‘OIOIC-n . - ‘ 1-in- , u, ,.;u4.-‘.'-" ‘ILO-U-m '° ' ' a I K.- ‘F-xJ 4 ’3' r‘: “ r t/ i, ii Alix-k x hiltfituél Shit: ‘ "“f-gUniverztty é; \Mflm A This is to certify that the dissertation entitled RAPID-SCANNING STOPPED-FLOW STUDIES OF THE REDUCTION OF CYTOCHROME g OXIDASE presented by Folim G. Halaka has been accepted towards fulfillment of the requirements for Ph.D degreein Chemistry .0 Z g Major professor Date October 22, 1981 MS U Lt an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LIBRARIES 4—3—- RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. - RAPID-SCANNING STOPPED-FLOW STUDIES OF THE REDUCTION OF CYTOCHROME g_OXIDASE By Folim G. Halaka A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPY Department of Chemistry 1981 ABSTRACT RAPID-SCANNING STOPPED-FLOW STUDIES OF THE REDUCTION OF CYTOCHROME §_OXIDASE By Folim G. Halaka Rapid scanning and fixed-wavelength stopped-flow spectropho- tometry were used to study the anaerobic reduction of cytochrome c oxidase by 5, lO-dihydro 5-methylphenazine (MPH) and by sodium dithio- nite. 'In both cases the decay of the oxidized Soret band of the protein was not uniform. With MPH, a neutral molecule, the reduction of the cytochrome a_component of the oxidase preceded that of cytochrome 33. The kinetics of the reduction were found to be triphasic. The fast phase is a second order reaction between the oxidase and MPH. This is followed by two first order processes, which were interpreted as intramolecular electron redistribution between the oxidase four metal centers. Analysis of kinetic data showed that during the fast phase, the decay at 830 nm, due to the reduction of the copper ion asso- ciated with cytochrome a, lags the growth of absorbance at 605 nm (due to cytochrome a). The method of weighted principal component analysis (PCA) was used to resolve the three dimensional data surface obtained by the scanning stopped-flow method. By using PCA, the wavelength- absorbance-time data surface was resolved to its independent Folim G. Halaka components, and the spectral shapes and time courses of those com- ponents. Time courses obtained by PCA confirmed the assignments of cytochrome a_as the site of reduction by MPH. The properties of MPH and its oxidized form (MPMS) were also studied. The kinetics of the reduction of MPMS to MPH by NADH and the oxidation of MPH by oxygen were investigated. The anaerobic photoreaction of MPMS at pH = 7.4 was found to produce pyocyanine and MPH at nearly equal concentrations. The reduction of cytochrome oxidase by sodium dithionite was found to involve the $02 anion radical as the reducing agent. In contrast to MPH, SD; was shown to reduce preferentially the cyto- chrome 33 site of the oxidase. This was interpreted on the basis that the cytochrome a surface must be negatively charged. The reduction of cyanide-bound cytochrome g_oxidase by MPH was found to be monophasic, and involving only the cytochrome g_site and its associated copper. The reduction of the CN-complex by sodium dithionite was biphasic. ACKNOWLEDGMENTS I would like to thank Dr. James L. Dye for proposing this work on the anaerobic stopped-flow system and for his encouragement, enthusiasm and stimulating discussions. I am most grateful to Dr. Gerald T. Babcock for his support and ideas, and for acquainting me with the properties of cytochrome oxidase and its biological importance. 1 wish to thank the National Science Foundation, National Institute of Health and the Department of Chemistry for financial support as research and teaching assistantships. I would like to thank Dr. Tom V. Atkinson and Dr. Tom H. Pierce for their help with the computer work that appears through- out this work, and for assisting with computer graphics. Finally, I am grateful to my wife, Carole, for her encourage- ment and help in the typing of this manuscript. ii TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES Chapter I. II. III. INTRODUCTION . A. Cytochrome c Oxidase . A. 1 Structure and Subunits . A.2 Physico- -Chemical Properties A. 3 Redox Properties . A. 4 Kinetic Studies . Scanning Stopped- -Flow Technique for the Study of Multicentered Enzyme Reactions Treatment of Scanning Stopped-Flow Data C.l Rate Equations--Program KINFIT4 . . C.2 Weighted Principal Component Analysis . Reducing Agents for the Study of Electron Trans- fer Reactions of Cytochrome Oxidase EXPERIMENTAL METHODS A. B. Materials Methods . . . B. l Preparation of Cytochrome Oxidase . . B. 2 Activity Assays of Cytochyrome c Oxidase B. 3 Preparation of Anaerobic Solutions B. 4 Anaerobic Titrations The Stopped- -Flow System . C. 1 Handling and Performance of the Stopped- -Flow System . . . Data Handling, Computations and Computer Graphics . . . . . . PROPERTIES OF 5- METHYL PHENAZINIUM METHYLSULFATE: REACTION OF THE OXIDIZED FORM WITH NADH AND OF THE REDUCED FORM WITH OXYGEN. . . . A. Introduction iii Page vi vii 43 43 Chapter IV. VI. B. Experimental Section . C. Results and Discussion C. l Anaerobic Titration and Transient Kinetics . of the Reduction of MPMS by NADH C.2 Reaction of MPH with Oxygen . . C.3 The Reaction of MPMS with NADH in the Presence of Oxygen . . C.4 Effect of Room Light on the Spectra of MPMS D. Conclusions . REDUCTION OF CYTOCHROME OXIDASE BY MPH A. Spectral Shape Analysis . 8. Kinetics of the Reduction of Cytochrome Oxidase . by MPH. . 8.1 Analysis of the Fast Phase 3.2 Kinetics of the Slow Processes . . . C. Reduction of the Cyanide-Bound Cytochrome Oxidase . . . . . C. 1 Spectral Shape Analysis C. 2 Kinetics . D. Conclusions . PRINCIPAL COMPONENT ANALYSIS OF THE REDUCTION OF CYTOCHROME c OXIDASE BY MPH. A. Introduction B. The Method of Principal Component Analysis (PCA): B. l Preparation Steps B. 2 Number of Components . . B. 3 PCA Determination of Real Components . C. Principal Component Analysis of the Reaction of . Cytochrome Oxidase by MPH C. l PCA on the Reduction of the Resting Enzyme . by MPH. . C. 2 PCA on the Reduction of the Oxygenated Enzyme by MPH D. Conclusions . REDUCTION OF CYTOCHROME OXIDASE BY SODIUM DITHIONITE. A. Introduction . . 8. Spectral Shape Analysis . . C. Kinetics of the Reduction of Cytochrome by Sodium Dithionite . . . D Effect of Cholate iv Page Chapter Page E. Reduction of the Cyanide-Bound Cytochrome Oxidase by Dithionite . . . . . . . . . . T42 F. Discussion . . . . . . . . . . . . . l45 VII. SUGGESTIONS FOR FUTURE WORK . . . . . . . . 151 A. Effect of Detergents . . . . . 151 B. Partial Reduction of Cytochrome c Oxidase . . . l52 C. Aerobic Experiment . . . . . . . . . . 152 D. Cytochrome £552 . l53 E. The Application of Principal Component Analysis to Other Systems . . . . . . . . . l54 REFERENCES . . . . . . . . . . . . . . . . . 155 Table IV.l IV.2 IV.3 LIST OF TABLES Analysis of the fast phase of the reduction of cyto- chrome oxidase by MPH in HEPES buffer containing 0.5% Tween 20, pH = 7.4 at Zl i l°C. k is the second order rate constant for the proces . . . . Summary of studies on the I'slow phase" of the reduc- tion of cytochrome oxidase by MPH in HEPES buffer containing 0.5% Tween 20, pH 7.4, T = 2l i l°C Rate constant for the reduction of the cyanide-bound cytochrome oxidase by MPH in 50 mM HEPES buffer con- taining 5% Tween 20, pH = 7.4 at 2l i l°C. Data are collected in fixed-wavelength experiments . Rate constant for the reduction of cytochrome oxidase by sodium dithionite in 50 mM HEPES buffer containing 0.5% Tween 20, pH = 7.4, T = 21°C. Rate constants are the observed first-order constants from the analysis of data by three exponentials . . Page 73 80 84 136 Figure 1.1 1.2 11.1 11.2 11.3 111.1 111.2 III.3 111.4 111.5 IV.l LIST OF FIGURES Structure of heme 1 Optical absorption spectra of cytochrome oxidase: , oxidized, and ----- , fully reduced Absorption spectra of cytochromes a, a3 and the CN- complex of cytochrome a3; . . . . . . EPR reductive titration data for g = 2, 3, and 6 signals . . . . . . . The Cu (a3) -- Fe (a3) and Cu (a)--Fe (a) centers in cytochrome oxidase Cells used for preparing anaerobic solution (A) and for anaerobic titrations and preparation of air- sensitive reductants (B) Flow-velocity profile as a function of pressure . Test of mixing efficiency in the stopped- -flow apparatus . . . . . . . . Structure of N-methyl phenzaine at different pro- tonation and oxidation levels . . . . . Anaerobic titration of 7 ml of 15.9 pM MPMS by successive additions of 0.026 ml increments of 0.527 mM NADH in 50 nm HEPES buffer, pH = 7.4 Spectral changes during the anaerobic reduction 3l.9 uM MPMS by 39 pM NADH . Spectral changes which result from mixing NADH with MPMS in the presence of atmospheric oxygen Effect of room fluorescent light on the spectra of 2l.2 uM aerobic MPMs . Absorbance-time—wavelength surface in the 330-530 region for the anaerobic reduction of cytochrome oxidase by 5, lO-dihydro-S—methyl phenazine (MPH) vii Page 10 ll 14 T7 34 38 4O 44 49 50 53 56 61 Figure IV.2 IV.3 IV.4 IV.5 IV.6 IV.7 IV.8 IV.9 IV.lO IV.lI IV.lZ V.l V.2 V.3 V.4 Absorbance- time wavelength surface for the anaerobic reduction of cytochrome oxidase by MPH in the 400 - 630 nm region . . . . Selected spectra from Figure IV.l Selected difference spectra constructed from Figure IV.l by subtracting the oxidized oxidase spectrum Time dependence of absorbance at 430, 4l2 and 388 nm; taken from the spectra shown in Figure IV.l Selected spectra in the reaction of MPH with cyto- chrome oxidase, taken from the data shown in Fig- ure IV.2 . . . . . . . . . . . . . . (A) Fit of the fast phase for the reduction of .cyto- chrome oxidase by MPH . . . (A) The decay at 830 nm and 605 nm during the reduc- tion of l0.8 uM cytochrome gag by 27 pH MPH Fit of the "slow phase" to one exponential (a); to a second order process,(b); and to two exponentials (C) Residual plots for the fit to the data in Figure IV. 9 . . . Spectral changes that resulted on the reduction of 2.4 uM CN-oxidase by 6.5 pM MPH Time courses for the reduction of the CN- oxidase by MPH followed at 430 nm (A),. 605 nm, (B) and at 444 nm (C) . . . Block diagram for main steps in factor analysis . Experimental absorbance-wavelength-time surface for the anaerobic reduction of 3.78 pM cytochrome oxidase by 8.4 uM MPH . . . . . Principal component analysis (PCA) reconstructed surface using three eigenvectors in M-analysis Reconstructed surface using only two M-analysis eigenvectors for the data in Figure v.2 viii Page 62 64 65 67 68 71 75 77 78 81 83 93 104 105 106 Figure Page v.5 Residual surface, resulting from subtracting the matrix in Figure v.3 from the data in Figure v.2 . 107 V.6 Residual surface . . . . . . . . . . . . 108 v.7 M-analysis fit of oxidized cytochrome 3 spectra from the data presented in Figure v.2 . . . . . 110 v.8 M-analysis fit of oxidized cytochrome 1 using three eigenvectors . . . . . . . . . . . lll v.9 M-analysis fit of the reduced oxidase + MPMS con- tribution . . . . . . . . . . . . . . 112 v.10 Concentration-time profile for the disappearance of oxidized cytochrome a_ . . . . . . . . . 114 V.ll Experimental wavelength-absorbance-time data sur- face for the anaerobic reduction of the oxygenated form of the oxidase by 26 pM MPH . . . . . . . 116 v.12 Reconstructed surface for the data presented in Figure V.ll using two eigenvectors . . . . . . 117 V.l3 Reconstructed surface for the data in Figure V.ll using three eigenvectors . . . . . . . . . 118 v.14 Residuals of the data presented in Figure V.ll . . 119 V.l5 Residuals of the data presented in Figure V.ll . . 120 V.l6 M-analysis fit of the combined oxidized cytochromes a_and g3 spectrum for the oxygenated oxidase . . . 122 v.17 Three eigenvectors M-analysis fit to the spectrum of reduced cytochrome a_ . . . . . . . . . 124 v.18 Three eigenvectors M-analysis fit to the spectrum of reduced cytochrome-a3 . . . . . . . . . . 125 v.19 Concentration- time profile for the growth of reduced cytochrome a, (B); reduced cytochrome a (C), and for the disappearance of the combined oogidized (133) . . . . . . . . . . . 126 V1.1 Spectral shapes during the anaerobic reduction of 3. 78 pH cytochrome oxidase by 47 uM sodium dithionite . . . . . . . . . . . . . 132 ix CHAPTER I INTRODUCTION A. Cytochrome §_0xidase Cytochrome g_oxidase (ferrocytochrome 9: oxygen oxido- reductase, EC 1.9.3.1) is the terminal oxidase in the respiratory chain of all aerobic organisms. The enzyme, referred to as cyto- chrome 333 or cytochrome oxidase, is a membrane-bound heme protein which catalyzes the reduction of molecular oxygen to water by ferro- cytochrome 9, according to Equation 1.1 4 Cytochrome 22* + 02 + 4 H+ + 4 Cytochrome 33* + H20 1.1 The energy released is used for the synthesis of adenosine triphos- phate (ATP), which, in turn, is used as the source of energy in other cell activities (see, for example, Lemberg, 1969). It is estimated that about 90% of the oxygen consumption in biological systems is achieved through cytochrome oxidase (Malmstrom, 1973). The protein is, therefore, prominent in tissues where energy requirements are high (Challoner, 1968; Tucker, 1966). Although this study will focus on the electron transport properties of the protein, the proton trans- location across the mitochondrial membrane as a result of Equation 1.1 has been observed and suggests that cytochrome oxidase can act as a proton pump (Wikstrom and Krab, 1979). 1 The work of Keilin (see, for historical background, Keilin, 1966) conclusively demonstrated the heme nature of the prosthetic group of the protein and its link to the cell respiratory chain. Due to the obvious importance of the protein, it was, and currently is, the subject of active investigation by every conceivable physico- chemical technique. Various comprehensive reviews have appeared in the literature which survey the state of understanding of the protein at the time of their publication. Examples are the articles by Lemberg (1969), Malstrom (1973), Wharton (1974), Nicholls and Chance (1974), Erecinska and Wilson (1978), and Malmstrom (1979). A.1 Structure and Subunits A.1.1 Metal components.--It is now agreed that all functional preparations of cytochrome oxidase contain iron and copper as essen- tial components (Volpe and Caughey, 1974). It is also well established that the iron to copper ratio is l (Griffith and Wharton, 196la,b). Caughey et a1. (1976), reported that most preparations contain about 11 nanomoles of iron per mg protein. More recent procedures (Komai and Capaldi, 1973; Hartzell and Beinert, 1974) contain iron ratio as high as 14 nanomole/mg protein. Keilin and Hartree (1938a,b; 1956) demonstrated that the iron of the oxidase is present as heme a (Figure 1.1). They also showed, from reactions with C0 and other inhibitors that cytochrome oxidase is actually composed of two heme a_components: Cytochromes a_and a3 (hence the name cytochromeaa3 for the oxidase). Only cytochrome a3 is found to react with exogenous ligands such as cyanide, azide, C0 heme 9 Figure 1.1.--Structure of heme a. and oxygen. However, only one heme compound with a particular struc- ture, heme a, (Fig. 1.1), is isolated from cytochrome oxidase (Caughey et al., 1975). This classical cytochromeaa3 picture was confirmed by recent investigations of the stoichiometry of the C0 binding, showing that only one-half of the oxidase heme binds CO (Wharton and Gibson, 1976; Toshikawa et al., 1977). Thus, the cyto- chrome oxidase molecule must contain two iron and two copper ions. Although the immediate environments of the iron of the oxidase are now accepted to be heme a, the environments of the two copper atoms are less well understood. The copper is apparently well shielded by the protein, which is evident from the fact that it is not readily extractable by common Cu2+ ligands. Griffith and Wharton (1961) showed that there is no exchange between added Cu2+ and the copper ions of the oxidase. When Cu was removed from cytochrome oxidase, the activity was irreversibly lost (Wharton and Tzagoloff, 1964). From electron paramagnetic resonance (EPR) studies, it was concluded that there exist two different functional copper ions in the oxidase (see Section A.3), only one of which is EPR detectable. The EPR detect- able Cu, normally associated with cytochrome a_is designated Cua (sometimes Cud). The EPR undetectable Cu is designated Cua3 (or Cuu). Based on spectroscopic properties, the copper sites in copper proteins are put into one of three major classifications (Malkin and Malmstrom, 1970). Type I or "blue" copper sites exhibit absorption 1 1 cm"), 600 nm (e = 3-10 mM’ cm"), 1 bands near 450 nm (e = 0.3-1.0 mM' and a near infrared band at around 800 nm (e = 0.3-3.0 mM- cm'l). Type 1 copper exhibits narrow EPR hyperfine structure (Peisach and Blumberg, 1974) and an oxidation-reduction potential of about 0.3 - 0.4 V. Type 11 copper sites (non-blue copper) have less well-defined optical pr0perties, they have a weak absorption band at 600 - 700 nm (e = 0.3 — 0.4 mM'1 cm") and their EPR spectra are characterized by a broad hyperfine structure (Peisach and Blumberg, 1974). In pro- teins having both type I and type II c0pper sites, the blue copper is more quickly reduced, while the non-blue copper has more affinity toward anions. Type III copper sites are EPR non-detectable. They are thought to be diamagnetic and consist 11f either Cu+ or a spin- paired, Cu2+ - Cu2+. Their oxidation-reduction potentials are greater than 0.5 V (Malkin and Malmstrfim, 1970). A.1.2 Paptide subunits.--Isolated cytochrome oxidase con- tains several subunits. The number and composition of these subunits, as well as their positions relative to one another and in the membrane, are still uncertain. The hypothesis that the enzyme complex consists of seven different subunits was develOped on the basis of studies on microbial enzymes (Sebald et al., 1973; Poyton and Schatz, 1975) and was subsequently extended to include mammalian enzyme (Downer et al., 1976; Tracy and Chan, 1979; Hochli and Hackenbrock, 1978). These subunits are given the Roman numerals I - VII. However, several other groups have prepared bovine enzymes with only six subunits (Rubin and Tzagoloff, 1973; Briggs et al., 1975; Penttila et al., 1979). The subunits are reported to be present in 1:1 ratio (Downer et al., 1976; Yu and Yu, 1977). There is, however, substantial disagreement about the molecular weights of the individual subunits. The hydrophobic nature of many of these peptide subunits adds another complication, since most conventional methods of determining molecular weights cannot be applied to them. Instead, gel electrophoresis was used. Recent preparations (Hartzell and Beinert, 1975) give good agreement between the molecular weight calculated on the basis of heme: protein ratio (14 nanomole/mg) and the sum of the individual molecular weights of the subunits, 140,000 Daltons. Since the number of subunits is greater than the number of metal ions in the enzyme complex, some of the polypeptides may have a role in arranging the complex in the membrane (Steffens and Buse, 1979) and/or a proton pump function (Wikstrdm and Krab, 1979). Phan and Mahler (l976a,b) reported experiments in which they prepared a four-subunit enzyme that retained electron transfer activity (toward cytochromeaz+ oxidation). These results have been extended by Fry et a1. (1978) who claim to have separated the protein into an electron transfer complex and an ion (H+) transfer portion, which contains no metal ions. The metal binding polypeptide subunits are subject to a great uncertainty. Almost all subunits have been reported to bind either copper or iron, or both (Tanaka et al., 1977; Gutteridge et al., 1977). Also, the results of experiments that have been designed to identify which subunit interacts with cytochrome a_are quite variable (Briggs and Capaldi, 1978; Bisson, et al., 1978; Birchmeier et al., 1976; Seiter et al., 1979). Recently Winter et a1. (1980) reported that subunit II bound most of the Cu and that heme a was found in equal amounts in subunits I and 11. They suggested that subunit II is the site for cytochrome a_binding, while subunit I is the binding site for oxygen. Isolated cytochrome oxidase has not been crystallized in a form suitable for X-ray diffraction studies, although one account of a crystalline ly0philized preparation (Yonetani, 1961) and a crystal- line cytochrome aecytochrome oxidase complex (Ozawa et al., 1980) have appeared. Two dimensional "crystals" of membranes containing cyto- chrome oxidase have, however, been obtained. Henderson et a1. (1977) have shown that the oxidase molecule is asymmetrically placed in the membrane. The picture that emerges from these studies, though far from perfect, is that of a multi-subunit enzyme spanning the membrane with two "heads,9 one in the cystolic side and the other in the matrix side. A.1.3 Ljpids.--Depending on the method of preparation, iso- lated cytochrome oxidase contains variable amounts of lipids. Certain amounts of lipids are essential in order to maintain electron trans- port activity. To solubilize the protein, the use of several non- denaturing detergents has been implemented. The importance of differ- ent phospholipid head-groups to cytochrome oxidase activity has been examined by several research groups (Awasthi et al., 1971; Yu et al., 1975). A study on the decrease in activity of the oxidase upon lipid depletion has been reported by Vanneste et a1. (1974). However, the activity was restored by the addition of phospholipids or detergents. Robinson and Capaldi (1977) found important differences in the mode of action of non-denaturing detergents. Their study on the displacement of the protein phospholipids by several ionic and non- ionic detergents showed that some phospholipids do not exchange with detergents. Their studies also suggest that the oxidase is in a dimer form (4 heme ajoxidase complex) and that the oxidase activity is greatly affected by the nature of the hydrocarbon portion of the detergent. Rosevear et a1. (1980) have examined the interaction of the detergent lauryl maltoside with cytochrome oxidase. Their results indicate that this detergent, in addition to maintaining high electron transport activity (oxidation of cytochrome a), also has the advantage that the protein solution is monodisperse (probably as the dimer). A.2 Physico-Chemical Properties It is almost impossible, due to their diversity, to cover in this introduction all the physico-chemical studies which have been done on cytochrome oxidase. Hence, this section will cover only recent studies relevant to the electron transfer properties on the protein. A.2.1 Optical absorption of gytochrome oxidase and its derivatives.--Simple and more convenient than many other physical methods, optical absorption spectroscopy provides a way to monitor changes in the oxidase induced by various chemical reactions. Indeed, spectral observations were instrumental in the pioneering experiments that characterized this important protein. Cytochrome oxidase has several absorption peaks that character- ize its redox and/or ligation state. Figure 1.2 shows the spectral characteristics of the oxidized (resting) and the fully reduced protein. The electromagnetic absorption in the near UV (the "Soret" or a-band) and that in the visible (the o-band) is due to the porphyrin ring of the heme a moieties. The structure of heme a is shown in Figure 1.1. A theoretical discussion of the origin of prophyrin spectra, based on group theoretical calculations of the 04h symmetry of the polyene and allowing for configuration interaction is given by Gouterman (1959). Assignments of wavelengths and extinction coefficients for hemes a_and a3 and copper (Vanneste, 1966) have been rather widely accepted (Lemberg, 1969). These assignments are summarized in Figure 1.3. It is, however, risky to synthesize spectra for cytochromes a and a3 on the assumption that the properties of one heme are independent of the oxidation or ligation state of the other metal component(s) (Caughey et al., 1976). This comment takes into account the fact that there is ample evidence of facile electron and magnetic exchange between the metal centers in the oxidase (Hartzell et al., 1974; Babcock, et al., 1978). It is obvious that any absorption, or in the same sense, change in absorption, due to copper in the Soret or the a-regions will be obscured by the much larger absorptions of the heme chromo- phores (see Section A.1.1 for the magnitudes of protein copper extinctions). 10 E, mM" cm" Olllllllllllllllllll 500 550 600 650 700 omtsmm [’4' W— 650 700 750 800 850 Wavelength, nm Figure I.2.--0ptical absorption spectra of cytochrome oxidase: , oxidized, and ----, fully reduced. Molar absorptivities are expressed per unit containing two hemes and two copper ions. 11 1TlT|1I1IVIII1 — o"3 - CN --- o'; - CN 100 —O '20- '1‘ ___0203 _ 100- ,1, _20 IE 80” ii -10 0 so .' 1 TE 40- 1‘ £5 22C>F m. 100 80 60 4O 2O 1111411444111 400 480 560 640 Wavelength, nm Figure I.3.--Absorption spectra of cytochromes a, a and the CN- complex of cytochrome a ; , oxidizgd and ----, reduced; taken from VanNeste (1966). 12 Griffiths and Wharton (1961a) drew attention to a weak absorption band (e ~ 4mm"1 cm") in the near 1.R. region (800 - 900 nm). This band was attributed mainly to copper, Cua.absorption (Wharton and Tzagoloff, 1963). However, it was argued that the oxidase hemes may contribute in this region (Greenwood et al., 1974). Based on extended X-ray absorption fine structure (EXAFS), Powers et a1. (1979) proposed that the absorption in the near IR region is due to both Cua and Cu However, this pr0posal was countered by Beinert a 3 et a1. (1980) who presented a survey of data over many years (mainly EPR and Optical reflectance spectra), and concluded that Cua does 3 not have significant absorption in the 800 - 900 nm region, in agree- ment with the conclusions of Babcock et a1. (1978). A.2.2 EPR and magnetic suscgptibility.--Several EPR signals have been detected for cytochrome oxidase (Hartzell and Beinert, 1974). The spectrum of the oxidized oxidase has contribution from a?+) at g values = 3.0, 2.2, 1.5. A narrow signal at g = 2 is generally attributed to C3 low-spin heme (cytochrome + (see, how— a . ever, Hu et al., 1977, and Peisach and Blumberg, 1974, for another interpretation of the "c0pper" signal). Assa et a1. (1976) inte- grated the intensities of the various signals and showed that the low-spin heme signal corresponds to only one heme. These authors also showed, by simulation, that the signals from the low-spin heme and copper correspond to two magnetically isolated centers (>10A° apart). The copper g = 2 signal corresponds to only about one-half of the total copper. 13 On partial reduction, a high spin heme signal at g = 6 appears. Babcock et a1. (1978) studied reductive titrations of the oxidase by sodium dithionite followed by Optical absorption and EPR measurements under argon. They suggested that the high spin signal is due, at least partially, to cytochrome a, Their EPR data are summarized in Figure 1.4 for the g = 2,3 and 6 signals. Magnetic susceptibility measurements provide information about the spin state of metal centers which is complementary to the EPR data. The susceptibility of the fully oxidized oxidase was measured at room temperature (Falk et al., 1977) and at 7-200°K (Tweedle et al., 1978). The data from the two groups indicate that 3+ the oxidized enzyme has two S = 1/2 centers; cytochrome a_ and Cu and one S = 2 center, an antiferromagnetically coupled cytochrome 2 a3+ - Cua + pair. The reduced oxidase has an S = 2 center, which is 3 attributed to high spin cytochrome a§t,with all other centers being diamagnetic. A.2.3 Other spectroscgpic studies.--Babcock et al. (1976, 1978) studied the magnetic circular dichroism (MCD) of cytochrome oxidase during the course of reductive titration. They also studied the M00 characteristics of several inhibitor complexes of the oxidase. Comparison with the MCD spectra of model compounds led these authors to confirm that cytochrome a_is low-spin, while cyto- chromea3 is high-spin. The authors also suggested, from MCD properties at various redox levels, that there is no interaction between cytochromes a_and a3 apparent in the M00 spectra. 2+ 14 CYTOCHROME OXIDASE=ARGON ATMOSPHERE ’ 1.0 1 l 1 1 ' .. 09=2 09:3 .. . v ‘ Q A936 18 0'1 0 ,2 .6— — 6 \ In .E Q -- .- m .2-- .. t ‘ A C) e'loxidose Figure I.4.--EPR reductive titration data for g = 2,3, and 6 signals (from Babcock et al., 1978). 15 Resonance Raman (RR) spectroscopy is currently an active tool for the study of the environments of the hemes in cytochrome oxidase. In principle, RR measurements should be capable of dis- tinguishing the two heme environments. The problem of photoreduction of the oxidized protein in the laser beam (Adar and Yonetani, 1978) has been minimized by using a flowing sample (Babcock and Salmeen, 1979). By careful choice of the excitation frequencies, Babcock et a1. (1981) were able to enhance selectively and assign vibrations to a particular heme in the oxidized, reduced or inhibitor-complexed cytochrome oxidase., 0n the basis of comparison to RR spectra of heme a model compounds, these authors concluded that cytochrome a is six coordinate low spin in both oxidized and reduced states. Cytochrome a3, however, was shown to be six coordinate and high-spin in the oxidized,but five coordinate and high-spin in the reduced enzyme. A correlation of these results to the porphyrin core size was attempted (Babcock et al.,1981; Callahan and Babcock, 1981). The conclusions drawn from these studies agree closely with earlier deduc- tions of the spectra of the separate components of the oxidase by Vanneste (1966). Another recent RR study on mixed valence cyto- chrome oxidase, compound C (Chance et al., 1975) was done at low temperatures (-70° C) (Yang et al., 1981). This study suggests that compound C has reduced cytochrome a and Cuaswhile cytochrome a3 - Cua3 are oxidized, as suggested in the original assignments of this com- pound. The EXAFS method is emerging as a useful tool for the study of the environments of metal centers in metalloproteins. Recent 16 EXAFS studies on cytochrome oxidase (Powers et al., 1981) gave inter- esting insights into the structure of the oxygen reducing site, the a3 - Cua3 pair. The results indicate the presence of a sulfur bridge in the resting enzyme, with a Cua - Fe (as) distance of 3.75A9. Chance and Powers (1981) suggested that as this site becomes reduced (a:+ - Cu; ), an oxygen molecule can replace the sulfur bridge to form the Eiroxide. To account for the high turnover number of the protein, the authors suggest that the sulfur bridge does not reform during turnover. This picture requires the bridging sulfur to be a weak ligand, since cytochrome a3 is high-spin. The model derived from these measurements is shown in Figure 1.5. A.3 Redox Properties Electrochemical titrations of cytochrome oxidase monitored by optical measurements were reported by Shroedl and Hartzell (1977a, b,c). Their interpretation of the results was that there are two different potentials for the two cytochromes, referred to as high and low-potential hemes without assignment of potentials to specific hemes. They also concluded that two different copper redox poten- tials, similar to those of the hemes, must be present. Nicholls and Hildebrandt (1978) concluded that binding of a single HCN molecule to the oxidase prevents the reduction of both cytochrome a3 and the copper associated with it. It should be emphasized, however, that these measurements suffer from two major difficulties: N\ S y \/ _'§;.. N— Fem-Ar?" Cyt o3”. Cuzas N\ /S Cu 3’ ‘3 Cu (0) Figure I.5.--The Cu (a ) - Fe (a in cytochE 17 oztcm I? S N—l—N \ / / + Cu N T \N I s N Cyt. as". Cu’a 3 T /NT=;N\ N—I'_N N Cyt.(o) ) and Cu (a) - Fe (a) centers ome oxidage (from Powers et aTI, 1981). 18 l. Electrochemical titrations usually contain high con- centrations of organic and inorganic redox couples and electron mediators that may interfere with the redox properties of cytochrome oxidase (Lanne et al., 1977). 2. Optical spectra cannot clearly discriminate between the isolated vs. interacting models of the metal redox centers at equilibrium. Magnetic techniques were also applied to the study of the oxidase redox properties. In particular, EPR measurements during redox titrations were very useful in monitoring redox centers of the protein (Hartzell and Beinert, 1975). Babcock et a1. (1978) studied the optical, MCD and EPR properties during redox titration of the oxidase with sodium dithionite. Their EPR data (presented in Figure 1.4) and optical data indicate that the two cytochromes titrate together at all points, as seen, for example, from the g = 3 signal. Their data also indicate that the EPR detectable copper, Cua, lags the reduction of the two cytochromes. These results are different from earlier results (van Gelder et al., 1973; Tiesjama et al., 1973) in which cytochromes a_anda3 and the detectable copper were assigned equal redox potentials of 280 mV. The data of Babcock et a1. (1978) provide evidence for heme-heme interaction via redox potentials, and indicate that the reduction of one cytochrome makes reduction of the other more difficult (negative cooperativity). It should be men- tioned, however, that EPR measurements are usually done at low tempera- tures and at higher protein concentrations than are optical 19 measurements. The redox potentials are, of course, temperature dependent and electron redistribution may occur during freezing. It is becoming more widely accepted, as shown by Lanne and Vanngard (1978), that a non-interacting four-redox center cannot describe the available EPR and optical data. Their analysis, on the other hand, showed that if interaction is introduced, the available data cannot discriminate between various models. In other words, assignments of equilibrium redox potentials to the four metal centers is still far from established. A.4 Kinetic Studies As is the case with other enzymatic systems, the study of the kinetics of cytochrome oxidase reactions plays an important role in understanding and characterizing its action. Studies of enzymatic reactions usually start by determining the steady state parameters in turnover experiments involving the enzyme and its natural substrate(s). Since turnover experiments are usually complex and involve the adjust- ment of many parameters, the study of transient state kinetics of pre- or post-steady state steps usually follows. These studies become important in determining accurate rate constants and in discriminating between mechanisms. The present study is devoted primarily to the investigation of the kinetics pathways which lead to reduction of the four redox centers of the oxidase. Hence, a summary of the litera- ture data on the reaction of cytochrome oxidase with different sub- strates is presented below. 20 A.4.l The reaction with cytochrome_Q.--The reaction of cytochrome a oxidase with its natural substrate, cytochrome a, has obviously received a great deal of attention. The kinetics of this reaction have been studied both anaerobically and in aerobic steady- state. The steady-state reaction mixtures usually contained ascorbate, cytochrome g, cytochrome oxidase and oxygen, with ascorbate serving as the electron source and oxygen as the sink. The kinetics of the steady-state reaction have been extensively studied (Slater, 1949; Yonetani, 1962; van Buuren et al., 1971; Petersen et al., 1976; Errede and Kaman, 1979; and Petersen and Cox, 1980). Most of these studies have found that Minnaert mechanism IV, (Minnaert, 1961), equation 11.2 fit the data. k1 k1 E + S -—————+ ES + E + P 1.2 “FT-1 “ET More recent experiments, in which wider ranges of cytochrome a_con— centrations were used,have shown non-linear Lineweaver-Burk and Eadie- Hofstee plots (Ferguson-Miller et al., 1976; Errede et al., 1976). In other words, the simple hyperbolic behavior of k(obs) expected from equation 1.2 is not found. A rate equation of the form given by equation 1.3 has been shown to describe the rate dependence on the cytochrome a concentration (Errede et al., 1976; Errede and Kamen, 1978). a1+a2[c] Velocity = - ——~ [Oxidase] [Ferrocytochrome 53 1.3 l+BIICJ+BZICI 21 Where a], a2, 8], 82 are constants and [c] is the total cyto- chrome a concentration. The quadratic term in total cytochrome a concentration indicates that under certain conditions two molecules of cytochrome a may be bound to the oxidase at the moment of electron transfer. The stoichiometry of cytochrome a_binding (Ferguson-Miller et al., 1976; Erecinska, 1975) supported this mechanistic model. Errede and Kamen (1979) discussed several models that can lead to the observed rate law behavior. Transient state kinetic studies of the reaction between cytochrome oxidase and cytochrome a_have been of great importance in understanding the nature of the interaction between the two enzymes. Under anaerobic conditions, the reaction is multiphasic (Gibson- et al., 1965; Antonini et al., 1973; Andreasson et al., 1972; Andreasson, 1975). Most authors report a second order rate constant 7 1 1, depending on the ionic for the initial phase of about 8 x 10 M- s- strength (Gibson et al., 1965; Andreasson, 1975; Wilms et al., 1981). It was also found that added ligands such as CN— or N5 do not affect this fast phase (Gibson et al., 1965; Andreasson, 1975), which sug- gests that this phase involves the reaction of cytochrome _c_ with the cytochrome a site of the oxidase. The slower phases are not as well defined as the initial phase. These phases, particularly the last phase, are too slow to be of significance in catalytic cycles. The presence of oxygen is thought to shift the redox potentials,such that these slow phases are much faster (Andreasson et al., 1972) to account for turnover numbers of the protein as high as 400 s'1 (Petersen et al., 1976). Andreasson et a1. (1972), who monitored the reaction 22 kinetics at 830 nm and at 605 nm,observed similar kinetics at these two wavelengths. They interpreted their data by assigning about 40% of the absorbance at 830 nm to cytochrome a, A.4.2 Reaction with oxygen.--The reaction of the reduced cytochrome oxidase with oxygen is also multiphasic (Greenwood and Gibson, 1967). The first phase is diffusion controlled (k = 1 x 108 M'is'I) and the flash-flow technique, where the reduced form of the oxidase was generated photochemically from the reduced oxidase-- C0 complex, was used to measure the reaction rate (Gibson and Green- wood, 1963). To slow the reaction with oxygen and to identify the possible intermediates leading to the reduction of molecular oxygen to water, Chance et a1. (1975) used low temperature flash photolysis technique. Chance and co-workers. (Chance at al., 1975; 1978; Chance and Leigh, 1977) have identified several intermediates in this reaction pathway. Their results have been confirmed by Clore and E. M. Chance (Clore and Chance, 1978a,b,c; 1979). It appears that the proximity of Cua3 and Fe(a3) plays an important role in the reduction of molecular oxygen. The high redox potential of the oxidase (Malmstrom, 1973) in addition to the unfavor- ably high energy for the formation of superoxide radical lend support to the idea that both Cua3 and Fe(a3) must be reduced before forming the peroxide. The bound peroxide intermediate could then be reduced via two one-electron steps. The intermediates formed are thought to be stabilized by electron delocalization within the Cua - Cyt a3 - O2 3 unit. 23 A.4.3 The "oxygenated" and the "oxygen—pulsed" forms.--When reduced cytochrome oxidase is reoxidized by oxygen, a new form of the protein,which is spectrally and functionally different from the resting enzyme,is produced (Okunuki et al., 1958). This form is called the "oxygenated" cytochrome oxidase, to distinguish it from the resting protein (the enzyme as prepared). The terminology origi- nated from the belief that this form contains an enzyme-oxygen complex (Sekuzu et al., 1959). Subsequent studies have shown, however, that this form does not contain oxygen bound to its metal centers, but instead, it repre- sents a conformational variant of the enzyme (Tiesjema et al., 1972; Nicholls and Petersen, 1974; Brittain and Greenwood, 1976). The "oxygenated" oxidase is characterized by an absorption maximum at 425 nm compared to 418 nm in the Soret region of the resting protein. Antonini et a1. (1977) introduced the term oxygen-pulsed oxi- dase, which is formed by exposing the reduced pretein to a "pulse" of oxygen. This form was shown to be more active in reaction with cytochrome a and during turnover experiments (Antonini et al., 1977; Petersen and Cox, 1980), which led these authors to suggest that this conformation is the catalytically significant form of the oxidized protein. However, the claim that this catalytically active species can form only after fully reducing the protein (Antonini, 1977) should be taken with caution, since the natural reducing agent, cytochrome a, cannot fully reduce the oxidase (4 electron) anaerobically (unpub— lished results; see also Andreasson, 1975). This is also disproved 24 by the fact that turnover is achieved after the addition of about two electrons to the oxidase. 8. Scanning Stopped-Flow Technique for the Study of Multicentered Enzyme Reactions Stopped-flow spectrophotometry has been extensively used in the study of the kinetics of enzymatic reactions. The method most commonly used is to monitor changes in the absorbance at a certain wavelength as a function of time (for historical background, see Sturtevant, 1964). Advances in computer-controlled data acquisition made scanning wavelength experiments practical for many enzymatic reactions (Papadakis et al., 1975; Coolen et al., 1975; Suelter et al., 1975; June et al., 1979; Cox and Holloway, 1977; Halaka et al., 1981a). In a scanning wavelength experiment, a spectral region is repeatedly scanned and the absorbance (or other spectrophotometric response) is measured for every scan. The data are stored as a function of time for every wave- length "channel." If the time of scan is short compared to the half- time of the reaction(s) studied, these data can be considered as a matrix a composed of N consecutive spectra (each essentially instan- taneous at time t), measured at p wavelength channels. The element Aij of this matrix is the absorbance measured at wavelength channel 1 at the time of scan j. The advantages obtained by collecting this time-wavelength- absorbance surface for a spectral region are important when studying enzymes with two or more interacting chromophores. The power of this 25 method becomes apparent from the consideration that one can have information about the kinetics of each chromophore at many wave- length channels in the aama experiment. This, of course, eliminates long time base-line drift problems and minimizes problems which result from lack of reproducibility from one experiment to another. As can be seen from the discussion on cytochrome oxidase, rapid-scanning stopped-flow methods seem ideal for the study of the kinetics of this enzyme. The results promise to provide information about possible pathways of electron transfer and sites of interaction with substrates. C. Treatment of Scanning Stopped-Flow Data Scanning stopped-flow experiments produce massive amounts of data for every reaction studied. A general procedure for treating these data is discussed below. C.l Rate Equations--Program KINFIT4 The first step in studying the kinetics of a certain reaction is to find the equation(s) that fit the time course at the wavelength channels of interest. This, in itself, can lead to important infor- mation about the interactions between chromophores in multicentered reactions and the mechanism of their reactions with certain reagents (e.g., sequence of steps and cooperativity). The computer program that is used to extract the rate and equilibrium constants is the program KINFIT4, which is a modified version of program KINFIT (Dye and Nicely, 1971). This program 26 utilizes a simultaneous fit of the equations to a number of data sets and computes estimates of the marginal standard deviations. These uncertainties include the effect of coupling among rate constants and other adjustable parameters. If the model proposed for the reaction is correct, these statistics can give reasonable confidence in the rate and equilibrium constants which result. C.2 Weighted Principal Component Analysis One of the ultimate goals of studying chemical kinetics is to propose and understand a mechanism that accounts for the known facts about a certain reaction. An essential step toward this goal is to know the number of interacting species in the reaction mixture. Enzyme reactions usually involve transient intermediates whose spectra are unknown. In fact, binding to proteins modifies the spectra of many substrates. To determine the number of light-absorbing species in a chemi- cal reaction, the method of weighted principal component analysis, PCA, was developed (Cochran and Horne, 1977, 1980; Cochran et al., 1980). The method and its history are discussed in detail in the Ph.D. dissertation of R. Cochran (1977). The PCA starting point is the matrix model mentioned in the previous section. No mechanistic assumptions are needed to apply PCA; the only assumption required is that the absorbances of species in the reaction mixture are linear functions of their concentrations (Beer's law). Two kinds of PCA, the second moment matrix principal com- ponent analysis (called M analysis) and the sample covariance matrix 27 principal component analysis (S analysis) are useful for kinetics experiments. Each requires only the matrix a_(with the proper weights; see Chapter V) and each gives an estimate of the minimum value of_g, the number of independent chromophores in the reaction. M analysis gives for q a lower bounds estimate that is sensitive to any linear dependences of the concentrations of the various detectable species. S analysis gives an estimate that is sensitive to the linear dependence of the time rates of change of the concentrations. The two estimates of q are not necessarily the same, and the applica- tion of both analyses enables one to discriminate between alternate stoichiometries during reaction (Cochran and Horne, 1977). 0. Reducing Agents for the Stady of Electron Transfer Reactions of Cytochrome Oxidase Scanning stopped-flow experiments provide information not only about the kinetics of a chemical reaction, but also about the spectral properties of the intermediates. Therefore, a reducing agent with characteristic spectral shape that does not interfere strongly with those of the oxidase,is preferred for these studies. Such a reducing agent will also have the advantage of providing estimates of the number of reducing equivalents which have been added to the oxidase at any extent of reaction. Cytochrome c2+, the natural reductant, was discussed in Section C.3.1. There are two difficulties involved in reduction by cytochrome.a: 1. The large Soret absorption of cytochrome a_obscures changes in the cytochrome oxidase spectrum in this region. 28 2. As shown by anaerobic equilibrium titrations (this study), only two electrons can be transferred from cytochrome a to cytochrome oxidase, in agreement with previous observations by stopped-flow (Andreasson, 1975; this study). Sodium dithionite is widely used as a reductant in biological systems (Lambeth and Palmer, 1973; Mayhew, 1978). This reductant has also been used to study reduction of the oxidase (Lemberg and Mansley, 1965; Orii, 1979). Sodium dithionite was used in the present study in scanning experiments and was valuable in providing information and in discriminating between the two hemes of the oxidase, since it is the only negatively charged reductant known to reduce the oxidase which does not require another electron mediator. The implication of this property will be discussed throughout this study. However, because of the complex redox chemistry of dithionite and the appar- ently variable (and small)molar absorptivity (6.2 mM'1 cm"1 at 320 nm [Mayhew, 1978]), it is difficult to use the dithionite absorption band to quantify the electron transfer. Metal ion redox couples have also been used for the study of the oxidase reduction. Greenwood et a1. (1977) have used hexaaquo- chromium (II) as the reductant. Scott and Gray (1980) used ruthenium (II) hexaamine. Both of these studies showed that these positively charged reductants preferentially reduce cytochrome a.of the oxidase in the first phase followed by slow electron redistribution. Reduced nucleotide adenine dinucleotide (NADH) coupled to N-methyl phenazinium methyl sulfate (MPMS) is also widely used. When 29 mixed anaerobically, these compounds form 5, 10-dihydro 5-methy1 phenazine (MPH), which is the reducing species. (See Chapter III). The difficulty with using this couple with excess NADH to cycle MPMS between the oxidized and reduced forms is that the reaction between NADH and MPMS is,itself,rather slow (Halaka, Babcock and Dye, 1981b) so that fast reduction processes involving MPH and the oxidized oxidase are obscured. MPH is the reductant used most extensively in this study. It was pre-formed by the anaerobic titration of MPMS by NADH. A sharp absorption band of MPMS at 388 nm (not present in MPH) (Halaka et al., 1981b) can be used to monitor the number of electrons transferred (see Chapter 111). CHAPTER II EXPERIMENTAL METHODS A. Materials Beef hearts were obtained fresh from Michigan State University Meat Lab. Sodium dithionite was a Virginia Smelting Co. product. Cytochrome a (Type 111 or Type VI), NADH, MPMS, EPES (see Chapter III), N-2-hydroxy ethyl piperazine N-2-ethanesulfonic acid (HEPES), and the detergents: octyl phenoxy polyethoxy ethanol (Triton X-lOO), poly- ethoxy ethylene sorbitan monolaureate (Tween 20), and cholic acid were purchased from Sigma Chemical Co. The detergents Triton X-114 and Tween 20 were kept refrigerated as 20% (v/v) solutions in water. Cholic acid was purified by recrystallizing from 95% ethanol. The crystals were mixed with equivalent amounts Titrant 12 cn1 123cn1 11— Figure 11.1.--Cells used for preparing anaerobic solution (A) and for anaerobic titrations and preparation of air-sensitive reductants, (B). (a) Kontes high-vacuum valves, (b) Fischer-Porter joint, used to connect cell to vacuum line, argon gas, or to mount cell onto the stopped- flow apparatus, (c) quartz l-cm path-length optical cell. 35 spectrophotometer to cover the cell shown in Figure II.lb during titrations. Spectra of MPH prepared by the above method were checked before mounting onto the stopped-flow system. No changes in these spectra were detected for periods as long as 24 hours. When- ever MPMS or MPH were used, the container was wrapped in aluminum foil, as was the stopped-flow buret which conveyed the MPMS or MPH solutions to the pushing syringes (see Chapter 111). C. The Stgpped-Flow System A double-bema, vacuum—tight, rapid-scanning stopped-flow apparatus was used throughout this study (Papadakis et al., 1975; Coolen et al., 1975; Suelter et al., 1975). Detailed descriptions of this system are given in the Ph.D. dissertations ofN. Papadakis (description of the flow system, glassware, and mechanics of the system) and R. B. Coolen (computer interfacing, synchronization of scans, and signal averaging schemes in both scanning and fixed-wave- length modes). The system was operated under positive pressure (~1 psig) of purified argon gas instead of under vacuum (vapor pressure of solu— tions). This was a necessary modification for two reasons: 1. Protein solutions are not stable under reduced pres- sures, and in the presence of detergents, as most of the solutions in this study contained detergents, solu- tions foam and create bubbles 2. Operating under positive pressure has the advantage that, if there is a slow leak, air will not enter the 36 apparatus. This is particularly important for experi- ments requiring long times. 0.1 Handling and Performance of the Stopped-Flow System Since the system is all-glass, it always required special care in handling and operation. Solution handling and mounting of reagents were carried out under argon. After all solutions (except protein) had been mounted, the system was evacuated to < 0.001 mm Hg and flushed with argon. This cycle was repeated at least four times. After the last cycle, while the apparatus was still under vacuum, water was delivered to the observation cell (to avoid bubbles). Argon was then introduced to 1 psig positive pressure throughout the entire run. When the protein was ready to be pushed, the protein solution was mounted onto the system. The buret delivering the protein solution, which can be isolated from the rest of apparatus by Kontes valves, was evacuated and flushed with argon separately several time to insure anaerobicity. The performance of the stopped-flow system was checked peri- odically by various tests, some of which will be discussed below. C.l.l Flow velocity and flow profile.--In stopped-flow experi- ments, the flow of solutions should be fast enough to cause turbulence for efficient mixing. However, pushing the solution too fast may cause cavitation. Cavitation occurs when the external pressure on the liquid is less than its vapor pressure. An important parameter, 37 which gives information about the kind of flow is the Reynolds Number, R: R = d v p/c II.2 Where d is the diameter of the tube, v is the average velocity of the liquid, p is the density, and g is the viscosity. Empirically, it has been found that turbulent flow occurs at r = 2000 - 3000 for fully regular tubes. However, it can be as low as 10 for jet-type mixers (see Wiskind and Berger, 1964). The mixer used in this study is a four-jet one (Papadakis et al., 1975). Figure 11.2 shows the velocity profile of the solution flow. With Equation 11.2, the critical velocity to get turbulent flow for water at 20° C is given by v = 200/d, where d is the mm. Since 2 mm diameter tubes were used for constructing the mixing and observation cell, the critical velocity is about 100 cm/s. The velocities shown in Figure 11.2 are then high enough to insure turbulent flow. By using an equation given in the reference mentioned in this section, cavitation should be expected at velocity around 1000 cm/s for a 2 mm tube. Experimentally;cavitation should be easy to detect from the response of the photomultiplier tubes. C.l.2 Mixing efficiengy.--Besides using a four-jet mixer and measuring the flow velocity, the mixing efficiency was tested by a specifically-designed reaction. A non-buffered acid solution of the acid-base indicator methyl orange was pushed against a base to give a final pH = 4.5, which is enough to give the color of the basic form 38 650- .... T’. 8 600- U) E; 13 5C1 ' '1 p51 ‘9 5555()" ti 3 :47; >1 .2: U 2 500- O) :> 8 450. 2 £2 F’814C1psi o 400- . 1* " 1, 350 - 3000.0 50 40 60 8'0 10b :20 I40 time, ms Figure 11.2.--Flow-velocity profile as a function of pressure. The plateau region of uniform velocity indicates that data could be collected at any point in this plateau. 39 of the indicator. A non-buffered solution is used in this test because if the final solution were strongly buffered, it might give the final color change even if the solutions were not efficiently mixed. Since protonation reactions of acid-base indicators are diffusion-controlled, the acid form in that experiment should go to the basic form during the dead time of the instrument (~7 ms), and slow changes should not be observed if the mixing were complete. Figure 11.3 presents a fixed-wavelength push of such a test (at 465 nm). The absorbance change at this wavelength, as measured on a Cary 17 spectrophotometer, was 0.62. Since no significant change in the absorbtion was observed, it was concluded that mixing was complete before the solution reached the observation point. C.l.3 Anaerobicity tests.--The anaerobicity of the stopped- flow apparatus was tested by several reactions that are air-sensitive. One such test was pushing MPH against buffer and collecting spectra as a function of time. In the presence of oxygen, the oxidized form, MPMS, which has a sharp peak at 388 nm.is produced (see Chapter 111). This test proved to be useful and economical,since MPH was the reductant in most of the reactions mentioned in this text. Another test that had been used earlier, when sodium dithionite was used as the reductant, was pushing reduced 1umof1avin-3-acetate against buffer. The reduced form of the reagent is colorless, while the oxidized form is yellow (peak at 460 nm). In each case, the system showed anaerobicity for periods as long as 36 hours. 4O . T' [0 [Sll'lJCXM .. . 1 c) 1p“. 0.00 0 0 -00. 0 - 00 0 .00 00 00 00 00 00 00 0 -.- C--. O. C -‘C. . I C C O. C O O .0 I O C 0 IO... .0 b) 8 g 08 0000000000000000000000oo00.00000000-00 e 0.7 o 0.6 B | 1 1 1 1 1 T 1 1 j I <1 0 100 zoo 300 400 500 INTEL nns 0.75 - a) 0.50 - 0.25 *- CK)‘ 1 1 l 1 1 1 1 1 134K) 2M5C> ‘4CXD ‘44ND 445M) £52K) wavelength, nm Figure 11.3.--Test of mixing efficiency in the stopped-flow apparatus. (a) Spectral changes that occurred upon mixing of the acid-base indicator methyl orange (at pH = 2.5) with base to give final pH of 4.2, (b) time progress curve of the absorbance at 465 nm collected in fixed-wavelength mode, (c) same as in b, but enlarged 200 times. 41 The anaerobic reduction of cytochrome oxidase, in addition, is a very sensitive measure of any oxygen present. Concentrations of oxygen as low as a fraction of micromolar cause a steady state to be observed in the reduction. The steady state portion of the curve becomes more pronouned at higher oxygen concentrations. 0. Data Handling, Computations and Computer Graphics Data acquisition was done by a PDP8/I computer interfaced to the stopped-flow system. The stored data were displayed on a Tektronix model 610 storage-display scope connected to a Tektronix model 4601 hard-copy unit, which was used for preliminary examination and production of figures,such as that presented in Figure 11.3. Data were transferred to Michigan State University CDC 750 computer, where kinetic and principal component analsyes were per- formed. Correction for finite scan times and conversion of channel numbers to wavelength were also performed as described in Appendix E of the Ph.D. dissertation of R. Cochran. Computer graphics of the type shown throughout this work were accomplished by using program MULPLT, written by Dr. T. Atkinson at Michigan State University for the Chemistry Department PDP-ll Computer. Computer programs to transfer calibrated stopped-flow data to be used for MULPLT from the CDC 750 computer to the POP-11 were written in collaboration with Dr. T. H. Pierce. Data representing wavelength vs. absorbance (see, for example, Figure 111.5) were subjected to a smoothening spline. However, the original data were displayed along with the "smoothened" data to assure that the spline smoothening did 42 not affect the shapes of the curves (see Figure V1.8). Three dimen- sional plots were constructed by program GEOSYS from the Michigan State University HAL routines. CHAPTER III PROPERTIES OF 5-METHYL PHENAZINIUM METHYLSULFATE: REACTION OF THE OXIDIZED FORM WITH NADH AND OF THE REDUCED FORM WITH OXYGEN A. Introduction 5-methyl phenazinium methyl sulfate (MPMS), often referred to as "phenazine methosulfate," is widely used as an electron-transfer catalyst in a variety of biological reactions and as a redox buffer in potentiometric titrations of protein-bound, oxidation-reduction cofactors. For example, MPMS has been used in the study of photo- phosphorylation (Zaugg et al., 1964; Vernon et al., 1963), and in redox reactions involving the mitochondrial electron transport chain (Kearney and Singer, 1956; Low and Vallin, 1963; Singer and Kearney, 1957; Dutton and Wilson, 1974) and with other biological systems (Dearman et al., 1974; Nishikimi et al., 1972; Ohno et al., 1975). The spectral and redox properties of MPMS are, however, fairly complex in that both protonation reactions and semiquinone formation can occur and that these processes are often significant in the physiological pH range. Zaugg (1964) studied the spectral proper- ties of the S-methyl phenazinium ion, MP+, of the two-electron reduced species, MPH, and of the intermediate semiquinonoid forms, + + MPH‘ and MP' (see Figure 111.1) as a function of pH. While MPH' 43 44 CH3 .4 7. N 3 8 ”2 ‘9 MD 1 MP’ (3113 m rs N ”N ,5 MPH m N <9 MPH‘ -+ - . - [013050; .MPMS Figure 111.l.--Structure of N—methyl phenzaine at different protona- tion and oxidation levels. 45 is stable below pH = 3.5, MP' and MPH? apparently disproportionate to MP+ and MPH at neutral and higher pH values. For the deprotonation of MPHT, pKa values of 6.8 (Rao and Hayon, 1976) and 5.7 (Rubaszewska and Grabowski, 1975) have been reported. However, these are apparent pKa values since both deprotonation and disproportionation could be involved. It is well known that MPMS is photosensitive (McIlwain, 1937; Marzotko et al., 1973; Rubszewska and Grabowski, 1975). In recent comprehensive photochemical studies (Chew and Bolton, 1980; Chew et al., 1980), electron paramagnetic resonance (EPR) and optical methods were used to determine a number of rate and equilibrium con- stants. They showed that irradiation at pH below 7 with excitation wavelength below 500 nm yields two products: the 5-methyl lO-hydro- phenazine cation radical (MPH?) and 1-hydroxy-5-methyl phenazinium cation, pyocyanine (PYH+) in a stoichiometric ratio 2:1. At higher pH values which favor disproportionation, photolysis would thus ultimately lead to the formation of equimolar amounts of PYH+ and MPH. Because of problems with photolability of MPMS in enzyme assays, Ghosh and Quayle (1979), suggested the use of 5-ethyl phenazinium ethylsulfate (EPES) as the electron mediator. They based their conclusion on the non-enzymatic reaction of MPMS and EPES with 2,6-dichloro indophenol and on EPR measurements. Other authors (Hisada and Yagi, 1979; Nakamura et al., 1980) have suggested the use of l-methoxy-S-methyl phenazinium methyl sulfate. The reduced nicotinamide adenine dinucleotide (NADH)/MPMS couple is often used as the reducing substrate in the study of 46 biological redox reactions. Generally, NADH is present in substrate amounts and MPMS serves to mediate electron transfer between NADH and the oxidizing cofactor (Nakano et al., 1975; Bergmeyer and Bernt, 1974). In two independent recent studies (Farrington et al., 1980; Anderson, 1980) NADH was shown to be essentially a two-electron donor, and that the one-electron transfer from NADH will only occur with oxidants for which E0. is more positive than about 300 mV. Thus MPMS with a reduction potential of 80 mV at neutral pH (Jagendorf and Margulies, 1960), is likely to serve as a two electron shuttle in most assay systems. Our interest in the NADH-MPMS system involved its use as the reductant couple in the study of aerobic (steady state) and anaerobic reduction of cytochrome a oxidase (Halaka et a1, 1981). In order to understand the mechanism of this reaction and other reactions in which the NADH-MPMS couple are used, we studied the reaction of NADH with MPMS both anaerobically and in the presence of oxygen as well as the reaction of MPH with oxygen. These studies made use of scanning and fixed-wavelength stopped-flow spectrophotometry. We also report the spectral changes which follow anaerobic and aerobic photolysis of MPMS and EPES. B. Experimental Section MPMS, EPES and NADH were from Sigma Chemical Co. Unless otherwise mentioned, all solutions were prepared in 0.05 14 N-2- hydroxy ethyl piperazine N-2-ethane sulfonic acid (HEPES) buffer, pH = 7.4, using glass-distilled water. Argon gas was purified by 47 passing it through a one meter BASF catalyst column heated to 115°C. Time-independent spectra and spectrophotometric equilibrium titra- tions were measured with a Cary 17 spectrophotometer equipped with a temperature controller. Equilibrium anaerobic titrations were per- formed in the cell shown in Figure II.lb. The same cell was also used for the preparation of MPH for stopped-flow studies. During anaerobic titrations the cell was wrapped with aluminum foil to prevent light- induced decomposition and the solution was kept under a positive argon pressure of ~3 psig. Photochemical decomposition of MPMS or EPES during st0pped-flow experiments was prevented by turning off the room lights and wrapping the MPMS (or EPES) burets and containers with aluminum foil. An anaerobic rapid-scanning double-beam computerized stopped- flow spectrophotometer (Coolen et al., 1975; Papadakis et al., 1975; Suelter et al., 1975) was used to collect up to 150 spectra/s over the desired wavelength region. Alternatively, it could be used in a fixed-wavelength mode if desired. Anaerobicity of the stopped-flow apparatus was achieved by at least three evacuations (<0.001 torr) followed by filling with purified argon. The system was kept under positive argon pressure (~l psig) during the anaerobic experiments. The anaerobicity was checked by mixing MPH with anaerobic buffer in the stopped-flow system. Since MPH reacts with oxygen to give MPMS, which has a characteristic sharp peak at 388 nm (extinction coeffi— 1 cient = 26.3 mM' - cm'], Zaugg, 1964), the presence of oxygen could have been easily detected. Since no change in absorbance at 388 nm 48 after mixing MPH solution with buffer was detected over a time span of many minutes, we conclude that the oxygen concentration in the buffer was less than 0.3 pM. C. Results and Discussion C.l Anaerobic Titration and Trans- ient Kinetics of the Reduction ofF MPMS by NADH Figure 111.2 shows the spectral changes which occurred in a 7 ml solution of 15.9 uM MPMS following the addition of 0.026 ml increments of 0.527 mM NADH. The isosbestic point at 358 nm and the uniformity of the relative absorbance changes at all wavelengths suggest that there are no detectable intermediates formed when MP+ is reduced to MPH (Equation 111.1). + + MP + NADH -———————+ MPH + NAD 111.1 A detailed examination of the spectra shows that even at the midpoint of the equilibrium titration, one-electron intermediates account for less than about 7% of the total starting concentration at this pH (7.4). The absence of such intermediates was further supported by studying the kinetics of the reaction represented by Equation 111.1. Figure III.3 displays the spectral changes which resulted when MPMS was rapidly mixed with NADH anaerobically at pH 7.4 to give final concentrations of 31.9 and 39 uM respectively. The change in absorbance at 388 nm with time (data from the same experiment) is shown in the inset to Figure 111.3. A generalized non-linear 49 Absorbance Wavelength, nm Figure III.2.--Anaerobic titration of 7 ml of 15.9 pM MPMS by successive additions of 0.026 ml increments of 0.527 mM NADH in 50 mM HEPES buffer, pH = 7.4. The cell (Figure 11.1) was wrapped with aluminum foil to prevent photodecomposition of MPMS and was kept under 1 psig argon pressure throughout the titration. Temperature = 21 i 1° C. 50 .N.HHH mesmwm cw me msmm ecu mew meowpwucou .maee mEem one Eocm :mxmu .mewu cpwz E: man we mocencomae cw mmmcmcu we» mzozm ummcH .Eo mm._ o sumcmp came PPmo .m.mm .n.wp .Nm.m .mo.e .mm.P .epo.o "mop we» seem .mcm mcwomam umuompmm one cow mucoomm cw .mmEFP .cowpowm pmucmswcmaxm on» cw nonwcummu Empmzm zopetumaqopm mew he cmpomppoo mew: ecuuoam .AmcwxPE cmumev zo3 m2az z: m._m mo composume o_aocmm:m on» mcwgzu mwmcmso _mcpomamrt.m.HHH mc:m_d arzc ctwmgvns. O¢+. Gui. ooea 10mm“ own. 10*» _ _ _ - hi n — h P . 00° 08“ .230 naz‘ P 00% L - P P h, - L h 03.06.» 2 o. o . . a: I Oofl N.— I owmw l t. 1, - .... 3 dt d1 - d1 d1 d a '71 I . d J l 1 q . _ . 01« Na . ed od ad 0.» AYIQ. x “thexy.nxzcacu>s= oBOCooc< 51 least-squares program (Dye and Nicely, 1971) was used to analyze the rate data. When a second order rate equation was fitted to these data, it gave a fairly well-determined second order rate constant. However, the residuals (calculated-observed absorbances) varied sys- tematically. Recalling that MP+ forms a complex with MPH (Zaugg, 1964) a rate equation was derived from the following scheme: + k1 + MP + NADH-————————+ MPH + NAD 111.1 MPH + MP+ K2 (MP- MPH)+ 111.2 The fit of these equations to the data gave random residuals. It was assumed that the equilibrium step (Equation 111.2) is fast, that the spectrum of (MPH MP)+ is the sum of the spectra of MP+ and MPH and that the complex does not react at an appreciable rate with 3 -1 -1 NADH. This analysis gave k] = (3.8 i 0.5) x 10 M s and K2 = 4 M-]. The constants reported here were averaged (1.3 i 0.2) x 10 from the analysis of rate data of several experiments at 388 and 340 nm. The value of K2 represents only about 14% bound complex at [MPH] = [MP+] = 15 pH. This accounts for the fairly good fit with only a simple second-order process. An alternative explanation is the removal of up to 14% of the MP+ concentration by the formation of semiquinoid intermediates according to: + K3 + MP + MPH +————;———=-* MPH' + MP' III.3 52 The data would be equally well fit by this scheme provided that neither MP“ nor MPH? react at an appreciable rate with NADH. The difficulty with this scheme is that it is known that the spectra of MP' and MPH? are different from those of MP+ and MPH (Zaugg, 1964), so that 14% of the material bound as the semiquinoid form should have been detectable. C.2 Reaction of MPH with Oxygen The reaction of MPH with oxygen, Equation 111.4, appears to be second order (first order in each reactant) k3 + _ MPH + 02 ————-——-—+ MP + H02 111.4 The spectral changes, as expected, are the opposite of those shown in Figure 111.2. A pseudo first-order rate equation gave very small and random residuals when fitted to the data collected for this reaction. Under atmospheric conditions the solubility of oxygen in water is 0.25 mM (Wilhelm et al., 1977) so that the mixture with anaerobic MPH solution had half this concentration. The second-order rate constant for the oxidation of MPH by oxygen was then calculated to be 180 M'1 5']. This is only an approximate value, since it is based upon the solubility of oxygen in pure water rather than in the buffer solution. C.3 The Reaction of MPMS with NADH in the Presence of Oxygen In the presence of oxygen, MP+ acts as a catalyst for the oxidation of NADH by oxygen. This should be kept in mind when studying 53 biological reactions that use the MPMS-NADH couple under aerobic conditions. By using the rate constants derived above for the reduction of MPMS by NADH and the oxidation of MPH by oxygen, one can predict the behavior of the NADH-MP+-02 system. The scheme repre- sented by Equations 111.1, 111.2, (or 111.3) and 111.4 can be used to describe this sytem. In a scanning experiment, digitized data are stored for the entire wavelength region spanned as a function of time. These data describe the kinetics of absorbance changes for every wavelength channel, at essentially the same time. This enables one to check the validity of a certain kinetic scheme at more than one wavelength. Figure 111.4 represents the spectral changes in an experiment in which 125 uM NADH solution reacted with 22 pM MPMS in the presence of atmospheric oxygen. The simultaneous changes at 388 nm and at 340 nm obtained from these spectra could be completely described by the scheme mentioned above. The resulting two coupled differential equa- tions which represent the rate of change of [MPMS] and [NADH] with time, were solved numerically by using the program KINFIT4 (Dye and Nicely, 1971) for the full time course shown. In this case, since all solutions were aerobic, the concentration of oxygen was taken to be 0.25 mM. The rate constants for the reduction of MPMS by NADH and for the oxidation of MPH by oxygen were adjusted. This analysis gave values for the rate constants, k], k4, (in Equations 111.1 and 111.4) 3 m"1 5'] and 160 M'] 5-], respectively, which are in of 3.6 x 10 satisfactory agreement with the values obtained for the separate reactions. 54 Wavenumber (cm"1 x 10'4) 2L0 257’ 2:5 2L3 2L1 T I 1 J 1 1 J l 1 1 1 J 4 . 1 l J 1.6'% t“\\ 1.4 .,_ 1 31.4 '. 1.2— gal -. §1.o- 20‘ . . - ' ' ' D '1 01 I «Q .1 .. 0 a 0.3— ...‘ U) 3 l ‘ _ 1 . . 0-5" " TI}. .21}. .25.; ..'. 3:7... 2...; .1... ... .1 ftp-no.1“ 0.4- A 1 . 012-“ (10% ‘ l ' 1 T 1 ' 1 T 1 P 1 i Tfi j 1 4340 1360 1380 ‘400 ‘4201 4401 14801 48G) 5KK1 5H!) Wavelength (nm) Figure 111.4.--Spectra1 changes which result from mixing NADH with MPMS in the presence of atmospheric oxygen. Con- centrations after mixing were 125 and 22 uM, respec- tively. Times, in seconds (from top to botton at 340 nm) are: 0.015, 1.47, 25.9, 48.6, 77.7, 97.1, 146, 204, 292. Inset shows the changes in absorbance at 388 (dots) and 340 nm (circles) with time from the same data. Medium is the same as in Figure 111.2. 55 0.4 Effect_ngoom Light on the §pectra of MPMS Figure 111.5 shows the spectral changes which occur when a solution of 31.2 uM aerobic MPMS solution at pH = 7.4 is exposed to room (fluorescent) light. From these spectral changes and the isosbesth2point at 345 nm, we conclude that the main, and apparently the only, product of the photodecomposition of MPMS in the presence of oxygen at this pH is pyocyanine (absorption maximum at 310 nm). This was also observed by McIlwain (1937) and by Chew and Bolton (l980) to be a major product of the irradiation of MPMS by light at pH = 7. It is worth mentioning in this context that EPES, the ethyl analog of MPMS gave nearly identical results. Not only was EPES found to be similar to MPMS in photosensitivity, but also in all the other reactions mentioned in this text. Thus the conclusion by Ghosh and Quayle (1979) that EPES is less photolabile than MPMS at pH = 9.7 cannot be extended to lower pH values (pH = 7.4 in these experiments). When anaerobic solutions of MPMS were exposed to room light, the spectral changes were different from those shown in Figure 111.5. The photodecomposition apparently produced MPH as well as PHY+. The two products were formed in nearly equal amounts since upon exposure of the final products to atmospheric oxygen, the absorbance of MPMS at 388 nm rose to about half its starting value. In addition, the absorbance of pyocyanin at 310 nm reaches only about half the value anaerobically that it attains aerobically. These observations are consistent with the scheme proposed by Chew et a1. (1980) for high pH 56 Absorbance Wavelength, nm Figure III.5.--Effect of room fluorescent light on the spectra of 2l.2 pM aerobic MPMS. Time between successive spectra was about 10 minutes. Buffer was the same as in Figure 3. 57 values as well as with the observation by Zaugg (1964) that most of the semiquinonoid (one-electron) form undergoes disproportionation at pH values between 7 and 8 in aqueous solutions. Thus the stoichiometry of the anaerobic photodecomposition of MP+ at pH = 7.4 may be represented by Equation 111.5 h 2MP+ + H20 ——"— MPH + mm+ + H+ 111.5 In the presence of oxygen, the reaction represented by Equation 4 will take place, and the photoreaction of MP+ will produce only pyocyanine, as suggested by the spectra shown in Figure 111.5. The results described here would appear to eliminate one of the possible mechanisms presented by Chew et al. (1980). Their value 4. for the disproportionation constants of MPH' and MP+ (Equations 11 and 12 of the mechanism proposed in their paper) are 2.5 x 10'8 and 6.3 x 105; respectively. If these values were correct, then at the midpoint of the reduction of 20 uM MP+ by NADH, at pH = 7, the con- centrations of MP' and MPH? would be 5 pM and 10 uM, respectively. This would lead to [MP+] = [MPH] = 2.5 uM, a drop of 75% from the radical-free case. MP+ and MPH? are known to have different spectra (Zaugg, 1964; and Chew et al., 1980) so that conversion of one of these forms to another would result in an appreciable change in the absorbance at 450 nm (where MPH? absorbs with an extinction coeffi- cient of l0 mM.1 cm'1) and at 388 nm. This would lead to different rate profiles at 450 nm (MPHT peak), 388 nm (MP+ peak) and 340 nm (NADH peak) during reduction by NADH. In fact, the rate profiles 58 showed no detectable deviation from clean stoichiometry. Therefore, the equilibirum constants for the disproportionation mechanism, presented as one of several possibilities by Chew et al. (1980) cannot be correct. 0. Conclusions The data presented here should be useful for those who use the MPMS-NADH couple in enzyme assays, particularly in the presence of oxygen. The reduction of MPMS by NADH is slow compared to the rate of many enzymatic reactions. Thus, the former reaction could be rate limiting and add complications to the enzymatic reaction being studied. The reaction of MPH with oxygen is also slow, and for many practical purposes can be ignored compared to the much higher rate of the reduction of oxygen by biological systems (for example, the rate of reaction of oxygen with reduced cytochrome_g oxidase is 8 1 about 10 m’ s" (Greenwood and Gibson, 1967). CHAPTER IV REDUCTION OF CYTOCHROME OXIDASE BY MPH Redox reactions of cytochrome g_oxidase are of central importance to the understanding of the catalytic activity of this protein in electron transport. The reduction of the protein, either under anaerobic conditions or during turnover in the presence of oxygen, has been studied extensively, particularly by using its natural substrate, cytochrome 9 (see, for example, Gibson et al., 1965; Andreasson et al., 1972; Andreasson, 1975; Wilms et al., l98l). The reduction of the oxidase has also been carried out under anaerobic conditions with the positively charged metal ion complexes hexa- aquochromium (11) (Greenwood et al., 1977) and hexaamine ruthenium (11) (Scott and Gray, 1980). It was concluded from those studies that the charge type of the reductant plays an important role in the kinetics of the reduction as well as in the site of the interaction with the oxidase. The reductant, 5, 10- dyhydro—S-methyl phenazine (MPH) was used in the present work to study the anaerobic four-electron reduc- tion of the oxidase by rapid scanning stopped-flow spectrophotometry (see Experimental). MPH was chosen because of its spectral proper- ties (see Chapter 111), because it is a neutral molecule, and because it has a low reduction potential at neutral pH (80 mV, Jagendorf and 59 60 Margulies, 1960), which is sufficiently negative to drive the four electron reduction of the oxidase. The implications of each of the properties mentioned above are discussed in the text. A. Spectral Shape Analysis The visible and near ultraviolet electronic absorption prop- erties of the MPH/MP+ couple are discussed in Chapter 111. Figure IV.l displays a three dimensional plot of the overall spectral changes from 330 to 530 nm which occur when oxidized cytochrome oxidase is mixed anaerobically with excess MPH. This wavelength region was chosen to insure the simultaneous observation of the spectral changes due to MPH, MP+ and reduced and oxidized cytochrome oxidase in the Soret region, without significant loss of wavelength resolution. When the a-band of the oxidase was studied, another wavelength region, usually ranging from 400 to 650 nm, was scanned. Owing to the fact that neither MP+ nor MPH absorb appreciably in this region, the three dimensional surface of such an experiment (Figure IV.2) has spectral contributions only from the oxidase chromo- phores. The principal changes of absorbance in Figure IV.l are the decay of the oxidized cytochromes_a and 33’ with maximum total absorbance at 418 nm, and the growth of the reduced heme band with a narrower peak at 444 nm. IIIaddition, the peak of MP+ grows in at 388 nm. Comparison of the initial and final spectra with those of MP+, MPH and oxidized and reduced cytochrome oxidase shows that the reaction follows the overall stoichiometry: 62 .F.>H mgzawm cw mm msmm mzp mom meowpwucou cmsuo .2: mm .xmz “z: mm.m .mmmuwxo "mew: mcwst Lmu$m meowumcucmocou .cowmmc e: ome - ooe msu cw Ia: x3 mmmuwxo meogsuouzu mo :o_uo:umg ownocmmcm mg» Low mumwcam gucmpm>mz mewuumucmngomnH mczm_m 255 £05.03? ova o8 own 8 .P . . _n ome owe one _0.0 . 0.0 9.0 aauoqmsqv (3) emu Nam 63 2MPH + Cytochrome Oxidase (oxidized)-————————+ 2MP+ + Cytochrome Oxidase (reduced) IV.l Figure IV.3 displays, in two dimensions, spectra selected from Figure IV.l as a function of time, as well as the spectrum of the oxidized cytochrome oxidase (against buffer, same conditions). When the spectrum of the oxidized oxidase was subtracted from the spectra shown in Figure IV.l, the resulting "kinetic difference" spectra (Figure IV.4) show a striking feature of this reaction. Two minima in the difference spectra are observed at around 410 and 430 nm and suggest that the Soret region of the oxidized oxidase consists of two bands that decay at different rates upon reduction. Further- more, these spectra suggest that during the early stages of the reduc— tion, the absorbance change at around 430 nm, where cytochrome a_has its absorbance maximum (Vanneste, 1966; Babcock et al., 1981; Halaka et al., 1981a), is greater than that at 410 nm, where cytochrome as has its absorbance maximum. In the initial fast process, the dif- ference spectra also indicate that there may be a shift (~5 nm) of the cytochrome-g3 band to a longer wavelength (or of the cytochrome a band to a shorter wavelength). The last phase of the reaction shows a greater change on the short wavelength (33) side of the Soret band. The latter effect is large enough to halt the growth in absorbance at 388 nm due to MP+ formation, as expected by comparison of the As values for MP+ and cytochrome 33’ which are approximately equal at this wavelength. The shift hitheisosbestic point from 437 to 428 nm as the reaction proceeds supports this observation. Examination of 64 .m Now .om mm m.mm .me mm mm.p .NN ”we OFF .m “we m.mp ._ “mew mgpumam mzowgm> to» mmswp .cowuuwmg o: saw: Acowumgucoocou msmmv mmmcwxo «sogcuouau umNPumxo mcu $0 yoga m? .xo .zu cmpmamp Ezcuumam och .—.>H mcam_m soc» mguumam umuummeuu.m.>H mgzmwm «EC» £93333 8' 2V on E on... 0.0 _ _ p _ _ _ _ _ _ 0.0 .6 .d «.0 «d ad ad 46.3 Qd .8 8 s - fl 5 0.0 J ad m q 3 s 3 w 3 0.0 06 a ad ad 0.. m- 0.. .a . w..; w q. 1d—qd.qqd.uq-qddq-u1dd—qdqu—uddq—dfldq-qqufiwqd no. 0.“ «.0 *8 ON 3 $19 x F55$ $083533 65 .Nmm .mm mm mm .ce mm m. .mm mm mm.. .mm "me o—. .m ”we m.mp .P "men mos.» .Am.>H mgamwm :. xo xu um—mnmpv Escuumqm mmmuwxo um~wuwxo asp mcwpumgunzm an _.>H mgzmwu 20L. umaozgpmcou ocpumgm mucmcmmmwu nmuum.mm--.v.>~ mcammu «E5 £05363 0.0 00» 2V 09 00V 0.. 00... 0h... 0w... 8.0: _ _ _ _ L _ _ _ _ _ _ L _ _ _ L 8.? etc: aid: 8.0! 8.0.! m and - and nu D # L T v 8.0 l r. 0nd 0. H m S . m 1 T 90° I q 4 I u I ' O 00.0 L n: 00.0 9 . H h 8 . q — u q u q q 4 fi 14 4 — q .— d 4 — 4 q q d — d 41d H d 4 ii 54 # q q u 4 q 11“ d — d d 8.- od «a ed 0N nu «vlo. x .IEo» LoQEacosoi 66 time courses at several wavelengths (Figure IV.5) also shows the different kinetic profiles of cytochromes a_and 93 (see below). Other features of this reaction are apparent in Figure IV.6 which displays spectra in both the Soret and the a-bands. Inspec- tion of the difference spectra in the o-band region shows that most of the changes here are completed in the early stages of the reduc- tion (see below for quantitative estimates). It is also important to note the shift in the peak position to shorter wavelength in the final spectrum, a phenomenon which, although small, is reproducible. This implies that the component that is reduced in the last stage of the reduction (cytochrome 93) has its peak around 595 nm compared to 604 nm for cytochrome a. This observation is further confirmed by the study, to be discussed in Chapter VI, of the reduction of the oxidase by sodium dithionite. 8. Kinetics of the Reduction of Cytochrome Oxidase by MPH The kinetics of the anaerobic reduction of cytochrome oxidase by MPH were studied by stopped-flow spectrophotometry in scanning and fixed-wavelength modes. Although scanning experiments performed with the stopped—flow apparatus described in Chapter II produce data with high signal to noise ratio and could be used to obtain rate parameters, kinetic analyses were usually done on data collected in a fixed-wavelength mode. This is due to the fact that, in a fixed- wavelength experiment, it is possible to collect more data points for the particular wavelength under study, which owing to computer memory limitations, cannot be obtained in scanning experiments. 67 .» w>F omam . new o. An uwpmwzm mew x vcm a go. mmqum we.“ as. .—.>H mg:m.m.c..wzo;mmmcuwmmm mcu so». :mxmu .Aev E: mwm new on Npe .Axv ome um mucmngomnm .o mucmccmamc we..--.m.>H mgszm doom» oENe own owN o¢m oom om. om. om ow AV — _ _ . — _ _ p _ . p . _ _ _ _ _ p _ _ — _ _ _ _ _ _ _ — _ _ _ — moo eouquOSQV x x x 68 .Nmm .mm.mm .epm.o .emm.o .opp.o .e.o.o ”men .5: com um mop cu Eouuon soc. mucoomm :. mgpumam mo mmswh .gumcm.m>m3 gwpgosm m 00 Ezgpumnm awn. ms» c. xmmaua mg» .o pwwsm mg» muoz .~.>H mg=m_u c. czogm mpmu mgu soc. cmxmp .mmmuwxo oEogsoouzu gawk :m: .o cowpummg mg“ c. mguumam umpum.mm--.m.>~ mesmwm «E5 50:336.; 000 000 0N0 00¢ 0.1V 00¢ 00.0 . p F p . p _ _ p p — p _ L _ p b . — . . . 0.0 I N0 00.0 - e 1 I. .V.O UV no 1 an n OF.O I an” a I. 0 O J .nru . a. o .. w s 9.0 .. .. m8 9 b 3 A l I. T 0.. 0N0 1 . ., .. a; ”No6 — . q d _ 1 . q — . q q _ a .1. 1 _ q . q 0.. h... 0.. .N ”N 0N «v10. x .183 cmnEzco>o§ 69 Scanning experiments, besides providing rate information, are used in spectral shape and principal component analyses. The reproduci- bility in fixed-wavelength experiments was such that the same rate constants to within 1% could be obtained in different experiments. Non-linear least squares fitting was carried out by using program KINFIT4 (a modified version of program KINFIT, Dye and Nicely, l971) and showed that the anaerobic reduction of the oxidase by MPH is triphasic. This was concluded from analysis of absorbances at 444 nm as well as several other wavelengths. The initial phase is best described by a second order reaction between MPH and the oxidase. This is followed by two first order processes, as discussed below. 8.1 Analysis of the Fast Phase A generalized second order rate equation (Equation IV.6) was used to fit second order processes, and is derived by considering that any second order reaction can be represented by nL + M-————£——~+ products where n is a stoichiometric ratio. If [Lo] and [Mo] are the initial concentrations of L and M, respectively, then the rate of this reaction can be represented as {1% = k[Mo] (R-nx)(1-x) IV.2 where x is the extent of the reaction, k is the rate constant and R is given by 70 Integrating Equation IV.2, leads, for R f n, to 53:21- = (R-n)k[MO]t +1n R IV.3 ln 1 Solved for x equation IV.3 becomes e'k't - l x = . IV.4 fl-e'k t - l R where k' = k (R-n) [M0]. Recall that for a stoichiometric reaction, x, in absorbance form is given by At ' Ao x = --———--—-— 1v.5 Am - A0 where At’ A0, and A00 are the absorbances at time t, zero, and infinity, respectively. Substituting x from Equation IV.5 into Equation IV.4 gives -k't _ l - e ‘8 The term (A0° - A0) corresponds to the absorbance changes for a par- ticular phase and can be adjusted as a parameter to give the extent of that phase. Figure IV.7 displays a typical fit of the second order equation to the initial phase of the reduction (in a fixed-wavelength 70 Integrating Equation IV.2, leads, for R f n, to R-nx _ 1n 1-x - (R-n)k[Mo]t + ln R IV.3 Solved for x equation IV.3 becomes x _ e-k't _ 1 ' n -k't fie -] IV.4 where k' = k (R-n) [Mo]' Recall that for a stoichiometric reaction, x, in absorbance form is given by At ' Ao x = -———-———- 1v.5 Am - AO where At’ A0, and Aw are the absorbances at time t, zero, and infinity, respectively. Substituting x from Equation IV.5 into Equation IV.4 gives -k't e-k't _ l - A - A0 + (Am - A0) 1 IV.6 zflzcb The term (Aco - A0) corresponds to the absorbance changes for a par- ticular phase and can be adjusted as a parameter to give the extent of that phase. Figure IV.7 displays a typical fit of the second order equation to the initial phase of the reduction (in a fixed-wavelength 71 0.7% m 0.6- U C _ D f 0.5— o A a) s .0 << 0A.— 0-3 I ' I I ‘ I T I O 50 100 150 200 T'irn e (rn s.) x _ x x x A 10 N) - x X B O h “_ X X x x x x \v _‘ X X x X 2 00- xx x x x x x x ‘ x 0 x X x x x xx x x “ x 0 xxx x b .w_ XX 0- x X g - x x 0: x '80- ] I I Y I I I 7 I 0 U in W a Thne (ms) Figure IV.7.--(A) Fit of the fast phase for the reduction of cytochrome oxidase by MPH. X's are the experimental data points and the solid line is the calculated curve. Data collected in a fixed-wavelength mode. Concentrations after mixing were: Oxidase, 3.l pM and MPH, 19 pM. (B) residuals of the fit shown in (A). Residuals are defined as A (calculated)--A (observed). Other conditions are the same as in Figure IV.l. 72 experiment at 444 nm). This phase was easily analyzed by simply choosing data points covering about the first 200 ms of the reaction, although analysis of the full time course, consisting of a second order phase followed by two first order processes was also done. Analysis of the full-time course becomes easier when MPH concentration is high compared to that of the oxidase (pseudo-first order condi- tions), where a three exponential equation representing the three processes was used. The rates of the three processes are sufficiently different so as to allow their easy separation. Note that the residu- als in Figure IV.7b are small and random which is an indiction that the equation used describes the data. Table IV.l summarizes results of the analysis of the fast phase of the reduction. It is worth mentioning here that the analysis gave, for absorbance changes at 444 nm, a delta absorbance equal to about 50% of the total change, while at 605 nm, the contribution of the fast phase was about 80% at this wavelength, in agreement with Scott and Gray (1980) on stopped-flow results obtained on the reduction of the oxidase by hexaamine ruthenium (II). This implies that cytochrome a contributes a major portion of the overall enzyme absorbance in the 605 nm region. 8.1.1 Kinetics of the fast phase at 830 nm.--The kinetics of the reduction of the oxidase by MPH in the near I.R. region were studied by an experiment that was designed to follow the anaerobic absorbance changes at 830 and 605 nm. The kinetics of the growth at 605 nm showed the expected second order behavior in the fast phase, 73 TABLE IV.l -- Analysis of the fast phase of the reduction of cytochrome oxidase by MPH in HEPES buffer containing 0.5% Tween 20, pH = 7.4 at 21 i 1°C. k1 is the second order rate constant for the process. [aa3], pM [MPH], pM ki X 10’? M-1 -1 11.2 30.0 2.8 i 0.3a’0 3.7 23.0 3.6 1 0.3a’d 3.3 23.0 4.1 i 0.4b’d 2.59 23.0 3.2 i 0.4"”1 1.1 8.4 2.9 i 0.4b’d Avg = 3.2 i 0.5 Analysis done on a fixed-wavelength experiment. Scanning experiment. Wavelength of analysis, 605 nm. RQB‘D' Analysis at 444 nm. 74 with essentially the same second order rate constant as that deter- mined at 444 nm (see Table IV.l). However, the rate of the 830 nm band decay, under identical conditions, showed a detectable lag (Figure IV.8). Although only one set of concentrations was used, this lag was reproducible. A simple scheme that accounts for the observed lag at 830 nm and agrees with the kinetic measurements at 605 nm is presented below f kl 2+ 2+ 4. MPH + [Cyt 33+ . . 012+] ———-—» MPH' [Cyt a. . . Cua ] (A) (B) kf 2 ————. MP+ [Cyt 32+. . Cug] IV.7 (C) The observed rate at 605 nm would be proportional to the formation of species B and C (defined in Equation IV.7) since both of them have reduced cytochrome 3, Also this rate is second order, as discussed above. We here make the assumption thatlfldfl'would transfer an elec- tron fast so that, in effect, it will keep the heme reduced while the latter transfers electrons to its associated copper. The concentration of B + C at time t (measured at 605 nm) would then be given by [LOJII-e'k't) -k't I-IELOJ - [Move ([B] + 1C1)t = IV.8 75 0-0 0-1 0-2 0-3 0-4 0.5 0.6 l J l l l 1 l L 1 1 l . 0.55— . . . ° ° °—o.12 a a; 0 ° . 0 . > E ' 0 ° 9 c 1:. i . . . . . 3 m 0.50 " If" 0 o o o _ 0.11 g D " ”’3 .0 ® I- O to II. °o° a V 0-45 "‘ ‘x o . o o ” .1 0": ’— 0010 . 0 0° ,5. " A C o xx Q 2 0.40 — 0° ’. . u 0 x .. - .° ~ 009 ° 0 00 ”six . b 3 2 0.35 ._. : xx'xxn“ ‘ 3 < _ O x xx x ‘ , \— 00 "‘x",x"""l1_0.08 0.30 I I 1 . I I I I I I I I O 0 01 0 2 0-3 0 4 O 5 O 6 0-12 '- Q) 0 0-11 - C o .. -Q L 0-10 - 0 m .1 ~Q < 0.09 — 0.08 . . . . , I l , 0.0 0.1 0.2 OS Tfinie (sec) Figure IV.8.--(A) The decay at 830 nm and 605 nm during the reduction of 10.8 uM cytochrome £93 by 27 pH MPH. (8) Fit of Equation IV.8 to the data at 830 nm, x's are experimental points and the solid line is the calculated curve. Other conditions are the same as in Figure IV.l 76 where [Lo] and [MD] are the initial concentrations of MPH and oxidase, respectively, and k' = k1 ([Lo] - [Mo]). At 830 nm, the rate of change of absorbance is pr0portiona1 to the production of species C, whose rate is given by 95%1 = k; ((131 +021) -'[c1) IV.9 We then substitute the term ([8] + [C]) by its value from Equation IV.8. Equation IV.9 was solved numerically and kg was adjusted. The fit (see Figure IV.8) gave k; a value of 17.8 i 0.5 5-]. 8.2 Kinetics of the Slow Processes The slow phase of the reduction is best fitted by two first order processes which are independent of MPH concentration (Halaka et al., 1981a). This is not in agreement with the suggestion of Scott and Gray (1980) that the slow phase is described by a single second order process (in enzyme). Their interpretation came about because of the better fit of their data for this phase to a second order equation than to a single first-order process. It is not sur- prising that two first order processes can be better fitted by a second order process than by a single exponential. A fit of the second order equation to the slow phase gave slightly less systematic resi- duals (Figure IV.9a, 10a) than a fit by a single exponential (Figure IV.9b, 10b). However, a fit of the same data by two first order processes gave the smallest and most random residuals (Figure IV.9c, 10c). 77 FIT 0r SLOW PHASE cam Influential) 1.1 -I ’(q Absorbanco 8 j I I I I I I I I 0 ‘0 20 w ‘0 50 w 70 I) 00 Time (sec) FIT 0F SLOW PHASE Second Order) 1 '1’ .“A—r‘ 1,1 -- .1 0 1 l"" I 2 . 3"". o I .r O 10 -‘ ' ‘ I o . 1 I: a I < I 09 J ——W_-1""fi'—_’Y-"'T‘"—Tm_1‘—_l_—_1—-" O 10 2O 30 4O 50 60 7C '0 .0 T i m e t s e- c ) FIT 0F SLOW PHASE fl-o EWUO’” r". ’ o If,” u c '- 0 a 10 .. h o C: U) a < 0.9 ~ I I I I I 1 1 O ‘0 20 30 ‘0 50 0° 70 w ’0 m Ti 9 (sec) Figure IV.9.--Fit of the "slow phase" to one exponential (a); To a second order process,(b); and to two exponentials (c). Conditions are the same as in Figure IV.7. 78 RES/DUAL PLOT OF SLOW PHASE (One Exponential) 4o — I A I V) ' .I l 0 2.0 - I III I u 2 I I I x V "I . a 00 d I "III I ‘ ' ' I 3 g - ' 0 l l .‘3 4° — . . a a , a. . 1 e 1 a {I 4.0 ~ p h A A A i A A A 6‘ TIM E t S E C ) RESIDUAL PLOT OF SLOW PHASE (Second Order Ecuation) h - 201 n o 1 "‘ 1.0 ~ . . . k s. 1 . b Residuals 8 l l 7T Y T ' I T 40 50 60 70 O W TIME (SEC) 3 3 84 RESIDUAL PLOT OF SLOW PHASE (T no E xponen Mom 10" to“ i c 3 as— - - ' . x :l' ' I. u I a ' " ~ .II I ' ' I I 1 ll ' n 00" H‘I"l‘. l I ‘ l' ' I I I 3 w -. . 2 .05-1" " I I U) l I . I 0 I t -10- Y T T I 7 7 rfiifi‘ TIME (SEC) Figure IV.lO.--Residual plots for the fit to the data in Figure IV.9. (a) One exponential, (b) Second order, and (c) Three exponentials. 79 In addition to inspecting the randomness of the residuals, the distinction between intermolecular (second order) and intra- molecular (first order) electron transfer in the slow phase was tested by varying the enzyme concentration. Over more than a 4-fold range of protein concentration, a fit of the slow phase by a second order equation did not show the expected dependence of the rate on the pro- tein concentration, whereas a two-exponential fit gave essentially the same first order rate constants over that range. Table IV.2 summarizes results of this study of the kinetics of the slow phase. C. Reduction of the Cyanide-Bound Cytochrome Oxidase As mentioned in Chapter I, ligands such as HCN react with cytochrome-g3 of the oxidase (and possibly with the copper associated with it; (Van Gelder and Muijsers, 1966). When bound to cyanide, the peak of the oxidized cytochromeg3 shifts from 414 to 428 nm in the Soret region, the result is a narrower absorption band for the oxidized CN- oxidase which peaks at 429 nm with an extinction coeffi- cient greater than that of the native oxidized protein-(180 mM'1 cm-1 for the CN- bound oxidase at 429 nm compared to 160 mM'1 cm'1 at 420 nm for the native protein). C.l Spectral Shape Analysis Figure IV.ll displays the spectral changes on mixing 2.4 uM CN- oxidase with 6.5 uM MPH (concentration after mixing). Note that the reaction is fast and that the reduced Soret peak of the oxidase does not grow to the same absorbance as in the case with the native protein, which is due to the fact that when complexed with HCN, 80 o.HH mooo. H oomo. moo. H mom. o.mm oH.H m.m mooo. H mono. moo. H ohH. o.mm oo.~ N.m vooo. H ooHo. H.o H mH.o v.o om.m m.m Nooo.o H MHmo.o moo.o H moH.o o.mm on.m m z oH x x o H m .mx 0 H m .Hx mm: mom HI HI on HI m HI m IumUHOIncooomII IIIIIIIIII mHoHucmcomxm 038 IIIIIII IIIIZ: .GOHumuucmo:OUIIII .ooH H Hm I a .e.e n ma .om :0039 wm.o mcHCHmucoo Hommsn mmmmm :H mm: wn ommcon mEounooumo mo eoHoosoou 030 «0 eomoeo sone 0:0 co moHosbm oHoocHx mo suoessm I- m.>H mamas 81 .F.>H mtzowu :. mm 02mm asp mew meowpwocoo nguo .m.- .Pm._ .F..o .Fm.o .opo.o ”men 2: ome pm soppon on no“ sot. .mocoowm c. .mmewh .Aocwxwe twp.m mcowpmepcmucouv :a: z: m.o xa mmmowxouzo z: e.m .o :owuozomt on“ go umppsmmc was“ mmocmzo Potpomam--....>H «gnome the. cwocooosok om» omv omv on» o.v oon ohm Dob — . — p _y p h h _ » L1 _ coo T. ‘73 ° aouoq?osqv ° ‘0 0.0 - _ H a q — . a a u q H H ..w 7w n.~ AH «vac. x .I=:: LooEaco>o§ Ins xm mmH :oHuosom scum .xe .:oHHMH>ov oumccmum HMCHmHmSo o.m om. H o.~H m.mv o.v moo o.~ omo.o H mv.oH m.mv m.v vow m.m ANmo.o H oH.H m.o v.~ omv Hum H12 .AmIOonx ehoH .x 2: Mai e: ._mm£_ ._ mm_ .numcme>mz .mucoEHHomxm numcmHm>m3Ioome CH vmuomHHoo one oboe .ooH H Hm Ho e.e n we .Hommsn mmmmm 2 mod eH mm: so ommcon mEounoouuo UGSOQIoUHcmmo 03H 00 coHuoswwH 03H H00 ucmumcoo wumm II m.>H mqmfle 85 The data presented here are distinct from those reported by Scott and Gray (l980) on reduction of the CN-oxidase complex by ruthenium (II) hexaamine, in which they observed biphasic kinetics. A possible resolution of this apparent discrepancy may lie in the fact that [Ru(NH3)6]2+ is a positively charged one-electron donor, so that addition of two electrons to the CN-bound oxidase may require the dissociation of an electrostatically formed complex between [Ru(NH3)6]3+ and the negatively charged cytochrome a_site. Such a situation could lead to biphasic kinetics. An alternative explana- tion for the observed biphasicity would be the outer sphere electron transfer from [Ru(NH3)6]2+ to the "bound" [Ru(NH3)6]3+. MPH, on the other hand, is a two electron donor (See Chapter III), and even in the case of a one-electron radical formation, it probably remains bound to the cytochrome a_site for a second electron addition after the redox equilibration between cytochrome a_and its associated copper; alternatively the free radical may donate directly to the Cua. D. Conclusions MPH, a neutral molecule, as well as positively charged metal ion complexes such as [Cr(H20)6]2+ (Greenwood et al., l977) and [Ru(NH3)6]2+ (Scott and Gray, 1980), reduce preferentially the cytochrome a_site of the oxidase. These reductants can thus be used as artificial donors that imitate the biological substrate, cytochrome c, For kinetic studies, particularly for those in which the scanning stopped-flow technique is used, MPH has an advantage over the above mentioned metal ion complexes. As MPH oxidizes, MP+ forms with a 86 characteristic peak at 388 nm, which has a suitable extinction coefficient, so that absorbance changes at this wavelength can be used to monitor the extent of the reduction as well as to measure the concentration of oxygen, if present. The ability of MPH to react with oxygen is important in testing for anaerobicity and in studying turnover experiments, where one can quantitatively measure the number of turnovers. Our data on kinetics at the near I.R. region (830 nm) show that the Cu component lags in the anaerobic reduction with respect to the cytochrome a iron. These data agree with observations by Wilson et al. (1975) who showed that, during the reduction of the oxidase by cytochrome c in the presence of oxygen, the bleaching of the 830 nm band either lagged or coincided with the growth at 605 nm, depending on the reductant concentration. Our data are also in agree- ment with rapid-freeze EPR experiments (Hartzell et al., l975) which show rapid equilibration between cytochrome a_and its associated copper, Cu. The present work, however, appears to be in disagreement with experiments on the anaerobic reduction of the oxidase by cytochrome c (Andreasson et al., 1972), in which the 830 nm decay did not show a detectable lag relative to the 605 nm growth. Since the second order rate constant for the reduction of the oxidase (cytochrome a) by cytochrome.g(~l08 M'1 5'], Gibson et al., 1965) is much faster than 5 l -l that by MPH (3 x lO M- s , present work), it appears likely that approximately one-half of the cytochrome a had been reduced during the 87 dead time in the reduction by cytochrome g_(~5 ms under the conditions described by Andreasson et al., l972). This is comparable to the dead time in a typical stopped-flow apparatus. Extrapolating the absorb- ance at 830 nm to zero-time would then obscure any lag which could have occurred, and one would observe a "parallel" reduction of the c0pper and iron of cytochrome a. Incidentally, these authors inter- preted the absence of a lag at 830 nm by proposing that cytochrome a must contribute to the absorbance at this region. This latter con- clusion is at odds not only with our own stopped-flow results, but also with a number of equilibrium titration experiments which have been carried out subsequent to the Andreasson et al. (1972) stopped- flow work. The consensus opinion from these experiments is that Cua is the principal absorber at 830 nm and that there is negligible con- tribution to this feature from either cytochrome a_or‘aa. The presence of three distinct phases in the anaerobic reduc- tion of the oxidase by MPH can be interpreted in more than one way. The fast initial phase, which is a second order process, accounts for about 50% of the total absorbance changes at 444 nm and 80% of the change at 605 nm. These observations, along with the fact that on binding to cyanide, the rate constant of the fast phase remains essentially unchanged, suggest that the fast phase monitored at 444 or 605 nm can be assigned to the reduction of the cytochrome a_site and its associated copper by MPH. The second phase, a first order process with a rate constant of about 0.2 s.1 contributed only about 10% of the total changes at 444 nm, with the third phase accounting 88 for the rest of the changes. One way to interpret these observations is by the scheme represented below (Equations IV.ll - IV.13), which takes into account the negative cooperativity between the two iron centers of the oxidase (Leigh et al., l974; Babcock et al., 1978). Another explanation of the observed multiphasic kinetics is that it may be due to heterogeneity in the enzyme preparation. In a recent study, Brudvig et al. (198l) have shown that the oxidized oxidase can exist in at least three conformations, depending on the method of preparation used and whether the protein was subjected to any catalytic cycles. 89 +3U . mF.>H m . +3U m A as 2 w 3 NP.>H + u m . . . l..u%u 6:0 +mm +~ w . o . m..u.>.u FP.>H +30 +N m . . l.. %0 mzu . +mm H +N N . . l.. ho Nx + 0:0 . +Nm p w +N m collo>.U Amzu. muv m..p»o +¢_ +N N .. l.>.U mx mzu . +Nw H m..psu + +N mm . . . ma..bsu :u +m . m..b»uv +N F .. s mx mau . . . +mm p u . m-.H%U + +N a: ma . . . m..p»u .Illuuu +N=u +m + In: In: _ a . . . m..p»o c; :0 +m CHAPTER V PRINCIPAL COMPONENT ANALYSIS OF THE REDUCTION OF CYTOCHROME §_OXIDASE BY MPH A. Introduction Optical absorption spectroscopy has been of fundamental importance in the characterization of cytochrome g_oxidase (see Lemberg, l969). The protein spectrum has peaks in the near UV, visible, and near IR regions that are sensitive to the oxidation and ligation states of its four metal centers. The absorption spectra of cytochromes a_and 33 of the oxidase overlap strongly in the Soret and a-bands. Also, in these two regions absorbance changes due to the oxidase copper centers may be obscured owing to the relatively small extinction coefficient expected for c0pper complexes compared to those of the heme moieties (Malmstrom, 1970). The classical 9-33 picture of the oxidase, including assignments of wavelength and extinction coefficients for the two hemes (Vanneste, l966) has been rather widely accepted. It was argued (Caughey et al., l976), however, that it is unwise to synthesize spectra of cytochromes §_and.§3 on the assumption that the properties of one heme are not affected by changes in the oxidation or ligation states of the other metal components, since there is strong evidence for electron and magnetic interactions between the metal components of the oxidase (Hartzell et al., l973; Babcock et al., l978). 90 91 We have used data obtained by rapid scanning stopped-flow spectrophotometry to resolve the individual spectra of cytochromes a andg3 by the method of principal component analysis (PCA). In the scanning wavelength experiments described here, a selected wavelength region is rapidly and repeatedly scanned as a function of the time after mixing of the "scan." The result is a p x N data matrix A composed of N consecutive spectra (essentially instantaneous if the time of scan is short compared to the half-time of the reactions studied), measured at p wavelength channels. There are good reasons for sampling this three dimensional space of absorbance-wavelength- time in the study of the kinetics of cytochrome oxidase instead of the two-dimensional space obtained in fixed-wavelength experiments; they are the need to characterize the kinetics of the strongly inter- acting components of the oxidase and to separate the contributions to the spectra of species with overlapping absorption bands. 8. The Method of Principal Component Analysis (PCA) Data in the form of the matrix A, described above, are suit- able for the application of principal component analysis (PCA), also known as principal factor analysis (PFA) (see, for example, Malinow- ski and Howery, l980). Although PCA is used for a wide range of phy- sical techniques (Sylvestre et al., 1974; Bulmer and Shurvell, 1975; Valasdi, l974; Ritter et al., 1976), only its application to scanning stopped-flow data is briefly discussed here. The method is discussed in detail in the Ph.D. dissertation of R. Cochran (1977, MSU; see also Cochran and Horne, l977; l980; Cochran et al., l980). A block diagram 92 of the main steps in factor analysis is shown in Figure V.l. PCA, if applied to suitably weighted data, can give the number of components (absorbers) in the system and, in favorable cases, can provide com- plete resolution of the absorbers' spectra and their time courses. The weighting scheme is discussed in Section B.l.l. PCA does not require any mechanistic assumptions. The only assumption required is that the absorbance at every wavelength channel be a linear func- tion of the concentration of the absorbing species (Beer's law). In other words, the elementlkkj(absorbance at wavelength channel i at time of scan j) of the matrix A is represented as MD f V.l ij ‘ k=1 ik ij where fik is the molar absorptivity (times the path length of the absorbance cell) of absorber k at wavelenth channel i, ckj is the molarity of k at time of scan j, and q is the number of absorbers. Accordingly, the matrix A_can be "factored out" into two matrices, one of which describes the spectral shapes of the absorbers (E matrix), while the other contains information on the concentration as a function of time, (§_matrix); Equation v.2 v.2 |3> II I'm In II H MD where_[ is a (p x q) matrix defined by [_= (jH’ jé, . . ., fq). The vector :d’ called the static spectrum of absorber j, is a p 93 .AmcoPCMUHCwooe now: mommp .xgmzo: ucm Fxmzocwpmz Eocwv mwmapmcm gouum; cw mamum :wms Low Emcmmwu xuopmnu._.> mgamwm 3.40 262 .302 sum *3 3:2.» a 32.22:“. _ mFZmZOQSOu 4 mg:m_m cw mm mamm as» mew: mcowpwucou gmsao .Imz :1 mm an A2: mm.mv ammuwxo as» $0 Ecow umpmcmmxxo asp we cowpuzumg ownocwmcm as» Low mummgzm mumv mewuimucmngomnwispmcwFm>m3 _mgcmswcmaxm--.F_.> wgzmwd 2.5 595.303 80 08 08 our. 00.? 01?? 1F _ _ _ . . aouquosqv 117 Figure v.12.-—Reconstructed surface for the data presented in Figure V.ll using two eigenvectors. 118 Figure V.l3.--Reconstructed surface for the data in Figure V.ll using three eigenvectors. 119 \\_‘ H “ 1% -..~_ ‘23 x 1 fi._&,w. -" /;’,' \‘\-"‘— ”A‘ .- '1 \‘I\ I x59, “" ‘ @\ Figure V.l4.--Residuals (A(2) - A) of the data presented in Figure V.ll. 120 Figure v.15.--Residuals (A Figure V.ll. - A) of the data presented in (3) 121 function for the fourth eigenvector was 0.67, indicating that r = 3. S-analysis of the same data indicated that there are only two com- ponents that independently change their concentrations with time, r5 = 2. This, again, was concluded from examination of the recon- structed absorbance surfaces and residuals. This value of rS was further confirmed by the values of Qr/(N—r)(p-r) for rS = 1, 2 and 3, which were 19, 1.8 and 0.7, respectively, indicating rS 2. The target absorbers for the fitting of the eigenvectors in this case (M-analysis) are four: The oxidized and reduced cytochromes A_and.33. C.2.2 Spectra of individual absorbers.-—Unlike the case dis- cussed for the shorter wavelength setting (Section C.l.l), the fit to the oxidized cytochrome g3 (from Vanneste, 1966) was unsatisfactory. This is explainable, however, by postulating that cytochrome_§§+ in the "oxygenated" enzyme must have its peak shifted closer to that of cytochrome 33+. This is supported by the fact that the "oxygenated" enzyme has a combined peak for cytochromesg;+ and-g3+ at 424 nm com- pared to about 418 nm in the resting enzyme. The oxidized cytochrome A_spectrum was found to fit as one of the M-analysis eigenvectors (Figure v.16). It should be pointed out, however, that this really represents both oxidized cytochromes (combined). The other two suspected absorbers (reduced cytochromes g and 33) were found to fit to the M-analysis eigenvectors. The reduced forms of the two cytochromes were resolvable in this case because of the better wavelength resolution in the Soret region, as well as the 122 .Ezguowam cmpmswamm we“ Anv .Ezgpomgm mo comeEmem mgoaom>cmmwm omega any mun m.o1w:m mpcwoa ummoaocg mew mew m.x .mmmvwxo umumcmmxxo on“ Low Eaguumam a van m mmsogcoopxo umNPquo nmcwnsoo we“ mo “we mwmapmcmiz Amvii.oF.> mezmwm «E:~ :.m:m.m>o§ ova can can and. an» ovv c0» Coo ‘POP n P — p p p — p p — — P.p L — p n P — F p p cog . I o o o o o o o o o o 0 90.000.00.00 ...| . v N o 1 . 1 N cow 3 . 1 I *.o v o 1 . .\ 1 m bog I. . .039... H. beam. ad 1 . . r «em 1 o i a a; 1 .. 9.. # u a o n o o p o o - one-anonun i o .1 u u n u 1 . v N a 1 . 1 N onv . S 1 1 l V.c we 1 . K 1 m . . c l . O. n a 11 he... 1 m cm «6 .1 .___. .. 1 ads 1 no .1 a Q.“ 1 .e 0.. 1 Ia . — q A . _ q u q _ q q q — 4 a J N.“ m.~ ..N M.N n.N «vie. x fi1E3 canacm>o§ bo~=oEcoZ < 190 Qmobmmm 1.5 ZOESSHMM <01 123 contribution of the a-band. It is important to note that reduced cytochrome A (Figure v.17) has a much higher absorption at around 600 nm than does reduced cytochrome g3 (Figure v.18). This has been suspected for some time, but the present data provide experimental proof that the contribution of heme g in this region strongly pre- dominates. C.2.3 Concentration profiles.--The concentration-time pro- files of the three absorbers are in general agreement with those obtained for the shorter wavelength setting. Again, cytochrome g§+ reduction lags the reduction of cytochrome §?+, again confirming that MPH reduces gé+ first (Figure v.19). Note that in this wavelength setting, we have the concentration-time profile for the appearance a2+ of cytochromes and 23: which can be compared to those of the dis- appearance of the oxidized forms of the two cytochromes in the shorter wavelength setting. Once again, we can conclude unequivocally that 3+ a . the fast phase represents reduction of cytochrome In spite of strongly overlapping bands, PCA has provided both wavelength and time resolution of the contributions from the separate chromophores. 0. Conclusions The work presented here, although starting with known or postu- lated spectra1 shapes of the oxidase components, provides resolution of the independent spectra1 shapes of the components of the protein. We have not only proven that the oxidase has two different cytochromes, and given a "hands-off" estimation of their spectra1 shapes, but also 124 .1 .sacpomam nmpms_pmm Any mucwca umums_pmm um.o .mucwoa umgmamca um.x . m msogzoouau vacuum; mo Ezguomgm as» on Hem mmmzpmcmiz mgouom>=mmwm omegh Anvii.Np.> «gnaw; «ES £32233: Owe cam Own own Qm§ 0%? 60% DUO .P.P -l p — n h 1- — P P - — b P p. — n n h _ h b - eta c l coo-co .111 o v we 1 . .\ 1 N oq Ta 1 .. r You 1 o l J O I. . o O! Q Q 1 11 b GO «6 L .1 Row. 1 o o i 9 Oefi If. . o. T. °.~ 1. q o o u a a a u a a a a o a a cacao-uuuunu .. «6 1 “a .1 «SW Tc 1 H. i Yam. 1 n u I J o o o o O- b O I . H. b 00 1 o a U Q.Q I . . I n.09 1 o o i. 3 Q41 .u. l 0.. q .d u —1 q d d - u d ‘1 fi 1- d d — a d I1 h.“ m.» . ~.N n.N m.N «vies x _1Eu» cmnEaco>o§ . boNaoEgoZ 9 E 808% 00 293228 <8 125 .vmpmemwm c.° N.c V.Q o.n Q.c a.» N.o v.9 o.o m.c o.» m hnv + m nm~_uwxo we Eaguumam 65H 0“ “we mwmxpmcmiz mcouom>emmwm awash Amv11.mp.> mczmwd .schomam .mucwoa cmumpzupmu as» men m.o new mucwon vmgmamca asp mew m_x they c.m:u~o>a§ Owe can own own on» ovv. Dev - I P—P P I — I _ P_rrh — P hip — PPF coo hi 0.30.. .11 o. con-no-oooo-o-uoooo-u r Como .1 .. m H rl . x w “no-cue ”I ‘1 11 q - u H q d1 .1 d d —i 1 q q —d1a1 h.“ m." s.N H.N m.N «via. x hiEuy ConEzcm>a§ « bogaEcoz .omoExO bogocouxxo . A? 1. < o to Emacs 1.5 E SE 0.0 . ov N baa You 1 m.omw m.c~a 9.. . uv N Dav To“ J Bow U n.03 c.~ 126 ' TIME COURSES OF OXIDASE COMPONENTS (Oxygenated. Wavelength Region 400-650 nm ) l Concentration (normalized) 1 al.-.- D II t.- H j I 1 I . 1 ' 1 . l 1 1 1 1 4O 80 120 160 200 240 280 Time (sec) 1.0 — a . I t s 0.8 _ ‘ . . U . . . . 3:" ‘A ... . k ‘ . 5°, 0'6 — ‘AI .A _ S ' u a :5 0-4 -I e. “‘ E “ co - . ‘ ‘ U . A A 5 A . s - . U 0‘2 - c c ‘ - c c_ c .. f: ¢c“ oo-— 3 49“ 1 1 fi fi 1 1 1 1 1| 0 1 2 3 5 Time (sec) Figure V.19.--Concentration-time profile for the growth of reduced cytochrome g, (B); reduced cytochrome 33 (C); and for the disappearance of the combined oxidized (333), (A). Bottom figure is the first 5 seconds of the reaction. 127 resolved their separate reduction-time profiles by MPH. This work shows that principal component analysis isaipowerful tool toward the goal of separating strongly overlapping spectral components, which show different temporal characteristics. From data obtained in the long wavelength region (Figure V.l7) PCA shows that the reduced cytochrome A_spectrum has a shoulder in the Soret region and also a higher contribution to the a-band than does reduced cytochrome 33, in agreement with previous suggestions. The analysis in this wavelength region also showed that in the "oxy- genated" enzyme, the spectral shapes of the oxidized cytochromes g and 23 are similar, and thus become difficult to separate. The concentration-time profiles of the components of the oxidase, which were deduced from PCA, without any mechanistic assump- Igiggg, agree well with our previous assignments for the reduction of the oxidase by MPH. We have previously assigned cytochromeg‘g+ to be the primary reduction site by MPH (Halaka et al., 1981a; see also Chapter IV). That was based on difference spectral shapes during the reduction. PCA not only confirmed this observation, but indicated a clean separation between the time-courses of the reduction of a3+ and g§+. In the two wavelength regions examined, we have found the same concentration-time profiles: Cytochromeigg+ reduction lags that of cytochrome 33+. This, of course, explains the multiphasic kinetics of the growth in absorbance at 444 nm upon reduction because growth at this wavelength results from contributions from Q93A_cytochromes. In the M-analysis for each of the two wavelength regions studies here, three independent absorbers were the minimum number of 128 components that allowed reproduction of the absolute experimental data surfaces. S-analysis gave only two components that change concen- trations independently with time. For the short wavelength setting the M-analysis components were interpreted to be the reduced peak of both cytochromes A_andig3 (combined, + MPMS) and oxidized cytochromes A_and 33. Those for the long wavelength setting, when "oxygenated" protein was used, were interpreted as the shapes of the separate reduced cytochromes A_and g3 and a combined shape for the two oxidized cytochromes. The rS = 2 results can be interpreted by a simplified scheme for both cases: For example, in the long wavelength region, there is the condition that the disappearancecfl’the (combined) oxidized peak must appear in either the reduced cytochromes A_or g3. A similar situation was illustrated by Cochran and Horne (1977) for the mechanism: ki A ————————+-B k2 For which they illustrated that the rank of M_= 3 and that of.§ = 2. The previously published spectra of cytochromes A_and 93 (Vanneste, 1966), which were deduced from ligand binding studies and have been questioned (Caughey et al., 1976), proved to be good approxi- mations to the spectral shapes of these components. Note that in the scanning stapped-flow data discussed here, only reductions of these chromophores takes place; there is no ligation at either site. The 129 power of the method of PCA is exemplified by the fact that it gives the spectral shape for the whole wavelength region of the experiment, even though only a limited range of wavelength channels is used for the proposed spectra. CHAPTER VI REDUCTION OF CYTOCHROME OXIDASE BY SODIUM DITHIONITE A. Introduction Sodium dithionite (Na25204) is a widely used reducing agent in biology. The dithionite ion dissociates in aqueous solutions to give 50; (Equation V1.1). s o 2' ki 2 507 2 4 +::=+' 2 V1.1 k-1 Lambeth and Palmer (1973) studied the reduction of several biological molecules by dithionite by using stopped-flow techniques. Their findings indicate that although 3203 can act as a reductant, SD; is generally more reactive than 5202'. They reported a value of 1.4 e 0.4 x W9 M for the equilibrium constant in Equation V.1, K1 = kl/k-l from EPR measurements. The value of the rate constant for the dissociation of 5202', k], was found by the same authors to be 1.7 s']. Mayhew (1978) calculated a value for the mid-point redox poten- tial, Eo', for the 50%/H50; couple at pH = 7 and 25°C of -0.66V. The theoretical potential for the redox couple SZOZVHZSO3 at pH = 7 was estimated to be -0.386 V. Studies on the reduction by dithionite of 130 l3l metmyoglobin derivatives (Olivas et al., l977) and methemerythrin (Harrington et al., l978) showed that $0; is the reducing species for those oxidants. The fact that sodium dithionite, as $0; or 520%., carries negative charge is of importance in understanding the role of electronic charge of the reductant on the kinetics and site of the reduction of the oxidase. This is particularly true owing to the fact that the interaction between cytochrome oxidase and its physio- logical reductant, cytochrome c, is thought to be controlled by electrostatic phenomena (Wilms et al.. 1981). Sodium dithionite has been sporadically used to study the reduction of cytochrome oxidase (Lemberg and Mansley, 1965; Lemberg and Gilmour, l967). Orii (l979) recently reported scanning stopped-flow experiments on the reduction of the oxidase by dithionite in an air-saturated system. Also, dithionite was used for anaerobic reductive titration studies of the oxidase (Babcock et al., l978). B. Spectral Shape Analysis The spectral changes as a function of time on anaerobically mixing cytochrome oxidase with sodium dithionite are presented in Figure VI.l. Concentrations of the oxidase and dithionite after mixing were 3.78 and 47 uM, respectively. These spectra were selected from 57 spectra collected by using the scanning stOpped-flow system described in the experimental chapter. The spectra represent the "simultaneous" spectral changes of both the Soret and the a-bands due to the reduction of the oxidase. 132 .Nmm .N.mm .w.wm .m.m~ .N.¢ .m¢.o .Po. .o .mucoumm cw .mgm Agog may op Eouuon mg» Eocw E: ewe umv mcpumqm umpumme as» Low mucouwm cw .mmewp .uopm u h .20 mm.p u sumcmp neon Ppmo .¢.m u :a .om cmmzh xm.o mcwcwmp 1cou gmwwan mmam: 2E om :_ Amcwst gmuwm cowgmgucmucouv mpwco_;g_u Ezwwom 2: me .3 0.0 Nd v.0 0.0 O.— N.» V.» mmchxo wsogcuopxu z: mm.m mo cowpuaumg ownogmmcm on» mcwgzu mmnmnm Fmguumam11.~.fi> mesmwd A85 50:30.63 9ND 80 can own own own con 09. 0.2. o: emu 8v _ _ _ _ _ _ _ _ _ _ _ /(1 1111111. 1111-11. 1111.11.11.41 \ _ _ _ _ _ _ _ _ _ *4 OK.» 8.5 cm.— OON Os .N QNN 8N Ovd OWN 9.10. x FIE3 canaco>O3 133 On subtracting the first spectrum collected (which, due to the slowness of the reduction by dithionite, is virtually identical to the sum of the spectra of the reactants), the resulting difference spectra are displayed in Figure V1.2. These show, in contrast to the reduction by MPH (Chaper IV), that the short wavelength side of the oxidized Soret region decays first. This can be taken as an indication that the reductant in this case reacts with cytochrome-33+ faster than 3+. The shift of the "isosbestic" point it does with cytochrome a from ” 420 nm to ~ 432 nm at later times during the reduction supports this conclusion. Another interesting feature of this reaction is also apparent in Figure V1.2. Examination of spectra in the 590 - 605 nm (a-absorp- tion band) shows that in the beginning of the reduction the o-band has its peak at around 595 nm, which shifts to 604 nm at later stages of the reduction. The opposite effect, faster reduction of the gé+ site, was observed in the case of the reduction of the oxidase by MPH (Chapter IV). Therefore, we conclude thatgg+ peak in the a-region 2 is blue shifted with respect to the a + peak. C. Kinetics of the Reduction of Cytochrome by Sodium Dithionite The kinetics of the reduction of the oxidase by sodium dithio- nite were found to be triphasic. resembling the kinetics of reduction by MPH. However, since the electrons enter the _a_3(Cua ) site faster 3 3+ than they reduce the a site, the intramolecular electron flow should be in the opposite direction to that which occurs upon reduction by MPH. 134 .Nmm .opp .mo .e.~m .Fm.mp .mm.e .Fp.o "men s: eee um Eoupon soc» .umm cw .mcpumam on» Low mmswh .P.~> mgzmwu cw mm mamm use mew mcowuwvcou PP< .Amcpumam mm umcwmucou chwmwcov P.H> acumen cw czocm mumu mg» Eocw nmpumgunam mm: emuum—Fou Ezguumum 9mg?» on» .mpwcowguwu anvom An mmmuwxo wsogcuouzu wo cowuuzvmc mcp mcwgzn mgpomam mucmcmeewu nmuowpmm11.m.a> mg:m_u 255 £05.90; cum 00m 0mm 0mm ovmowmoomomvomvovwomv 00v 7 _ _ _ _ _ _ _ _ _ . . 1 v.01 000.. .. m - 1 No1 w W. 0001 ‘l -1 111 1\\/1 0.0 W HV. 1, m .- ~.o W m 1 v0 0 m - w . 100 a 0.0 1 . 1 md 2... 8.. om. 8m o_.~ 8m 8m $10. x .183 .353533 135 The conclusion that the cytochrome a_site is being reduced by dithionite, but at a slower rate, is drawn from the data presented above and from the fact that the CN- complex of the oxidase undergoes reduction by sodium dithionite with a slower rate than that observed for the native protein in the absence of the inhibitor (see below). Since the sodium dithionite concentration was much higher than that of the oxidase in all of the experiments reported here, full time courses were analyzed by using three exponentials which gave, for example, very small and random residuals for the decay at 4l0 nm (Figure V1.3). Table VI.l summarizes results on the rate constants of the reduction by sodium dithionite. Analysis of the fast phase, which is interpreted as a bimolecu- lar encounter of the reductant with the oxidase, has revealed that the actual reducing species is 50;, since the observed pseudo-first order rate constant for that phase varied linearly with the square root of the dithionite concentration (Figure V1.4). The second order rate constant for the reduction of the oxidase (33 site) by 50; was cal- 6 M.1 s']. This value was obtained K from the slope of Figure V1.4 and the equilibrium $20: +::::;:?'250; culated to be k = 3.4 i 0.8 x 10 with an equilibrium constant, K] = 1.4 1 0.4 x 10'9M (Lambeth and Palmer, l973). The rates of the slow processes showed slight increase on increasing the dithionite concentration (Table IV.l), which may be attributed to some reduction at the a_site by dithionite. No system- atic study was done on the dependence of the rate of these processes on the protein concentration. However, by analogy to the reduction by 136 ._.H> meamwe cw mm msmm men men meowmwncou emcuo PF< .mpmwpcmcoqu mmegp xn m>e=m umpmpzmpmu me» we mcwp vaom .25: Pm ucm : wu.m memz mcwst emuem cowumeucmucou .muwcowgmwu e:_com an mmmu_xo msoezmouxu mo cowuuzume me» em 5: ope pm mama mewu1mucmneomnm mg» mo a_m11.m.H> meamwe «come 0E:- omw 3N 00m on F on. on 0e 0 r _ . _ _ _ . _ . e _ _ _ _ _ _ I 00.0 I 05.0 eouquosqv I 0m.0 137 .cowumeucmmcom mpwcownwwu Ezwuom mo Home memzcm me» so F.H> anmp c_ Fe .mmmza mmme me“ we mcmumcom mane emueoiumewm ocammq me» ya mmcmucmamo11.e.H> mesmwe «z. o. x 222.5% - m. e._ m. m._ ___ o__ m m A. mmd 1 0.. 1 N._ ”I 13.14 . .s x 10. e. 1 m.— 1 ON 138 mcoo.o H mmao.o eooo.o H mmo.o m.o H mm. mn.m we mooo.o H omao.o eoo.o H omo.o m.o H He.a m>.m Hm mooo.o H oeHo.o moo.o H mmo.o m.o H m.H mn.m Ame mooo. H mwao.o mo.o H mmo.o ~.o H m.H m>.m an . m e N H .2: 2: Him x H1m e H1m . e .emmaeone .emoHcoHemHme .mamwucmcomxm owns» an mumw mo mflmwamam mzu Eoum mucmumcoo mumu HmUHOIHmHHm ©m>Hmmno mnu mum museumcoo mumm .anm n B .e.n H mm .om cmm39 wm.o maficflmucoo Hmmmsn mmmmm CH muwcoflnuflw EDH©Om >3 mmmnwxo mEOH500H>U mo coflposwmn map How mucmumcoo mumm 11 H.H> mamas 139 MPH, one might expect the two slow processes to be independent of the enzyme concentrations. D. Effect of Cholate When cholate was used as the detergent instead of Tween 20, the reduction of the oxidase by sodium dithionite was slowed consid- erably. Figure V1.5 presents the spectral changes in the wavelength range from 400 to 620 nm that occurred on anaerobically mixing cyto- chrome oxidase with sodium dithionite in HEPES buffer containing 0.5% cholate. Concentrations after mixing of the oxidase and dithionite were 0.92 uM and 3l7 pM, respectively. The spectra have qualitatively the same shape as those in Figure VI.l (where Tween 20 was present as detergent). However, these spectra show that the reduction is incom- plete, even after more than five minutes. The difference spectra (Figure V1.6) taken from Figure V1.5, show that 33 is the principal reduced species with some reduction at the a site, as indicated from the shift in the ”isosbestic" point in the Soret region. The indi- cation that cytochrome E3 is more quickly reduced than_a results from the minimum in the difference spectra (Figure V1.6), which occurs at about 410 nm, near the "classical" a§+ peak (Vanneste, 1966). This can also be seen from the spectra of Figure V1.5, where the decay on the short wavelength side of the oxidized peak (~4l0 nm) is much greater than that at the peak itself (420 nm). The rate data were analyzed in a manner similar to the case 1 when Tween 20 was used as the detergent (Section VI.C). The analysis was based upon program KINFIT4 (Dye and Nicely, 1971) and showed that 140 .Nmm .o.me .m.mp .N.e .eé .me.o .em.o .Po.o ”mew .mucoumm cw .5: eee am no» my Eoupon Eoee .mmewh .om :mmzh mo nmmHmcw Hammempmc ms» mm Hemmmea mm: mumposm am.o Hazy Hammxm P.H> mezmwe cw mmogu mm mamm meH memz mcowumccou .Amcwxwe emummv muwcomsuwc Ezwuom z: NFm cue: mmmu_xomwmsoe;mouzm z: mm.o we cowummme msp mcweau meaumam umpmmpmm11.m.H> meamwe A55 59.226: 8. 8» 8n 8... or» own 8n 3.. 8.. 2... on. 8.. 8.0 . _ _ _ . e . _ . _ _ _ _ e . e . e . e . 86 86 1 1 36 1F 1 , 1 e.- 26 1 . . 1 26 A I. 26 1 1 26 1 e 9.3 1 I 86 1 e and 1 1 and 80° 1 — q - q - u _ u - ‘1 — J ‘ d - 1 30° 3 3 3 3 ma .N «.w 3 3H 3 neloe x p153 £5263 «E0 m< H.396 gag» mmonO mEoEootG m> mtcoEtQ 141 .wa .o.we .m.N_ .~.e .e.F .me.o men 5: eee um now on Eouuoa Eoemv mucommm cw «meek .meuumnm Hcmzcmmnzm eoee mmummemnzm mm: umHmmppou EzeHumam Hmewe mzp .m.H> meamwe cw czogm mumu mEmm mew eee; meummam mucmemmwwu cmuumpmm11.o.m> mezmwe e5:.§€12§: 2o 8... men on» on» 2» 8.. 9? on» on. 2.. on» 0.1? — p _ e — p —1 e — b — e — p p p _ e _ . — _ Ob.°- 1. 3.°l. m T. 00.0 m m 1.an 1.ofim q - a — 1- ‘ 1 — u — u _ d - d 1‘ -d e; 9. 9. ad nu «a mu ed ad Avlop x 7.53 $350.63 assatmqu.83£mazeaE& mmoExO mete-230:0 m> mtcoectm 142 the time dependence of the absorbance at 410 nm and 444 nm is biphasic. A two exponential fit to the data at 4l0 nm (Figure V1.7) gave two 1 1 first order rate constants of 0.38 i 0.04 s' and 0.0l34 1 0.0004 s' . Similar analysis at 444 nm (Figure V1.7) gave rate constants of 0.42 i 1 l 0.05 s' and 0.0l2l i 0.0003 s' . The rate constant for the first phase, a pseudo first order rate constant, and the equilibrium con- stant for dithionite disproportionation (Equation VI.l) were used to calculate the second order rate constant for the reduction of the oxi- dase (as) by $02. The second order rate constant was found to be 5.3 i 0.5 x 105 M45"1 when cholate was used as the solubilizing agent 6 1 1 compared to a value of 3.5 i 0.8 x l0 M' s- when Tween 20 was present as the detergent. E. Reduction of the Cyanide-Bound Cytochrome Oxidase byiDithionite Figure V1.8 displays spectral changes as a function of time when 2.75 pH cytochrome oxidase solution was mixed anaerobically with 350 pM sodium dithionite (concentrations after mixing). The major changes in the absorbance are those due to the reduction of cytochrome 23+. The oxidized cytochrome a_band (maximum at 430 nm) decays as the reduced 1 absorbance band (at 444 nm) grows in. The reduced a- band also grows in at 605 nm due to the reduction. Fig- ure V1.9 shows kinetic difference spectra, in the wavelength range 370 - 510 nm, under conditions similar to those of Figure IV.8. Analysis of the absorbance-time data at 444 nm and 430 nm shows, in contrast to the reduction of the cyanide-bound enzyme by MPH which followed monophasic kinetics, by the criteria of small and random 143 .m.H> me:m_e cw czocm Hcmswemaxm mcwccmum mEmm me» see» mums .mm>e=u umpmpzmpmu pmwucmcoaxm oz» mew mmcwp nepom we: eee Hm m.o .E: ope pm mama men m.x .Amcwx_s emuemv muwcowsuwu Ezwuom x3 mmmuexo memesuouxm mo cowuuaume ownoemmcm mg» eoe E: eee new ope Hm mpmu mappimmcmneomnm mzu oH HF» _c_pcmcoquiozh11.N.H> mezmwe «come 05:. omN 0¢N 00m 00 F 0.0. p on 0* 0 . — p b p p p b 1? — p — p — L — m P.o 1 0N0 .. 1 NWO I end eouquosqv .- .I 0N.O 0N0 ' 1 .8928 8 3296 .E: 23.2. .o “emcee HEQEE xm mmedexo msomloo-Zo m0 ZQRQDQMQ 144 .mpewoe emeoum _Hzmua me. me. m.1 .Nmm .me. .m_e .m.wm .m.~. .m.e .e.. .me.o .Po.o .mvcommm :. .mem s: eee Hm aoH op :ouuon Eoe. mmewe .50 mm._ 1 :umcm. span _qu .ee.e 1 zav om came. em.o newewaocou mmam: zs om mm: emwmzm .Amc.x.s emH.mv mp.:o.;p.u E=.uom z: omm xn mmmu.xo mEoegmopxu canonimo.:mxm z: m..~ .o co.pm:cme meg no me.“ .o co.Hmc:. m we mmmcmgm .memmmam11.m..> mezm.e ¢5¢.§E12§z 000 000 000 000 0..» own 000 00% 00v 0: one 00v °.° P p — p _ p — e — e b e _ e _ e h _ — e P . 0.0 N .0 v.0 0.0 0.0 q — d ‘ d - d .- d - d — 1- ‘ d — d 0.. 5. 0.. 0.. 0d. ..N NN MN QN 0N :10. x .153 395.8263 «225850 mmoExO 0:96 m> meecoEtQ 145 .Nmm .om mwe. .om mode .Ne mm.m..om mm.e .mm mS.o .. "mew mucommm :. .mme.. .mepmmam Hcmacmmnzm soe. cmpumeuazm mm: umpmm..om Eaeuumam Hme.. me. .Am:.x.s empmmv ma.:o.;p.u 5:.uom z: omm an mmmu.xo msoegmoHAU eczon mu.cmxm z: m..~ mo co.Hu:ume mgu m:.e=u co.mme Hmeom meu :. meummqm mm:mem...u m.um:.e umHmm.mm11.m..> me:m.e ¢§¢.£r!2§3 9.» can one one on» new one ovv one ome oev cm» can can asn n61 _ e e _ _ _ _ r1 _ p e _ _ _ n61 «6| .61 0.0 ..0 N6 as H _ _ _ e _ . n6 0“ . .N NN HN QN 0N ON 5 .N Avlo. x .1335 3.32.26.» 032235 mmoExO 0:96 m> 9:22.05 146 residuals, that the reduction by dithionite is biphasic. A two exponential fit for the data at 444 nm gave two observed first order rate constants (Figure v1.10) of 0.21 s 0.04 s" and 0.0176 1 0.006 s']. Treating the first of these as a pseudo first order rate con- stant as in Sections VI. 8 and C gave a second order rate constant for the reduction of the cyanide-bound cytochrome oxidase by $0; of 5 l -l 2.9 x 10 M' s F. Discussion The anareobic reduction of cytochrome oxidase by sodium dithio— niheprovides several interesting insights into the spectral and electron transfer properties of the oxidase. The data presented here show that dithionite, reacting as the negatively charged 50;, reduces the cytochrome 33+ site faster than it does the; site under our condi- tions. This can be viewed as resembling the reactions of ligands such as cyanide, formate or oxygen with thea3 site. It can be argued that there is some "steric hindrance" in the accessibility to the cytochrome as site, since, judging from the redox potentials of the 3+ a3 (-0.66V, Mayhew, l978) one would expect this reaction to be much Fe /Fe§; couple (~ 0.35V, Babcock et al., l978) and the SOS/H503 faster. This argument is supported by the fact that the neutral mole- cule MPH, reacts primarily at the a_site. This is also supported by the fact that MPH is a bigger molecule than 50;. It is known that positively charged donors react mainly at the cytochrome a site of the oxidase. Examples are the reduction by cytochrome_c__2+ (cytochromegz+ carries a net positive charge at neutral 147 .w..> mezm.e :. mm mEmm mga mew mco.H.ucou .mm>e:m nmum.=m.mu mew mmc.. u..om use .mueu E: ome mew m.o .5: eee mew m.x ..Amc.x.s emu.mv mu.co.;u.c 5:.uom 2: 00m an mmmu.xo ecsonimu.:mxu z: m..m .o co.um:cme meg eo. E: ome mam eee um Hume m2.p-mmcmneomnm ms» .0 0.. m.m.p:mcoaxm1oz.11.o..0> me:m.e «come 0E... 0N0 00w 0eN 00m 00 . 0N . 00 0e 0 _ . e . _ . _ . _ . _ . e . e . e . 1 010 1 00.0 1 00.0 1 00.0 1 1 00.0 1 000 aouoq103qv 1 05.0 1 00.0 E: 00v 0:0 eee .0 E .0055qu 9... 00200000 .0 00008120 .00 20000000 148 pH; Koppenol et al., l974), ruthenium (II) hexaamine (Scott and Gray, l980), and hexaaquochromium (11) (Greenwood et al., 1977). Combina- tion of this fact with the present work, which shows that MPH also reacts with the cytochrome a while 50; reacts at thea3 site suggests that cytochrome a has some net negative charge near the active site and is more accessible than cytochrome a Any reaction of $05 with _3. the cytochrome a site could then be viewed as largely electrostatic, and can be accelerated by increasing the ionic strength of the medium, since both reactants carry negative charges. Thus, Lemberg and Mansley (l965) who worked at higher ionic strength (approximately 0.2 M compared with about 0.025 M in the present work) and with high lipid content cytochrome oxidase, observed that cytochrome a reacts with dithionite faster than cytochrome a_3 does. It may also be concluded that this presumed negative charge on the a site is the reason that many negatively charged donors are not effective as reducing agents for the oxidase in the absence of a mediator. Examples include ascorbate and NADH (both are negatively charged at neutral pH). A report on the reduction of the oxidase by hydrated electrons also shows there is no appreciable reduction in the absence of cytochrome .a. (Van Buuren et al., l974). The spectral shapes during the reduction are in agreement with the hypothesis mentioned above and confirm the classical assignments of the a_and a_3 bands (Vanneste, 1966). The spectra also show com- plementary results to the reduction by MPH, in that the reduced 01- band of cytochrome a_3 has a peak at 595 compared to 604 for the reduced-a. 149 The observed multiphasicity in the reduction of the oxidase by dithionite (triphasic when Tween 20 was present as the detergent) can be interpreted as a consequence of intramolecular electron transfer. This may be complicated, however, by direct reduction at the cyto- chrome a site. It is interesting to note that the two intramolecular 1 and 0.013 s") rate constants for the reduction by dithionite (0.03 s- are slightly smaller than those observed in the case of the reduction by MPH (0.l9 and 0.02 5.1). No attempt is being made, however, to determine equilibrium constants for separate intramolecular steps, since there is no evidence that electrons follow the same intermediate steps. However, by comparing the intramolecular rate constants for the two cases, one can argue that these equilibrium constants are not far from unity. In the presence of cholate as the detergent instead of Tween ; 20, the reaction becomes slower and involves mainly the reduction of the-a3 site. Cholate is known to inhibit the activity of the protein (as measured by activity assay) (see Experimental chapter). The pres- ent results on the reduction by dithionite suggest that cholate may hinder the intramolecular electron transfer froma3 to a, The reduction of the CN-bound oxidase by dithionite was found to be biphasic in contrast to the case of the reduction of the complex by MPH where it was found to be monophasic. Thre are two possible explanations for this biphasicity. (l) 50; is a one electron electron donor, unlike MPH which is essentially a two electron donor (see Chapter 111), so that addition of two electrons to the oxidase may be 150 biphasic. This was observed with another one electron donor reduction of the CN-complex using Ru(NH3)62+ (Scott and Gray, l980); (2) There might have been some uncomplexed a§+. This is less likely, since the same procedure of preparing the CN-oxidase was followed in both cases. CHAPTER VII SUGGESTIONS FOR FUTURE WORK The work presented here has shown the usefulness of artificial electron donors in the study of the reduction of cytochrome a_oxidase. MPH, in particular, has proven to be a useful reductant because of its low redox potential and spectral properties. Extension of this work and application to other systems are feasible in the following areas: A. Effect of Detergents The kind of detergent used to maintain the protein in solution is found to be extremely important in determining the stability and the activity of the oxidase. There are suggestions that the state of aggregation of the protein affects its activity (Robinson and Capaldi, l977). Rosevear et al. (l980) have shown that the detergent lauryl maltoside is very effective in keeping a uniform (mono-disperse) solu- tion of the protein, while retaining high activity. It is, in particu- lar, the state of homogeneity of the cytochrome g oxidase solution that is important in the study of the kinetics. A preliminary experiment was done (in collaboration with Zexia Barnes) on the reduction of oxidase solution prepared in lauryl maltoside by MPH. Although the kinetics of the fast phase seem similar to those discussed in the text for solutions prepared in Tween 20, analysis of the full time 151 152 course is not complete. Study of the detergent effects are important in determining if the state of aggregation of the oxidase plays a role in its inter- and intra-molecular electron transfer. 8. Partial Reduction 0f Cytochrome c.0xidise The work presented in this text involved only the fall reduction of cytochrome a oxidase (4 e/molecule of oxidase). A study of the anaerobic partial reduction (less than 4 electrons) should be very helpful in understanding the electron redistribution among the protein metal centers. Comparing the transient state spectra to those obtained under equilibrium conditions (Babcock et al., l978) should reveal possible electron pathways through the protein. Preliminary work on partial reduction (particularly 2 e/oxidase molecule) produced some interesting results. We have found that with only two electrons, there are still slow phases of electron redistribution. Careful analysis of the extent of these phases and the exact counting of electrons (which could be obtained by comparing final "kinetic” spectra to those obtained under equilibrium titrations) have not been done. C. Aerobic Experiment Although data on many "aerobic" experiments on the reduction of the oxidase by MPH were obtained by accident or by design, only the anaerobic kinetics were studied in detail. Data from experiments with oxygen present should have an enormous amount of information about the 'rates of the intra-molecular electron transfer in the protein. 153 Specifically, when the MPH concentration is higher than that of oxygen, the continuation of the reduction of the oxidase after oxygen exhaus- tion should reveal information on the effect of "oxygenation" (directly following catalytic cycles) on the intramolecular electron transfer. These are to be compared with the rates in the strictly anaerobic cases discussed in the present work. It has been suggested that the oxidized form can exist in three (or more) different conformations (see, for example, Brudvig et al., l981). One of these forms is the “oxygenated" oxidase (Okunuki et al., l958), which is formed when the reduced protein is subjected to reoxidation. The experiments (when oxygen is present) can then provide an answer to another question: Does this conformational change occur simultaneously upon reoxidation, or is it a slow process? The answer to this question is to be found by comparing the experiments mentioned above to those of anaerobic experiments on previously prepared (anaerobic) "oxygenated" protein. 0. Cytochromea552 Cytochrome 9552: flavocytochromea552 is a heme protein that has a molecular weight of about 72,000 D and contains two moles of heme and one of flavin (FAD) per mole of enzyme (Bartsch and Kamen, l960). The protein is believed to play a role in bacterial H25 oxidation (Dickerson and Timkovich, l975; Strekas, 1976). A number of EPR and Raman studies have appeared on the properties of this protein (Ondrias, l979; 0ndrias et al., l980). An important observation from the limited kinetic studies on this enzyme (Vorkink, l972) is that the rates of the reduction of the hemes anadifferent from that of the 154 flavin. A preliminary study of the anaerobic reduction of the pro- tein with sodium dithionite revealed that the reduction is fast, with t% ~ 50 ms when dithionite is present in slight excess. This t1 appears to be smaller than that reported earlier (Vorkink, l972) for the same system. A study of this reduction at lower temperatures (to make use of the scanning mode) should be useful in resolving spectral shapes of the hemes and the flavin. Also, the use of other reductants, especially those of biological interest such as sulfide, may prove helpful in understanding the role of this protein. Flavo- cytochrome is analogous to cytochrome a oxidase in that it 2552 accumulates a total of four reducing equilvalents, shows evidence of interaction between redox centers and can react with exogeneous ligands, such as CO. E. TheaApplication of Principal Component Analysis to Other Systems In Chapter V, a scheme for the statistical weighting of prin- cipal component analysis (PCA) was discussed. The application of PCA to resolve the spectral shapes and time courses of the strongly interacting chromophores of cytochrome a_oxidase proves the power and the applicability of this method. The application of PCA to other enzymatic systems--particularly heme proteins--may prove to be useful in understanding the spectral and kinetic properties of these enzymes. PCA, when supplied with reasonable target spectral shapes or concentration profiles, can completely resolve scanning stopped-flow kinetic experiments. 155 REFERENCES REFERENCES Adar, F., and Yonetani, T. (1978). Biochem, Biophys. Acta, §Q2, 80. Anderson, R. F. (1980). Biochem. Biophys. Acta, gag, 277. Andreasson, L. E. (1975). Eur. J. Biochem., 52, 591. Andreasson, L. E., Malmstrom, B. G., Stromberg, C., and Vanngard, T. (1972) FEBS lett., 22, 297. Antonini, E., Brunori, M. Colosimo, A., Greenwood, C., and Wilson, M.T. (1977). Proc. Natl. Acad. Sci. U.S.A., 24, 3128. Antonini, E., Brunori, M., Greenwood, C., and Wilson, M. T. (1973). Biochem, Soc. Trans. 1, 34. Assa, R., Albracht, S. P. J., Falk, K. E., Lanne, B., and Vanngard, T. (1976). Biochem. Biophys. Acta 422, 260. Awasthi, Y. G., Chuang, T. F., Keenan, T. W., and Crane, F. L. (1971). Biochim. Biophys. Acta gag, 42. Babcock, G. T., Callahan, P. M., 0ndrias, M. R., and Salmeen, I. (1981). Biochemistry 29, 959. Babcock, G. T., and Salmeen, I. (1979). Biochemistry 12, 2493. Babcock, G. T., Vickery, L. E., and Palmer, G. (1978). J. Biol. Chem. Q, 2400. Babcock, G. T., Vickery, L. E., and Palmer, G. (1976). J. Biol. Chem. 251, 7907. Bartsch, R. G., and Kamen, M. D. (1960). J. Biol. Chem. 225, 825. Beinert, H., Shaw, R. W., Hansen, R. E., and Hartzell, C. R. (1980). Biochim. Biophys., Acta 521, 458. Beinert, H., and Shaw, R. W. (1977). Biochim. Biophys. Acta 442, 121. Bellman, R. (1970). Introduction to Matrix Analysis, 2nd ed., McGraw-Hill, New York. 156 157 Bergmeyer, U. H., and Bernt, E. (1974). in Methods of Enzymatic Analysis (Bergmeyer, U.H., Ed.) Vol. 2, pp. 579-582, Academic Press, New York. Birchmeier, W., Kohler, C. E., and Schatz, G. (1976). Proc. Natl. Acad. Sci. 0.5. 7_3_, 4334. Bisson, R., Azzi, A., Gutweniger, H., Colonna, R., Montecucco, C., and Zanotti, A. (1978). J. Biol. Chem. 252, 1874. Briggs, M. M., and Capaldi, R. A. (1978). Biochem. Biophys. Res. Commun. 29, 553. Briggs, M., Kamp, P. F., Robinson, N. C., Capaldi, R. A. (1975). Biochemistry 14, 5123. Brittain, T., and Greenwood, C. (1976). Biochem. J., 4§§, 453. Brudvig, G. W., Stevens, T. H., Morse, R. H., and Chan, 3. I. (1981). Biochemistry 29, 3912. Blumer, J. T., and Shurvell, H. F. (1973). J. Phys. Chem, 22, 256. Callahan, P. M., and Babcock, G. T. (1981). Biochemistry 29, 952. Caughey, W. S., Wallace, W. J., Volpe, J. A., and Yoshikawa, S. (1976). in "The Enzymes," V01. 13 (Boyer, P.D., ed.), pp. 299-344, Academic Press, New York. Caughey, W. S., Smythe, G. A., O'Keefee, D. H., Maskasky, J. G., and Smith, M. L. (1975). J. Biol. Chem. 259. 7602. Challoner, D. R. (1963). Nature, 212, 72. Chance, 8., and Powers, L. (1981). Biophys. 122,, 95a. Chance, B., Saronio, C., Waring, A., and Leigh, J. 5. Jr. (1978). Biochim. Biophys. Acta, 492, 37. Chance, B., Saronio, C., and Leigh, J. S., Jr. (1975). J. Biol. Chem. 259, 922b. Chance, B., and Leigh, J. S., Jr. (1977). Proc. Natl. Acad. Sci. 0.5., 14, 4777. Chew, V. S. F., and Bolton, J. R. (1980). J. Phys. Chem. 24, 1903. Chew, V. S. F., Bolton, J. R., Brown, R. G., and Porter, G. (1980). J. Phys. Chem. 24, 1909. 158 Clore, G. ., and Chance, E. (1979). Biochem. J., 422, 613. Clore, G. ., and Chance, E. M M Clore, G. M., and Chance, E. M. (1978). Biochem. J. 422, 799. M M (1978). Biochem. J-.lZ§: 811. M M Clore, G. ., and Chance, E. (1978). Biochem. J. 122, 709. Cochran, R. N., and Horne, F. H. (1980). J. Phys. Chem., 24, 2561. Cochran, R. N., Horne, F. H., Dye, J. L., Ceraso, J., and Suelter, C. H. (1980). J. Phys. Chem. 24, 2567. Cochran, R. (1977). Ph.D. Dissertation, Michigan State University, East Lansing, MI. 48824, USA. Cochran, R. N., and Horne, F. H. (1977). Anal. Chem. 42, 846. Coolen, R. B., Papadakis, N., Avery, J., Enke, C. G., and Dye, J. L. (1975). Anal. Chem. 42, 1649. Cox, R. P., and Hollaway, M. R. (1977). Eur. J. Biochem. 24, 575. Dearman, M., Rao, N. A., and Yagi, K. (1972). Biochem, Biophys. Res. Comm. 44, 849. Dickerson, R. E., and Timkovich, R. (1975). in "The EnzymesJ'Vol. 11 (P. D. Boyer, ed.) Academic Press, New York, pp. 497. ' Downer, N. W., Robinson, N. C., and Capaldi, R. A. (1976). Bio- chemistry 1a, 2930. Dutton, P. L., and Wilson, 0. F. (1974). Biochim. Biophys. Acta £6, 165. Erecinska, M. (1975). Arch. Biochem. Biophys. l§2§ 199. Errede, B., and Kamen, M. D. (1979). in "Symposium on Cytochrome Oxidase" (King, T. E., Orii, Y., Chance, B., and Okunuki, K., eds), pp. 269-280. Elsevier/North-Holland Biomedical Press, Amsterdam. Errede, B., and Kamen, M. D. (1978). Biochemistry, 12, 1015. Falk, K. E., Vanngard, T., and Angstrom, J. (1977). FEBS Lett. 75, 23. Farrington, J. A., Land, E. J., and Swallow, A. J. (1980). Biochim. Bi0phys. Acta gag, 273. 159 Ferguson-Miller, S., Brautigan, D. L., and Margoliash, E. (1976). J. Biol. Chem., 221, 1104. Fry, M., Zande, H. V., and Green, D. E. (1978). Proc. Natl. Acad. Sci. u.s. _7_5_, 5908. Ghosh, R., and Quayle, J. R. (1979). Anal. Biochem. 22, 112. Gibson, 0. H., Greenwood, C., Wharton, D. C., and Palmer, G. (1975). .1. Biol. Chem. 212, 888. Gibson, 0. H., Palmer, G., and Wharton, D. C. (1965). J. Biol. Chem. 242, 915. Gibson, 0. H., Greenwood, C. (1963). J. Biol. Chem. 22, 541. Gouterman, M. (1959). J. Chem. Phys. 22, 1139. Greenwood, C., Brittain, T., Brunori, M., and Wilson, M. T. (1977). Biochem. J. 222, 413. Greenwood, C., Wilson, M. T., and Brunori, M. (1974). Biochem. J. 3g, 205. Greenwood, C., and Gibson, 0. H. (1967). J. Biol. Chem. 242, 1782. Griffiths, D. E., and Wharton, D. C. (1961). Biochim. Biophys. Res. Commun. 4, 199. Griffiths, D. E., and Wharton, D. C. (19618). J. Biol. Chem. 222, 1850. Griffiths, D. E., and Wharton, D. C. (1961b). J. Biol. Chem. 222, 1857. Gutteridge, S., Winter, 0. B., Bruzninck, W. J. A., and Mason, H. S. (1977). Biochem. Biophys. Res. Commun. 22, 945. Halaka, F. G., Babcock, G. T., and Dye, J. L. (1981a). J. Biol. Chem. _2_5_§, 1084. Halaka, F. G., Babcock, G. T., and Dye, J. L. (1981b). J. Biol. Chem., in press. Harrington, P. C., DeWaal, D. J. A., and Wilkins, R. G. (1978). Arch. Biochem. Biophys. 121, 444. Hartzell, C., and Beinert, H. (1975). Biochim. BiOphys. Acta, 222, 318. Henderson, R., Capaldi, R. A., and Leigh, J. S. (1977). J. Mol. Biol. 112, 631. 160 Hisada, R., and Yagi, T. (1977). Biochemistry 22, 1649. Hochli, L., and Hackenbrock, G. R. (1978). Biochemistry, 42, 3712. Hu, V. W., Chan, 5. 1., and Brown, G. S. (1977). Proc. Nat'l Acad. Sci. USA 7_4, 3821. Jacobs, E. E., Jacob, M., Sanadi, D. R., and Bradley, L. B. (1950). J. Biol. Chem. 6223, 147. Jagendorf, A., and Margulies, L. (1960). Arch. Biochem. Biophys. Q, 184. June, 0. S., Kennedy, 8., Pierce, T. H., Elias, S. V., Halaka, F. G., Behbahani-Nejad, 1., El-Bayoumi, A., Suelter, C. H., and Dye, J. L. (1979). J. Am. Chem. Soc. 424, 2218. Kearney, E. B., and Singer, T. P. (1956). J. Biol. Chem. 242, 963. Kellin, 0., and Hartree, E. F. (1938a). Proc. R. Soc. (B), 422, 171. Keilin, D. (1966). The History of Cell Respiration and Cytochromes, Keilin, J., ed., Cambridge Univ. Press. D. D. Keilin, , and Hartree, E. F. (1954). Proc. R. Soc. (B) 422, 167. Keilin, , and Hartree, E. F. (1938b). Nature 444, 870. Komai, H., and Capaldi, R. (1973). FEBS Lett. ag, 273. K0ppenol, W. H., Van Buuren, K. J. H., Butler, J., and Braams, R. (1976). Biochim. Biophys. Acta 442, 157. Lambeth, D. O., and Palmer, G. (1973). J. 8101. Chem., 242, 6095. Lanne, B., and Vanngord, T. (1978). Biochim, Biophys. Acta, 224, 449. Lanne, B., Malmstrom, B. G., and Vanngord, T. (1977). FEBS Lett., 22, 146. Leigh, J. S. Jr., Wilson, 0. F., 0wne, C. S., and King. T. E. (1974). Arch. Biochem. Bi0phys. 422, 476. Lemberg, M. R. (1969). Physiol. Rev. 42, 48. Lemberg, R., and Gilmour, M. W. (1967). Biochim. Biophys. Acta., 143, 500. Lemberg, M. R., and Mansley, G. E. (1965). Biochim. Bi0phys. Acta, 96, 187. 161 Low, H., and Villain, I. (1963). Biochim. Biophys. Acta, 22, 361. Malinowski, E. R., and Howery, D. G. (1980). in Factor Analysis in Chemistry, Wiley--Interscience. Malkin, R., and Malmstrom, B. G. (1970). Adv. Enzymes, 22, 177. Malmstrom, B. G. (1979). Biochim. Biophys. Acta, 242, 281. Malmstrom, B. G. (1973). 0. Rev. Biophysics, 2, 389. Marzotko, D., Wachowiak, R., and Warchol, J. B. (1973). Acta His- tochemica, 22, 150. Mayhew, S. (1978). Eur. J. Biochem. .22, 535. McIlwin, H. J. (1937). Biochem. Soc. Part 2, 1704. Minnaert, K. (1961). Biochim. Biophys. Acta, 22, 23. Nakamura, S., Arimura, K., Ogawa, K., and Yagi, T. (1980). Clinica Chimica Acta, 424, 321. Nakano, S., Kimura, K., Inskuchi, H., and Yagi, T. (1975). Biochem. 18, 1347. Nicholls, P., and Hildebrandt, V. (1978). Biochim. Bi0phys. Acta 213, 457. Nicholls, P., and Chance, B. (1974). in Molecular Mechanisms of Oxygen Activation (Hayaishi, 0., ed.) pp. 149, Academic Press, New York. Nicholls, P., and Petersen, L. C. (1974). Biochim. Biophys. Acta 222, 462. Nishiki, M., Rao, N. A., and Yagi, K. (1972). Biochem. Biophys. Res. Comm. 42, 849. Ohno, A., Kimura, T., 0ka, S., 0hnishi, Y., and Kagami, M. (1975). Tetrahedron Letters 22, 2371. Okunuki, K., Hagihar, B., Sekuzu, 1., and Horio, T. (1958). Proc. Int. Symp. Enzyme Chemistry, Tokyo and Kyoto, Maruzen, Tokyo, p. 264. 01ivas, E., DeWaal, D. J. A., and Wilkins, R. G. (1977). J. Biol. Chem. 222, 4038. 0ndrias, M. R., Findson, E. W., Leroi, G. E., and Babcock, G. T. (1980). Biochemistry 42, 1723. 162 0ndrias, M. (1979). Ph.D. Dissertation, Michigan State University, East Lansing, Michigan. Orii, Y. (1979). in "Symposium on Cytochrome Oxidase" (King, T. E., Orii, Y., Chance, 8., and Okunuki, K., eds.) pp. 331-340, Elsevier/ North-Holland Biomedical Press, Amsterdam. Ozawa, T., Suzuki, H., and Tanaka, M. (1980). Proc. Nat'l Acad. Sci., USA, 22, 928. Papadakis, N., Coolen, R. B., and Dye, J. L. (1975). Anal. Chem. 41, 1644. fieisach, J., and Blumberg, W. E. (1974). Arch. Biochem. Biophys. 65, 591. Penttila, T., Saraste, M., and Wikstrom, M. F. K. (1979). FEBS Letts. 424 295. Petersen, L. C., and Cox, R. P. (1980). Biochim. Biophys. Acta 222, 128. Petersen, L. C., and Cox, R. P. (1980). Eur. J. Biochemistry, 422, 321. Petersen, L. C., Nicholls, P., and Degn, 11. (1976). Biochim. Biophys. Acta 45_2, 59. Phan, S. H., and Mahler, H. R. (l976a). J. Biol. Chem. 224, 257. Phan. S. H., and Mahler, H. R. (1976b). J. Biol. Chem. 224, 270. Powers, L., Chance, B., Ching, Y., and Angiolillo, P. (1981). Biophys. .1. 3_3_, 95a. Powers, L., Blumberg, W. E., Chance, 8., Barlow, C. H., Leigh, J. S., Jr., Smith, J., Yonetani, T., Vik, S., and Peisach, J. (1979). Biochim. Biophys. Acta 242, 520. Poyton, R. 0., and Schatz, G. (1975). J. Biol. Chem. 222, 752. Rao, P. S., and Hayon, E. (1976). Anal. Chem. 42, 564. Ritter, G. L., Lowry, S. R., Isenhour, T. L., and Wilkins, C. L. (1976). Anal. Chem. 42, 591. Robinsin, N. C., and Capaldi, R. A. (1977). Biochemistry 42, 375. Rosevear, P., Van Aken, T., Baxter, J., and Ferguson-Miller, S. (1980). Biochemistry, 42, 4108. 163 Rubaszewska, W., and Grabowski, Z. R. (1975). J. Chem. Soc., Perkin Trans. 2, 417. Rubin, M. S., and Tzagoloff, A. (1973). J. Biol. Chem. 242, 4269. Scott, R. A., and Gray, H. B. (1980). J. Am. Chem. Soc. 422, 3219. Sebald, W., Machleidt, W., and Otto, J. (1973). Eur. J. Biochem. 22, 311. Seiter, C. H. A., Margolit, R., and Perreault, R. A. (1979). Biochem. Biophys. Res. Commun. 22, 473. Sekuzo, 1., Takemori, S., Yonetani, T., and Okunuki, K. (1959). J. Biochem., Tokyo, 42, 43. Shroedl, N. A., and Hartzell, C. R. (1977a). Biochemistry, 42, 1327. Shroedl, N. A., and Hartzell, C. R. (l977b). Biochemistry, 42, 4961. Shroedl, N. A., and Hartzell, C. R. (1977c). Biochemistry,_42, 4966. Singer, T. P., and Kearney, E. B. (1957). in Methods of Biochemical Analysis (Glick, 0., ed.) Vol. 4, pp. 307, Interscience, New York. Slater, E. C. (1949). Biochem. J. 44, 305. Smith, L., in "Methods of Biochemical Anslysis," Glick, D. E., ed., Interscience, New York, 1955, Vol. II, p. 427. ‘ Steffens, G. J., and Buse, G. (1979). in "Cytochrome Oxidase" (King, T. E., Orii, Y., Chance, 8., and Okunuki, K., eds.) pp. 79- 90, Elsevier/North-Holland, Amsterdam. Strekas, T. C. (1976). Adv. Enymol. Relat. Areas Mol. Biol. 42, 287. Sturtevant, J. M. (1964) in Rapid Mixing and Sampling Techniques in Biochemistry, Chance, B., Eisenhardt, R. H., Gibson, 0. H., and Lonberg-Holm, K. K., eds. Academic Press, New York, pp. 89. Suelter, C. H., Coolen, R. B., and Dye, J. L. (1975). Anal. Biochem. 69, 155. Sylvestre, E. A., Lawton, W. H., and Maggio, M. S. (1974). Techno- metrix 42, 353. 164 Tanaka, M., Hanin, M., and Yasunobo, K. T. (1977). Biochem. Biophys. Res. Commun. 22, 1014. Tiesjema, R. H., Muijsers, A. 0., and van Gelder, B. F. (1973). Biochim, Biophys. Acta, 222, 19. Tiesjema, R. H., Muijsers, A. 0., and van Gelder, B. F. (1972). Biochim. Biophys. Acta 222, 32. Tracy, R. P., and Chan, S. H. P. (1979). Biochim, Biophys. Acta, 222, 109. Tucker, A. (1966). Science, 424, 150. Tweedle, M. F., Wilson, L. J., Gracia-Iniguez, L., Babcock, G. T., and Palmer, G. (1978). J. Biol. Chem. 222, 8065. Vadasdi, K. (1974). J. Phys. Chem. 22, 816. Vanneste, W. H., Yseback-Vanneste, M., Mason, H. S. (1974). J. Biol. Chem. 242, 7390. Van Buuren, K. J. H., van Gelder, B. F., Wilting, J., and Braams, R. (1974). Biochim. Bi0phys. Acta 222, 421. Van Buuren, K. J. H., van Gelder, B. F., Wilting, J., and Braams, R. (1974). Biochim. Biophys. Acta 222, 421. Van Gelder, B. F., Tiesjema, R. H., Muijesers, A. 0., and Van Buuren, K. J. H. (1973). Fed. Proc. 22, 1977. Van Buuren, K. J. H., Zuurendonk, P. F., van Gelder, B. F., Muijsers, A. 0. (1972). Biochim. BiOphys. Acta 222, 243. Van Buuren, K. J. H., van Gelder, B. F., and Eggelte (1971). Biochim. Biophys. Acta 224, 468. Vernon, L. P., Zaugg. W. S., and Shaw, E. R. (1963). in Photosy - thetic Mechanisms of Green Plants, pp. 504, National Academy of Science, National Research Council, Washington. Volpe, J. A., and Caughey, W. S. (1974). Biochem. Biophys. Res. Commun., 24, 502. Vorkink, W. P. (1972). Ph.D. dissertation, University of Arizona, Tuscon. Wharton, D. C., and Gibson,(J.H.(1976). J. Biol. Chem. 224, 2861. 165 Wharton, D. C., and Tzagoloff, A. (1964). J. Biol. Chem._222, 2036. Wharton, D. C., and Tzagoloff, A. (1963). Biochem. Biophys. Res. Commun. 42, 121. Wikstrom, M. K. F., and Krab, K. (1979). Biochim. Biophys. Acta gig, 177. Wilhelm, E., Battino, R., and Wilcock, R. (1977). Chemical Reviews, 77, 219. Wilms, J., Veerman, E. C. 1., Konig, B. W., Dekker, H. L., and van Gelder, B. F. (1981). Biochim. Biophys. Acta 222, 13. Winter, 0. B., Bruyninckx, W. J., Foulke, F. G., Grinich, N. P., and Mason, H. S. (1980). J. Biol. Chem. 222, 11408. Wiskind, H. K., and Burger, R. L. (1964). Appendices I and II in Rapid Mixing and Sampling Techniques in Biochemistry, Chance, B., Eisenhardt, R. H., Gibson, 0. H., and Lonberg-Holm, K. K., eds., Academic Press, New York. Yang, E. K., Adar, F., Leigh, J., and Chance, B. (1981). Biophys. .1. aa, 949. - Yonetani, T. (1962). J. Biol. Chem., 222, 550. Yonetani, T. (1961). J. Biol. Chem. 222, 1680. Yoshikawa, S., Shoc, M. G., O'Toole, M. C., and Caughey, W. S. (1977). .1. 8161. Chem. _2_5_2_, 5498. Yu, C., and Yu, L. (1977). Biochim. Bi0phys. Acta 422, 248. Zaugg. 11. S. (1964). .1. 8161. Chem. gag, 3964. Zaugg. W. S., Vernon, L. P., and Tripack, A. (1964). Proc. Nat'l Acad. Sci. USA, 24, 232.