome c 5.59 298 292 62 -57 108 6.01 298 289 60 -77 110 6.56 298 284 60 -96 114 7.00 278 287 62 -113 110 283 286 60 -116 115 288 283 66 -118 112 293 281 63 -121 108 298 279 60 -123 120 302 277 59 -126 122 307 274 62 -129 106 7.73 298 274 62 -155 112 H i s 3 9 C y s cytochrome c 5.80 298 261 62 - -6.00 298 262 63 - -6.42 298 262 62 - -6.53 298 260 60 - -7.03 278 272 64 - -284 270 64 - -167 Prote in p H T ( K ) F ° heme F e 2 7 F e 3 + ( m V vs S H E ) A E h e m e ( m V ) ^ flavin O x / R e d ( m V vs S H E ) A E ^ ( m V ) H i s 3 9 C y s cytochrome c 7.03 288 265 65 - -294 262 63 - -298 259 60 - -303 255 58 - -308 252 58 - -7.54 298 255 62 - -8.08 298 251 63 - -A c f l a v i n modif ied H i s 3 9 C y s cytochrome c 5.49 298 259 60 -19 108 5.77 298 261 60 -55 109 6.01 298 261 62 -63 113 6.54 298 259 63 -91 116 7.04 278 269 64 -102 108 283 265 64 -105 110 288 262 65 -108 107 293 260 63 -110 115 298 257 60 -115 110 303 253 59 -117 108 307 250 60 -120 104 7.86 298 257 62 -140 109 A s n 6 2 C y s cytochrome c 5.73 298 288 63 - -6.04 298 287 60 - -6.51 298 284 62 - -7.00 279 288 64 - -283 286 63 - -168 Prote in p H T ( K ) F ° h e m e F e 2 7 F e 3 + ( m V vs S H E ) A E h e m e ( m V ) C O f l a v i n O x / R e d ( m V vs S H E ) A E f l ^ ( m V ) A s n 6 2 C y s cytochrome c 7.00 288 284 63 - -293 281 62 - -298 280 60 - -303 278 60 - -308 276 60 - -7.53 298 275 61 - -8.05 298 272 62 - -A c f l a v i n modif ied A s n 6 2 C y s cytochrome c 5.61 298 286 63 -51 110 5.95 298 286 63 -71 110 6.59 298 282 62 -95 118 7.00 278 288 59 -107 122 283 286 60 -110 108 288 284 59 -113 107 293 282 66 -115 105 298 279 60 -118 110 303 277 58 -121 122 307 275 64 -123 109 7.10 298 278 62 -117 114 7.57 298 274 60 -143 116 L e u 8 5 C y s cytochrome c 5.64 298 287 64 - -6.00 298 284 65 - -6.42 298 292 63 - -7.10 278 284 65 - -283 282 66 - -169 Prote in p H T ( K ) ^ heme F e 2 7 F e 3 + ( m V vs S H E ) ( m V ) ^ flavin O x / R e d ( m V vs S H E ) A E f l ^ ( m V ) L e u 8 5 C y s cytochrome c 7.10 288 280 66 - -293 277 64 - -298 274 63 - -303 272 62 - -308 269 63 - -7.52 298 268 67 - -8.13 298 265 66 - -A c f l a v i n modif ied L e u 8 5 C y s cytochrome c 5.66 298 285 66 -59 110 6.06 298 282 65 -74 112 6.50 298 279 63 -91 115 7.00 279 283 59 -104 110 283 281 60 -106 98 288 278 61 -109 110 293 275 62 -112 106 298 272 59 -115 109 304 270 62 -118 111 308 268 62 -121 114 7.01 298 270 64 -115 120 7.67 298 266 60 -153 118 170 APPENDIX B - Intermolecular rate constants for the reaction of cytochromes c with flavin semiquinones Protein Flavin T (K) [protein] (uM) * o b s x 10'3 (s'1) horse heart Cytochrome c horse heart Cytochrome c horse heart Cytochrome c Riboflavin 7-acetyl-10-methylisoalloxazine 7cc-bromoacetyl-10-methylisoalloxazine 298 283 288 293 298 303 308 298 283 288 293 298 303 308 298 10 20 30 40 50 10 20 30 40 50 10 20 30 40 0.71(3) 0.99(3) 1.36(4) 1.86(4) 1.65(5) 1.90(4) 2.15(6) 2.46(6) 2.80(8) 3.05(9) 0.60(2) 1.11(4) 2.00(4) 2.61(5) 2.25(5) 2.55(6) 2.85(6) 3.12(4) 3.55(5) 4.20(10) 0.61(5) 1.08(9) 1.77(8) 2.42(6) 171 Protein Flavin T(K) [protein] (uM) ^ x l O - ^ s - 1 ) horse heart 7a-bromoacetyl-10-Cytochrome c methylisoalloxazine 298 50 2.88(6) Yeast iso-1- 7-acetyl-10-cytochrome c methyli soall oxazine 298 10 1.01(4) CC 20 2.06(5) c c 30 3.08(4) c c 40 4.11(6) 283 50 3.67(6) 288 3.86(7) 293 4.10(12) 298 5.14(16) 303 5.54(19) 308 6.25(18) Yeast iso-1- 7-acetyl-10-cytochrome c methylisoalloxazine 298 10 1.22(4) Thr8Cys CC 20 2.03(4) c c 30 2.96(8) c c 40 4.04(15) c c 50 5.13(17) Yeast iso-1- 7-acetyl-10-cytochrome c methylisoalloxazine 298 10 0.85(3) His39Cys c c 20 1.74(4) c c 30 3.40(8) c c 40 4.04(10) c c 50 4.81(16) Yeast 750-7- 7-acetyl-10-cytochrome c methylisoalloxazine 298 10 0.86(6) Leu85Cys c c 20 2.22(6) 172 Protein Flavin T(K) [protein] (uM) * o b s x 10"3 (s1) Yeast iso-1-cytochrome c Leu85Cys 7-acetyl-10-methylisoalloxazine 298 E. coli cytochrome c 7-acetyl-10-methylisoalloxazine E. coli cytochrome c 7-acetyl-10-Asn62Cys methylisoalloxazine 298 298 30 40 50 10 20 30 40 50 10 20 30 40 50 3.25(8) 4.30(12) 5.60(18) 1.06(8) 2.04(7) 3.14(5) 4.11(10) 5.29(18) 1.16(8) 2.19(8) 3.40(7) 4.45(8) 5.57(10) 173 APPENDIX C - Intra-molecular kinetic parameters associated with flavin modified cytochrome c variants. (Experimental error associated with ket values ±5%) Protein pH T ( K ) -AG" (eV) Acflavin modified Thr8Cys cytochrome c 5.16 298 0.337 1.48xl03 7.30 5.49 298 0.363 2.03xl03 7.62 5.77 298 0.384 2.73xl03 7.91 6.56 298 0.403 3.16xl03 8.06 7.04 285 0.420 3.34xl03 8.11 - 288 - 3.45xl03 8.15 - 294 - 3.71xl03 8.21 - 298 - 4.41xl03 8.39 - 303 - 4.98xl03 8.51 - 308 - 5.29xl03 8.57 Acflavin modified His39Cys cytochrome c 5.56 298 0.302 4.00xl62 6.00 6.00 298 0.336 6.40xl02 6.46 6.94 298 0.382 1.15xl03 7.05 7.00 284 0.390 8.04xl02 6.69 - 288 - 1.16xl03 7.05 - 293 - 1.18xl03 7.07 - 298 - 1.29xl03 7.16 - 304 - 1.32xl03 7.18 - 313 - 1.64xl03 7.40 7.26 298 0.400 2.17xl03 7.68 7.96 298 0.417 2.70x103 7.90 Acflavin modified Asn62Cys cytochrome c 5.33 298 0.343 1.30xl02 4.87 5.98 298 0.384 1.84xl02 5.22 174 P r o t e i n p H T ( K ) - A G 0 M O In ket ( e V ) A c f l a v i n modif ied A s n 6 2 C y s cytochrome c 6.98 283 0.411 1 . 8 3 x l 02 5.21 6.98 288 0.411 2 . 0 0 x l 0 2 5.30 - 293 - 2 . 1 8 x l 0 2 5.38 - 298 - 2 . 4 6 x l 0 2 5.51 - 303 - 2 . 7 1 x l 0 2 5.60 - 308 - 3 . 0 3 x l 0 2 5.71 7.17 298 0.420 3 . 6 1 x l 0 2 5.89 7.41 298 0.433 2 . 9 0 x l 0 2 5.67 A c f l a v i n modif ied L e u 8 5 C y s cytochrome c 5.24 298 0.353 4 . 4 0 x l 02 6.08 5.85 298 0.375 5 . 0 0 x l 0 2 6.21 6.69 298 0.401 7 . 2 7 x l 0 2 6.59 7.01 283 0.410 6 . 1 1 x l 0 2 6.42 - 286 - 7 . 3 4 x l 0 2 6.60 - 293 - 8 . 1 8 x l 0 2 6.71 - 298 - 8 . 5 1 x l 0 2 6.75 - 303 - 9 . 5 5 x l 0 2 6.86 - 308 - L l O x l O 3 7.00 7.38 298 0.425 1 . 2 0 x l 0 3 7.09 175 APPENDIX D - Fe(EDTA)2'reduction kinetic data Protein T(K) [FefEDTA)2-] (mM) (s1) Met80Ala cytochrome c 298 2.9 0.055(21) a 5.7 0.152(23) c c 8.6 0.196(10) c c 11.4 0.259(11) a 14.3 0.342(22) a 17.1 0.444(43) 288 20.0 0.281(10) 293 c c 0.402(25) 298 c c 0.512(68) 303 c c 0.670(70) 308 c c 0.937(98) Met80Ala/Tyr67Phe 8.25(8) cytochrome c 298 6.4 CC 8.0 12.5(12) CC 11.2 14.9(14) c c 14.4 20.5(18) c c 17.6 24.6(22) 288 20.0 20.9(20) 293 c c 24.2(24) 298 c c 29.7(27) 303 c c 40.5(30) 308 c c 51.1(34) Met80Ala/Phe82Ser cytochrome c 298 8.0 0.0128(13) c c 11.2 0.0162(15) c c 14.4 0.0215(20) 176 Protein T (K) [Fe(EDTA)2"] (mM) kobs (s'1) Met80Ala/Phe82Ser cytochrome c 298 17.6 0.0241(25) 288 20.0 0.0128(16) 293 " 0.0200(24) 298 " 0.0272(28) 303 " 0.0310(34) 308 " 0.0406(40) 177 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088692/manifest

Comment

Related Items